Identification of Synaptic DGKθ Interactors That Stimulate DGKθ Activity

Abstract

Lipids and their metabolic enzymes are a critical point of regulation for the membrane curvature required to induce membrane fusion during synaptic vesicle recycling. One such enzyme is diacylglycerol kinase θ (DGKθ), which produces phosphatidic acid (PtdOH) that generates negative membrane curvature. Synapses lacking DGKθ have significantly slower rates of endocytosis, implicating DGKθ as an endocytic regulator. Importantly, DGKθ kinase activity is required for this function. However, protein regulators of DGKθ's kinase activity in neurons have never been identified. In this study, we employed APEX2 proximity labeling and mass spectrometry to identify endogenous interactors of DGKθ in neurons and assayed their ability to modulate its kinase activity. Seven endogenous DGKθ interactors were identified and notably, synaptotagmin-1 (Syt1) increased DGKθ kinase activity 10-fold. This study is the first to validate endogenous DGKθ interactors at the mammalian synapse and suggests a coordinated role between DGKθ-produced PtdOH and Syt1 in synaptic vesicle recycling.

Keywords: APEX2; Syt1; lipids; neurotransmission; proximity labeling; synapse; synaptic vesicles.

FIGURE 1

DGKθ-GFP-APEX2 is expressed ubiquitously throughout neurons and labeling of interactors is dependent on H₂O₂. (A) DGKθ-GFP-APEX2 was cloned into the neuronal expression vector pCAG. (B) Cortical neurons (DIV21) expressing DGKθ-GFP-APEX2 were incubated with biotin-phenol (BP, 500 μM, 30 min). Depolarized neurons were treated with a 30 s incubation with 50 mM KCl; control neurons were treated with aCSF. The addition of H₂O₂ (1 mM, 1 min) catalyzes the reaction. Figure made with BioRender.com. (C) DIV21 cortical neuron lysates express near equal amounts of DGKθ-GFP-APEX2 with and without depolarization. (D) DGKθ-GFP-APEX2 biotinylated proteins with similar efficacy in depolarized and non-depolarized samples, and labeling is dependent on BP and H₂O₂. (E) In cortical neurons (DIV19), DGKθ-GFP-APEX2 is localized throughout the cell, in spines, and colocalizes with vGlut1 and DGKθ overexpression. Scale bars represent 20 and 5 μm (cropped region).

FIGURE 2

Differential labeling of synaptic proteins by DGKθ-GFP-APEX2 with KCl. (A) Biotinylated tyrosine residues within DGKθ-GFP-APEX2, corresponding to peptides identified in one (*), two (#), or three (†) experiments. (B) Mass spectrometry experiments 1 and 3 consisted of three KCl-depolarized samples and three control samples, while experiment 2 consisted of two of each. Each sample contains ∼50 e⁶ cells. The three experiments produced similar numbers of total biotinylated peptides and their corresponding proteins. (C) The differential labeling of peptides with KCl is illustrated for mass spectrometry experiment 2. All biotinylated peptides identified are graphed by their log₂(peptide abundance). Peptide abundance is calculated as (relative abundance in KCl-depolarized samples)/(relative abundance in control samples). (D) Of the total number of peptides identified from our three mass spectrometry experiments, 51% were found in all three experiments, 22% were found in two of the three experiments, and 27% were found in only one experiment.

FIGURE 3

Selection of candidate DGKθ proteins of interest. (A) Criteria for the selection of candidate DGKθ-interacting proteins. (B) The selected candidate DGKθ interactors and one or two biotinylated peptides (all of which were identified in all three mass spectrometry experiments), with their PSMs in each condition. Site of biotinylation, PSMs, and % protein coverage from mass spectrometry experiment 1.

FIGURE 4

Validation of candidate DGKθ-interacting proteins. (A) Streptavidin beads were used to isolate biotinylated proteins from DGKθ-GFP-APEX2 labeled cortical neuron lysates (DIV21), and biotinylation was confirmed biochemically for dynamin-1, synaptogyrin-1, Munc18-1, clathrin heavy chain 1, Syt1, CaMKIIα, Hsc70, and syntaxin-7. (B) DGKθ overexpressed in HEK cells is immunoprecipitated with a myc antibody. (C) DGKθ interacts with Hsc70, CaMKIIα, Syt1, synaptogyrin-1, dynamin-1, syntaxin-7, and Munc18-1 when both are overexpressed in HEK cells. Hsc70 interacts with DGKθ at its endogenous expression level. (D) Clathrin heavy chain 1 does not interact with DGKθ when overexpressed in HEK cells. All blots are representative of at least three separate experiments.

FIGURE 5

Syt1 increases DGKθ activity 10-fold over DGKθ alone. (A) List of candidate proteins for which an interaction with DGKθ was validated and their function. Syt1 has the highest percent basicity of the confirmed interactors. (B) DGKθ kinase activity was measured for DGKθ alone and for DGKθ (10 ng/μL) plus the addition of each confirmed interactor (500 ng/μL). Syt1 increased DGKθ activity 10-fold, while syntaxin-7 increased DGKθ activity four-fold. Error bars indicate SEM. Graph summarizes data from 2 experiments with three replicates each, n = 6, *p < 0.05, ***p < 0.0005, ****p < 0.0001 against DGKθ only, ratio paired t-test.

FIGURE 6

Model for DGKθ’s interaction with Syt1. Syt1 is a transmembrane protein on synaptic vesicles and the presynaptic membrane. Perhaps to aid in the membrane deformation required in the fusion and fission of vesicles, Syt1 binds to DGKθ and stimulates its kinase activity, producing local hotspots of PtdOH that generate negative membrane curvature. AZ = active zone, EZ = endocytic zone. Figure made with BioRender.com.