What is MicroScale Thermophoresis (MST)?

MicroScale Thermophoresis, or MST, is a biophysical technique that measures how molecules interact with each other by tracking their movement through a tiny temperature gradient. When a focused infrared laser heats a small spot in a sample-filled capillary, molecules drift either toward or away from the warm zone, and the speed and direction of that drift change when a molecule binds to a partner. Because this thermophoretic behavior is sensitive to a molecule’s size, charge, and hydration shell, MST can detect binding events ranging from large protein-protein complexes down to interactions with small drug-like molecules or even ions. The technique has gained traction across academic labs and pharmaceutical companies since its commercialization in the early 2010s, largely because it works in solution, requires very little material, and can handle messy biological samples that trip up older methods.

The Physical Phenomenon Behind MST

Thermophoresis is the directed movement of molecules along a temperature gradient. It has been studied since the 19th century, but its molecular basis was only clarified in detail more recently. In simplified terms, a molecule’s tendency to move toward or away from heat depends on how it interacts with the solvent molecules around it. The key factors are the molecule’s size (larger molecules experience more friction), its surface charge (ions redistributing around the molecule create an entropic pull), and the structure of its hydration shell (how tightly water molecules cling to its surface).¹ When any of these properties change, the molecule’s thermophoretic behavior shifts measurably.

This is exactly what happens during a binding event. When a protein binds a ligand, its effective size increases, the charge distribution on its surface changes, and water molecules rearrange around the newly formed complex. Even if the binding partner is tiny compared to the protein, the alterations in charge and hydration can produce a strong thermophoretic signal. That is why MST can pick up interactions between two similarly sized proteins with the same sensitivity as interactions involving small molecules or ions.²

Interestingly, the direction of thermophoretic movement can reverse depending on conditions. At lower temperatures, some molecules actually move toward the heat source instead of away from it. Research on DNA and polystyrene beads showed this sign change and attributed it to a shift in the entropy of hydration at different temperatures.³ For practical MST experiments, this underlying complexity is handled by the instrument’s software, but it helps explain why the technique is so sensitive: thermophoresis integrates multiple physical properties of a molecule simultaneously, giving it a rich signal to work with.

How the Instrument Works

An MST instrument is essentially a fluorescence microscope paired with an infrared laser. The infrared laser focuses through the microscope objective onto a thin glass capillary containing your sample, creating a localized temperature increase over a spot roughly 25 micrometers across. The same objective simultaneously excites and collects fluorescence from labeled molecules in the capillary.² This all-optical design means there are no moving parts touching the sample and no surfaces the molecules need to stick to.

The capillaries themselves are remarkably small, typically about 100 micrometers in diameter, holding a total volume of roughly 500 nanoliters. That translates to very low sample consumption, which matters when you are working with proteins that took weeks to purify or are available only in small quantities.² In a typical experiment, you prepare a dilution series of one binding partner, mix each concentration with a fixed amount of the fluorescently labeled molecule, load each mixture into a capillary, and let the instrument scan through them. The whole process can be done in under an hour for a straightforward binding curve.

One detail worth noting: the capillaries come in different surface chemistries. Standard glass capillaries work for many proteins, but some molecules adsorb to the glass walls, which drains material from solution and distorts the signal. Premium capillaries coated with a proprietary polymer minimize this problem.⁴ Choosing the right capillary type is one of the first troubleshooting steps when setting up a new assay.

What MST Actually Measures

The primary output of an MST experiment is a binding curve from which you can extract the dissociation constant, or Kd. This number tells you how tightly two molecules hold onto each other: a low Kd means high affinity (the molecules bind strongly), while a high Kd means weaker binding. To get there, the instrument measures how much the fluorescent molecule’s concentration changes in the heated zone at each ligand concentration. Plotting these concentration changes against the amount of added binding partner produces a sigmoidal curve, and fitting that curve yields the Kd.²

Beyond simple affinity, MST data can reveal cooperativity, which tells you whether binding of one ligand makes it easier or harder for a second ligand to bind. The technique can also flag practical problems during drug screening: if a compound causes the target protein to aggregate or unfold, the MST traces will look obviously different from clean binding, giving you a built-in quality check.⁵ That kind of information usually requires separate experiments with other techniques.

Advantages Over Older Binding Methods

MST entered a field already populated by well-established techniques for measuring molecular interactions, so it had to offer something genuinely different to gain adoption. The two most common comparisons are with isothermal titration calorimetry (ITC) and surface plasmon resonance (SPR), each of which has its own limitations that MST sidesteps.

ITC measures the heat released or absorbed when molecules bind. It is considered a gold standard for thermodynamic characterization because it gives you enthalpy and entropy values in addition to affinity. But it requires large amounts of purified protein, often hundreds of micrograms per experiment, and each run takes a long time. Throughput is low.² MST, by contrast, uses nanoliter volumes and can measure a full binding curve in minutes.

SPR works by immobilizing one binding partner on a gold-coated sensor chip and flowing the other partner over it, measuring changes in light refraction as binding occurs. It is powerful and widely used, but the immobilization step can be a problem. Some proteins lose activity when stuck to a surface, and the orientation of the immobilized molecule can block the binding site you care about. MST avoids this entirely because both binding partners float freely in solution. Both MST and SPR have been used to study the same interactions, for instance measuring the affinity of the immune protein perforin for calcium ions, and both yield concordant results.⁶ The choice between them often depends on whether immobilization is practical for your particular molecule and whether you need the kinetic information SPR excels at providing.

Working in Complex Biological Samples

One of MST’s most distinctive strengths is its ability to measure binding in crude biological fluids rather than just purified buffer solutions. Most biophysical techniques require your sample to be clean and well-behaved. MST can work in cell lysates, blood serum, and other complex mixtures because its readout depends on fluorescence of a labeled molecule rather than on bulk optical properties of the solution.

This capability has produced some eye-opening results. When researchers measured the binding between the signaling protein interferon-gamma and its antibody in cell extract, the apparent affinity was roughly 165 nanomolar, compared with about 10 nanomolar in clean buffer.² That 16-fold difference matters enormously: it suggests that the crowded, competitive environment inside a cell substantially weakens some interactions that look strong in a test tube. For drug development, this is crucial information, because a drug candidate needs to work under real biological conditions, not just in purified buffer.

The approach has also been extended to detect antibodies directly in human serum. By adding a fluorescently marked antigen to diluted serum and running MST, researchers could quantify both the concentration and affinity of antibodies in their native environment, without any surface attachments or purification.⁷ The practical value here extends to diagnostics and vaccine development, where understanding antibody behavior in actual blood is more informative than measuring it under artificial lab conditions.

Drug Discovery and Fragment Screening

The pharmaceutical industry has embraced MST as a tool for early-stage drug discovery, particularly for fragment-based screening. Fragment screening starts with very small molecules, typically weighing less than 300 daltons, and looks for weak but genuine binding to a drug target. These fragments serve as starting points that medicinal chemists then grow into more potent drug candidates. The challenge is that fragment binding is inherently weak and can be hard to detect.

In one campaign targeting the kinase MEK1, automated MST screening identified multiple fragment hits that were later confirmed by X-ray crystallography. Some of these hits were missed entirely by other screening methods, suggesting that MST’s sensitivity to charge and hydration changes picks up binding events that size-based techniques overlook.⁵ The same study noted that MST’s ability to flag aggregation and protein unfolding helped researchers filter out false positives early, saving time downstream.

Large collaborative efforts like the European Lead Factory have incorporated MST into their hit-characterization pipelines, using it to triage compounds identified through high-throughput screening campaigns. The technique complements other biophysical methods by providing quick, low-consumption confirmation of binding before more resource-intensive follow-up experiments.⁴

Newer Developments and TRIC Technology

MST technology has not stood still since its initial commercialization. One significant extension is the ability to measure binding kinetics, not just equilibrium affinity, in a single experiment. Traditional MST tells you how tightly something binds, but not how fast the molecules come together or fall apart. A newer approach called kinetic MST exploits what happens after the infrared laser switches off: as the temperature gradient dissipates, the system re-equilibrates, and the rate of that re-equilibration contains information about association and dissociation rates.⁸ This is useful because two drug candidates can have identical Kd values but very different kinetics, and the one that stays bound longer usually performs better in the body.

A related technology called TRIC (Temperature-Related Intensity Change) takes a different approach to the measurement altogether. Instead of tracking actual molecular movement through the temperature gradient, TRIC measures how the fluorescence intensity of a labeled molecule changes when it is heated. This intensity change is itself sensitive to the molecular environment and therefore to binding events. TRIC can be run on newer instruments designed for higher throughput, making it practical to screen large chemical libraries. A recent study used the Dianthus platform, which employs TRIC, to screen compound libraries against the immune checkpoint receptor CD28, identifying 50 initial hits from thousands of compounds in a high-throughput format.⁹ Moving from hundreds of measurements per day with classic MST to thousands per day with TRIC-based instruments has significantly expanded the technique’s reach in industrial drug screening.

Studying Difficult Targets

Some of the most interesting MST applications involve molecules that are notoriously hard to study with other methods. Membrane proteins, for example, normally sit embedded in cell membranes and lose their structure and function when extracted and purified in isolation. MST has been used to measure the binding affinity of membrane proteins directly in host membrane preparations, without the need to fully purify them first. One study evaluated the human Hedgehog receptor Ptch1 while it was still sitting in yeast plasma membranes, measuring its interaction with a small-molecule inhibitor.¹⁰ This ability to study proteins in something close to their native lipid environment is a meaningful advantage over surface-based methods that require purified, solubilized protein.

Intrinsically disordered proteins present another challenge. These proteins lack a fixed three-dimensional shape and are often involved in diseases like Parkinson’s, where the protein alpha-synuclein aggregates into toxic clumps. Because disordered proteins are floppy, traditional structural methods struggle to characterize their interactions. Thermophoresis, however, does not care about structure per se; it responds to physical properties like charge and hydration. Researchers used quantitative thermophoresis to study alpha-synuclein monomers and oligomers binding to a nanobody, observing changes in thermophoretic behavior that went beyond what simple electrostatic effects could explain.¹¹ The technique offered a window into the physical properties of these aggregation-prone species that other methods could not easily provide.

MST in Structural Biology

Beyond drug discovery, MST has become a common supporting technique in structural biology, where researchers combine it with methods like cryo-electron microscopy or X-ray crystallography to build a complete picture of how protein complexes form and function. A structure can show you the physical shape of a complex, but it does not tell you how tightly the components hold together or how binding strength varies between different states of the complex.

A good example comes from research on the RAS-RAF signaling pathway, one of the most intensively studied pathways in cancer biology. Investigators used MST to measure the affinity of the KRAS protein for different forms of the BRAF kinase, finding that KRAS bound full-length BRAF in both its active and inactive states with similar affinity, around 106 to 127 nanomolar, while binding to an isolated domain was tighter at about 26 nanomolar.¹² These measurements helped confirm that the binding site on BRAF remains accessible even in the autoinhibited state, a finding with direct implications for cancer drug design.

Similarly, in studies of the CST-Polα/primase complex involved in DNA replication, MST revealed two distinct binding events between the protein components, with affinities of roughly 3.3 micromolar and 122 micromolar. The two events produced opposite changes in the fluorescence signal, allowing researchers to tease apart the contributions of different interaction surfaces.¹³ This level of detail would be difficult to obtain from structure alone.

Characterizing PROTAC Molecules

A particularly timely application for MST involves PROTACs, a new class of drug molecules that work by simultaneously binding a disease-causing protein and a cellular disposal enzyme, forming a three-way complex that leads to the target protein being destroyed. Characterizing PROTACs requires measuring not just simple two-molecule binding but the formation of these ternary complexes, along with the cooperativity between the two binding events.

MST and TRIC have been adapted to measure both the binary interactions (PROTAC to target, PROTAC to disposal enzyme) and the ternary complex in a single workflow.¹⁴ This is valuable because a PROTAC might bind both partners individually with decent affinity but fail to bring them together effectively, or the cooperativity might be negative, meaning the presence of one partner actually makes binding to the other weaker. Understanding these nuances is essential for optimizing PROTAC drug candidates, and having a rapid solution-based method that can measure all three interactions streamlines a process that would otherwise require multiple separate experimental setups.

Practical Limitations and Where MST Falls Short

MST is not a universal solution, and experienced users are candid about its limitations. The technique requires at least one binding partner to be fluorescent, which usually means attaching a chemical label. Labeling can occasionally interfere with the binding site, though controls for this are straightforward. Some instruments can also detect intrinsic fluorescence from tryptophan residues in proteins, avoiding the need for a label, but this approach limits which wavelengths and buffer conditions you can use.

The throughput of classic MST, while better than ITC, is still modest compared with plate-based assays that measure thousands of interactions per day. Each capillary holds one condition, and scanning a full binding curve means loading 12 to 16 capillaries. Automated systems and TRIC-based platforms have improved this, but for screening libraries of hundreds of thousands of compounds, MST typically serves as a secondary confirmation method rather than the primary screen.

Reproducibility can also require care. As mentioned, protein adsorption to capillary walls is a common source of artifacts, and choosing the wrong capillary coating or LED power setting can produce misleading results. The quality of the fluorescence signal before the infrared laser turns on matters too: uneven loading, air bubbles, or aggregated protein can all create noisy data. Published guidance from groups like the European Lead Factory emphasizes systematic optimization of these parameters before trusting the binding curves that come out.⁴

Finally, MST gives you an equilibrium dissociation constant and, with the kinetic extension, on- and off-rates. It does not give you thermodynamic parameters like the enthalpy and entropy of binding, which ITC provides. If a researcher needs a complete thermodynamic profile of an interaction, MST alone will not suffice, and ITC or differential scanning calorimetry would be needed as a complement.

Why Buffer Conditions Change the Answer

One underappreciated lesson from MST work is how dramatically binding affinity can shift depending on what is in the solution. The interferon-gamma example, where affinity dropped by more than an order of magnitude in cell extract compared with clean buffer, is not an isolated case. Differences in pH, salt concentration, viscosity, and the presence of competing molecules all influence binding measurements. This is true for any binding technique, but MST’s ability to work in crude samples makes the discrepancy visible in a way that buffer-only methods never reveal.

For researchers developing therapeutics, this has a practical takeaway: a binding constant measured in phosphate-buffered saline may not reflect what the drug encounters in the bloodstream or inside a cell. MST in serum or lysate gives a closer approximation, though even that is a simplification of the real in vivo environment. The technique does not solve the problem of translating lab measurements to clinical reality, but it narrows the gap more than most competing methods do.

1
PubMed Central. Molecular interaction studies using microscale thermophoresis
2
Nature Communications. Protein-binding assays in biological liquids using microscale thermophoresis
3
PubMed Central. Why molecules move along a temperature gradient
4
SLAS Discovery. Using Microscale Thermophoresis to Characterize Hits from High-Throughput Screening: A European Lead Factory Perspective
5
PubMed Central. An Automated Microscale Thermophoresis Screening Approach for Fragment-Based Lead Discovery
6
PubMed Central. Surface plasmon resonance and microscale thermophoresis approaches for determining the affinity of perforin for calcium ions
7
PubMed. Direct detection of antibody concentration and affinity in human serum using microscale thermophoresis
8
PubMed Central. Kinetic Microscale Thermophoresis for Simultaneous Measurement of Binding Affinity and Kinetics
9
PubMed. Temperature-related intensity change (TRIC)-based high-throughput screening enables the discovery of small molecule CD28 binders
10
PubMed. Microscale Thermophoresis to Evaluate the Functionality of Heterologously Overexpressed Membrane Proteins in Membrane Preparations
11
Scientific Reports. Quantitative thermophoretic study of disease-related protein aggregates
12
Nature Communications. Cryo-EM structure of a RAS/RAF recruitment complex
13
Nature Structural & Molecular Biology. Cryo-EM structure of the human CST–Polα/primase complex in a recruitment state
14
PubMed. MST and TRIC Technology to Reliably Study PROTAC Binary and Ternary Binding in Drug Development