📄 bioengineering

Structure-Guided Discovery of High-Affinity Methotrexate Aptamers for Electrochemical Drug Monitoring

This study presents a structure-guided workflow that combines high-stringency SELEX, biophysical characterization, and molecular modeling to discover and optimize two distinct methotrexate-binding DNA aptamer families, which were successfully integrated into electrochemical sensors capable of sensitive, selective, and reagent-less therapeutic drug monitoring in undiluted human blood.

Original authors: Sequeira, M. B., Pimenta, T. C., Vieira, R., Beliaev, A. S., Alexandrov, K., Sousa, S. F., Silva, S. M., Eremeeva, E.

Published 2026-10-07

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Sequeira, M. B., Pimenta, T. C., Vieira, R., Beliaev, A. S., Alexandrov, K., Sousa, S. F., Silva, S. M., Eremeeva, E.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

In the world of medicine, some drugs are powerful enough to save lives but dangerous enough to cause severe harm if the dose is slightly off. This is the case with methotrexate, a medication used to treat cancers and autoimmune conditions. Because the difference between a helpful dose and a toxic one is so small, doctors must carefully monitor how much of the drug remains in a patient's bloodstream. Currently, this monitoring requires drawing blood and sending it to a central laboratory, a process that takes time and limits how often a doctor can check on a patient. Scientists have long sought a way to measure these drugs instantly at the bedside using tiny sensors, but finding a molecular "key" that fits perfectly into the drug's lock has proven difficult. The ideal key would be a piece of genetic material, known as an aptamer, that can grab onto the drug quickly, let go when the drug is gone, and work reliably inside the complex environment of human blood.

A team of researchers has now discovered two such keys, creating a new path toward instant drug monitoring. They started by generating a vast library of random DNA strands and filtering them through a rigorous selection process designed to find only those that stick tightly to methotrexate. They did not just stop at finding the binders; they treated the discovery like a structural engineering project. By analyzing the shape of the DNA strands and testing how they behaved under different conditions, the scientists identified two distinct families of DNA molecules that recognize the drug. One family, which they named the 3315 family, proved to be remarkably sturdy, maintaining its grip on the drug even when the chemical environment changed or when mixed with human blood. The other family, the 930 family, behaved differently, requiring a specific type of mineral ion to hold its shape and function correctly.

The researchers then trimmed these long DNA strands down to their smallest, most efficient forms, much like a sculptor removing excess stone to reveal the essential statue. They found that a short version of the 3315 family, containing just 55 building blocks, could bind the drug with high precision, while a 65-block version of the 930 family performed even better, provided the right mineral ions were present. To ensure these molecules were practical for real-world use, the team tested them in conditions that mimic the human body, including temperatures of 37 degrees Celsius and mixtures containing human serum. The 3315 variant held up impressively well in these challenging conditions, whereas the 930 variant showed that its performance is tightly linked to the presence of magnesium ions. This difference suggests that the two families use completely different structural tricks to catch the drug, offering scientists multiple options for building future sensors.

Crucially, the team did not stop at understanding how the molecules work; they built a working sensor to prove their utility. They attached the shortened DNA strands to a gold electrode and coated them with a dye that changes its electrical signal when the drug binds. When they exposed these sensors to methotrexate, the electrical signal shifted in direct proportion to the amount of drug present, allowing them to detect concentrations ranging from 100 to 1,000 micromolar. The sensors were able to distinguish methotrexate from folic acid, a chemically similar molecule that often confuses other detection methods. Perhaps most significantly, the sensors worked directly in undiluted human blood, a complex fluid that usually interferes with delicate molecular measurements. This success demonstrates that the 3315 and 930 families are not just theoretical curiosities but functional tools capable of operating in the messy reality of a patient's bloodstream.

While the new sensors show great promise, the study also revealed a specific limitation that must be addressed before they can be used in all clinical scenarios. The researchers found that both DNA families struggle to tell the difference between methotrexate and a specific breakdown product of the drug called MTX0, which appears in the blood after patients receive a rescue treatment to clear the drug from their system. Because the sensors bind to both molecules with similar strength, they cannot currently distinguish between the active drug and this harmless byproduct. This means that while the technology is ready for monitoring drug levels during standard treatment, it would need further refinement to be used during rescue therapies where that breakdown product is present. Nevertheless, the work establishes a clear, step-by-step method for turning raw DNA sequences into high-performance sensors, linking the initial discovery of a binding molecule directly to a working electronic device. By combining computer modeling, physical testing, and sensor engineering, the researchers have provided a blueprint for creating the next generation of tools that could one day allow doctors to monitor critical medications instantly, anywhere, without the delay of a central laboratory.

Technical Summary: Structure-Guided Discovery of High-Affinity Methotrexate Aptamers for Electrochemical Drug Monitoring

Problem Statement
Methotrexate (MTX) is a critical antiproliferative agent for cancer and autoimmune diseases, but its narrow therapeutic index necessitates precise therapeutic drug monitoring (TDM) to prevent severe toxicity. Current TDM relies on centralized laboratory methods (e.g., HPLC, LC-MS/MS, immunoassays) that lack the temporal resolution and portability required for high-frequency, point-of-care monitoring. While electrochemical aptamer-based (EAB) sensors offer a promising alternative for continuous monitoring, existing MTX aptamers suffer from poorly understood molecular recognition mechanisms, variable binding affinities, and a lack of structural optimization for sensor integration.

Methodology
The authors employed a comprehensive, structure-guided workflow to discover and engineer MTX-binding DNA aptamers:

High-Stringency Capture-SELEX: A library-immobilized selection strategy was used to isolate DNA aptamers. Stringency was progressively increased over 14 rounds by decreasing MTX concentration, increasing washing rigor, and employing folic acid (FA) as a counter-target to eliminate non-specific binders. Human serum was introduced in later rounds to select for robustness.
Biophysical Characterization: Selected aptamers were characterized using Isothermal Titration Calorimetry (ITC) to determine dissociation constants ( $K_D$ ) and thermodynamic parameters across varying temperatures (20–37 °C) and buffer conditions (including Mg²⁺ dependence and serum presence). Biolayer Interferometry (BLI) was used to assess binding kinetics ( $k_{on}$ , $k_{off}$ ), and strand displacement assays evaluated conformational switching.
Computational Modeling and Mutagenesis: Secondary structures were predicted using NUPACK, and 3D models were refined via molecular dynamics (MD) simulations. Site-directed mutagenesis and truncation studies were performed to identify minimal functional domains and validate recognition architectures.
Sensor Integration: The optimized aptamers were labeled with Methylene Blue (MB) and immobilized on gold electrodes. Electrochemical performance was evaluated using Square-Wave Voltammetry (SWV) in PBS and undiluted human whole blood. Selectivity against folic acid and MTX degradation products was also tested.

Key Results

Discovery of Two Distinct Aptamer Families: The study identified two novel, structurally distinct DNA aptamer families, 3315 and 930.
- 3315 Family: The full-length 3315-88 bound MTX with a $K_D$ of 89 nM. Truncation yielded 3315-55, which retained high affinity (63 nM) and demonstrated broad buffer tolerance, maintaining binding even in 10% human serum. This family functions largely independently of Mg²⁺.
- 930 Family: The full-length 930-88 bound MTX with a $K_D$ of 408 nM. Truncation produced 930-65, which achieved a $K_D$ of 44 nM in the presence of 10 mM Mg²⁺. This family exhibits a strict dependence on divalent cations for binding, suggesting a Mg²⁺-stabilized binding pocket.
Thermodynamic and Kinetic Properties: Binding for both families was found to be strongly enthalpy-driven. BLI confirmed rapid and reversible target recognition. Notably, 930-88 showed a more pronounced target-induced conformational change in strand displacement assays despite its lower initial affinity, indicating potential utility for structure-switching sensors.
Specificity and Limitations: Both aptamers showed high selectivity against folic acid and unrelated small molecules. However, both families cross-reacted with MTX0 (DAMPA), a degradation product of MTX, with affinities comparable to MTX itself. This limitation is noted as relevant for glucarpidase rescue therapy contexts but less critical for standard clearance monitoring.
Electrochemical Performance: MB-labeled variants (3315-55 and 930-65) were integrated into EAB sensors. These sensors demonstrated concentration-dependent responses to MTX in the 100–1000 µM range in PBS, successfully discriminated MTX from folic acid, and functioned directly in undiluted human whole blood.

Significance and Claims
The paper claims to establish a rational, structure-guided workflow that links aptamer discovery with mechanistic validation and electrochemical translation. The primary contributions are:

New Molecular Recognition Elements: The identification of two high-affinity MTX aptamer families with distinct structural mechanisms (one Mg²⁺-independent and robust, the other Mg²⁺-dependent and highly sensitive).
Optimized Scaffolds: The generation of compact, functional variants (3315-55 and 930-65) suitable for sensor integration, overcoming the size and stability limitations of full-length aptamers.
Proof of Concept for Point-of-Care Monitoring: The demonstration that these engineered aptamers can operate in complex, undiluted biological matrices (human blood) to detect MTX, providing a foundation for future reagent-free, point-of-care therapeutic drug monitoring platforms.

The authors emphasize that while the current aptamers do not yet discriminate between MTX and its degradation product MTX0, the established workflow and the distinct structural insights gained provide a critical foundation for further engineering and the development of next-generation biosensors.

More like this