Innovative OpenFISH Technique Enables Integrated Spatial Molecular Imaging in Tissues

Spatial transcriptomics (ST) shows which genes are being expressed and where within a tissue. Spatial metabolomics (SM), on the other hand, maps the location of metabolites - small molecules produced or used by cells - within tissue.

Now, in an important advance, researchers have developed an open, low-cost imaging-based ST platform that can also integrate SM on the same slide.

Called OpenFISH, the new technique addresses two major ST challenges: high cost and incompatibility with SM.

The study, led by Prof. DUAN Lihui at the Institute of Genetics and Developmental Biology (IGDB) of the Chinese Academy of Sciences (CAS), was published in Neuron on September 29.

A biological system can be thought of as a multistory building. Information flows from the genome to transcripts to proteins to metabolites, and finally to the phenotype - and every layer interacts with the others. Measuring just one layer can miss the full picture and the connections between them.

Scientists can now detect molecular features within tissues at single-cell resolution, but existing ST technologies remain expensive and difficult for many laboratories to use. Furthermore, integrating imaging-based ST with the state-of-the-art untargeted SM method MALDI-MSI for the same tissue section presents an additional challenge.

To overcome this challenge, the researchers used a modular probe design to cut probe synthesis costs, along with a simple coding system for genes that eliminates the need for a microfluidic system. They also optimized the experimental procedure, bringing the wet-lab time down to no more than 13 hours. A standard 20× widefield fluorescence microscope is sufficient to capture clear signals in situ. Together, these efforts reduce the total cost of OpenFISH by about 95% compared with leading commercial platforms.

The researchers then used OpenFISH in two neuroscience applications: examining cell-type-associated transposable element (TE) elevation during inflammation and neuronal cell lamination distortion after Reln gene knockout. Their observation of reproducible TEs upregulation implicates TEs as active drivers or modulators of neuroinflammatory pathways. Beyond the known anatomical changes caused by Reln knockout, they also observed a decrease in D1-type inhibitory striatal neurons.

ST and SM are commonly performed on serial sections. However, inherent differences between two adjacent tissue sections can compromise data interpretation. To address this problem, the researchers modified conductive glass slides for MALDI-MSI. Through polyacrylamide gel embedding, protein digestion, and lipid removal, OpenFISH signals could still be readily detected even after the harsh laser processing used in MALDI-MSI.

Ion feature signals and transcript detection were barely affected by the integration. With this combined pipeline, the researchers revealed cell-type-associated metabolites in mouse brain cells and found that merging the two modalities improved anatomical depiction.

They also applied the pipeline to 5xFAD mice, a model for Alzheimer’s disease (AD). Microglia showed the strongest changes in AD mice, and multiple metabolites associated with specific cell types were elevated in these mice compared with healthy controls.

The researchers report that this work introduces a cost-effective ST tool that pairs seamlessly with SM platforms to better untangle complex molecular interactions in biological systems.