Mical1 MO-CH-LIM domains form an intramolecular complex with its bMERB domain.

(a) A schematic diagram of human Mical1: Mical1 contains an N-terminal monooxygenase (MO) domain, calponin homology (CH) domain, Lin11, Isl-1, and Mec-3 (LIM) domain, and a C-terminal coiled-coil bMERB (bivalent Mical/EHBP Rab binding) domain. PRM: proline-rich motif. (b-f) Size exclusion chromatography (Superdex 200) coupled with multi-angle light scattering (SEC-MALS) was performed for full-length Mical1, MO-CH-LIM, MO-CH, and MO, while size exclusion chromatography (Superdex 75) coupled with SEC-MALS was conducted for the bMERB domain. The Rayleigh scattering is represented in blue, and the molecular weight distribution across the peak is shown in gray. The measured molecular weight closely corresponds to that of a monomer. (g-i) MO-CH-LIM domain construct interacts with the bMERB domain. Binding affinities were measured by titrating the bMERB domain (600 µM) with the MO-CH-LIM/MO-CH/MO domain (40 µM). Integrated heat peaks, corrected for the offset by subtracting the control experiment, were fitted to a one-site-binding model, yielding the binding stoichiometry (N), enthalpy (ΔH), entropy (ΔS), and dissociation constant (KD). Clear complex formation was observed for the MO-CH-LIM, whereas no complex formation was observed for the MO-CH and MO. The data represents results from at least three independent experiments. N.D. denotes not detected.

Full-length bMERB domain is essential for interaction with MO-CH-LIM domains.

(a-c) The full-length bMERB domain is required for MO-CH-LIM interaction. Binding affinities were measured by titrating the bMERB/bMERBH1-2/bMERBH2-3 domain (600 µM) with the MO-CH-LIM domain (40 µM). Integrated heat peaks, corrected for the offset by subtracting the control experiment, were fitted to a one-site-binding model, yielding the binding stoichiometry (N), enthalpy (ΔH), entropy (ΔS), and dissociation constant (KD). The data clearly show that the full-length bMERB construct is necessary for interaction with MO-CH-LIM. The data are representative of at least three repetitions. N.D. denotes not detected. (d) The Coomassie blue-stained SDS-PAGE analysis of purified Mical1 and the MO-CH-LIM:bMERB complex proteins, incubated in the absence or presence of the crosslinker disuccinimidyl dibutyric urea (DSBU), demonstrates efficient DSBU-dependent crosslinking. (e-f) Intramolecular crosslinking of Mical1. (e) Schematic visualization showing inter- and intra-domain crosslinks identified by mass spectrometry analysis of purified full-length Mical1 crosslinked by DSBU. Crosslinks are shown in light purple. (f) Schematic diagram showing inter and intra-domain crosslinks between purified MO-CH-LIM and bMERB domain crosslinked by DSBU. Schematic diagram illustrating inter- and intra-domain crosslinks between the purified MO-CH-LIM and bMERB domains, crosslinked by DSBU. Self-crosslinks are shown in light purple, whereas crosslinks between the isolated MO-CH-LIM and bMERB domains are shown in light blue.

AlphaFold2 model of auto-inhibited Mical1.

(a) The cartoon representation of the AlphaFold2 Mical1 model. The MO domain, CH domain, LIM domain, the helix connecting the CH and LIM domains (HelixCHL), and the bMERB domain of Mical1 are colored in orange, pink, blue, gray, and green, respectively. The insets highlight the various binding interfaces between the MO:bMERB, CH:bMERB, LIM:bMERB, and HelixCHL:bMERB domains, with hydrogen bonds and salt bridges indicated by gray dashed lines. (b-c) The cartoon representation of the bMERB domain depicts the MO domain, CH domain, LIM domain, and HelixCHL binding interfaces in orange, pink, blue, and gray dashed lines, respectively. The hydrophobic patches are shown in red dashed lines on the bMERB domain. (d) Structural alignment was performed between the oxidized mouse Mical1 MO domain (light orange, PDB: 2BRY) and the reduced mouse Mical1 MO domain (slate, PDB: 2C4C), and both were compared with the human Mical1 AlphaFold2 model (MO domain in orange and bMERB domain in green). (e) Schematic illustration showing the movement of the isoalloxazine ring of the FAD head during the catalytic cycle upon NADPH treatment. Insets depict the conformational changes of several active site residues upon NADPH treatment. (f) Results from a systematic analysis of the binding interfaces between various domains of the AlphaFold2-predicted Mical1 auto-inhibited model using PISA64.

Crucial bMERB interface residues for MO-CH-LIM interaction.

(a) Schematic illustration of the interactions between the MO-CH-LIM and the bMERB domain. Hydrophobic interactions are represented by light orange dashed lines, while hydrogen bonds (H-bonds) and salt bridges are depicted with gray dashed lines. Residues that have been mutated are highlighted in red dashed boxes. (b–i) Mutational alanine screening of the bMERB domain via ITC measurements. The binding of different bMERB mutants (600 µM) with MO-CH-LIM (40 µM) was systematically tested and affinities were determined by ITC experiments. Integrated heat peaks, corrected for the offset by subtracting the control experiment, were fitted to a one-site-binding model, yielding the binding stoichiometry (N), enthalpy (ΔH), entropy (ΔS), and dissociation constant (KD). N.D. denotes not detected. No binding was detected for the mutant F925A or the mutant L1047A_F1050A. However, a more than 5-fold reduction in binding affinity was observed for all other mutants, except for K918A_E921A, which exhibited a marginal defect. The data represent at least three repetitions.

Allostery between the Rab binding domains.

(a) A schematic diagram of human Mical1. (b) Full length Mical1 binds to Rab8 in 1:2 stoichiometry. Binding affinities were measured by titrating the Rab8 (800 µM) to full-length Mical1 (40 µM). Integrated heat peaks were fitted to a two-site-binding model yielding the binding stoichiometry (N), the enthalpy (ΔH), the entropy (ΔS), and the dissociation constant (KD). (c) Cartoon representation of the bMERB:Rab10 complex structure (PDB: 5LPN)28, with the bMERB domain colored in green and the Rab10 domain in wheat. (c-d) Schematic illustration of the interactions between the bMERB domain and Rab10. Rab10 binds to the Mical1 bMERB domain via its N-terminal regions, switch regions, and inter-switch region. Hydrogen bonds and salt bridges are depicted with gray dashed lines, while light orange dashed lines indicate hydrophobic interactions. RabSF1, RabSF2, RabF1, RabF2, RabF3, and RabF4 motifs are colored in orange, gray, green, pink, purple, and brown, respectively. Additionally, analytical size exclusion chromatography results are provided from a systematic analysis of interactions between the bMERB domain mutants and increasing concentrations of Rab8 (refer to Supplementary Fig. 4 and 5). (e-g, m-n) The bMERB domain (green), Rab8 (gray), and a mixture of both (blue) were loaded onto a Superdex 75 10/300 GL column to monitor for complex formation. (h-l, o-p) Binding affinities were measured by titrating the Rab8 (600 µM) to mutants bMERB domain (30 µM). Integrated heat peaks were fitted to a one-site-binding or two-site-binding model yielding the binding stoichiometry (N), the enthalpy (ΔH), the entropy (ΔS), and the dissociation constant (KD). The data are representative of at least three repetitions.

Crystal structure of the bMERBV978A_V985A:Rab10 complex.

(a) Cartoon depiction of the bMERBV978A_V985A:Rab10 complex. A single Rab10 molecule (gray) binds to the bMERBV978A_V985A domain (blue) at the high-affinity C-terminal Rab-binding site. Switch I and Switch II of Rab10 are shown in red and blue, respectively. GppNHp and Mg2+ are depicted as sticks and a green sphere, respectively. (b) Structural superposition of the bMERBV978A_V985A:Rab10 complex and bMERB:Rab10 complex (PDB: 5LPN)28. (c) Structural alignment of the bMERB domains of bMERBV978A_V985A:Rab10 (blue) with the bMERB:Rab10 (PDB: 5LPN)28 and the bMERB domain (PDB: 5LEO)26.

Phosphomimetic mutation of bMERB S960D does not affect Rab8 family member binding.

(a) The schematic diagram of human Mical1 shows an N-terminal monooxygenase (MO) domain, a calponin homology (CH) domain, a Lin11, Isl-1, and Mec-3 (LIM) domain, and a C-terminal coiled-coil bMERB (bivalent Mical/EHBP Rab binding) domain, along with PAK1 phosphorylation sites S817 and S960. (b) The surface electrostatic potential of bMERB was calculated in PyMOL using the APBS-PDB2PQR plugin and visualized with colors ranging from red (-5 kT/e) to blue (+5 kT/e). AlphaFold2 predicted the Mical1 amino acids 807–820, with S817 shown in stick representation. The side chain of S817 points towards a negatively charged patch on the Mical1 bMERB domain. (c) Cartoon representation of the Mical1 bMERB domain (PDB: 5LPN)28, with S960 shown in stick representation. Rab10 binding sites are indicated with dashed circles. (d, e, h-k) Binding of Rab8a (gray)/Rab10 (wheat)/Rab15 (red) with different Mical1 bMERB domain (green) was systematically tested onto a Superdex 75 10/300 GL column. (f, g, l-o) Binding affinities were measured by ITC experiments. Binding affinities were measured by titrating the Rab8/Rab10/Rab15 (600 µM) to wild type/mutant bMERB domain (30 µM). Integrated heat peaks were fitted to a two-site-binding model yielding the binding stoichiometry (N), the enthalpy (ΔH), the entropy (ΔS), and the dissociation constant (KD). Phosphomimetic mutation of S960 of the bMERB domain does not affect interaction with Rab8/Rab10/Rab15. The data are representative of at least three repetitions.

Impact of Rab binding site mutations in the bMERB domain on MO-CH-LIM interaction.

(a) Schematic illustration of the interactions between the MO-CH-LIM domain and the bMERB domain. Hydrophobic interactions are indicated by light orange dashed lines, and hydrogen bonds are indicated by gray dashed lines. (b-i) Effect of RBS mutants on MO-CH-LIM:bMERB domain interaction. The binding of MO-CH-LIM with different RBS bMERB mutants was systematically tested, and affinities were determined by ITC experiments. Integrated heat peaks, corrected for the offset by subtracting the control experiment, were fitted to a one-site-binding model, yielding the binding stoichiometry (N), enthalpy (ΔH), entropy (ΔS), and dissociation constant (KD). N.D. denotes not detected. The data are representative of at least three repetitions.

Structural basis of Mical1 activation upon Rab8 binding.

(a) Structural superposition of the Mical1 auto-inhibited AlphaFold2 model and the bMERB:Rab10 complex (PDB: 5LPN)28. (b) Structural overlay clearly shows that the binding of the first Rab molecule at high affinity stabilizes the helix 3 conformation, which structurally interferes with the CH domain and HelixCHL binding site. Additionally, the movement of helices 1-2 strictly hinders the CH binding site. (c) The superimposition of the bMERB domain of Mical1 in the auto-inhibited AlphaFold2 model with the bMERB domain of the bMERB:Rab10 complex (PDB: 5LPN)28 reveals the movement of bMERB helices upon Rab binding. (d) Structural overlay of the bMERB domain of Mical1 in the auto-inhibited state Alphafold2 model with the bMERB domain of the bMERB:Rab10 complex (PDB: 5LPN)28, with an inset highlighting the conformational changes of interface residues upon Rab10 binding. (e) The results of systematic analysis of interactions between full-length Mical1 or different Mical1 deletion constructs with F-actin via co-sedimentation experiments reveal that the MO domain is the minimal domain required for interaction with F-actin, whereas the full-length Mical1 exhibits perturbed interaction with F-actin. These experiments were independently repeated at least three times with consistent results.

The activation mechanism of Mical1 by Rab8 family members.

In the absence of active Rab8, Mical1 exists in an auto-inhibited state, where the MO, CH, HelixCHL, and LIM domains interact with its bMERB domain. Rab8 is recruited to the membrane by its GEF molecule (e.g., Rabin 8 or GRAB). The first molecule of active Rab8 binds to the high-affinity C-terminal binding site, leading to a change in the conformation of helix 3, which displaces the CH domain and HelixCHL. The first binding of the Rab molecule also changes the conformation of helix 2 and helix 3 at the low-affinity N-terminal Rab binding site, further interfering with CH binding, and this conformation leads to a change in the LIM and MO binding interface on the bMERB domain, resulting in the displacement of the LIM and MO domains. The second Rab8 molecule binds to stabilize the active state of the enzyme, allowing the MO domain to be available for F-actin interaction, leading to the depolymerization of F-actin.