14-3-3 binding maintains the Parkinson's associated kinase LRRK2 in an inactive state

Abstract

Leucine-rich repeat kinase 2 (LRRK2) is an essential regulator in cellular signaling and a major contributor to Parkinson's disease (PD) pathogenesis. 14-3-3 proteins are critical modulators of LRRK2 activity, yet the structural basis of their interaction has remained unclear. Here, we present the cryo-electron microscopy structure of the LRRK2:14-3-3₂ autoinhibitory complex, revealing how a 14-3-3 dimer stabilizes an autoinhibited LRRK2 monomer through dual-site anchoring. The dimer engages both phosphorylated S910/S935 sites and the COR-A/B subdomains within the Roc-COR GTPase region. This spatial configuration constrains LRR domain mobility, reinforces the inactive conformation, and likely impedes LRRK2 dimerization and oligomer formation. Structure-guided mutagenesis studies show that PD-associated mutations at the COR:14-3-3₂ interface and within the GTPase domain weaken 14-3-3 binding and impair its inhibitory effect on LRRK2 kinase activity. Furthermore, we demonstrate that type I LRRK2 kinase inhibitor, which stabilizes the kinase domain in its active conformation, reduces 14-3-3 binding and promotes dephosphorylation at pS910 and pS935. Together, these findings provide a structural basis for understanding how LRRK2 is maintained in an inactive state, elucidate the mechanistic role of 14-3-3 in LRRK2 regulation, inform the interpretation of PD biomarkers, and suggest therapeutic strategies aimed at enhancing LRRK2-14-3-3 interactions to treat PD and related disorders.

Fig. 1. Structure of the LRRK2:14-3-3₂ complex.

a Schematic representation of LRRK2 domain organization. Residues S910 and S935, which serve as 14-3-3 binding sites upon phosphorylation, are highlighted in red. b Cryo-EM density map at a resolution of 3.96 Å (left), with the corresponding structural model shown on the right. The model is colored according to the domain color code in (a) and shown in two different orientations for clarity.

Fig. 2. Detailed interactions and mutational effects at the LRRK2:14-3-3₂ binding interfaces.

a Overview of the LRRK2:14-3-3₂ contact regions in LRRK2:14-3-3₂ complex. Insets detail the primary and secondary interaction sites, supported by the corresponding cryo-EM densities. b Close-up view of the primary interactions, where LRRK2 phosphorylation sites pS910 and pS935 engage with the 14-3-3 substrate binding grooves. Assignment of this interaction is supported by integration of structural fitting, mass spectrometry, and biochemical validation. c Close-up view of the secondary interactions, showing LRRK2 COR-A and COR-B subdomain residues contacting the α−9 helices of the 14-3-3 dimer. d Quantitative analysis of LRRK2/14-3-3 interactions through Co-IP experiments of LRRK2 with endogenous 14-3-3, comparing WT LRRK2 with mutants at the secondary interface, as well as primary interface mutants (S910A/S935A). Data illustrate the impact of mutations on the interaction strength. Refer to Supplementary Fig. 16 for representative membrane images and source data for complete membrane images. Data are mean ± SEM (n  =  3 independent experiments), significance of difference was quantified using one-way Brown–Forsythe and Welch ANOVA test and reported with the exact p values in the source data file. e Binding affinity between (WT or mutant) LRRK2 and (WT or mutant) 14-3-3 proteins was determined by MST. Mutations at the primary (left) and secondary (right) binding sites were analyzed. Data are mean ± SEM (n  =  3 independent experiments), significance of difference was quantified using one-way Brown–Forsythe and Welch ANOVA test and reported with the exact p values in the source data file. Refer to Supplementary Fig. 11 for full binding curves.

Fig. 3. Overlap of the LRRK2:14-3-3₂ and LRRK2 homodimer interfaces.

a Structural comparison of the inactive LRRK2 homodimer (PDB: 7LHW, in gray, left) and the LRRK2:14-3-3₂ complex (colored as in Fig. 1), showing the overlapping interfaces mediated by the COR-B domain (right). Loops containing the S910 and S935 sites, flexible and unresolved in the dimer structure, are illustrated with dashed lines. b Surface representations of the COR-B domain. Left: residues involved in the LRRK2 dimer interface shaded in tan. Middle: residues involved in the LRRK2:14-3-3₂ interface shaded in gray. Right: Overlay of the two interfaces showing steric clash, suggesting mutual exclusivity between LRRK2 dimerization and 14-3-3 binding. c SEC-MALS chromatogram of the pre-formed LRRK2 dimer in the presence and absence of 14-3-3, indicating changes in molecular weight distribution.

Fig. 4. 14-3-3 binding maintains LRRK2 in an inactive conformation and inhibits its kinase activity.

a Structural representation of the kinase domain in the LRRK2:14-3-3₂ complex, showing an inactive conformation characterized by an outward αC helix and a broken regulatory (R-) spine (inset). b Overlay of the LRRK2:14-3-3₂ complex (colored as in Fig. 1) with the active LRRK2 monomer from a LRRK2 tetramer (PDB: 8FO9, in gray), highlighting the conformational differences in the LRR domain, depicted as a cartoon. c In vitro kinase assay showing inhibition of LRRK2 activity by increasing concentrations of 14-3-3, measured by Rab10 phosphorylation levels (pRab10) in western blots, normalized to LRRK2 protein levels. Effects of interface mutations on kinase activity inhibition by 14-3-3, with results for LRRK2 mutations shown in (d) and 14-3-3 mutations in (e). Data in (c–e) are mean ± SEM (n  =  3 independent experiments), with representative blots for inhibition assays shown in (c). See source data for membrane images and for significance of difference with the one-way Brown–Forsythe and Welch ANOVA test with the exact p values when applicable.

Fig. 5. Structural comparison of the LRRK2:14-3-3₂ complex with active LRRK2.

a Side by side comparison of the LRRK2:14-3-3₂ complex (colored as in Fig. 1) and the active conformation of LRRK2 derived from a tetrameric structure (PDB: 8FO9, in gray), highlighting the Roc-COR domain rotation (indicated by an arrow). 14-3-3 proteins are omitted for clarity. b Superimposition of the COR domain from the LRRK2:14-3-3₂ complex (shown in coral) with the COR domain from the active LRRK2 (shown in gray). Alignments were performed by the COR-A domain (top and bottom left) and by the COR-B domain (bottom right), to illustrate the structural shifts and effects. c Binding affinity measurements of LRRK2 and 14−3-3γ in the presence of kinase inhibitors, as determined by MST in vitro. Data in (c) are mean ± SEM (n  =  3 independent experiments), significance of difference was quantified using one-way Brown–Forsythe and Welch ANOVA test and reported with the exact p values in the source data file. Refer to Supplementary Fig. 17 for full binding curves.

Fig. 6. Impact of PD-associated mutations in Kinase and GTPase domains on LRRK2:14-3-3₂ interaction.

a In vitro LRRK2 kinase activity assay comparing wild-type (WT) and various PD-related hyperactive mutants in the kinase and GTPase domains, showing increased Rab10 phosphorylation levels indicative of enhanced kinase activity. b Comparative analysis of 14-3-3 inhibition of kinase activity across WT and PD mutants, illustrating that mutations differentially impair LRRK2 regulation by 14-3-3. c Co-IP assays quantifying cellular interaction between LRRK2 WT/ PD-related mutations and endogenous 14-3-3 in cells, revealing altered binding affinities caused by specific mutations. LRRK2 kinase activity and co-IP data (a–c) are mean ± SEM (n = 3 independent experiments). Refer to Supplementary Fig. 16 and source data for membrane images and for significance of difference with the one-way Brown–Forsythe and Welch ANOVA test with the exact p values when applicable.