Welcome to the age of proteomics 

Proteomics, a (very brief) primer 

In 1995, the word proteomics first appeared in a scientific paper, referring to the "protein complement" encoded in microbial genomes. In other words, the proteome is the collection of all proteins in a sample, whether that sample is a microbe, a cell, or a tissue biopsy. Proteomics is the study of the composition, structure, and activity of these proteins. Since proteins are often called the 'workhorses of the cell', proteomics gives us a unique and detailed look under the hood of fundamental life processes. ​ 

Since those first challenging steps (proteins, after all, are complex), the field of proteomics has bloomed into a thriving research avenue. 

Now, with the advent of improved computational methods – not in the least AI tools – and ever more sensitive mass spectrometry (and other) machines, proteomics seems poised to provide many breakthroughs. ​ 

Time for a chat with Simon Devos, head of VIB's Proteomics Core. The core began as a spin-off of the Gevaerts lab at the VIB-UGent Center for Medical Biotechnology in 2015. Eight experts strong, the VIB Proteomics Core offers cutting-edge proteomics services, such as shotgun proteomics, affinity purification, and post-translational modification analysis. ​ 

Hello, Simon! Thanks for taking the time to guide us through the world and future of proteomics. First, how did you get into the field? 

I got the proteomics and mass spectrometry (MS) bug during my master's thesis in the lab of Prof. Bart Devreese. I acquired hands-on experience using nano-flow liquid chromatography coupled to Fourier transform-ion cyclotron resonance mass spectrometry (let’s call it FTICR…shall we), a beast of a machine. At the time, this was still pursued with 2D-PAGE protein separation, phosphoprotein staining, and collecting the protein gel spots for mass spectrometry analysis. This was technically challenging on all fronts, but that triggered me to stay in the field and become an expert. ​ 

I then embarked on several proteomics projects during my PhD and postdoc years, developing methods and technology for various projects. I also led a project on single-molecule peptide sequencing, first with imec and later with the US company Quantum-Si. Every project had its challenges, but they all fueled my passion for proteomics. Eventually, it led to the opportunity to take over the leadership of the VIB Proteomics Core. This puts me at the forefront of the proteomics field, a seat I’m happy to have. ​ 

How did you see the field change over the years?  

The greatest improvements in mass spec technology relate to speed and sensitivity. In the decade after my thesis, MS systems gradually improved, but an important milestone in MS innovation happened at ASMS in 2023, where companies launched systems that truly pushed the boundaries in sensitivity and speed. ​ The leap in sensitivity even enables MS-based single-cell proteomics, which was long thought to be impossible.

But speed and throughput are for sure impactful. I remember analyzing complex proteomes using 3-hour runs per sample, sometimes even with up-front fractionation (2D-LC). Now we can dig deeper into the proteome with up to 10x shorter run times. But the improvements are not limited to MS technology. Sample preparation methods, LC separation technology, MS acquisition methodology, data analysis… significant improvements happened on all fronts that jointly pushed the proteomics field forward. ​ 

Proteomics also showed its true potential in spatial biology. With laser-capture microscopy and sensitive, fast MS systems, we can perform proteome analysis on differential cell-types or structures directly from fresh or fixated tissue (sorting from tissue if you will). This is a very powerful technique to study biology at the protein level. For example, in the context of a VIB Grand Challenges project (BE.Amycon) we are fine-tuning amyloidosis typing by dissecting amyloid from patient tissue and analyzing the samples with ultra-sensitive MS. Spatial proteomics meets clinical proteomics. ​ 

Technically, the term proteomics is thirty years old (but protein analysis is older). Yet, there seems to be the feeling that the field is now poised to make a big splash. Do you agree? Why (not)?  

It did, at least for MS-based proteomics. However, there are a lot of other innovations cooking. Affinity-based proteome assays use either antibodies or aptamers carrying DNA barcodes that can be read out through sequencing. These assays greatly improved in terms of reproducibility and precision but remain focused on the clinical proteomics field (you need binders against a big part of the proteome after all), but it very well might find its niche there.

Another field that is slowly maturing is single-molecule proteomics. Companies are working on ways to either capture peptides or proteins or on single-molecule sensors, or pulling them through biological or solid-state nanopores and scanning the amino acid sequence. The holy grail is to truly sequence proteins and to enable digital quantification. At this point, it is unclear if that big splash will happen soon. ​ 

Beyond software, are there other advances that make you think breakthroughs could be on the way?  

Ion mobility is a powerful and orthogonal separation technique in proteomics. However, having an ion mobility analyzer in MS instruments, either integrated or as an upfront add-on, can deliver an extra boost in depth and throughput, and could even replace LC separation in the future. ​ 

Another thing to watch is de novo sequencing in proteomics. Currently, MS-acquired peptide fragmentation spectra are matched to theoretical spectra generated in silico from a protein sequence database. Hence, you only see what is in the database. With de novo sequencing, peptide sequences are derived directly from the fragmentation spectra without the need for a database. This will be useful in exploring unique mutations, other protein variants, and non-standard protein products (the ‘dark proteome’). And it is already finding its way in the field of immunopeptidomics and metaproteomics. Several machine learning algorithms already exist but are still maturing. ​ 

Finally, let's get a bit speculative. Think ten years ahead. Where does proteomics stand? What technologies will have changed the field? Which new insights, speaking broadly, might we have arrived at? ​ ​  

MS-based proteomics is still stuck to a one-by-one sample analysis setup and is thus missing parallelization and multiplexing as seen in the nucleomics field. I think MS technology will keep on making improvements in speed, sensitivity, and reproducibility, however affinity-based proteome assays might very well become true players in the field of clinical proteomics and even single cell proteomics, as this highjacks the benefits of DNA sequencing technology. It will also make proteomics more accessible (MS systems are complex), and likely more affordable. ​ 

Putting my passion for mass spectrometry aside for a moment, I think the technology will be around for a long time as it is still the only way to assess the proteome in an unbiased way, regardless of its origin. MS can identify and quantify proteins and proteoforms across the Kingdom of Life. 

Thank you, Simon! 

Gunnar De Winter

Science Communications Expert, VIB

• gunnar.dewinter@vib.be