Elettroimaging Brain Organoids: a new frontier of cellular model characterization

1. Overview

Understanding the structural and functional organization of neural networks is a central challenge in neuroscience. Whole-brain organoids represent a powerful in vitro model that recapitulates key aspects of human brain development and complexity. Current approaches fail to simultaneously capture the spatial architecture and the dynamic electrical activity of these systems in a comprehensive and quantitative manner.

To address this gap, this proposal is grounded in the integration of advanced imaging techniques with electrophysiological recordings in living whole-brain organoids. This combined approach enables the correlation of structural connectivity with functional activity, providing a more complete and reliable representation of neural network organization. By studying organoids in their intact, living state, the method preserves physiological relevance and allows for longitudinal analyses of network development and dynamics. This approach is general purpose, and can be implemented for several applications:

· Monitoring organoids culture wellbeing over time, from the initial generation steps

· Obtaining integrated and synergistic information about structure and function to describe physio-pathological features of the in vitro constructs, which can in turn be used for validating them against existing information about the human brain

· Identifying reliable functional metrics without neglecting the structural information a priori, for disease modelling and organoid intelligence (i.e., exploiting learning protocols and monitoring the construct over time)

The importance of this work lies in its potential to establish a novel framework for mapping and characterizing the connectome in human-derived brain models. Such a framework could significantly enhance our understanding of brain development, network formation, and dysfunction. Furthermore, it may provide critical insights into neurological and neurodevelopmental disorders, offering new opportunities for disease modeling, drug screening, and the development of targeted therapeutic strategies. This project provides a proof of concept that could be further extended in the future using HD-MEAs to achieve a more refined characterization of the results. The estimated timeline for this project is 12 months and will fit within the objectives of track 3.

The research will be carried out at University of Pisa (UNIPI) facilities, in particular in the labs of the Centro di Ricerca E.Piaggio, and the Imaging Facility of the Center for Instrumentation Sharing (CISUP). The facilities are already equipped with standard instruments for cells and organoids culture, advanced imaging platforms as confocal microscopes and devices and equipment for in vitro electrophysiology.

Our group is advocate to the use of Reduction, Replacement and Refinement (3R) principles in biomedical research, and promote the implementation of New Advanced Methods (NAMs), specifically serving cellular neuroscience.
This project is gender balanced, as the two principal investigators are a male and a female scientists. They are under-35 early-career researchers, extensively working on brain organoid technology, particularly focusing on the structural and functional characterization for disease modelling and sleep analysis.

All pipelines and customized scripts developed in this project will be deposited in public repositories (e.g. GitHub, Zenodo), to support transparency and replicability. Moreover, all results will be published respecting open-science principles.

Team expertise:

· Federico Tozzi is a biomedical engineer, currently attending the third year of PhD program at UNIPI. His main fields of expertise are cellular imaging, image processing, signal analysis.

· Sonia Cerchio is a biotechnologist, actually attending the third year of PhD program at University of Firenze and Pisa. Her main fields of expertise are cell and brain organoids generation and culture, metabolic features extraction and in vitro electrophysiology.

2. Introduction

Connectome refers to the comprehensive map of connections within the brain, encompassing both structural and functional interactions among neural cells ¹. This network-based perspective is of particular interest because cognitive processes, information transfer, and system-level functions are thought to emerge from patterns of connectivity rather than from the activity of individual neurons alone^2,3. Investigating the connectome therefore provides a framework to understand how network organization develops, adapts, or deteriorates, offering insights into mechanisms underlying neurodevelopment, functional integration, and disease-related dysfunction³.

However, studying the human brain connectome at the microscale remains extremely challenging. Owing to its size and complexity, it is difficult to investigate large-scale connectivity without disrupting its native structure and consequently altering the underlying neuronal networks. In this context, brain organoids offer a promising alternative model. These three-dimensional constructs are generated from human pluripotent stem cells and can self-organize into neural tissues that recapitulate key aspects of early brain development, including elements of spatial organization and emerging functionality⁴. Organoids display heterogeneous populations of neural progenitors, neurons, and glial cells arranged into – albeit exemplified - layered or region-specific architectures⁵, as observed in vivo.

Due to technical limitations - such as the restricted diffusion of antibodies and fluorophores within the organoid - analyses are often limited to sample surface characterization, or require organoid slicing⁶. Moreover, studies focus on either structural or functional connectivity, neglecting the pivotal role of their integration in fully understanding how the organoid network works^6,7. These constraints hinder the reconstruction of a comprehensive connectome, as they either exclude morphological features or disrupt native neuronal connections through tissue sectioning, thereby preventing the full characterization of the underlying network architecture.

Here, we present a pipeline that enables the simultaneous acquisition of functional and morphological features from a living organoid, allowing the extraction of a connectome that integrates both sources of information (weighted connectome). Samples are longitudinally acquired using confocal microscopy and microelectrode arrays (MEAs) during the maturation process. The resulting connectivity measures are evaluated alongside morphological descriptors, standard electrophysiological metrics and new structure-functional parameters defined ad hoc to quantify the emergence of coordinated signal propagation and network synchronization within the developing organoid.

Hypothesis: The integration of structural and functional measurements in living brain organoids enables the reconstruction of a weighted connectome. This last reflects the maturation of network organization, with increasing morphological complexity supporting the emergence of coordinated signal propagation and synchronized activity.

Impact statement: By enabling the reconstruction of a connectome starting from a living brain organoid, this work provides a new tool to study the emergence of network organization during development and its alterations in neurological disorders.

3. Methods

Sample

Organoids generation: Whole-brain organoids will be produced with an adaptation of the Lancaster’s protocol⁴. The starting cells are commercially available human Induced Pluripotent Stem Cells (hiPSCs) with a MAP2 differentiation reporter: those cells will spontaneously express GFP fluorescent protein together with MAP2 protein, a well-known marker of terminally differentiated neurons – LINK: https://www.angioproteomie.com/commerce/ccp25491-human-ipsc-with-map2-differentiation-reporter-28gf-cap-0500-001map2-gfp.htm.

Briefly, hiPSC will be detached and 9000 cells will be plated in each well of an ultra-low attachment 96 well round bottomed plate. Through the administration of specific factors, the cells will be induced to become embryoid bodies, neuroepithelia and then brain tissue. Following a power analysis on previously generated datasets, we established a minimal sample size of 10 organoids per batch, and 3 batches to obtain reliable results.

Materials and Procedures

Organoids plating: 1 month old Whole-brain organoids will be plated on Multichannel System Multi-Electrode-Array (MEAs) – LINK: https://www.multichannelsystems.com/products/microelectrode-arrays/60hexamea4010ir-ito.

MEAs will be previously coated with Poly-D-Lysine and Laminin, through an already established procedure. MEAs will be selected for their transparent tracks and contact pads, for the arrangement and density of electrodes in the space and for their compatibility with the recording system already present at UNIPI. When the organoid will reach the optimal maturation stage – expression of fluorescent GFP associated to MAP2 – the connectome will be investigated. One time per week, per approximately 3 months, organoids will be imaged through confocal microscopy and electrophysiological recording will be performed through Multichannel System device.

Data Acquisition: Datasets will be acquired using standard confocal microscopy (Confocal Nikon A1 Eclipse T1) and spinning disk microscopy (Nikon Eclipse Ti2) systems available in our laboratory. Electrophysiological recordings will be obtained using a Multichannel MEA2100mini System already present in the facility. All datasets will be processed and analyzed using ad hoc scripts developed in MATLAB.

Connectome analysis: The main goal of the analysis will be the extraction of two connectivity surface maps: the structural and the functional.

The structural map will be defined through a two-level graph representation. At the mesoscale, each cell will be represented as a node, while edges will be established only between cells located within a predefined synaptic interaction distance in 3D space. At the microscale, each node will be further described to quantitatively describe the morphology of the corresponding neuron. This hierarchical configuration enables not only the identification of potential intercellular connections, but also the localization of these interactions across the cellular structure, thereby improving the morphological characterization of the sample surface. The pipeline comprised 3 steps:

1. Extraction of a global map summarizing cell and electrode position within the well. This will be performed through a low-magnification stitched image of the entire well superimposed to the electrode tracks geometry. This map will therefore serve as a spatial reference to performed and link high-resolution structural measurements with the functional activity recorded from specific electrodes.

2. Single cell isolation using SENPAI⁸ and ad hoc algorithms to remove spurious branches. This strategy enables the reconstruction of single cells from volumetric (3D) image stacks and the incorporation within the two-level graph structure.

Generation of the functional map from MEAs recordings. Multi-Unit Activity (MUA) will be extracted from the electrophysiological recordings, followed by spike sorting to isolate the activity of individual neurons. Functional relationships between neuronal signals will then be estimated using information-theoretic metrics.

4. Extraction of the weighted graph will be extracted starting from both structural and functional information. Graph-theoretical features (e.g., small-worldness, modularity, and related metrics) will be evaluated⁹.

Organoid characterization: At each timepoint, some organoids will be fixed and immunoassayed to identify the neuronal populations present on their surface. Immunostaining will be performed to identify cortical neurons (CTIP2, SATB2), excitatory neurons (VGLUT1, VGLUT2), inhibitory neurons (VGAT), and astrocytes (GFAP).

Additional parameters about single cell complexity and network maturation will be extracted to assess both the developmental stage and overall sample wellbeing.

Planned Analyses

This is the planned GANNT chart for this project: months 1, 2 and 3 would be dedicated to the generation of the organoids and the establishment of MEA-organoid ensemble, months 3, 4 and 5 to the acquisition of combined electrophysiological and imaging data, to data analysis and extraction of information about the model. The same pipeline would be repeated 3 times, per 3 different batches. Lastly, quantitative information about connectome would be produced from month 6 till the end of the project.

IF-THEN contingencies:

IF IPSC, once induced to neural tissue and differentiated in whole-brain organoids do not express GFP and consequently, we are not able to image alive organoids plated on MEA chips, THEN we will fix them directly on the chip and perform classical immunocytochemistry (protocol adapted from¹⁰). The maturation stage would be frozen at the moment of fixation; nevertheless, fixing organoids at diverse maturation stages, would enable the extraction of quantitative information on the neural networks and on their modifications over time.
IF the number of recording electrodes is not enough to capture network complexity and maturation over time, THEN we will implement spike sorting algorithms^11,12 to increase the recorded signals.
IF transparent MEA chips do not allow the reference electrode clear visualization, and consequently we cannot ensure the correct registration of imaging and electrophysiological acquired data, THEN we will insert references on the chip via inkjet.

Statistical tests: in this project we will perform linear mixed model to compare inter-organoids electrophysiological and structural acquired data, implementing softwares as GraphPad and R studio. Moreover, information related to the connectome will be extracted over time and linear mixed model statistic will be performed to evaluate the network maturation.

Hypothesis specific predictions: we expect that measured organoids will exhibit concordant electrophysiological and structural features, showing a progressive increased complexity of the sample.

Ethics and Data Management

Organoids will be generated from commercial cell lines, following the good practices and procedures under the EU directive 2010/63.

4. Pilot Data (Optional)

Figure 1. Example of neuron reconstructed from 3D stack of brain organoids.

We already have experience in organoid image acquisition and analysis. Using the SENPAI segmentation algorithm together with ad hoc pruning scripts developed in MATLAB, we are able to extract morphological information from single neurons by isolating them from surrounding cells and reconstructing their shape (Figure 1). During the course of the project, these algorithms will be further refined and extended to enable the extraction of multiple individual neurons and their subsequent integration into a graph-based representation of the sample, allowing the reconstruction of its structural connectivity.

Figure 2. An example of multitaper spectrogram obtained from raw electrophysiological signal, first filtered in the low-frequency range

We implemented a customized algorithm for the analysis of electrophysiological data acquired with Multichannel System devices. The algorithm filters the raw data in the high and in the low frequency ranges and extracts quantitative information from both bandwidths. In particular, Multi-Unit Activity parameters such as Spikes and Bursts metrics are obtained from the high frequency range, while Local Field Potential parameters, including spectrograms – an example is reported in Figure 2 –, and individual brain waves characteristics are elicited from the low frequency range.

5. Budget

· Personal: two partial Fellows - $17000

· MEAs: Multichannel System 60HexaMEA40/10iR-ITO-gr - $3000 (Acquisition system already present at UNIPI)

· iPSC and Organoid culture media: $5000

· Labware: Standard tissue culture plasticware already available at UNIPI,

· Imaging Facility already at UNIPI

6. References

1. Sporns, O., Tononi, G. & Kötter, R. The Human Connectome: A Structural Description of the Human Brain. PLoS Comput. Biol. 1, e42 (2005).

2. Bassett, D. S. & Sporns, O. Network neuroscience. Nat. Neurosci. 20, 353–364 (2017).

3. Bullmore, E. & Sporns, O. Complex brain networks: graph theoretical analysis of structural and functional systems. Nat. Rev. Neurosci. 10, 186–198 (2009).

4. Lancaster, M. A. & Knoblich, J. A. Generation of cerebral organoids from human pluripotent stem cells. Nat. Protoc. 9, 2329–2340 (2014).

5. Koo, B., Choi, B., Park, H. & Yoon, K.-J. Past, Present, and Future of Brain Organoid Technology. Mol. Cells 42, 617–627 (2019).

6. Sharf, T. et al. Functional neuronal circuitry and oscillatory dynamics in human brain organoids. Nat. Commun. 13, 4403 (2022).

7. Trujillo, C. A. et al. Complex Oscillatory Waves Emerging from Cortical Organoids Model Early Human Brain Network Development. Cell Stem Cell 25, 558-569.e7 (2019).

8. Cauzzo, S. et al. A modular framework for multi-scale tissue imaging and neuronal segmentation. Nat. Commun. 15, 4102 (2024).

9. Ullo, S. et al. Functional connectivity estimation over large networks at cellular resolution based on electrophysiological recordings and structural prior. Front. Neuroanat. 8, (2014).

10. Ciani, C., Pistorio, G., Mearelli, M. & Falcone, C. Immunofluorescence protocol for localizing protein targets in brain tissue from diverse model and non-model mammals. STAR Protoc. 4, 102482 (2023).

11. Yger, P. et al. A spike sorting toolbox for up to thousands of electrodes validated with ground truth recordings in vitro and in vivo. eLife 7, e34518 (2018).

12. Quiroga, R. Q., Nadasdy, Z. & Ben-Shaul, Y. Unsupervised Spike Detection and Sorting with Wavelets and Superparamagnetic Clustering. Neural Comput. 16, 1661–1687 (2004).

Elettroimaging Brain Organoids: a new frontier of cellular model characterization

Proposal

Proposal

Proposal

Proposal

Elettroimaging Brain Organoids: a new frontier of cellular model characterization

Funding Progress

Nonprofit Recipient

Peer Reviews

Topics

DOI

Proposal

Proposal

Funding Progress

Nonprofit Recipient

Peer Reviews

Topics

DOI

Funding Progress

Nonprofit Recipient

Peer Reviews

Topics

DOI