The Human Genome Project, which was declared complete in 2003, stimulated the expectation that great medical benefits were imminent. But years passed, and the benefits seemed to recede into the future. What went wrong? Well, nothing—except that the value of knowing the genome’s contents had been overstated, however inadvertently. The word “genome” includes the suffix “ome,” which betokens completeness. That is, it suggests that all that needs to be known is known, at least in one field of endeavor. But even the genome needs context.
Recently, the genome and other “omes”—including the epigenome, the transcriptome, and the metabolome—started to be contextualized in various ways. For example, instead of compiling bulk omics information, many researchers began to collect single-cell omics information. These researchers showed that one needn’t be satisfied with broad averages. Instead, one could explore cellular heterogeneity.
Essentially, researchers demonstrated that they could acquire contextualized views by “zooming in” on single cells. And now researchers are “zooming out,” too. While preserving the granularity of single-cell profiling, researchers are looking at cellular environments. In other words, researchers are putting cellular heterogeneity in spatial contexts.
Although spatial context may be sought at the subcellular level, it is, for now, mostly sought among collections of cells, or in specific tissues. Let’s look at just one example. At the University of Queensland, researchers used a spatial and molecular profiling platform to survey the pulmonary transcriptional landscape of COVID-19, pandemic H1N1 influenza, and uninfected control patients (Kulasinghe et al. Eur. Respir. J. 2021; 2101881). Doing so allowed the researchers to dissect virus-specific host responses and gene signatures.
The researchers described one of their key observations as follows: “SARS-CoV-2 was non-uniformly distributed in lungs (emphasizing the advantages of spatial transcriptomics) with the areas of high viral load associated with an increased type I interferon response.” The researchers concluded that “spatial transcriptomics is a powerful tool to identify novel gene signatures within tissues, offering new insights into the pathogenesis of SARS-CoV-2 to aid in patient triage and treatment.”
Work of this sort—that is, work that puts single-cell omics in spatial contexts—is becoming more common. It is starting to characterize disease states in unprecedented detail. It is also expanding the scope of research into disease mechanisms. Such research is bound to inform drug development. Most important, it promises to realize many of the hopes originally inspired by the Human Genome Project.
Molecular biology gets a slider bar
One of the University of Queensland researchers involved in the study cited earlier is Arutha Kulasinghe, PhD. He leads the “Clinical-O-Mx” group at the University of Queensland’s Diamantina Institute.
“Spatial biology is the next generation of tissue imaging combined with genomic readout,” he says. “It enables researchers to profile proteins and genes from tissue samples down to single cell and subcellular resolution.
“It’s the ‘Google maps’ approach, where you can profile diseased tissue at the city level (a tissue region), the street level (a few cells), or the house level (a single cell). This is enabled through advances in molecular barcoding technologies and enables scientists and clinicians to profile tissue samples from patient biopsies with a greater level of understanding of the underlying tissue biology.”
Visualizing genomic organization
“Studies of spatial biology of organs and tissues have emerged because it has been possible to sequence the output from genes (the mRNAs) in single cells,” says Niels Tommerup, MD, DMSc, a professor of medical genetics at the University of Copenhagen. “This has allowed researchers to classify specific RNA profiles for each cell type, and hence to both identify new cell types and to potentially identify which specific cell type that are affected in a given disease.”
The key benefit, Tommerup emphasizes, is the way spatial analysis links interactions at the molecular level to tissue biology. Spatial biology, he suggests, is a concept that can be elaborated across several scales—from individual cells to tissues and organs. For example, if one were to use spatial biology to study genomic organization, one could focus on changes within a single cell, assess which conformations are more common in cell populations, or determine how conformations may vary among cells that occupy certain locations.
Spatial biology across scales has improved, Tommerup says, thanks to techniques such as deep sequencing technology and chromosome conformation capture (3C) technology. According to Tommerup, 3C techniques such as HiC and MicroC assays can help researchers map chromatin interactions genome wide.
“Following fixation, fragmentation, and re-ligation, DNA regions that are physically closer to each other can be co-sequenced, and this has revealed that the genome is organized into structural domains and chromatin loops, bringing specific genes physically together with their potential myriad of regulatory elements that may be situated far from the gene.”
An environmental perspective
At Cold Spring Harbor Laboratory, researchers led by Assistant Professor Je Hyuk Lee, MD, PhD, study how cells interact with their microenvironment to regulate gene expression during development. His online profile states that “single-cell heterogeneity in gene expression can result from spatial differences in cell-cell and cell-extracellular matrix interactions.”
Lee declares that spatial biology and spatial genomics is about “visualizing biological information.” These technologies, he continues, allow formerly dimensionless information to be put in a three-dimensional context and given functional meaning. Such meaning, he insists, is “central to linking genetic information to form and function.”
Lee’s group focuses on the role of noncoding RNA in chromatin remodeling and tumor progression using mouse and organoid models of human cancer. The group’s long-term goal is to improve our understanding of the tumor microenvironment, and to use understanding to support the development of tumor classification tools and anticancer therapeutics.
Enabling diverse applications
As the University of Queensland study cited earlier indicates, spatial techniques are already being used to improve our understanding of infectious diseases. Spatial techniques are also being applied to generate insights into cancer and neurodegenerative diseases. According to Kulasinghe, spatial techniques are broadly applicable because they enable researchers to map cell types to the genes and proteins that are being expressed.
“In cancer, one of the current unmet clinical needs is the need for predictive biomarkers of response for immunotherapies such as immune checkpoint blockade,” he elaborates. “Spatial biology approaches have led to a greater understanding of the tumor and tumor microenvironment.
Besides showing where certain immune cell types are located within the tumor microenvironment, spatial biology can also give clues about the degree of immune activation in a tumor. This kind of information, Kulasinghe notes, can help predict how patients respond to immunotherapies.
Another researcher using spatial techniques in cancer research is Simon Gregory, PhD, Professor and Vice Chair of Research, Department of Neurology, Duke University School of Medicine. He is also applying the techniques to improve our understanding of neurological diseases.
For example, in multiple sclerosis, Gregory is applying spatial biology to study the impacts of lesions in the central nervous system. “We’re hoping to characterize lesions in different parts of the central nervous system,” he says. Doing so could help Gregory and his colleagues elucidate the role of immune cell function in lesion development. It could also help them distinguish between benign and progressive lesions.
“In Alzheimer’s, we’re characterizing not only what happens within amyloid-β plaques and neurofibrillary tangles, but also what is happening in the peri-plaque region,” he adds. “And in brain tumors, we’re trying to characterize the heterogeneous landscape of the tumor.” This landscape isn’t limited to the cellular and transcriptomic topology of the tumor itself. It extends to the surrounding stroma that is invaded by the tumor.
Clarifying disease mechanisms
“Spatial genomics is still in its infancy, but it has huge potential,” says Tommerup. He is particularly optimistic about using spatial genomics to expand the number of genetic factors that are known to determine common and complex diseases. Many such factors, probably the majority of such factors, have yet to be identified. But Tommerup thinks he knows where more might be hiding: “Conceivably, the missing inheritance factors are of regulatory nature, and hence determined by the 3D genomic organization.”
“Spatial biology of organs and tissues will undoubtedly have the power to refine biological studies,” Tommerup declares. This view is shared by Gregory, who says that in future, spatial methods will offer researchers the ability to fully understand diseases and therapeutic response from the molecular to the whole organism level.
“I see spatial approaches as occupying the intersection of genomics and pathology,” Gregory elaborates. “For more than 100 years, pathology has allowed us to define cellular phenotyping/morphology as it relates to a disease state, but we haven’t been able to understand the underlying biological mechanisms.”
“Using spatial or in situ transcriptomic approaches, we’re not only about to characterize the aberrant transcriptomic profiles of these disease-related cells, but we’re also going to be able to describe the transcriptomic changes that preempt disease-related cellular changes.”
Spatial techniques in combination with other cutting-edge analytical methods can help scientists understand disease mechanisms and develop novel interventions. In biopharma, spatial techniques could identify and localize diseased cells that currently lack biomarkers (for example, disseminated tumor cells). Spatial techniques could also reveal how immune cells react around diseased cells.
According to Lee, being able to use spatial techniques to characterize the response to “different types of cell or immunotherapies in a robust, scalable, and high-throughput manner would be a game changer, especially in combination with detailed tumor pathology and artificial intelligence–based data processing.”
Lee states that there are also potential applications in tissue engineering. He predicts that spatial techniques will be embraced by cell therapy developers and the wider regenerative medicine sector. “Currently, genetic and phenotypic analyses of engineered cells are done in vitro in a Petri dish,” he relates. “Spatial genomics technologies could potentially upend this. For tissue biology and pathology, the impact may be lower.
“I am not sure that looking at gene expression signatures alone will lead to dramatic advances in our understanding of complex tissue biology or cell-cell interactions. At the end of the day, I believe that spatial biology will have to provide information on genetic mechanisms, molecular functions, and cell engineering solutions to truly transform translational research, biopharma development, and medicine.”