Transcriptomics has been an invaluable field of study, yielding key information about DNA and RNA interactions and the initial transcription and translation into protein chains. But it doesn’t say as much about proteins themselves—it can’t reveal information about post-translational modifications, the intended destination of proteins, or translation that never occurs. It is even further removed from interactions between cells, and protein-protein interactions.
Too often, many researchers stop with transcriptomics, making the mental leap that this should reveal what’s needed to understand proteomics. To some degree, transcriptomics can do that—to date, mass spectrometry of cell lines and human tissues identified more than 85 percent of proteins encoded by human protein-coding genes.
But, as this recent study conducted by Stanford and University of Washington researchers and published in Cell shows, relying on transcriptomics misses key proteins. Their study, which quantified protein levels from 12,000 genes in 32 healthy human tissues, missed more than 3,000 proteins that were not enriched at the RNA level (but were at the protein level). Particularly dramatic discrepancies between RNA and proteins were found in stomach, brain cortex and cerebellar tissues, the researchers found.
While this study did not use spatial biology tools, it does underscore the need for technologies like Syncell’s Microscoop®, which uses microscopy guided biotinylation to pinpoint the exact location of proteins in cells and tissues, and then mass spectrometry to identify the proteins themselves.
The researchers admitted that their study might have overlooked low-expression but tissue-specific proteins represented different mixtures of cell types even within single-tissue samples, and their techniques could not correct for unbalanced samples for each tissue. In contrast, spatial proteomics technologies can overcome these barriers to understanding protein structures and function. Spatial analysis at the cellular and subcellular (organelle) level can identify proteins stored in vesicles, and differentiate them from proteins elsewhere in the cell, and from proteins in the extracellular matrix.
Moreover, many of these proteins and their spatial interactions can explain the pathogenesis of genetic diseases, which transcript information alone cannot show.
Spatial proteomics has been instrumental in Alzheimer’s disease research, aiding in the identification of proteins associated with amyloid-beta aggregates in affected brain regions. By integrating this data with transcriptomic analyses, researchers can uncover how gene expression changes contribute to protein aggregation and disease progression. Similarly, in cancer research, spatial proteomics helps identify proteins involved in tumor microenvironments, aiding in the understanding of how genetic mutations and RNA expression changes drive cancer progression. In addition, certain tumor suppressor proteins, including retinoblastoma and p53, can be regulated by post-translational modifications, altering their structure (albeit temporarily) and therefore, function.
It’s important to note that just as transcriptomics doesn’t substitute for proteomics, protein analysis doesn’t negate the importance of understanding transcription and translation. In fact, by integrating these fields, scientists can:
- Validate transcriptomic data by confirming the presence and localization of corresponding proteins. Not all RNA transcripts translate into functional proteins, and spatial context can reveal regulatory mechanisms that influence protein expression.
- Gain functional insights, identifying the actual proteins present in specific cellular compartments. This elucidates the functional implications of genetic variations and RNA expression changes.
- Map protein networks by pinpointing the exact locations of proteins and their interactions within the cellular environment. This complements transcriptomic data, which often predicts potential interactions based on co-expression patterns.
So, while transcriptomics and genomics have been (and remain) very valuable molecular biology tools, high resolution spatial proteomics is already delivering on its promise to reveal cellular and subcellular interactions that help shape health and disease.
