Background

Since the completion of the Human Genome Project, the life sciences community has focused on elucidating how approximately 20,000 protein-coding genes support the precise biological activities of the human body. The classical central dogma of molecular biology, which posits that a single gene is expressed as a single protein through the transcription process, underwent one expansion with the discovery of alternative splicing. This was due to the proof that multiple messenger RNAs (mRNAs) are produced from a single gene, thereby increasing the number of protein variants.

Nevertheless, the gap between genomic sequence information and the actual protein clusters functioning within cells remains unbridged. It has been difficult to fully explain why cells with identical genetic information exhibit different functions and reactivities using only the existing genetic code system. Previous proteomic research relied primarily on bottom-up mass spectrometry. Because this method involves fragmenting proteins into peptides to reconstruct sequences, it possessed an inherent limitation: it could not fully preserve the complete structure of proteoforms, which are the total combinations of variations within a single protein molecule. There was an urgent need for a platform to systematically measure the total amount of actual functional diversity created by the combination of structural differences from genetic variations and non-genetic variations such as post-translational modifications (PTM).

Key Findings

In this study published in Nature, an international collaborative research team established a high-resolution multidimensional proteomic profiling platform to track the functional proteomic diversity within human cells at unprecedented resolution. The team quantitatively analyzed how the functional protein pool is formed, including not only amino acid sequence variations based on the genetic code but also non-genetic factors, including non-standard ribosomal translation and post-translational modifications.

Experimental results confirmed that functional protein variants, which form stable, unique three-dimensional structures and exhibit biochemical activity in actual cells, are at least five times more abundant than the basic protein list predicted by genomic information. Notably, the contribution of variants arising from non-genetic factors to the expansion of proteomic diversity was greater than that of isoforms derived from genetic code variations. Even for proteins produced from the same mRNA transcript, different binding affinities and enzymatic activities were observed depending on the combination of phosphorylation, glycosylation, and ubiquitination, as well as fine-scale regulation during the translation process.

Using a combination of Surface Plasmon Resonance (SPR) analysis and cryo-electron microscopy (cryo-EM) structural determination technologies, the research team verified approximately 40 types of representative immune signaling proteins and metabolic enzyme complexes. Among these, they identified that many protein variants processed through non-heritable modification pathways formed new protein-protein interaction (PPI) networks in addition to previously known binding partners. Based on empirical data, it has been demonstrated that even in the absence of mutations in the amino acid sequence itself, subtle deviations in phosphorylation cluster patterns can accelerate signaling speed by more than threefold or alter intracellular trafficking pathways.

Significance and Outlook

This achievement provides an opportunity to shift the focus of precision medicine, which was previously limited to gene sequence analysis, toward the analysis of complete protein isoforms, which are the actual functional units. It provides new theoretical clues to explain why specific pathological mechanisms operate in patients with intractable diseases where no genetic abnormalities are found. If disease-specific non-genetic proteomic variants can be identified, it could open new avenues in the field of early diagnosis, where differentiation using existing biomarkers is difficult.

Significant changes are also expected in the drug development pipeline. When targeting proteins in tumors or autoimmune diseases, it has become clear that it is difficult to overcome efficacy variations or resistance issues by assuming only a single-target model based on genetic sequences. This provides the background for the required development of next-generation targeted therapy strategies that bind selectively to specific variants.

At the same time, technical barriers to overcome are evident. The technology for high-sensitivity quantitative measurement of non-genetic protein variants existing in trace amounts within cells in routine clinical environments is still in the process of advancement. Furthermore, establishing a functional verification screening system that can clearly distinguish pathogenic variants that actually cause disease from simple biological noise among numerous proteoforms is identified as a task for follow-up research.