Interactome analysis of Caenorhabditis elegans synapses by TurboID-based proximity labeling

Murat Artan,Stephen Barratt,Sean M Flynn,Farida Begum,Mark Skehel,Armel Nicolas,Mario De Bono

doi:10.1016/j.jbc.2021.101094

Abstract

Proximity labeling provides a powerful in vivo tool to characterize the proteome of subcellular structures and the interactome of specific proteins. The nematode Caenorhabditis elegans is one of the most intensely studied organisms in biology, offering many advantages for biochemistry. Using the highly active biotin ligase TurboID, we optimize here a proximity labeling protocol for C. elegans. An advantage of TurboID is that biotin's high affinity for streptavidin means biotin-labeled proteins can be affinity-purified under harsh denaturing conditions. By combining extensive sonication with aggressive denaturation using SDS and urea, we achieved near-complete solubilization of worm proteins. We then used this protocol to characterize the proteomes of the worm gut, muscle, skin, and nervous system. Neurons are among the smallest C. elegans cells. To probe the method's sensitivity, we expressed TurboID exclusively in the two AFD neurons and showed that the protocol could identify known and previously unknown proteins expressed selectively in AFD. The active zones of synapses are composed of a protein matrix that is difficult to solubilize and purify. To test if our protocol could solubilize active zone proteins, we knocked TurboID into the endogenous elks-1 gene, which encodes a presynaptic active zone protein. We identified many known ELKS-1-interacting active zone proteins, as well as previously uncharacterized synaptic proteins. Versatile vectors and the inherent advantages of using C. elegans, including fast growth and the ability to rapidly make and functionally test knock-ins, make proximity labeling a valuable addition to the armory of this model organism.

Highlights

Proximity labeling provides a powerful in vivo tool to characterize the proteome of subcellular structures and the interactome of specific proteins
We optimize the recently developed TurboID-based proximity-dependent protein labeling approach for C. elegans, and show that it permits single neuron proteomics and characterization of the interactome of a synaptic protein expressed at endogenous levels
We create reagents that allow TurboID to be applied to different C. elegans tissues and identify >4000 proteins expressed in at least one tissue

Summary

Introduction

Proximity labeling provides a powerful in vivo tool to characterize the proteome of subcellular structures and the interactome of specific proteins. By combining extensive sonication with aggressive denaturation using SDS and urea, we achieved near-complete solubilization of worm proteins We used this protocol to characterize the proteomes of the worm gut, muscle, skin, and nervous system. Characterizing the interactomes of specific proteins, and the proteome profiles of subcellular structures, cells, and tissues, provides a powerful entry point to probe molecular. Each method has limitations that can include high false-positive rates, poor detection of transient or weak interactors, a low signal-to-noise ratio when detecting PPIs in specific cell types or subcellular compartments, artifacts created during tissue homogenization, and competing requirements for solubilizing proteins while keeping complexes intact [1, 2]. Proximity-labeling methods overcome many of these limitations [1, 6,7,8] and have allowed the proteomes of subcellular compartments [1]

Objectives

Results

Conclusion