Conserved LIR-specific interaction of Sigma-1 receptor and GABARAP

Abstract

Among its various functions, the sigma-1 receptor (σ1R) has been reported to modulate macroautophagy. It is currently unknown how this activity is mediated. We phylogenetically, structurally, and biochemically analyzed σ1R regarding its function in autophagy. We identified several putative LC3-interacting-regions (LIRs) that may mediate interactions with ATG8 proteins, which are known to promote autophagosome biogenesis, autophagic cargo reception, and lysosome fusion. Human σ1R comprises a LIR motif (hLIR5) typical for interaction with a specific ATG8, GABARAP. Biochemically, we uncovered a GABARAP-σ1R interaction depending on this motif via peptide array analysis and confirmed this via immunoprecipitation, co-localization, and proximity ligation assays. In addition, we verified a LIR-dependent presence of σ1R in isolated native autophagic vesicles. Excitingly, two point mutations within this LIR that have previously been reported to be associated with autosomal-recessive distal spinal muscular atrophy lack the ability to interact with GABARAP, highlighting the physiological relevance of the hLIR5-mediated σ1R-GABARAP interaction.

Keywords: Biochemistry; Molecular biology.

Graphical abstract

Figure 1

σ1R phylogeny Phylogenetic tree based on the amino acid sequences of σ1R in the Deuterostomia using the tunicate Phallusia mammillata as outgroup. Major vertebrate phyla are annotated with a specified color code. Schematic structures of corresponding σ1R orthologs are aligned at the tip of each branch. Homologous putative LIR motifs are highlighted within the structure. Putative LIRs corresponding to those of human σ1R are highlighted by color code.

Figure 2

In-depth analysis of LIR conservation and structural accessibility in σ1R (A) Diagram showing putative LIR motifs, their position, and their sequences in human σ1R. (B) Multiple sequence alignment of selected vertebrates shown in Figure 1. Representatives of major phyla are marked by specific background colors (mammals (red), birds (green), reptiles (gray), osteichthyes (blue), and chondrichthyes (dark gray)). Putative human LIRs (hLIR) and their conservation are highlighted in red boxes. Frequency plots of all hLIRs were calculated using skylign to visualize conservation. (C) Rotation of the monomeric human σ1R, modeled using AlphaFold2.^, Putative LIRs are highlighted by color code. (D) Crystal structure of the trimeric state of human σ1R (PDB:5HK2). Predicted hLIR5 and N-terminal acidic residues are highlighted.

Figure 3

Biochemical assessment of human σ1R LIR motifs (A) Peptide array using spotted oligopeptides of σ1R (15mers with three amino acid offset) after incubation with recombinant GST-tagged LC3A, LC3B, LC3C, GABARAP, GABARAPL1, or GABARAPL2. Colored boxes represent peptides harboring putative LIR motifs. Negative control for hLIR5 (mutant hLIR5, orange box) displays no interaction with GABARAP. Tables display all peptides containing the respective LIR motif. (B) One representative of at least three independent Western blots of GABARAP, LC3B, and σ1R following immunoprecipitation with anti-FLAG-antibodies from HeLa cells transfected with FLAG-tagged wild type σ1R (WT) or a ΔhLIR5 construct and treated with σ1R activating compounds. (C) One representative of three independent Western blots of total lysates for immunoprecipitations in B. Treatment parameters (B and C): alazocine (Ala, 2.5 μM, 2h), bafilomyin A1 (BafA1, 2 μM, 2h), blarcamesine (Blarca, 10 μM, 2h), H₂O₂ (50 μM, 2h), PRE-084 (Pre, 10 μM, 2h).

Figure 4

Evaluation of σ1R GABARAP interaction and merization state (A) Representative confocal images of untransfected HeLa cells obtained from three independent proximity ligation assays (PLA) of the interaction of endogenous σ1R and GABARAP (red), DAPI (blue), Magnification: 40×. Scale bar: 20 μm. (B) One representative of three independent Western blots of total lysates of GABARAP, σ1R, and tubulin (Tub) from wild type and σ1R-deficient (σ1R^−/−) HeLa cells. (C) Statistical analysis of B by two-way ANOVA. Post hoc p-values were calculated using Benjamini–Hochberg. Statistics are depicted as mean ± SD of three independent experiments; ∗∗∗p ≤ 0.001. (D) Representative confocal images obtained from three independent immunofluorescence stainings of wild type and σ1R-deficient (σ1R^−/−) HeLa cells, GABARAP signals are shown in magenta, p62 signals in green. Confocal images of single channels are depicted in Figure S1C. Magnification: 100×. Scale bar: 20 μm. (E) Quantification of proportion of p62 signals overlapping with GABARAP signals from D. Significance was statistically verified by two-way ANOVA. Post hoc p-values were calculated using Benjamini–Hochberg. Statistics are depicted as mean ± SD of three independent experiments with at least three randomly chosen optical fields containing at least three transfected cells; ∗∗∗p ≤ 0.001 between treatment, ###p ≤ 0.001 between wild type and σ1R^−/− cells. (F) One representative of at least three independent Western blots analyzing σ1R monomerization in HeLa cells transfected with FLAG-tagged wild type σ1R (WT) using an anti-FLAG-antibody antibody. A short exposure of the blot is depicted in Figure S1F. (G) Quantification of relative monomeric to oligomeric σ1R ratio from F. Statistics are depicted as mean ± SD, Significance was statistically verified by one-way ANOVA. Post hoc p-values were calculated using Benjamini–Hochberg. (A–G) Treatment parameters: alazocine (Ala, 2.5 μM, 2h), bafilomyin A1 (BafA1, 2 μM, 2h), blarcamesine (Blarca, 10 μM, 2h), PRE-084 (Pre, 10 μM, 2h).

Figure 5

Induction of σ1R GABARAP co-localization, clinical mutations in hLIR5, and autophagic vesicle analysis (A) Representative confocal images obtained from three independent immunofluorescence stainings of HeLa cells transfected with FLAG-tagged wild type σ1R (WT) or a ΔhLIR5 construct and treated with BafA1. Confocal images of DMSO control and single channels are depicted in Figures S1D and S1E. GABARAP signals are shown in magenta, FLAG signals in green; Magnification: 100×. Scale bar: 20 μm. (B) Quantification of GABARAP signals overlapping with FLAG signals normalized to total FLAG signals from A. Significance was statistically verified by three-way-ANOVA. Post hoc p-values were calculated using Benjamini–Hochberg. Statistics are depicted as mean ± SD of three independent experiments with at least three randomly chosen optical fields containing at least three transfected cells. (C) One representative of three independent Western blots of GABARAP and σ1R following immunoprecipitation with anti-FLAG-antibodies from HeLa cells transfected with FLAG-tagged wild type σ1R (WT) or F83L and V84L mutants treated with alazocine. (D) One representative of three independent Western blots of total lysates for immunoprecipitations in C. (E) Representative Western blot (out of three independent experiments) for LC3B, GABARAP, σ1R, SQSTM1 (p62), and Tubulin (TUB) of total lysates (TL) (10 μg) compared with isolated autophagic vesicles (AV) (2∗10⁶ vesicles) from HeLa cells treated with BafA1 (2 μM, 2h). (F) Representative Western blots for σ1R and LC3B from isolated autophagic vesicles (2∗10⁶) from HeLa cells transfected with wild type σ1R (WT) or a ΔhLIR5 construct, all cells were treated with BafA1 (2 μM, 2h) and with or without blarcamesine (Blarca, 10 μM, 2h); samples loaded were normalized to the numbers of FACS-sorted autophagic vesicles. (G) Quantification of autophagosomal σ1R in F. Significance was statistically verified by two-way ANOVA. Calculated p-values (Holm-Sidak post hoc test) are displayed above the compared column. Statistics are depicted as mean ± SD of three independent experiments. (A–G) Treatment parameters: alazocine (Ala, 2.5 μM, 2h), bafilomyin A1 (BafA1, 2 μM, 2h), blarcamesine (Blarca, 10 μM, 2h).