Sitting Down With… Michal Holčapek, Professor of Analytical Chemistry & Head of the Mass Spectrometry Group, Faculty of Chemical Technology, the University of Pardubice, The Czech Republic.
Did you always want to be a scientist?
Not at all. During my time at grammar school, it was still the communist era, and my family had no support for the Communist Party. As my grandparents owned quite a bit of property, it was very unlikely that I’d be allowed to attend university; so at the time, a research career was the last thing on my mind. Back then, I was more into listening to heavy metal, having long hair, and playing football – at a fairly high level. In fact, my dream was to become a professional football player, though the truth is I was quite far from ever being good enough for the Czech first league.
Then the Velvet Revolution happened during my final year at grammar school, and in a moment everything changed. At the last minute, I decided to study chemistry, and once I started, it wasn’t long before I’d completely fallen in love with the subject. I did hesitate when choosing between analytical and organic chemistry, but I went with analytical – which turned out to be a really good choice. That’s when I truly became interested in research, and that interest has stayed with me ever since.
What first sparked your interest in lipidomics and mass spectrometry?
It started during the first year of my PhD. Our group, led by Professor Pavel Jandera, had just received its first LCMS system, a single quadrupole instrument – which seems funny now, but back then it was the very first LC-MS installation in the Czech Republic. As a first-year PhD student, my task was to analyze biodiesel production from rapeseed oil.
Rapeseed oil is essentially a complex mixture of triacylglycerols (lipids), and when you produce biodiesel, you perform a transesterification with methanol. This results in an even more complex mixture containing tri-, di-, and monoacylglycerols, free fatty acids, and the final product: methyl esters. It’s a complicated system, but what fascinated me was how these lipid molecules have repeating structural units, meaning their retention and fragmentation behaviors are very regular.
For a researcher, that regularity is beautiful. It allows you to find consistent dependencies in retention times, fragmentation patterns, and so on. From there, I extended the work to other plant oils, and we built what was probably the largest collection of triacylglycerol analyses in plant oils ever done – possibly still to this day.
Over time, I moved into the analysis of phospholipids and sphingolipids, and eventually into true lipidomics. But it all began with that early work on triacylglycerols during my PhD.
Your work covers both the development of advanced lipid analysis methods and their use in biomedical applications, such as cancer research. What drew you to this combination – and what keeps you engaged?
There are essentially three kinds of people when it comes to liquid chromatography-mass spectrometry (LC-MS). Some come from a chromatographic background and focus entirely on separation, often referring to the mass spectrometer as merely a “detector.” Others come from the mass spectrometry side, and therefore view chromatography as nothing more than a “sophisticated injector.” Both views are wrong, because you can’t fully exploit the power of the technique unless you optimize both parts. I’ve always tried to give equal attention to chromatography and MS (mainly high-resolution); when you do that, you can truly develop top-quality analytical methods.
As for lipidomics, there came a point when I felt we already knew quite a lot about lipid analysis – naively thinking we could analyze almost everything. Of course, that wasn’t true, but it made me realize we needed to apply these methods in a way that demonstrated their true potential. I’ve always aimed to make my research understandable to those outside chemistry – the sort of explanation you could give to a grandparent. That perspective is what motivated me to apply lipidomics to disease research, particularly cancer.
Cancer, in simple terms, refers to uncontrolled cell division. For cells to divide rapidly, they need building material – and much of that material is lipid, because every cell and every intracellular organelle is surrounded by a lipid bilayer membrane. So if a tumor grows quickly, it requires a massive supply of lipids. With this in mind, it seemed obvious to me that cancer must be closely linked to lipid metabolism and composition.
That realization was the starting point for our move into cancer lipidomics. We began analyzing cancer cells, tissues, and body fluids to understand these lipid changes. From there, our work evolved into systematic studies, discoveries, patents, and publications – but it all began with that simple observation about the connection between lipids and uncontrolled cell growth.
What excites you most about the high-resolution MS and separation techniques you employ – and what are their current limitations?
For me, it comes down to how complementary these two technologies are. Chromatography gives you the ability to separate molecules that differ only slightly in structure – even isomers, such as those with different double bond positions or cis-trans configurations. The structural differences can be tiny, but their biological consequences can be significant, so it’s crucial to quantify them individually. Doing so helps reveal mechanisms of biological dysregulation, which is especially important in disease research.
Then comes mass spectrometry. It’s an extremely sensitive technique that allows you to determine the structure of the separated molecules and quantify them precisely. When you synchronize both technologies, you gain a truly powerful analytical workflow: high-throughput, automated, and capable of handling many samples efficiently.
That combination is what enables us to generate high-quality data for biological applications, including cancer research in our case; and beyond that, it’s what allows meaningful collaboration with clinicians and biologists. In many ways, that synergy represents the central role of analytical chemistry: to connect molecular-level analysis with real biological and medical insight.
Right now, the biggest limitation in these technologies is quantitation. With mass spectrometry, you always have to contend with effects like ion suppression, matrix interference, and other instrumental factors that influence quantitative accuracy. If you give the same biological samples to multiple laboratories and ask them to quantify the same lipids, you’d expect identical results – but that’s rarely the case. Even small variations in methods, calibration, or sample handling can lead to significant differences. This inconsistency becomes a major obstacle when you want to translate lipidomics into a clinical context, where reproducibility and harmonization are essential.
Many groups are aware of this issue now: the International Lipidomics Society (ILS), Lipid MAPS, and the Metabolomics Society to name a few. Within ILS, we’re running large ring trials to address the problem. In the current study of Clinical Lipidomics Interest Group (CLIG) in ILS, 24 research groups are analyzing a reference human plasma materials from NIST, and we’re working hard to harmonize the results across all sites. The project is divided into three stages, and we plan to publish a manuscript showing that this harmonization is achievable.
So while quantitation is still the bottleneck, it’s improving rapidly. We’re learning how to make lipid and metabolite quantification consistent across laboratories, and once that’s achieved, it will be a critical step toward true clinical translation.
What’s the biggest challenge currently facing lipidomics as a field?
As previously mentioned, the biggest challenge in lipidomics right now is translating our research into the clinical environment. And the key obstacle there is achieving reliable, robust, and high-throughput quantitation. This is, at its core, a more general hurdle in analytical chemistry.
Quantitative lipidomics demands methods that aren’t just fast, but also extremely consistent. That means using the right internal standards, performing complete analytical validation in line with FDA or EMA guidelines, and ensuring that methods are genuinely reproducible across labs. Certified standards and well-defined quality control samples are essential.
Finally, laboratories need to take part in interlaboratory or ring trials to verify that they can produce harmonized quantitative results. Until we demonstrate that kind of consistency and comparability, lipidomics will struggle to reach clinical translation. Many people in the community recognize this issue, and there’s a lot of discussion about it within professional societies – but I fear few truly grasp the magnitude of the challenge. For me, the critical frontier is achieving high-throughput, high-confidence quantitation that meets clinical standards. That’s what will move the field forward.
Has there been a mentor or collaborator who’s had a lasting influence on your career?
In general, I’m not the kind of person who looks for idols or follows specific figures, but there is one person who had a truly important influence on my career – my PhD supervisor, Professor Pavel Jandera. He was an exceptionally strong personality in the field of chromatography, and he taught me everything I know about that area, though his experience wasn’t in mass spectrometry itself.
What really made an impact was his attitude toward precision and detail. I remember early on in my PhD, I’d bring him what I thought to be complete, impressive results, expecting him to be delighted. Instead, he would calmly respond, “This is not a bad start, but we need to verify it further.” He would insist on additional experiments, different validation techniques, and further verification until he was absolutely confident in the data. As a student who was young and eager, I found it frustrating at the time, but I later came to fully appreciate his approach. Nowadays, I’ve adopted it completely in my own work and now apply it with my students. When preparing our first papers together, for example, they go through many revisions. Every figure, table, and paragraph is refined until it’s perfect.
What’s the biggest lesson you’ve learned in your career so far?
The biggest lesson I learned from him was to never compromise on confidence; if you’re not completely sure about your data, you shouldn’t publish it.
It’s easy to get excited after the first promising measurement – everyone does – but that excitement has to be followed by careful verification. You should never publish results unless you’re certain that they’re correct and reproducible. Unfortunately, in today’s landscape of open-access publishing and sometimes weak peer review, we see thousands of papers coming out that can’t be reproduced by anyone. Often, they’re based on very small datasets, and of course you can find apparent biomarkers in such small studies, but those findings rarely hold up. To make real progress, we need large cohorts, independent validations, and multiple analytical methods confirming the same result.
A good example from my own experience was a paper we published in Nature Communications in 2022. Initially, it was rejected by Nature Medicine for being too preliminary, and they suggested independent verification. So we did exactly that – we took over 500 plasma samples from cancer patients and healthy controls, divided them into aliquots, and sent them to two independent labs, one in Germany and one in Singapore. It was a huge logistical effort – we even shipped around 10 kilograms of dry ice to Singapore to preserve the samples! We didn’t tell those labs which samples were which; they just analyzed them and sent the data back. When we processed everything statistically, the biomarkers we identified were independently confirmed.
That experience reinforced my belief: if you want to publish something truly solid – especially in top journals – you must reach the highest possible level of confidence. That’s what underpins not only good science, but also our ability to translate research into practice. It’s a slower process, but it provides us with full confidence in what we report. And in science, that confidence and rigor are worth everything.
What’s your overall outlook on the future of lipidomics and its impact on biomedical research?
I’m absolutely convinced that the future of lipidomics is very bright. It’s one of the fastest-growing omics fields, with many researchers now recognizing its potential. We’re seeing strong evidence for the use of lipid biomarkers across a wide range of diseases – not only in cancer, but also in cardiovascular and metabolic disorders, among others.
If you open any modern biology textbook, you’ll see the “omics cascade”: DNA, RNA, proteins, then metabolites and lipids. Genomics is predictive; it tells you what might happen in an organism based on its genetic code, but when you move toward the metabolic or lipid level, you’re much closer to actual biological function. This is where the phenotype – what’s currently happening in the organism – is reflected, and that is why lipidomics and metabolomics are such powerful tools for disease biomarker discovery.
When analytical chemists collaborate closely with clinicians and biologists, combining rigorous measurement with biological interpretation, the impact can be tremendous. We can contribute to earlier diagnosis, better therapeutic monitoring, and ultimately more personalized medicine.
Provided we continue to focus on quality, harmonization, and strong interdisciplinary collaboration, I see rapid growth ahead for lipidomics. If we follow those principles, the potential for meaningful clinical translation is enormous.
Newsletters
Receive the latest analytical science news, personalities, education, and career development – weekly to your inbox.
