Our research has focused on understanding how synapses—the connections between brain cells—differ across brain regions, cell types, and activity states. We first developed methods to isolate and study specific types of synapses from different parts of the mouse brain. By combining advanced sorting techniques with protein analysis, we characterized the molecular makeup of 18 different synapse types. We identified thousands of proteins that are either shared between synapses or enriched in particular synapse types (
https://syndive.org/(opens in new window)). These data provide a resource for understanding how molecular differences contribute to the specialized functions of different synapses (Van Oostrum et al., 2023).
fsWe then created a detailed molecular map of the hippocampus - a brain region important for learning and memory - and its synapses (
https://syndive.org/(opens in new window)). We found that different hippocampal regions have distinct molecular profiles with differences in receptors, ion channels, metabolic proteins, and adhesion molecules. The protein content was partly linked to but also decoupled from the underlying (m)RNA pool. We concluded roles for protein trafficking versus local translation in establishing compartmental organization of neurons, with distal dendrites showing increased reliance on local protein synthesis, helping neurons to respond rapidly to changes in their environment (Kaulich et al., 2025)
A related study examined the outermost layer of the cerebral cortex, known as layer 1, which receives and integrates information from many sources and plays an important role in learning-related changes. We found that this region and its synapses are metabolically active and contain many RNA molecules that can be used to produce proteins locally. The molecular profile of layer 1 was distinct from deeper layers of the cortex and showed strong similarities to a comparable distal layer of the hippocampus. These findings suggest that local protein production is an important feature of these highly specialized brain regions and provide the first comprehensive characterization of the transcripts localized to layer 1 and its synapses (Spano et al., 2025).
We also investigated how mRNA molecules are distributed across different synapse types. We identified around 4,000 mRNAs enriched at synapses (
https://syndive.org/(opens in new window)) including some that are specific to particular synapse types and others that are widely shared. Their distribution was not simply determined by how much mRNA was present in the cell body, suggesting that neurons actively transport and sort mRNA to individual synapses. Together, these findings show that the transcriptome of synapses reflects their molecular and functional diversity and may help maintain synapse-specific properties (Juengling et al., 2025).
To study individual synapses rather than groups of synapses, we developed SynDrops, a new high-throughput method for analyzing the mRNA profile from thousands of individual synapses. This approach revealed previously hidden differences between synapses, including distinct subtypes of inhibitory synapses. We also showed that synapses can change their mRNA profiles when they undergo forms of plasticity—the process by which connections between brain cells become stronger or weaker. This provides a new way to study how individual synapses change during learning and other brain processes (Perez et al., publication in preparation).
We have started using our established methods for transcriptome and proteome analysis to study how synapses respond to changes in activity. Early results show that the RNA content of synapses changes when neuronal activity is altered, including during processes that help stabilize brain activity. We are now developing methods to identify and isolate synapses according to their activity level, which will allow us to study how individual synapses remodel during different forms of brain plasticity and behavior.
The methods developed through this work have also supported several studies of how proteins are made locally within neurons. We found evidence that ribosomes—the molecular machines that make proteins—can interact directly with postsynaptic proteins, suggesting that synapses may have privileged protein-production machinery. We also found evidence that ribosomes can mature locally within dendrites, potentially allowing neurons to rapidly adjust their ability to make proteins where they are needed (Fusco et al., 2025; Bourke et al., 2026). In addition, we developed a new mouse model that allows newly produced proteins to be tracked in specific cell types, providing a useful tool for future studies of how protein production changes in diseased synapses (Alvarez-Pardo et al., 2025).
To further address the nature and responsiveness of the translation machinery in neurons, we recently adopted cryo-electron microscopy together with biochemical methods to investigate how neurons respond to cellular stress. We found that stress can cause protein production to shut down by placing ribosomes into an inactive, protective state. These inactive ribosomes form pairs called disomes. We further showed that this pairing is important for cells to remain healthy and recover after stress, revealing a previously unknown mechanism for controlling protein production in mammalian cells (Schwarz et al., 2026).
Overall, this work provides new tools and molecular maps for understanding how individual synapses differ, how they produce proteins locally, and how they change in response to activity and stress. These approaches create a foundation for studying how synaptic changes contribute to learning, brain function, and disease.
