Simple and Efficient 3D-Printed Superfusion Chamber for Electrophysiological and Neuroimaging Recordings In Vivo

Abstract

In vitro and in vivo experimentation in the central nervous system are effective approaches to study its functioning. Manipulations in vitro are characterized by easy experimental control and stable experimental conditions. However, transferring these advantages to in vivo research remains technically and ethically challenging, preventing many research teams from acquiring critical recordings in their animal models. In order to transfer the benefits of in vitro experimentation to in vivo experimentation, we developed a suite of 3D-printed tools (a superfusion chamber with an independent brain presser and animal stand). Using the immature rat barrel cortex as a model, we show that our set of tools (further "superfusion preparation") provides stable conditions for electrophysiological and neuroimaging recordings in the neonatal rat neocortex in vivo Highly correlated intracellular and extracellular activity was recorded during spontaneous and evoked cortical activity, supporting the possibility of simultaneous long-lasting electrophysiological recordings from a single cortical column in vivo The optical intrinsic signal of evoked cortical responses was also recorded from the skull-free neocortex, suggesting the effective combination of the superfusion preparation with neuroimaging approaches. Modulation of immature activity by epicortical application of pharmacological agents via superfusion equally supports the use of the superfused cortex preparation in pharmacological screening. In addition to high efficiency (in affordability, reliability, and ease of use in vivo), the 3D-printed set of tools developed should reduce animal use, supporting the 3Rs principle (Replacement, Reduction, and Refinement) of ethical use of animals.

Keywords: 3D printing; drugs screening; electrophysiology; in vivo; neuroimaging.

Graphical abstract

Figure 1.

Electrophysiological extracellular recordings using the superfused cortex preparation. A, B, Scheme and picture of the superfused cortex preparation with a presser fixed on the rodent head. C, Evoked and spontaneous neuronal network activity recorded in the barrel cortex with (black trace) and without (gray trace) the superfusion chamber. Expanded episodes of evoked (left) and spontaneous (right) activity marked by a single and doubled asterisk on the traces are shown below. Lines correspond to the fluctuations of LFP, while vertical bars are MUA. D, Group data of spontaneous activity occurrence recorded with and without the superfusion chamber. E, F, Group data of MUA rate during episodes of evoked activity and its duration recorded with and without the superfusion chamber. G, Power spectral density of evoked activity recorded with and without the superfusion chamber. H, Group data of evoked activity spectral peak distribution in the β and γ frequency ranges recorded with and without the superfusion chamber. Filled circles represent the results of individual experiments; red circles, mean values; red whiskers, confidence intervals based on t Student’s criteria.

Figure 2.

Electrophysiological intracellular and extracellular recordings using the superfusion chamber. A, Picture of simultaneous intracellular and extracellular recording using superfusion chamber. B, Simultaneously recorded evoked EPSCs (red trace), IPSCs (blue trace), and corresponding extracellular LFP responses (black trace) recorded in one cortical column of the barrel cortex. Note that spontaneous activity is also observed both extracellular and intracellularly. C, Cross-correlation between extracellular and intracellular activity recorded using the superfusion chamber. Cross-correlation centered on LFP troughs. D, LFP (black line), EPSC (red line), and IPSC (blue line) spectrum of the evoked response during PW stimulation. E, Group data for the spectrum peak frequencies in the β (left) and γ (right) frequency range. Peak frequency values of the evoked and spontaneous response for the EPSC (light red filled and empty circles, respectively), IPSC (light blue filled and empty circles, respectively), and LFP (light gray filled and empty circles, respectively) are shown. Black filled circles represent mean values; black whiskers, confidence intervals based on the t Student’s criteria.

Figure 3.

Optical intrinsic signal recordings using the superfusion chamber. A, Experimental scheme of OIS recordings in vivo. B, Field of view of the cortical surface fixed under the presser recorded using the green illumination shown on the left. Expanded view of the presser with clearance (red) and crosspiece (blue). C, OIS evoked by whisker stimulation from the experiment shown in B. Note the crosspiece and clearance zones are over the barrel cortex. D, Comparison of the dynamics of the OISs recorded under the clearance (red), crosspiece (blue), and that recorded through the skull (gray). The shaded area corresponds to the SD of the OIS. E, Group data for the OIS amplitude under the clearance (red), crosspiece (blue), and recorded through the skull (gray). F, Group data for OIS onset in the zone of clearance (red), crosspiece (blue), and through the skull (gray). G, Group data for OIS duration under the clearance (red), crosspiece (blue), and through the skull (gray). Black filled circles represent mean values, black whiskers, confidence intervals based on the t Student’s criteria.

Figure 4.

Epipial pharmacological application using the superfusion chamber. A, Cortical granular layer activity recorded in P27 animal in control condition followed by consecutive application of glutamatergic blockers, wash-out and then application of an inhibitor of GABAergic transmission. LFP (black lines) and corresponding MUA (red bars) are shown for each pharmacological condition. B, Group data for the LFP spectral power in γ frequency range (30–100 Hz; calculated using wavelet decomposition) for each pharmacological condition (left panel). Gray-filled circles are the results of individual experiments. On each box, the central mark indicates the median, and the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively. The whiskers extend to the most extreme data points not considered outliers. Nonparameteric Wilcoxon test p-values checkboard (right panel) between LFP response spectral power during various pharmacological conditions. Blue color (significant difference) corresponds to p < 0.05, while red color (nonsignificant difference) to p > 0.05.