A healthy brain balances “activate” and “inhibit” signaling in its neural microcircuitry, with the inhibitory signaling often provided by cell receptors for the neurotransmitter GABA.

 

Unexpected variations in composition or relative arrangement of the subunits that make up these receptors, called Type-A GABA receptors (GABA_ARs), may provide new layers of regulation to fine-tune the inhibitory tone, according to research led by Andrija Sente of the Medical Research Council Laboratory of Molecular Biology in Cambridge, who now works at InstaDeep.

 

For this glimpse at the atomic structure of GABA_ARs, their diversity, and for structural insights into the long-sought mechanism of receptor assembly, Sente is the grand prize winner in cell and molecular biology of the 2024 Science & SciLifeLab Prize for Young Scientists.

 

Sente’s research used cryogenic electron microscopy to reveal the molecular structure of GABA_A receptors formed by various subunit combinations. He and his colleagues found that these subunits can assemble in unique ways, creating or removing sites where natural molecules or drugs can bind. They also discovered that a single neuron can assemble structurally and functionally distinct receptors to respond to combinations of neurotransmitter signals, Sente writes in his prizewinning essay published Nov. 15 in Science.

 

GABA_ARs are already targeted by drugs to treat anxiety, seizures and during general anesthesia. But the new research highlights how specific subunit combinations may create previously unrecognized receptor types, underscoring the need to account for these variations in drug development to avoid unintended binding sites. “To improve therapies that target GABA_A receptors, we really need to understand how these receptors are made in the cell,” Sente said.

 

When the neurotransmitter GABA binds to a GABAaR on a signal-receiving neuron, the receiving neuron becomes less active or inhibited. GABA is the primary inhibitory neurotransmitter in the brain, said Sente. “In addition, the roles of GABAaRs are increasingly recognized in peripheral systems, for example in pancreatic and immune signals or in the brain-to-gut communication.”

 

GABAaRs are made up of 19 molecular subunits, and different structures produced by different combinations of these subunits lead to different functions of the receptor. Researchers previously thought there might be about 800 different receptor types. With new information gleaned from their research, such as the importance of the relative arrangement of subunits, Sente and colleagues calculated there could be as many as 324, 727 possible receptor structures.

 

“I think it is important here to distinguish two related but different questions: which receptors can be assembled, and which do get assembled under physiological conditions,” Sente said.

 

“In neurons and other tissues expressing GABAaRs, assembly is constrained by a number of factors,” he explained. “For example, subunit expression is developmentally programmed, as well as tissue- and cell-type specific. It is also possible that in some cases certain subunits never ‘meet’ each other as they are translated in different parts of the cell.”

Sente said the assembly of different GABAaRs could be dynamically adjusted, possibly through mechanisms that regulate the production of specific subunits at the neuronal synapse.

There are known cases of longer-term adjustment in GABAaRs signaling, he added. Levels of the hormone progesterone increase during pregnancy and make GABA activate certain GABAaRs more efficiently. To maintain the balance between activation and inhibition, neurons somehow—Sente said the mechanism is unclear—reduce the number of GABAaRs containing a subunit that is sensitive to progesterone.

These receptor numbers usually rebound after pregnancy, but if they are slow to recover, they can cause an activation-inhibition imbalance that is thought be connected to postpartum depression.

“These insights have inspired one of the better examples of rational drug design in psychiatry: the development of brexanolone, a progesterone analog, which shows remarkable efficacy in treating postpartum depression,” Sente said.

Similar adaptations also result from chronic drug use. A deeper understanding of these adaptation pathways may allow the design therapeutics that do not induce tolerance or dependence. As a first step toward this goal, Sente and his colleagues determined the structure of a partially assembled GABA_AR bound to the assembly factor NACHO, offering the first glimpse into the mysterious process of GABA_AR assembly.

These findings are timely, given the growing interest in targeting receptor-associated proteins therapeutically rather than the receptors themselves. For instance, this research could help selectively enhance the neuronal expression of nicotinic acetylcholine receptors (whose assembly is also mediated via NACHO). Such an approach may offer new therapeutic avenues for Parkinson’s and Alzheimer’s diseases, where the loss of these receptors has been documented.

The Science & SciLifeLab Prize for Young Scientists acknowledges that global economic health is dependent upon a vibrant research community that needs to incentivize the best and brightest to continue in their chosen fields of research, as they begin their scientific careers.

 

2024 FINALISTS:

 

Gabriele Casirati is a molecular medicine finalist for his essay, “To target, to escape, perchance to cure.” Casirati received his medicine degree from Università degli Studi di Milano and completed his hematology residency at Università Vita-Salute San Raffaele. In 2020 he joined Dana-Farber Cancer Institute as a research fellow and, in 2023, he received a Ph.D. in molecular and translational medicine from Università Milano-Bicocca. He is a senior postdoctoral fellow at Boston Children's Hospital. His research focuses on immune engineering strategies to enhance targeted immunotherapies for hematological malignancies and hematopoietic stem cell transplantation.

Sabrina Rondeau is an ecology and environment finalist for her essay, “Digging below the surface.” Rondeau received an undergraduate degree from the Université du Québec à Trois-Rivières, an M.Sc. from the Université Laval, and a Ph.D. from the University of Guelph. She is a postdoctoral fellow at the University of Ottawa. Her research explores the far-reaching consequences of human activities on wild pollinators, including agricultural intensification, pesticide use and climate change.

Jarrod Shilts is a genomics, proteomics and systems biology finalist for his essay, “How to build a human.” Shilts received an undergraduate degree from Vanderbilt University and a Ph.D. from the University of Cambridge. He then transitioned to a spin-out biotechnology company specializing in the kinds of difficult-to-express proteins he worked with during his doctoral degree. The research group he leads develops technologies to advance the production of new therapeutic proteins.

In a Policy Forum, Chad Nelson and colleagues highlight the efforts of the U.S. Food and Drug Administration (FDA) in advancing alternative methods to reduce animal testing for regulatory use. Animal studies have been crucial for advancing disease understanding, developing therapies, and assessing the safety and effectiveness of consumer products. However, reducing animal use and developing effective alternatives is an ongoing priority. Although advances in biology, engineering, and artificial intelligence offer new opportunities to improve product safety assessments, these technologies require extensive development to meet regulatory standards. According to Nelson et al., the FDA is working to advance alternative methods – new approach methodologies (NAMs) – that improve the safety, risk, and efficacy understanding of regulated products while reducing animal use. Here, the authors highlight these efforts. Over the years, the FDA has advanced NAMs, such as in vitro ocular irritation tests, which have replaced traditional rabbit testing. The FDA has also advanced the use of computational models to assess drug safety. Additionally, the agency has developed qualification programs to evaluate and validate these alternative methods for regulatory use, ensuring their effectiveness in making regulatory decisions. The FDA encourages innovation through programs like ISTAND, which supports new tools like organ-on-a-chip technologies. Nelson et al. argue that by collaborating with stakeholders, hosting workshops, and conducting its own research, the FDA continues to lead efforts in validating and adopting alternative methods that reduce reliance on animal testing.

A new study presents “Evo” – a machine learning model capable of decoding and designing DNA, RNA, and protein sequences, from molecular to genome scale, with unparalleled accuracy. Evo’s ability to predict, generate, and engineer entire genomic sequences could change the way synthetic biology is done. “The ability to predict the effects of mutations across all layers of regulation in the cell and to design DNA sequences to manipulate cell function would have tremendous diagnostic and therapeutic implications for disease,” writes Christina Theodoris in a related Perspective. With a vocabulary of just four nucleotides, DNA encodes all the genetic information essential for life. Variations in the genomic sequence reflect adaptations selected for specific biological functions. These variations drive evolution by enabling organisms to adapt to new or changing environments. Advances in DNA sequencing technologies have allowed for genomic variations to be mapped at the whole-genome scale. These data, combined with novel machine learning algorithms, could enable the creation of a comprehensive model that can understand DNA, RNA, and protein functions and their interactions. But, while some researchers inspired by the success of large language models (LLMs) have attempted to model DNA as a "language" by applying similar techniques, current generative models tend to focus narrowly on individual molecules or DNA segments. Alongside computational limitations, this has constrained the scope of these models in capturing broader genomic interactions necessary for understanding complex biological processes.

 

Here, Eric Nguyen and colleagues present Evo – a large-scale genomic foundation model, equipped with 7 billion parameters and designed to generate DNA sequences up to whole-genome scale. Built on the StripedHyena architecture, Evo was trained on a dataset of 2.7 million evolutionary diverse microbial genomes. According to Nguyen et al., Evo excels in both predictive and generative biological tasks, achieving high accuracy in zero-shot evaluations for predicting mutation impacts on bacterial proteins and RNA, as well as in modeling gene regulation. Evo also grasps the intricate coevolution between coding and noncoding sequences, supporting the design of complex biological systems like CRISPR-Cas complexes and transposable elements. At the genomic scale, Evo can generate sequences over 1 megabase in length, a capability vastly surpassing prior models. “Future models may learn from diverse human and other eukaryotic genomes, using larger context lengths to capture distant genomic interactions over larger genomic scales,” writes Theodoris in the Perspective.