Spatial Transcriptomic Technologies

Abstract

Spatial transcriptomic technologies enable measurement of expression levels of genes systematically throughout tissue space, deepening our understanding of cellular organizations and interactions within tissues as well as illuminating biological insights in neuroscience, developmental biology and a range of diseases, including cancer. A variety of spatial technologies have been developed and/or commercialized, differing in spatial resolution, sensitivity, multiplexing capability, throughput and coverage. In this paper, we review key enabling spatial transcriptomic technologies and their applications as well as the perspective of the techniques and new emerging technologies that are developed to address current limitations of spatial methodologies. In addition, we describe how spatial transcriptomics data can be integrated with other omics modalities, complementing other methods in deciphering cellar interactions and phenotypes within tissues as well as providing novel insight into tissue organization.

Keywords: NGS-based spatial profiling; image-guided spatially resolved single cell sequencing; imaging-based spatial profiling; probe-based spatial profiling; spatial omics technologies.

Figure 1

Four main types of spatial transcriptomic methods. Sequencing-based methods (method (A)) use barcoded DNA arrays to capture polyadenylated RNA transcripts from tissues followed by next-generation sequencing. Probe-based methods (method (B)) capture user-defined targeted transcripts in manually selected regions of interest (ROIs), using corresponding, barcoded oligonucleotide-conjugated probes and can be demultiplexed accordingly afterwards. Imaging-based methods (method (C)), similar to probe-based methods, rely on in situ hybridization but with complimentary fluorescent probes, and the targeted transcripts can be detected in a cyclic manner. Image-guided spatially resolved scRNAseq methods (method (D)) can select spatially different single cells in ROIs (i.e. photoactivation of single cells in ROIs) followed by fluorescence-activated cell sorting and scRNAseq, thereby preserving their native spatial information as well as retaining in-depth profiling of scRNAseq.

Figure 2

(a) The pipeline of spatially annotated FUNseq. (b) Live HNSCC tumor tissue slice imaging on the UFO microscope. Left panel: SPY505 nuclear staining (green) and CD4 immunostaining (red) of the tissue were imaged; right panel: Phototagged image, where the regions with a higher density of CD4+ T cells (highlighted in blue circles in the left panel) were phototagged (higher intensity). Scale bar = 300 μm.