Interrogation and validation of the interactome of neuronal Munc18-interacting Mint proteins with AlphaFold2

Abstract

Munc18-interacting proteins (Mints) are multidomain adaptors that regulate neuronal membrane trafficking, signaling, and neurotransmission. Mint1 and Mint2 are highly expressed in the brain with overlapping roles in the regulation of synaptic vesicle fusion required for neurotransmitter release by interacting with the essential synaptic protein Munc18-1. Here, we have used AlphaFold2 to identify and then validate the mechanisms that underpin both the specific interactions of neuronal Mint proteins with Munc18-1 as well as their wider interactome. We found that a short acidic α-helical motif within Mint1 and Mint2 is necessary and sufficient for specific binding to Munc18-1 and binds a conserved surface on Munc18-1 domain3b. In Munc18-1/2 double knockout neurosecretory cells, mutation of the Mint-binding site reduces the ability of Munc18-1 to rescue exocytosis, and although Munc18-1 can interact with Mint and Sx1a (Syntaxin1a) proteins simultaneously in vitro, we find that they have mutually reduced affinities, suggesting an allosteric coupling between the proteins. Using AlphaFold2 to then examine the entire cellular network of putative Mint interactors provides a structural model for their assembly with a variety of known and novel regulatory and cargo proteins including ADP-ribosylation factor (ARF3/ARF4) small GTPases and the AP3 clathrin adaptor complex. Validation of Mint1 interaction with a new predicted binder TJAP1 (tight junction-associated protein 1) provides experimental support that AlphaFold2 can correctly predict interactions across such large-scale datasets. Overall, our data provide insights into the diversity of interactions mediated by the Mint family and show that Mints may help facilitate a key trigger point in SNARE (soluble N-ethylmaleimide-sensitive factor attachment receptor) complex assembly and vesicle fusion.

Keywords: AP3; APP; ARF3; AlphaFold; CASK; IQSEC1; LRP; Mint; Munc18; SNARE; STXBP1; TJAP1; X11; calsyntenin; neurexin.

Figure 1

The Mint1 MID interacts directly with Munc18-1 but not Munc18-3.A, schematic diagram of the human Mint proteins. B, pulldowns with GST-Mint1 MID show a direct interaction with purified Munc18-1 but not Munc18-3. Image shows Coomassie blue–stained gel. C, pulldowns with GST-tagged Mint1-truncated sequences identify residues 261 to 282 as sufficient and required for Munc18-1 binding. Image shows Coomassie blue–stained reducing SDS-PAGE gels. AHM, acidic α-heilcal motif; CID, CASK-interacting domain; GST, glutathione-S-transferase; MID, Munc18-1-interacting domain; Mint, Munc18-interacting protein; PTB, phosphotyrosine binding domain.

Figure 2

A conserved sequence in Mint1 and Mint2 binds Munc18-1 and is not influenced by phosphorylation.A, ITC of synthetic Mint1^261–282 peptide binding to purified Munc18-1. The top shows raw ITC data, and the bottom shows integrated and normalized data fit to a 1:1 binding model. B, sequences of peptides tested for Munc18-1 binding by ITC. The minimal and highly conserved sequence required for Munc18-1 binding is shaded. Mutated peptide residues are highlighted in blue. C, endogenous Munc18-1 is bound to GFP-tagged Mint1 but not the mutant GFP-Mint1^D269A/I270A. GFP-tagged Mint1 proteins were transiently transfected into PC12 cells, immunoprecipitated with GFP-nanobody–coupled beads, and the bound proteins probed by Western blot with anti-Munc18-1. ITC, isothermal titration calorimetry; Mint, Munc18-interacting protein.

Figure 3

Modeling of Munc18-1 in complex with the Mint1 AHM sequence.A, AlphaFold2 prediction of the complex between Munc18-1 and Mint1 AHM. The three top ranked models are overlaid and shown in backbone ribbon representation. The AHM is consistently modeled in an α-helical structure associated with the Munc18-1 domain3b (highlighted in blue). On the right, the predicted alignment error (PAE) is plotted for each model. Signals in the off-diagonal regions indicate strong structural correlations between residues in the peptide with the Munc18-1 protein. Fig. S4 shows predictions of the full-length Munc18-1 and Mint1 complex as well as models of other Munc and Mint homologs and orthologs. B, the top-ranked complex of Munc18-1 and the Mint1 AHM is shown in cartoon representation, with Munc18-1 domains highlighted. C, as in (B), but the surface of Munc18-1 is shown colored for sequence conservation as calculated by ConSurf (120). Movie S1 shows an animation of the structure and its key features. D, opposite views showing details of the Mint1 AHM bound to the Munc18-1 domain3b. E, ITC of synthetic Mint1^261–282 peptide binding to purified Munc18-1 and structure-based mutants. The top shows raw ITC data, and the bottom shows integrated and normalized data fit to a 1:1 binding model. Mutated residues are highlighted in D. F, cumulative release events over time for each data group were analyzed, and the number of release events at each 10 s interval from 0 to 290 s was determined, plotted as mean ± SEM. G, total evoked release events following stimulation were measured per μm². Nonparametric Mann–Whitney U test, p < 0.05, ∗p < 0.01. N = 15 cells (WT) and 17 (R388A) from independent experiments. AHM, α-helical motif; ITC, isothermal titration calorimetry; Mint, Munc18-interacting protein; NS, nonstimulated; S, stimulated with 2 mM BaCl₂.

Figure 4

Comparison of the predicted Munc18-1–Mint1 complex with experimental structures.A, overlay of the Munc18-1 complex with Mint1 AHM predicted by AlphaFold2 and the recent crystal structure of the Munc18-1–Sx1a–Mint1 complex (47) (Protein Data Bank code: 7XSJ). The inset shows details of the binding site modeled by AlphaFold2 and observed in the crystal structure. The two structures are identical in all key respects. B, overlay of the Munc18-1 complex with Mint1 AHM predicted by AlphaFold2 and the cryo-EM structure of the Munc18-1–Sx1a–VAMP2 complex (62). The Mint1 AHM is expected to bind Munc18-1 independently of the Sxa1 t-SNARE and VAMP2 v-SNARE proteins. AHM, α-helical motif; Mint, Munc18-interacting protein; Sx1a, Syntaxin1a.

Figure 5

Mint1 and Sx1a show allosteric effects on binding to Munc18-1.A, pulldowns with GST-Mint1 MID show that Munc18-1 domain3a hinge loop is important for binding. As Mint1 does not contact domain3a, this suggests an allosteric effect on the domain3b-binding site. Image shows Coomassie-stained gel. B, ITC of synthetic Mint1^261–282 peptide binding to purified Munc18-1 (red) and Munc18-1^Δ317–333 (black) confirms the requirement of domain3a for Mint1 interaction. C, pulldowns with GST-Mint1 MID show that Mint1 can bind Munc18-1 both alone and in the presence of Sx1a. Image shows Coomassie-stained gel. D, although Mint1 and Sx1a can bind Munc18 simultaneously, ITC of Mint1^261–282 AHM peptide binding to Munc18-1 in the absence (red) and presence (black) of Sx1a shows a reduction in binding affinity and enthalpy. E, ITC of Sx1a binding to Munc18-1 in the absence (black) and presence (red) of synthetic Mint1^261–282 peptide. Together, this shows there is a subtle allosteric inhibition of Sxa1 binding to Munc18-1 in the presence of Mint1. F, ITC of Sx1a binding to Munc18-1^Δ317–333 in the absence (black) and presence (red) of synthetic Mint1^261–282 peptide. GST, glutathione-S-transferase; ITC, isothermal titration calorimetry; MID, Munc18-1-interacting domain; Mint, Munc18-interacting protein; Sx1a, Syntaxin1a.

Figure 6

Interactions of the Mint PTB domains with canonical NPxY-containing peptide motifs predicted by AlphaFold2. Overlay of the top-ranked AlphaFold2-predicted structures of the Mint1 PTB domain (blue) in complex with various NPxY-related peptide motifs (green) shown in backbone ribbon representation. These sequences are predicted to bind the canonical binding groove of the PTB domain (with similar interactions predicted for Mint2 [not shown]). The right panel shows details of the different sequences derived from various Mint-interacting transmembrane proteins. Mint, Munc18-interacting protein; PTB, phosphotyrosine binding.

Figure 7

Interactions of the Mint PTB domains with a noncanonical peptide motif from TJAP1 predicted by AlphaFold2.A, overlay of the top-ranked AlphaFold2-predicted structures of the Mint1 and Mint2 PTB domains (blue) in complex with the APP NPxY motif (green) and the N-terminal peptide of TJAP1 (brown) shown in backbone ribbon representation. B, the top-ranked complex of Mint1 PTB domain bound to APP and TJAP1 is shown in cartoon representation. The lower panel shows the surface of Mint1 colored for sequence conservation as calculated by ConSurf (120). C, details of the Mint1 PTB domain interaction with the TJAP1 peptide. D, ITC of Mint1 PTB binding to the peptide motif from APP in the presence and absence of a peptide from TJAP1. E, ITC of Mint1 PTB binding to the peptide motif from TJAP1 in the presence and absence of a peptide from APP. The top shows raw ITC data, and the bottom shows integrated and normalized data fit to a 1:1 binding model. APP, amyloid precursor protein; ITC, isothermal titration calorimetry; Mint, Munc18-interacting protein; PTB, phosphotyrosine binding; TJAP1, tight junction–associated protein 1.

Figure 8

Interactions of the Mint N terminus and PDZ domains with novel binders predicted by AlphaFold2.A, overlay of the top-ranked AlphaFold2-predicted structures of the Mint1 and Mint2 tandem PDZ domains (blue) in complex with ARF3 and ARF4 (green) shown in backbone ribbon representation. The middle panel shows the Mint1 complex with ARF3 in ribbon representation, with the position of GTP and Mg²⁺ based on the previous crystal structure of active ARF3-GTP (79). The right panel shows the same image but with Mint1 surface colored for sequence conservation as calculated by ConSurf (120). B, overlay of the top-ranked AlphaFold2-predicted structures of the N-terminal Mint2 IQSEC binding motif (IQSECbm) (blue) in complex with the C-terminal Sec7 and PH domains of IQSEC1 (green) shown in backbone ribbon representation. The middle panel shows the top-ranked Mint2 complex with IQSEC1 in ribbon representation. The right panel inset shows the details of the Mint2 interaction with IQSEC1 PH domain. C, overlay of the top three-ranked AlphaFold2-predicted structures of the Mint1 YxxΦ motif (blue) in complex with the C-terminal μ-homology domain (MHD) of the AP3 μ3A subunit (green) shown in backbone ribbon representation. The middle panel shows the Mint1 complex with μ3A in ribbon representation. The right panel inset shows the details of the Mint1 YxxΦ motif interaction with μ3A. D, overlay of the top-ranked AlphaFold2-predicted structures of the Mint1 and Mint2 tandem PDZ domains (blue) in complex with the C-terminal PDZbm of NRX1 (green) shown in backbone ribbon representation. The middle panel shows the Mint1 complex with ARF3 in ribbon representation. The right panel inset shows the details of the NRX1 interaction with Mint1 PDZ2 domain. ARF, ADP-ribosylation factor; Mint, Munc18-interacting protein; NRX1, neurexin-1; PDZbm, PDZ binding motif.

Figure 9

Mint1 structural model and interactions.A, structural model of Mint1 derived from AlphaFold2 (113). B, schematic summary of Mint1-mediated interactions, and speculative model suggesting that at the cell surface, Mint1 may act to reduce the affinity of Munc18-1 for the autoinhibited Sx1a, thus enhancing the ability of Sx1a to associate with VAMP2 and SNAP25 to form the trans-SNARE assembly required for vesicle fusion. The C-terminal domains of Mint1 in contrast are associated with other proteins containing NPxY and PDZbm sequences and ARF small GTPases that may enhance Mint1 membrane recruitment and modulate transmembrane protein trafficking. ARF, ADP-ribosylation factor; Mint, Munc18-interacting protein; PDZbm, PDZ binding motif; Sx1a, Syntaxin1a.