How neutron science helps make blood tests more reliable
Blood-based biomarkers are currently used to diagnose and monitor diseases through simple blood tests. However, in some cases they can be partially “hidden”, becoming harder to detect accurately. This may occur when biomarkers interact with lipid assemblies. Researchers investigated, with molecular accuracy, how biomarkers interact with lipid membranes. Neutron reflectometry is uniquely suited to this task, as it provides information on the structure of interfaces at the nanometre scale.
Biomarkers: importance and challenges
A biomarker is a biological molecule found in the body (for example, in the blood or tissues) that is a sign of a certain process (normal or abnormal), condition or disease. Such molecules may be used to see how well the body responds to a treatment. Blood-based biomarkers play a central role in modern medicine. Indeed, they enable disease diagnosis and therapy monitoring in a minimally invasive way through simple blood tests. Certain tumour and metabolic biomarkers are routinely analysed in clinical practice – this is the case of alpha-fetoprotein (AFP) and glycated serum albumin (gSA).
However, detecting blood biomarkers reliably is not always straightforward. In some cases, biomarker molecules can become partially “hidden”, making them harder to detect accurately. This masking can occur when biomarkers interact with lipid assemblies naturally present in blood, such as lipoprotein particles (particles made of lipids and proteins present in the bloodstream; cholesterol and triglycerides are two types of lipids found in lipoproteins). These interactions are critical when patients are not in fasting before blood tests since lipids and lipoproteins vary with food consumption.
In two recently published studies, researchers addressed this challenge by directly investigating, with molecular accuracy, how clinically relevant biomarkers interact with lipid membranes.
How neutron reflectometry can help
The team made extensive use of the two ILL neutron reflectometers, D17 and Figaro. Neutron reflectometry is uniquely suited to this task, as it provides information on the structure of interfaces at the nanometre scale. It further allows the localisation of proteins within complex membrane architectures without perturbing the system.
The work was carried out as part of the PhD project of Beatrice Barletti, now a postdoctoral researcher at LMGP (UGA). The project was conducted at the ILL within the InnovaXN EU-funded PhD programme that connects academic research, industrial challenges and large-scale research infrastructures. The study involved researchers from Université Grenoble Alpes (UGA), Centre national de la Recherche Scientifique (CNRS) and the ILL, in collaboration with the industrial partner Surgical Diagnostics Pty Ltd (Australia), a leading company in lipid-based biosensor development. Complementary molecular dynamics simulations were performed with collaborators at the Heidelberg University Biochemistry Center.
The studies focused on alpha-fetoprotein (AFP), a glycosylated tumour biomarker involved in liver cancer, and glycated bovine serum albumin (gBSA), a model biomarker relevant to type 2 diabetes (owing to its structural similarity to human serum albumin, HSA, and its ready commercial availability, BSA is a widely used model protein). Their interaction with lipid bilayers of different compositions was studied. These artificial membranes were designed to mimic key features of biological membranes.
The instrument FIGARO at the ILL.
Native BSA was used as a reference protein, allowing the researchers to disentangle the effects of lipid composition (by comparing different lipid layer compositions) and protein modification (by comparing gBSA and BSA) in modulating the interaction. The experiments revealed that biomarker–membrane interactions are strongly dependent on both protein structure and lipid composition.
More specifically, AFP was found to exhibit pronounced interactions with charged membranes, remaining adsorbed on positively charged bilayers while inducing significant disruption in negatively charged systems. Moreover, it was found interacting significantly with membranes containing ordered lipid nanodomains. In contrast, native BSA was found to interact significantly only with negatively charged membranes. The fact that this interaction was markedly enhanced in gBSA (i.e. by chemically induced glycation of BSA) provides direct experimental evidence that glycation promotes lipid binding.
By providing the first direct nanostructural evidence of how disease-related protein modifications influence lipid interactions, these studies demonstrate the power of neutron techniques to tackle complex biomedical questions. The results open the way for improving analytical strategies for cancer biomarker detection and support the development of lipid-based biosensors, with potential impact on clinical diagnostics and disease monitoring.
Biological membranes and neutron reflectometry
All life forms contain cells and organelles, which are surrounded by membranes. The main constituents of these membranes are lipids—molecules made of hydrophilic (water loving) "heads" and hydrophobic (water fearing) "tails". In the presence of water, they form closed, hollow spherical structures called vesicles. Membranes are complex systems that control the movement of substances in and out of a cell. Understanding the interactions between the membrane and biomolecules, such as proteins, is fundamental in biomedical and pharmaceutical researcher.
The double layer of lipids that forms the basic structure of membranes tends to form differentiated patches, or nanodomains, that don’t easily mix with the rest of the membrane. While much is still do be understood, they are believed to be essential for the function of the cell. The outer surface of the membrane consists of the head groups of several different lipids, which can be positive, negative, neutral or zwitterionic (contain both positively and negatively charged groups, resulting in an overall neutral charge).
Cell membranes are approximately five nanometres thick, which makes detailed atomic-scale investigation extremely challenging. Neutron reflectometry offers a solution to measure the structure of these ultra-thin layers. So-called cold neutrons, with wavelengths comparable to X-rays, are perfectly suited for the task.
The advantage of neutrons is that they carry very little kinetic energy (a few meV) rather than the several keV of X-rays, preventing membrane damage during measurement.
Taking a closer look
The four graphs on the left present the volume fraction distribution—a measure of the concentration of proteins, lipids, and water perpendicular to the membrane—at varying distances from the surface for alpha-fetoprotein (AFP) interacting with different lipid membrane compositions. These diagrams illustrate the models that best fit the experimental neutron reflectometry data, revealing distinct interaction patterns depending on the lipid composition of the membranes (zwitterionic membranes, membranes with ordered nanodomains, positively and negatively charged membranes, see box above). The observed interactions range from no interaction to incorporation, adsorption, or even disruption of the lipid bilayer structure. On the right, protein snapshots derived from molecular dynamics (MD) simulations depict the structure of AFP, with the domains most involved in membrane interaction highlighted in blue.
Neutron reflectometry made it possible to experimentally measure the reflectivity (which depends on the scattering power of the material, in turn related to its density). Several structural models, assuming different average spatial arrangements of the protein relative to the bilayer (see scheme), were then tested against the experimental data.
References:
B. Barletti, N. Paracini, G. Fragneto, J.P. Alcaraz, A. Nelson, I. Vilgrain, D.K. Martin, and M. Maccarini, Glycation Enhances Protein Association with Lipid Bilayer Membranes, Langmuir (2025) 41, 31169−31178 (https://doi.org/10.1021/acs.langmuir.5c03975)
B. Barletti, M. König, N. Paracini, G. Fragneto, J.P. Alcaraz, A. Nelson, I. Vilgrain, D.K. Martin, F. Lolicato, M. Maccarini, How lipid composition shapes the nanostructural interaction of tumor biomarker alpha-fetoprotein and bovine serum albumin with model membranes, Journal of Colloid and Interface Science, Vol. 708 (2026) 139753 (https://doi.org/10.1016/j.jcis.2025.139753)
Institutions involved in the research: Université Grenoble Alpes, Centre national de la Recherche Scientifique (CNRS) , Surgical Diagnostics Pty Ltd, Heidelberg University Biochemistry Center