Deeper neuronal and glial proteomic insights using an optimized pipeline for proximity labeling proteomics
Proximity-based proteomics using TurboID has enabled cell-type-specific profiling without the need for cell purification, although major bottlenecks in sample lysis, biotinylated protein enrichment, digestion, and mass spectrometry (MS) parameters have limited depth of proteome coverage. Here, we systematically optimized these variables using TurboID-based labeling of BV2 microglia in vitro and brain astrocytes in vivo to define conditions that maximize proteome coverage. In microglia, the optimized protocol using 8 M urea lysis with on-bead S-Trap digestion and data-independent acquisition MS (DIA-MS) identified 4,016 proteins, double the depth of prior studies, and revealed metabolic, ribosomal, lipid-processing, autophagy, and trafficking signatures. Brain astrocyte proteomes were best recovered using SDS lysis with S-Trap digestion and DIA-MS, yielding a proteome of over 3,600 highly enriched proteins, twice the depth of prior astrocyte-TurboID studies. The expanded astrocyte proteomes captured canonical astrocyte markers as well as membrane-associated, vesicular trafficking, and presynaptic protein signatures, consistent with labeling of astrocyte-neuron interface regions, including proteins involved in receptor signaling, lipid metabolism, and plasticity at tripartite synapses, and several AD risk proteins. The increased peptide recovery following S-Trap digestion allowed the reduction of starting material to 20 {micro}g protein for DIA-MS, and enabled multiplexed tandem mass tag (TMT-MS) proteomics using even smaller samples. When applied to synaptosomes enriched from mouse brains with neuronal TurboID labeling, our pipeline identified a synapse-specific proteome of 2,529 proteins, revealing synaptic, mitochondrial and disease-relevant signatures not detectable in prior studies. By tackling critical bottlenecks from tissue processing to MS, our optimized pipelines enable cell-type and compartment-specific proximity-labeling proteomics to obtain comprehensive biological and disease-relevant insights across various biological fields.