Still climbing: FPLC and the peaks of modern biology

In 1975, César Milstein and Georges Köhler unveiled a way to produce unlimited identical antibodies. By fusing immune B‑cells with myeloma cells, they created hybridomas—immortal cell lines that secrete one specific antibody indefinitely. These monoclonal antibodies (mAbs), they concluded modestly, “could be valuable for medical and industrial use” (1).

This, of course, turned out to be an understatement. In 2025, the editors of Nature wrote that the discovery has “had some of the greatest impacts on science and health care in modern times” (2). Today, the mAbs market has reached $250 billion and counting, with tens of millions of patients treated. And the impact reaches far beyond medicine. As Nature noted, mAbs today are “the workhorse of research laboratories” (2). So, too, is the technology required to produce, purify, and study them.

How did that happen? Many recognized the promise of mAbs early on, but the road ahead was, as Milstein put it in his 1984 Nobel lecture, “full of pitfalls and difficulties” (3). Not least of those was building a body of knowledge around molecular structures and interactions, disease mechanisms, and the drug targets mAbs might be able to address. As remarkable as it was, Milstein and Köhler’s discovery was just one peak in a mountain range that spans structural biology, protein research, immunology, mechanical engineering, computational protein design, and molecular biology—among others.

Just as no two mountaineering expeditions are the same, every scientific breakthrough has a unique path. What they all have in common is their interdependency: the milestones, once reached, become stepping stones. Milstein and Köhler’s 1975 paper was both a culmination of—and the starting point for—decades of insights and technological advances. Here we’ll look back at some highlights.

Early footholds: the rise of molecular biology and protein science

Our story begins in the wake of a scientific revolution that put biological molecules at center stage. Between 1940 and 1965, a cascade of discoveries laid the foundation for molecular biology. Researchers established that DNA carries genetic information and uncovered a direct relationship between genes and proteins. These insights culminated in the description of the central dogma: the flow of genetic information from DNA to RNA to protein.

These conceptual leaps were propelled by technological innovation. X‑ray crystallography revealed three‑dimensional molecular structures; ultracentrifugation enabled finer separation of proteins by size and density; electrophoresis distinguished proteins and nucleic acids by size and charge.

The next wave of advances—creation of the first recombinant DNA molecules in 1972 (4), Milstein and Köhler’s hybridoma-derived mAbs in 1975, and recombinant human insulin in 1978—opened even more paths. But the ascent grew steeper. Recombinant insulin still had to be separated from cellular contaminants. mAbs from mouse or rat hybridomas were difficult to produce and risked serious side effects in humans. Likewise, studying enzymes, receptors, and other proteins depended on isolating them from complex mixtures. Protein purification technology could help address these issues, but its limitations slowed progress.

Early purification strategies such as solvent or salt precipitation and ultracentrifugation, many still used today, were instrumental. But they could be slow, and the yields and purity could vary widely from one run to another. Liquid chromatography could help here. In this technique, analytes in a mobile phase (such as a cell culture supernatant) move through a stationary phase composed of packed particles or beads. These beads separate the molecule of interest from the rest of the supernatant as it passes through. This can happen by size exclusion (where molecules that are too large to pass through get stuck in the beads), by affinity (where molecules bind to a ligand affixed to the beads), or by other mechanisms. Whatever the mechanism, the result is the same: separation of the protein of interest from the mixture. Early systems relied on gravity to pull the mobile phase through, resulting in slow, poorly reproducible separations that often had low resolution (6)—hardly ideal footing for researchers trying to maintain their grip on intact biomolecules.

High-performance liquid chromatography (HPLC), which emerged in the 1960s and 1970s (7,8), was a means to separate molecules faster and with more precision. By combining high-pressure pumps with small resin particle sizes, HPLC provided faster, more controlled separations with dramatically improved resolution. But its high pressures (tens to hundreds of bar) and use of organic solvents could denature sensitive biomolecules. For researchers climbing toward the isolation of intact, functional proteins—like those of future mAb therapies—a gentler route was needed.

A gentler incline: separation for sensitive molecules

A biocompatible alternative to HPLC was the goal of scientists at Pharmacia Fine Chemicals (now Cytiva) in Uppsala, Sweden, beginning in 1975. They sought a system that preserved HPLC’s high resolution while protecting fragile biological molecules.

After seven years of intensive research, testing, and incremental module releases, the first complete FPLC system debuted (Fig 1). Initially named fast performance liquid chromatography (FPLC) to distinguish it from HPLC, it was the world’s first integrated system designed specifically for preparative chromatography. Where HPLC focused on analytical identification, FPLC was created explicitly for biopurification.

FPLC enabled the separation of large macromolecules at gentle pressures (<5 bar). Columns packed with cross‑linked agarose beads provided larger particle diameters, allowing higher flow rates than those of gravity-based separations, at pressures still low enough to preserve samples intact.

Gaining altitude with FPLC

Once researchers had this new tool in their pack, progress quickened. As early as 1985, landmark publications used FPLC for protein purification (9,10). Bruce Beutler, for example, described in his Nobel Prize lecture the difficulty of preserving the primary structure of mouse cachectin purified from macrophage‑conditioned media. He turned to Pharmacia’s FPLC system to develop the method that finally yielded an accurate amino acid sequence (9). This sequence revealed that cachectin was the mouse ortholog to human tumor necrosis factor (TNF), an insight that proved foundational in the work for which Beutler won the Nobel Prize in 2011.

The FPLC system rapidly became a versatile tool across disciplines, purifying everything from B-cell growth factor (11) and mAbs (12,13) to plant cell–signaling molecules (14). It supported pivotal work by researchers such as Steven McKnight (15), Franz Ulrich Hartl (16), and others (17), who relied on FPLC to isolate nuclear binding proteins, folding related enzymes, and nucleotides. FPLC acted like a sturdy tether for these projects, supporting steady upward progress across widely varied scientific terrain.

An outrageous expedition: climbing Mount Ribosome

Throughout this period, another ambitious global quest was under way: solving the ribosome structure at atomic resolution. The ribosome—in bacteria, a massive RNA‑protein complex roughly 2.5 million Da—is essential for translating genetic information. It’s also an important drug target. Over half of clinical antibiotics bind to the ribosome, and solving the atomic structure would enable the design of new ones.

Yet crystallizing the ribosome posed extraordinary challenges. As one of the researchers said later, “the idea of crystallizing the ribosome was completely outrageous” (18). But there was progress. Ada Yonath at the Weizmann Institute of Science in Israel gained a foothold in 1980, tackling the problem of low resolutions in purified ribosomes. Other groups, including Thomas Steitz at Yale and Venkatraman Ramakrishnan at the MRC Laboratory of Molecular Biology in Cambridge, climbed still higher, obtaining crystals diffracting to 7–9 Å by the late 1990s (19).

As Ada Yonath told the Nobel Committee in 2009: “What we felt is that we are climbing mountains in order to reach the climax—and these mountains are like the Everest, the biggest most difficult to climb—only to find out that there is another mountain waiting behind it to be climbed afterwards” (20).

Purification was not the biggest challenge in this expedition, but instead one of the base camps. Ribosomes are fragile and prone to disassembly with shear stress or improper salt conditions. Crystallography demanded highly pure, homogeneous ribosomal subunits at high concentration—free of nucleases and proteases. The tool of choice for this work was FPLC using hydrophobic interaction.

In 2000, after decades of incremental progress, three groups reached the summit. Steitz’s team solved the 50S subunit, Yonath’s group independently solved the 50S, and Ramakrishnan’s lab solved the 30S subunit (18). The effort was a technological tour de force, earning Ramakrishnan, Steitz, and Yonath a Nobel Prize in Chemistry in 2009.

By then, FPLC had earned its place among key enabling technologies. As scientists continued finding ingenious ways to leverage the technology, Pharmacia’s FPLC system was, in turn, evolving to make purification more accessible to more users.

New routes and rising peaks: expeditions into discovery

Once one summit was reached, new peaks came into view. As structural biology expanded, FPLC remained central. Roderick MacKinnon used FPLC to purify junctional proteins for crystallography (21), work that earned him the 2003 Nobel Prize in Chemistry for elucidating ion channel structure and function. This understanding laid the foundation for understanding and treating diseases linked to ion-channel dysfunction, such as arrythmias, neurological disorders, and autoimmune disorders. FPLC technology supported purification for studies on transcription factors (22), mitochondrial biology (23), hepatitis C (24), and unique enzymatic mechanisms (25).

Purifying proteins for downstream use became just as important as purifying them for structural analysis. Frances Arnold had pioneered directed evolution in the 1990s and 2000s (26), creating enzymes with enhanced or novel functions. Her Nobel Prize-winning work was foundational in the life sciences and in drug development and manufacturing. In pharmaceutical production, for example, directed evolution has been used to optimize drug products and their manufacturing processes to improve efficacy and safety. Alex Winter, who shared the Nobel Prize in Chemistry with Arnold in 2018, used directed evolution in proteins to develop biotherapeutics. Directed evolution is used regularly today to change biopharmaceuticals to increase solubility, purifiability, or stability, and to decrease immunogenicity. The first marketed drug to be developed through this method was adalimumab, approved in 2002 for rheumatoid arthritis and other autoimmune conditions (26).

Realizing the potential of engineered enzymes and proteins required, as any expedition does, the right tools for the job. Scalability was key, and that demanded predictable, high-throughput purification of target molecules, using a method that would not disrupt their function. Here again, FPLC was the method of choice for many scientists (27–29).

By now, FPLC technology had become more sophisticated. Purification could be automated via systems like Pharmacia’s SMART chromatography system, introduced in 1990, and its next iteration, the ÄKTAexplorer system, launched in 1996. Automation and the standardization of control software further expanded scientists’ capabilities, enabling them to take advantage of higher throughputs, simplified protocols, and more reproducible results.

Ascending the impossible: team CRISPR storms the summit

In 2012, the world learned about CRISPR-Cas9, a bacterial immune system repurposed as a gene-editing tool, and biology hasn’t been the same since. But this was a demanding climb that required technical precision. The basic idea was bold: could researchers engineer a system in which the bacterial protein (Cas9), guided by a small RNA, could cut DNA at a desired location? If so, it would provide an easy way to edit genomes.

Jennifer Doudna and Emmanuelle Charpentier led the ascent (30,31). The tools they used along the way weren’t flashy, but they were dependable. The work involved overexpressing Cas9 in E. coli and purifying it using a multi-step chromatography protocol. RNA components were synthesized by in vitro transcription and purified (often by chromatography). When they mixed the purified Cas9 with the crRNA and tracrRNA and a target DNA, they saw a precise cut at the expected site.

The ability to easily cut DNA at chosen sites, once shown, unleashed an avalanche. As Doudna said later, “We had a sense in those very early days, in my work with Emmanuelle, that…we were onto something big, but I think we had no idea how big” (32). Within months, several groups showed CRISPR-Cas9 could edit genes inside living mammalian cells. The simplicity of the system made it accessible to many labs, which has led to countless advances: creating disease models, engineering crops, developing new therapies (CRISPR-based gene therapy is already showing promise to treat genetic diseases in clinical trials). It’s hard to overstate CRISPR’s impact. But none of this could have occurred if Charpentier and Doudna hadn’t first proven the biochemistry worked. And that proof rested on the tools of the trade: not just FPLC, but also gene cloning kits, expression vectors, and purification columns and protocols to get active Cas9.

Carrying discoveries down the mountain: scaling science to medicine

Elsewhere, climbers were conducting immunology research that would enable modern immunotherapies. This work, too, depended on isolating receptors, signaling molecules, and antibodies (33,34). Purification platforms grew increasingly important in antibody production, supporting therapeutic mAbs (35), bispecifics (36), and antibody-drug conjugates (ADCs) (37).

But reaching a summit is only half the journey; bringing discoveries safely back down to patients can be just as demanding. Purification systems proved as indispensable in biomanufacturing as in research labs. FPLC platforms became foundational tools in industrial bioprocessing, enabling workflows from bench scale to production scale. The scalability of enabling technology, like the ÄKTA™ systems that today are ubiquitous in labs and manufacturing suites, helped carry protocols seamlessly into manufacturing environments. Such platforms have supported purification of mAbs (38,39), bispecific antibodies (40), recombinant proteins (41,42), growth factors (43), and other biopharmaceuticals (44). Early FPLC systems contributed to developing rituximab, the first mAb approved for cancer and the first B cell–targeted cancer therapy (38).

Nucleic acid purification also transformed rapidly. During the development of mRNA vaccines in the 2010s, teams led by Katalin Karikó and Drew Weissman used ÄKTA™ systems to produce mRNA for preclinical and lipid nanoparticle studies (45–48). During the COVID‑19 pandemic, global production of purified mRNA demanded unprecedented scale-up—an effort made possible, in part, by chromatography systems designed for smooth climbing from benchtop to large-scale manufacturing.

Scanning the next ridgeline

In the era of synthetic and computational biology—where machine learning designs novel enzymes and artificial intelligence predicts protein structures—ÄKTA™ systems continue to serve as the bridge between in silico design and real‑world validation. They help ensure that synthetic constructs not only match predictions but also function as intended.

Nobel laureate David Baker has used ÄKTA™ systems to purify algorithmically designed proteins (49,50). Similarly, researchers at Google relied on an ÄKTA pure™ system to validate the activity of proteins created using its AlphaProteo machine learning system (51). Indeed, de novo protein design tools such as AlphaFold and RFdiffusion would not have been possible without FPLC. Such models are trained on structural protein data generated by the research community over the past several decades. The Protein Data Bank, as of February 2026, contains almost 250 000 experimentally determined protein structures (52). Most of these have been purified by FPLC.

It’s impossible to say, at the top of a mountain, that a given stretch mattered more than any other. In this landscape of scientific summits, every step proceeded from the decades of insight, persistence, and innovations that came before it. Still, when scientists were ready to push upward, the purification tools in their hands often determined how far and how safely they could climb. From the first hybridomas to today’s AI-designed proteins, systems like FPLC and ÄKTA™ chromatography systems helped transform insights into breakthroughs, and breakthroughs into therapies. As new challenges appear on the horizon, these technologies will continue to steady the climb.