Proteins interacting with cell membranes play a vital role in countless biological processes, from how cells communicate to how they respond to external signals like hormones or medications. Understanding these interactions at a molecular level is crucial for advancing medicine, especially in designing drugs that target these proteins.
A recent study, led by Lawrence Livermore National Laboratory (LLNL) scientists and published in The Journal of Chemical Physics, is offering new insight into modeling these complex interactions using a combination of detailed molecular simulations and large-scale models. Researchers said the work is a significant step forward in modeling protein-membrane interactions - resulting in a more accurate and scalable tool for studying cellular membranes, which could have potential applications in drug design and understanding fundamental biology.
"This new model will allow us to study complex interactions at a much larger scale and with greater fidelity than previously possible," said principal investigator Tim Carpenter, LLNL's Biochemical and Biophysical Systems deputy group leader. "Drug interactions with proteins in the membrane do not adhere to the same norms or conventions as soluble drugs, so these types of models will help advance study in that area of mostly untapped therapeutic potential."
A new perspective through anisotropic modeling
Cell membranes are intricate structures, composed of a double layer of lipids interspersed with proteins. These lipids come in hundreds of varieties, each with unique properties that can influence protein behavior. Likewise, proteins don't just passively sit in the membrane - they actively shape their immediate environment by organizing the surrounding lipids, creating what scientists call a "lipid fingerprint." These fingerprints are essential for the proteins to function properly.
Traditional models often simplify these interactions, treating them as uniform and isotropic (the same in all directions). While this approach works for simpler systems, it falls short when applied to the dynamic and complex environment of cellular membranes.
As described in the paper, LLNL researchers developed a novel model to better capture the complexity of protein-membrane interactions. Their approach is based on dynamic density functional theory (DDFT), a method that allows them to represent the distribution of lipids as a continuous field rather than as individual molecules. This method maintains the molecular-level details while enabling simulations on a much larger scale.
The team adapted their previous protein-membrane model, which was inherently one-dimensional, and expanded it to convey complex, two-dimensional details. What sets the new model apart is its ability to incorporate "anisotropic" interactions, meaning it accounts for directional differences in how proteins and lipids interact. This is critical for replicating the unique lipid patterns that form around different proteins and for understanding how these patterns influence biological functions.
To demonstrate the power of their model, the researchers applied it to two biologically significant systems: the RAS-RAF complex and G Protein-Coupled Receptors (GPCRs). The RAS-RAF complex plays a pivotal role in cell signaling and is often implicated in numerous forms of cancer. Anchored to the cell membrane, it interacts with specific lipids to regulate cell growth and division. GPCRs are integral membrane proteins that span the entire lipid bilayer and are involved in many critical processes, such as responding to adrenaline or other signals. These proteins are also the target of approximately 30% of all U.S. Food and Drug Administration-approved drugs.
The team's new model successfully replicated the distinct lipid patterns around the RAS-RAF protein complex, providing insights into how these interactions influence its function. And modeling the lipid environment around a specific GPCR allowed the researchers to explore how the protein's active and inactive states affect its surrounding lipids.
One of the biggest advantages of this new approach is the ability to bridge two traditionally separate scales of study: molecular dynamics simulations - which are highly detailed but limited to tiny systems and short timeframes - and continuum models, which can simulate much larger systems but often lack molecular detail, according to the team.
By combining the strengths of both, the model allows scientists to study protein-membrane interactions on scales that are biologically relevant, both in size and time, opening the door to phenomena that were previously out of reach using current computational methods, such as how proteins aggregate or how they recruit specific lipids, Carpenter said.
"Large-scale protein-protein interactions within the cell membrane is an area that the community is just starting to embrace and appreciate the importance of, so we are well positioned to contribute to the field," Carpenter said. "The code is in the process of being open-sourced, and we are encouraging other groups to use it. The model has been presented at several conferences, where it has attracted great interest."
Researchers said the insights gained through the model could lead to breakthroughs in drug development, as membrane proteins are key players in many diseases, and understanding their lipid environments and how a protein organizes its surrounding lipids might help scientists design drugs that disrupt harmful interactions or enhance beneficial ones.
Additionally, the work highlights the value of interdisciplinary approaches - by integrating advanced mathematical models with biology and data from molecular simulations, the researchers have created a tool that could enhance basic understanding of cellular dynamics.
Team members said there is still additional work to be done; while the current model focuses on flat membranes, real cell membranes curve, bend and have regions with varying stiffness. Future studies could extend this approach to incorporate these features, making the simulations even more realistic. The researchers added that as experimental techniques continue to improve, they will be able to test and refine these models against real-world data, likely leading to even greater insights into the complex dance of proteins and lipids within cells.
Co-authors included LLNL scientists Jeremy Tempkin, Tugba Ozturk and Helgi Ingólfsson, former LLNL computer scientist Tomas Oppelstrup, now of Cerebras, and former Lab scientist Liam Stanton, now an associate professor at San Jose State University.
