A mutant LC3b allosteric switch for autophagy regulation

Abstract

The present invention relates to a mutant LC3B allosteric switch, wherein three specific residues have been mutated to either reduce or enhance receptor binding, compared to the wild- type protein. In particular, the present invention provides a method for generating conformational ensemble of lipidated protein on the membrane and introducing specific mutations within the protein activation switch that is indirectly altering protein binding pocket without disrupting its functional domains. The identification of an activation site opens avenues for the development of small molecules or peptides to enhance or regulate protein activity, providing a new dimension for therapeutic interventions. Therefore, by linking the structural insights of lipidated forms to functional outcomes, the invention advances knowledge on LC3 activity regulation, which could further be utilized to modulate the autophagy pathways.

Description

[0001] P_IN105408

[0002] A MUTANT LC3B ALLOSTERIC SWITCH FOR AUTOPHAGY REGULATION FIELD OF INVENTION

[0003] The present invention relates to a mutant LC3B allosteric switch useful for regulating autophagy receptor binding. In particular, the present invention relates to LC3B protein mutants comprising of polypeptide sequence selected from SEQ ID No. 2 and SEQ ID No. 3, wherein one or more amino-acid substitutions are introduced at an allosteric site distal from the canonical LC3 -interacting region (LIR) binding pockets, thereby modulating autophagy receptor binding by reducing or enhancing receptor binding affinity compared to the wild-type LC3B protein represented by SEQ ID No. 1.

[0004] BACKGROUND OF THE INVENTION

[0005] Autophagy is an evolutionary conserved eukaryotic pathway which is required to maintain cellular homeostasis. It is a tightly controlled process that regulates the degradation and recycling of diverse cellular components such as damaged or superfluous organelles, protein aggregates, or toxic microbes. In macroautophagy (hereafter refers as autophagy), bulk degradation of cellular components takes place by enclosing them in a double membrane vesicle called autophagosome. The pathway initiates by the formation of a cup-shaped phagophore membrane which grows in length and size to form a double membrane autophagosome vesicles. The cargo gets simultaneously recruited to the growing autophagosome membrane. Once the double membrane autophagosome vesicle with the cytosolic entity is formed, it is transported to lysosomes for degradation. Since this is a complex pathway, the autophagosome biogenesis has been divided into four major steps: initiation, elongation, closure / maturation, and fusion. Membrane biogenesis and cargo recruitment during autophagy is regulated by approximately 40 core autophagy-related (ATG) proteins that actively participate in this complex process, working in close coordination with an array of lipids. The dynamic membrane remodeling events crucial for autophagosome formation and cargo recruitment heavily depend on the interplay between the peripherally interacting ATG proteins with specific lipids. One such key autophagy-related peripheral membrane protein is the LC3B protein that is used as a biomarker to monitor autophagosome formation in cells. LC3B is a small protein of 17 kDa that peripherally attaches to the autophagosome membranes via covalent interaction (Kabeya, Y et al. “LC3, a mammalian homologue of yeast Apg8p, isP_IN105408

[0006] localized in autophagosome membranes after processing.” The EMBO journal vol. 19,21 (2000): 5720-8). LC3B is initially synthesized as a soluble, cytosolic protein with an extended C -terminal. Upon autophagy induction, the C-terminal is cleaved, generating the LC3-I form. This is followed by lipidation of LC3B, anchoring it to the autophagosomal membrane and resulting in the formation of the LC3-II state. (Mizushima, Noboru, and Tamotsu Yoshimori. “How to interpret LC3 immunoblotting.” Autophagy vol. 3,6 (2007): 542-5). The lipidation of LC3B proteins serves as a hallmark event and it is utilized for monitoring autophagy flux under both normal and stress conditions (Mizushima, Noboru. “Methods for monitoring autophagy.” The international journal of biochemistry & cell biology vol. 36,12 (2004): 2491-502). LC3B specifically attaches to phosphatidylethanolamine lipid via conserved glycine on the autophagosome membrane. LC3, yeast ortholog of ATG8, has six different copies in humans. These homologs are divided into two subfamilies: the LC3 family (LC3A, LC3B, LC3C) and the GAB ARAP family (GAB ARAP, GABARAPL1, GABARAPL2). LC3B (from hereafter referred to as LC3) is the most studied protein among other homologs and studies have revealed that it directly interacts with 200+ binding partners present in different cellular processes like endocytosis, transport, metabolism, immune pathways etc.

[0007] Dysfunctional LC3 has been shown to be associated with several cancer types and neurodegenerative diseases including Alzheimer's and Huntington’s diseases. Therefore, LC3 represents a promising therapeutic target for several disease conditions, making it essential to investigate its underlying mechanism of action. The crystal structure of human LC3B revealed a highly conserved ubiquitin fold, connected to two N-terminal a-helices, which exhibit variations among members of the LC3 protein family. The structure of LC3 contains four a-helices and four P-sheets connected via several loops. Within these secondary structure elements, two canonical functional binding pockets called the hydrophobic pocketl (HP1) and the hydrophobic pocket 2 (HP2) are present (Sugawara, Kenji et al. “The crystal structure of microtubule-associated protein light chain 3, a mammalian homologue of Saccharomyces cerevisiae Atg8.” Genes to cells: devoted to molecular & cellular mechanisms vol. 9,7 (2004): 611-8). These pockets provide a platform for the receptor proteins like p62, NBR1, OPTN to bind (Ichimura, Yoshinobu et al. “Structural basis for sorting mechanism of p62 in selective autophagy.” The Journal of biological chemistry vol. 283,33 (2008): 22847-57). The receptor proteins contain a consensus sequence: “W / F / Y-XX-L / I / V” motif also called LC3 interacting region (LIR) which binds to the hydrophobic pockets. LC3 also contains a membrane interacting region which consists of four structural motifs i.e. N-terminal, Loop3, Loop6, andP_IN105408

[0008] C-terminal (Zhang, Wenxin et al. “Autophagosome membrane expansion is mediated by the N-terminus and cis-membrane association of human ATG8s.” eLife vol. 12 e89185. 8 Jun.

[0009] 2023). Cellular studies have identified the binding partners of LC3 while biophysical experiments have elucidated its 3D architecture. However, the structural details of membranebound LC3 remain elusive, as it is challenging to study the protein with higher resolution in lipid environment.

[0010] The present invention addresses this challenge by identifying a previously unrecognized allosteric regulatory site within LC3B that is functionally coupled to receptor binding upon membrane association.

[0011] As used herein, the term “allosteric site” refers to a region of LC3B spatially separated from the hydrophobic LIR-binding pockets (HP1 and HP2), wherein conformational perturbation of said site results in altered receptor binding affinity.

[0012] The terms “open conformation” and “closed conformation” refer to relative Loop-6-Helix-3 distances as determined by molecular dynamics simulations and correlated with experimentally measured binding affinities.

[0013] The present invention is not limited by any particular theoretical model, and the mechanistic explanations provided herein are intended solely to aid understanding of the disclosed LC3 variants.

[0014] However, traditional approaches to modulate protein activity by deleting segments of proteins or modifying binding site residues have been reported. Our group has also previously reported important residues involved in spontaneous membrane binding of lipidated LC3 protein using coarse-grain simulations. The study revealed a patch of basic residues (R65, R68, R69, and R70) at the base of protein is required for correct positioning of LC3 on membrane (Thukral, Lipi et al. “The Molecular Mechanism Underlying Recruitment and Insertion of Lipid-Anchored LC3 Protein into Membranes.” Biophysical journal vol. 109,10 (2015): 2067-78). All-atom MD simulations of cytosolic structure human LC3 and its homolog identified the distinct characteristics of six proteins in apo and receptor bound states (Jatana, Nidhi et al. “Human LC3 and GAB ARAP subfamily members achieve functional specificity via specific structural modulations.” Autophagy vol. 16,2 (2020): 239-255). However, these characterizations were either based on the cytosolic form, or the atomistic details of theP_IN105408

[0015] membrane bound form were missing. In addition, the role of peptide receptor bound to LC3 in membrane environment has also not been investigated.

[0016] To understand the biological activity of any given peripheral membrane protein, it is important to understand how it interacts with the membrane and how it orients itself on the membrane. One such way to investigate specific interactions between lipids and proteins is through allatom molecular dynamics simulations. The detailed molecular models also provide quantitative understanding of lipid binding processes and conformational ensemble of peripheral membrane proteins. Thus, there is a need in this field to identify LC3 variants capable of modulating receptor binding upon membrane association, which need is fulfilled by the LC3 variants disclosed herein.

[0017] OBJECTIVES OF THE INVENTION

[0018] The main objective of the present invention is to provide a mutant LC3B allosteric switch useful for regulating autophagy receptor binding.

[0019] Yet another objective of the present invention is to provide LC3B mutants comprising of polypeptide sequence selected from SEQ ID No. 2-3, that exhibits altered receptor binding properties for autophagy regulation.

[0020] Still another objective of the present invention is to provide LC3B mutants comprising substitutions at positions corresponding to residues 64, 89 and 91 of human LC3B, wherein the said substitutions modulate autophagy receptor binding via an allosteric mechanism without directly altering the canonical LC3-LIR binding interface.

[0021] Yet another objective of the present invention is to provide LC3B mutants having defined amino-acid substitutions at the a3-Loop-6 interface, which alter receptor binding affinity relative to wild-type LC3B protein.

[0022] Still another objective of the present invention is to provide functional LC3 variants and equivalents thereof, including conservative amino acid substitutions and homologous LC3 family members, which retain allosteric modulation of receptor binding.

[0023] Yet another objective of the present invention is to provide a method to prepare LC3B mutants to modify its receptor binding function and regulate autophagy.P_IN105408

[0024] Still another objective of the invention is to provide a novel allosteric site in LC3B protein, which plays a crucial role in receptor binding properties for targeting autophagy.

[0025] SUMMARY OF INVENTION

[0026] Accordingly, the present invention provides a mutant LC3B allosteric switch useful for regulating autophagy receptor binding. In particular, the present invention provides LC3B protein mutants comprising of polypeptide sequence selected from SEQ ID No. 2 and SEQ ID No. 3, wherein one or more amino-acid substitutions are introduced at an allosteric site distal from the canonical LC3 -interacting region (LIR) binding pockets, thereby modulating autophagy receptor binding by reducing or enhancing receptor binding affinity compared to the wild-type LC3B protein represented by SEQ ID No. 1.

[0027] In a preferred embodiment, the present invention provides LC3B protein mutants comprising one or more amino-acid substitutions at residues 164, V89 and V91, wherein the said substitutions modulate autophagy receptor binding through an allosteric conformational change involving the relative orientation of Loop-6 and Helix-3, distinct from the canonical LC3-LIR interaction interface.

[0028] In another embodiment, the present invention provides human LC3B mutants in which combinatorial substitutions at the a3-Loop-6 junction alter receptor binding affinity by stabilizing either an increased-affinity (open) or decreased-affinity (closed) conformational state. The present invention is limited to LC3B mutants supported by experimental or computational evidence demonstrating altered receptor binding.

[0029] In still another embodiment, the present invention provides a mutant LC3B allosteric switch useful in autophagy regulation comprising a polypeptide sequence selected from the group consisting of SEQ ID No. 2-3.

[0030] In yet another embodiment, the present invention provides the mutant polypeptide having SEQ ID No. 2 comprises amino acid substitution at positions selected from the group consisting of I64D, V89P and V91D.

[0031] In still another embodiment, the mutant polypeptide having SEQ ID No. 3 comprises amino acid substitution at positions selected from the group consisting of I64K, V89D and V91F.P_IN105408

[0032] In yet another embodiment the corresponding nucleic acid sequence encoding the polypeptide is selected from the group consisting of SEQ ID No. 4-5.

[0033] In still another embodiment, the mutations are located at a3-Loop-6 allosteric interface distal from the canonical hydrophobic LIR-binding pockets (HP1 and HP2).

[0034] In yet another embodiment, the present invention provides an in vitro method for regulating autophagy receptor binding comprising the steps of:

[0035] a) providing cDNA of the mutant polypeptide having SEQ ID No. 2-3, b) cloning the cDNA obtained in step (a) in a suitable vector;

[0036] c) delivering the cloned vector obtained in step (b) into a host cell line to assess receptor binding for autophagy regulation.

[0037] In still another embodiment, the present invention provides the mutant polypeptide having SEQ ID No. 2 exhibits reduced autophagy receptor binding.

[0038] In yet another embodiment, the present invention provides the mutant polypeptide having SEQ ID No. 3 exhibits enhanced autophagy receptor binding.

[0039] In still another embodiment, the present invention provides the use of LC3B mutants for identifying modulators of LC3 -mediated autophagy.

[0040] In yet another embodiment, the invention provides an allosteric site in human LC3B protein with distinct conformation in cytosolic and membrane bound form.

[0041] In still another embodiment, the LC3B mutant having SEQ ID 2 shows the disruption of structural orientation at the switch region 164, V89, and V91 by mutating them to D64, P89, and D91 which leads to a significant reduction in receptor binding properties as compared to the wild type LC3B protein represented by SEQ ID No. 1.

[0042] In still another embodiment, the LC3B mutant having SEQ ID 3 contains amino acids at 164, V89, and V91 positions converted to K64, D89, and F91 respectively wherein these modifications allow the activation switch to form with tighter binding, playing a crucial role in increasing receptor-binding properties of the LC3B protein.P_IN105408

[0043] In Yet another embodiment, the results were further validated using biophysical ITC experiments wherein the mutations caused 2-fold increase in LC3B-LIR interaction in SEQ ID 3, whereas a decrease of 1.5-fold in the binding affinity of LC3B-LIR in SEQ ID 2.

[0044] In still another embodiment of the invention is to provide a novel allosteric site in LC3B protein that is located away from the receptor-binding pocket but plays a significant role in receptor binding properties. Thus, the invention provides a base for further exploration of autophagy-related process and development of therapeutic strategies targeting the LC3 protein.

[0045] In yet another embodiment, the present invention provides experimental data demonstrating the efficacy of previously reported LC3 mutants described in the literature as enhancing interactions with the LC3 -interacting regions (LIRs) of optineurin (OPTN) and ATG13. The data generated in accordance with the present invention indicates that none of the known mutants exhibit a measurable enhancement in LC3 functional activity. Accordingly, the present invention establishes that the previously described LC3 mutants exhibit binding specificity toward the LIRs of optineurin (OPTN) and ATG13, and do not demonstrate enhanced interaction with a general LIR motif, such as that of p62, which is known to bind multiple LC3 homologs.

[0046] BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The embodiments of the present disclosure will be described with reference to the following drawing wherein:

[0048] Figure 1: shows the structural dynamics of the activation loop. The line plot shows the time evolution of the distance between L6 and a3 (allosteric site, AS) in cytosolic, apo, and LIR-bound LC3B. The snapshot shows the superimposition of the representative structure, highlighting the stabilization of the L6-a3 conformation upon membrane or / and receptor binding.

[0049] Figure 2: shows the isothermal titration calorimetry (ITC) data for LC3B and LIR binding, (a) the plot shows the binding of LC3B WT and to the p62 peptide, (b) the plot shows the binding of LC3BAS-Mutant 1and to the p62 peptide (c) the plot shows the binding of LC3BAS-Mutant2and to the p62 peptide.

[0050] (d) The bar plot displays the thermodynamic data, AH, AG, and -TAS values for WT, mutant SEQ ID 3, and mutant SEQ ID 2.P_IN105408

[0051] Figure 3: illustrates comparative efficacy of reported LC3 / GABARAP binding-enhanced variants, 3(a) is the plot shows the ITC binding data of LC3B8AAmutant and 3(b) LC3BK49Amutant with p62.

[0052] DETAILED DESCRIPTION OF INVENTION

[0053] The present invention provides a mutant LC3B allosteric switch useful for regulating autophagy receptor binding. In particular, the present invention provides LC3B protein mutants comprising of polypeptide sequence selected from SEQ ID No. 2 and SEQ ID No. 3, wherein one or more amino-acid substitutions are introduced at an allosteric site distal from the canonical LC3 -interacting region (LIR) binding pockets, thereby modulating autophagy receptor binding by reducing or enhancing receptor binding affinity compared to the wild-type LC3B protein represented by SEQ ID No. 1.

[0054] In an aspect, the present invention provides LC3 protein mutants, wherein three specific residues have been mutated to either reduce or enhance receptor binding, compared to the wildtype LC3 protein.

[0055] In another aspect, the present invention provides a method to prepare human LC3B mutants, via a novel allosteric site, to modify its receptor binding function.

[0056] LC3B protein becomes functionally active in autophagy after lipidation and subsequently attaches to the phagophore membrane; however, current knowledge lacks the understanding of how functional regions are modified due to lipids. Hence, specific lipid-induced changes that activate LC3B on the membrane remain an open question. To identify the structural changes induced by lipids in the function associated binding regions of LC3B and to modulate its structure, four key steps were involved.

[0057] In yet another aspect, the present invention provides a method to synthesize human LC3B mutants to modify its receptor binding function comprising the steps of:

[0058] a) Generating structural states of LC3 in water and membrane environment and performing microsecond-timescale molecular dynamics simulations. The membrane states also incorporated a covalently attached phosphatidylethanolamine (PE) chain. The input LC3 structure was taken from aprevious study (Jatana, Nidhi et al. “Human LC3 and GAB ARAP subfamily members achieve functional specificity via specific structural modulations.” Autophagy vol. 16,2 (2020): 239-255) and a covalentP_IN105408

[0059] bond between the carbonyl carbon (-COOH) of C-terminal G120 and the (-NH2) amine group of the PE head was created. The manual parametrization of the lipidated LC3B was done based on charmm36 force field. To mimic the receptor-bound state, crystal structure coordinates of p62-LIR were extracted and docked with the lipidated LC3B. Both apo and receptor bound lipidated LC3 was then placed in a 400 lipids membrane patch comprising of autophagosomal membrane like composition (DOPC: DOPE: DOPS: POPI in 65:20:5: 10 ratio). The cytosolic setup was also prepared along with the membrane bound states. The three states, cytosolic, membrane-bound apo, membranebound with LIR LC3B were then subjected to all-atom molecular dynamic simulations. Three replicas of each state were simulated for 1 ps and a cumulative simulation time of 9 ps data was generated.

[0060] b) Further understanding the conformational heterogeneity of LC3 in three states.

[0061] While the overall ubiquitin fold was consistently maintained in all three states, the loops displayed dynamic behavior. In comparison to water-soluble cytosolic state which showed more flexibility, attaching to membrane significantly limited LC3 conformations. The average root mean square deviation (RMSD) of LC3B protein was 4 A to 3 A and 2.6 A in cytosolic, membrane-bound apo, and membrane-bound with LIR respectively. This was also confirmed by geometric clustering of states resulting in an average number of 45, 21, and 12 clusters in cytosolic, apo lipidated, receptor bound lipidated LC3B respectively. Overall, the stabilization of LC3B structure on membrane by lipids and further by receptor binding was observed. We monitored two receptor binding pockets called the Hydrophobic pocket 1 (HP1) and Hydrophobic pocket 2 (HP2). The volume of LC3B binding pocketwas calculated. The volume of the canonical receptor binding pockets (HP1 and HP2) gradually increased from cytosolic to membrane-bound apo, and membrane-bound with receptor with a net pocket volume of 35.56 A3, 64.40 A3, and 87.72A3respectively followed by monitoring the residues regulating the binding pocket volume. Then precise orientations of the residues were calculated. The oscillatory movement of residues within HP1 (DI 9, 123, K51) and HP2 (135, F52, 163, R70) exhibits a positive correlation with the pocket volume. The mean distance of HP1 residues with pocket center in cytosolic LC3 was determined to be 4.72 A, which increased to 5.40 A in membrane-bound apo-LC3II and 6.15 A in LIR-LC3BII. Similarly, the average mean distance of HP2 residues, increased from 5.37 A to 5.54A and 6.07A in cytosolic LC3, membrane-bound apo-LC3, and LIR-LC3B, was observed respectively. This highlights the dynamic characteristics ofP_IN105408

[0062] LC3 binding pockets that were previously unobserved due to the absence of lipid interactions in the experimentally determined structures.

[0063] c) In addition to the local effects at the functional site (i.e., membrane targeting motif (MTM) and binding pocket), the membrane-binding or p62-binding resulted in numerous distal changes indicative of sites within the protein that also gets activated upon membrane binding. To investigate long-range protein dynamics, intraprotein contacts (within 5A) and %1 dihedral angle data were utilized to compute the net communication in different states of LC3B. To understand communication between functional regions and the rest of the protein, the dihedral data included the torsion angles, their rotamers, and ordemess were calculated. Further the communication strength of each residue in a molecular state using the resulting net-communication matrixwas calculated. Over 30% of residues exhibit a significant shift in sidechain rotamers (shift>0.3) upon membrane binding. It was observed that 60% of these residues do not belong to the MTM and binding pocket regions. The significant number of residues forming a distinct hub spanning a3, loop 5 (L5), P3, and loop 6 (L6) region, showed a higher communication strength in both apo and LIR bound states.

[0064] d) It was hypothesized that membrane interaction induced structural changes in the receptor binding pockets. To understand this, the path of the information flow between membrane targeting residues and receptor binding residues was computed. The information flow quantifies the betweenness centrality (BC) for each selected region. Any other region or residue exhibiting high BC is essential for interconnecting various regions of the protein and its removal would likely disrupt the communications between the different function regions, in this case MTM and BP. A strong intercoupling was observed between MTM and BP with communication traversing through residues near HP2 in the receptor-bound state of LC3B.BC (betweenness centrality) of these residues decreases as the distance from the HP2 pocket center decreases yet, significant long- range interactions persisted. Particularly, L6 residues M88, V89, and V91, situated roughly around 13.9 A from the HP2 core, demonstrated considerable BC, suggesting their involvement in facilitating communication beyond the local pocket area.

[0065] e) Upon monitoring the L6 conformation in the three-state system, it was revealed that L6 exhibits significant stability, with membrane or receptor binding enhancing its conformational stability. Substantial displacement of L6 from a3 was noted, with a distance extending up to 10 A in the cytosolic LC3B. This distance decreased to 3.9 A in both apo and receptor-bound membrane-anchored LC3B (Fig 1). This data suggestsP_IN105408

[0066] a new allosteric site located at the a3-L6 junction that undergoes a conformational change following lipid binding that may influence the binding pockets through long- range interactions.

[0067] f) Further modulating the allosteric site of LC3 at the interface of loop6 and a3. The first criterion for selecting the mutants at this site was to ensure that the residues were not located in the functional sites, such as the receptor binding and membrane interacting regions, to preserve the protein's function. The second criterion was to choose nonconserved residues to prevent any structural deformation. The third criterion was to select the residues showing high betweenness centrality.

[0068] g) The above criteria narrowed the search down to three residues: 164, V89, and V91. The conformational state of these three residues was studied carefully in all three states and probability density maps of distance between them confirmed that they exist are distinct in open and close conformations. The mutants were created such that they can mimic the movement of L6 close and far from a3 as observed from the simulation data. Moreover, according to the intraprotein network analysis, residues 164 and V89 were uniquely interacting with each other and the S90 and V91 were closer to M60 and S61 in the membrane-bound LC3B. The mutants were created such that these residues either moved closer or moved away from each other

[0069] h) In the allosteric site LC3B mutant SEQ ID 3, where the changes were introduced to enhance the local interaction network and decrease the distance of L6-a3, these residues were mutated to K64-D89-F91, by introducing a salt bridge at this location. On the other hand, in allosteric site LC3B mutant SEQ ID 2 where the changes were introduced to disrupt the hydrophobic interactions between these residues, and they were modified to D64-P89-D91.

[0070] i) The sequences as described in SEQ ID 2 and SEQ ID 3, contain mutations at three locations, I64D-V89P-V91D and I64K-V89D-V91F respectively.

[0071] j) To validate the findings, silico simulations of two mutants were performed. The mutated lipidated LC3B structures were subjected to all-atom MD simulations within a membrane environment, each simulated for 1 ps. The distances between L6 and a3 decreased to 2.7A in allosteric site LC3B mutant SEQ ID 3 whereas this distance increased to 7.5 A in allosteric site LC3B mutant SEQ ID 2 trajectories mimicking the closed and semi-open / open conformation of allosteric site in membrane anchored human LC3B.P_IN105408

[0072] k) Further validation of in-silico findings using biophysical ITC (isothermal titration calorimetry) experiments validated the receptor binding properties of the WT and mutated LC3B using the p62 LIR peptide. The ITC experiments were performed at 25 °C, and data are shown in Fig 2. The raw ITC data in the upper panel shows the heat release over time (p / mol) from the successive injection of LIR peptide into the protein solution. The bottom panel shows the integrated heat data (kcal / mol), plotted against the molar ratio of ligand to protein, showing a sigmoidal curve fitting the binding interaction model. The average parameters from three independent titration experiments revealed a dissociation constant (Kd) of 5.42 ± 0.19 pM,2.92 ± 0.3 pM, and7.72 ± 0.5 pMfor WT, allosteric site LC3B mutant SEQ ID 3, and allosteric site LC3B mutant SEQ ID 2, respectively (Fig 2a-c). The Kd data suggests a 2-fold increase in ligand binding in allosteric site LC3B mutant SEQ ID 3, whereas the LC3B-LIR binding affinity decreased by approximately 1.5-fold in allosteric site LC3B mutant SEQ ID 2 compared to WT. The stoichiometry (n) of LC3B-LIR binding remained approximately 1:1 in WT as well as in the mutants. The heat changes in WT and allosteric site LC3B mutant SEQ ID 3(Fig 2a and 2c) show smoother and more gradual transitions compared to allosteric site LC3B mutant SEQ ID 2(Fig 2b), where peaks are sharper, suggesting differences in binding kinetics. Interestingly, the positive entropy changes in mutants compared to the WT suggest a conformational shift in the protein upon LIR binding (Fig 2d). The comparison of the open conformation of allosteric site LC3B mutant SEQ ID 3 and the close conformation of allosteric site LC3B mutant SEQ ID 2 with the wild type suggests that the conformational change in the protein is influencing its binding to the p62 peptide positively and negatively, respectively. Together, this data demonstrated that allosteric site LC3B mutant SEQ ID 3has the highest affinity (Kd = 2.92 pM) with near 1:1 binding, suggesting an optimized interaction, making this ligand-protein pair the most favorable. Whereas allosteric site LC3B mutant SEQ ID 2appears weaker (Kd = 7.72 pM) with 1:1 binding but lower affinity, making it less favorable for binding.

[0073] l) Previously reported LC3 and GABARAP mutants have focused on enhancing interactions with selective autophagy receptors through targeted amino acid substitutions within or proximal to canonical LIR-binding interfaces. For example, mutations such as F7L-R11L-T12N-15QL-E18D-D19N-K49I-K51G in optineurin were shown to increase LC3-OPTN LIR binding by approximately 2000-foldP_IN105408

[0074] (Putyrski, M. et al. “Disrupting the LC3 Interaction Region (LIR) Binding of Selective Autophagy Receptors Sensitizes AML Cell Lines to Cytarabine.” Frontiers in Cell and Developmental Biology vol. 8 (2020): 208.), while the L50T-D54N-I55V mutation in GABARAP enhanced GABARAP-p62 interaction by ~80-fold. Similarly, the LC3A K49A mutation was reported to strengthen LC3A-ATG13 LIR binding in vitro (Suzuki, H. et al. “Structural basis of the autophagy-related LC3 / Atgl3 LIR complex: recognition and interaction mechanism.” Structure vol. 22, no. 1 (2014): 47-58.). These approaches primarily rely on direct reinforcement of LIR-LC3 interface contacts. In contrast, the present proposed novel LC3B open mutant exhibits a fundamentally distinct interaction profile. Isothermal titration calorimetry (ITC) measurements demonstrate that engineered mutations positioned away from the canonical binding pocket retain measurable interaction with p62. We attribute this to allosteric mechanism and hence change in protein specifically either to“ enhance” or disrupt the activation profile. Comparative data, however, shows that none of the previous known mutants are able to enhance the LC3 binding with p62 receptor. Specifically, ITC analysis shows that LC3B8AAmutant produces no detectable heat release upon titration, consistent with complete loss of binding, while LC3BK49Amutant yields weak, nonsaturating heat signals indicative of a substantially weakened interaction. These results confirm that the functional behavior of the LC3 open mutant is not attributable to simple enhancement of LIR affinity but instead reflects a distinct modulation of LC3 interaction properties, differentiating it mechanistically from previously reported binding-enhanced mutants.

[0075] Table 1: List of Sequences involved:

[0076]

[0077] P_IN105408

[0078]

[0079] P_IN105408

[0080]

[0081] EXAMPLES

[0082] The following examples are given by way of illustration of the present invention and therefore should not be construed to limit the scope of the present invention

[0083] EXAMPLE 1: Generation and simulation details of different states of human LC3B protein

[0084] The full-length structure of human LC3B (3vtu.pdb) was obtained from the previous work (Jatana, Nidhi et al. “Human LC3 and GAB ARAP subfamily members achieve functional specificity via specific structural modulations.” Autophagy vol. 16,2 (2020): 239-255) and the residues beyond Gly 120 were removed. A covalent link was established between the C-terminal Gly (-COO) and the headgroup (-NH2) of phosphatidylethanolamine (PE), utilizing the preexisting parameters of glycine and l-palmitoyl-2-oleoyl-sn-phosphoethanolamine (POPE) in the CHARMM36 force field. After the lipidated state was obtained, it was subjected to a short MD simulation for minimization of 10 ns. The lipidated structure of LC3B was then manually placed in a 400-lipid membrane. The membrane contained ER-like lipid composition of DOPC, DOPE, DOPS, and POPI in 65, 20, 5, and 10 ratios respectively (Fracchiolla, Dorotea et al. “A PI3K-WIPI2 positive feedback loop allosterically activates LC3 lipidation in autophagy.” The Journal of cell biology vol. 219,7 (2020): e201912098). The membrane was prepared using charmm GUI webserver and was simulated before placing it with the lipidated LC3B.

[0085] To obtain the receptor bound membrane anchored state of LC3B, crystal structure coordinates of p62 LIR bound to LC3B were taken (2zjd.pdb) (Ichimura, Yoshinobu et al. “Structural basis for sorting mechanism of p62 in selective autophagy.” The Journal of biological chemistry vol.

[0086] 283,33 (2008): 22847-57). The 2zjd.pdb was superimposed with lipidated LC3B and the coordinated of p62 LIR was saved with the lipidated state. This receptor bound lipidated LC3BP_IN105408

[0087] was then placed with the ER-like membrane as explained above. As a control, the full-length cytosolic LC3B was also simulated along with the above-mentioned states. The in-sillicoallosteric site LC3B mutant SEQ ID 3 and allosteric site LC3B mutant SEQ ID 2 proteins were prepared by introducing these mutations in WT protein structure using Chimera software. The mutant proteins were then lipidated and inserted in the ER membrane as explained above and subjected to MD simulations.

[0088] The cytosolic, apo, and LIR bound LC3B with / without ER membrane were positioned in a rectangular box sufficiently large to accommodate both the protein and the membrane. Water molecules were incorporated using the TIP3P model, and Na+ ions were introduced to achieve system neutrality. MD simulation was conducted with GROMACS version 2018.3and 2021.4employing the CHARMM36 all-atom force field for the membrane anchored states whereas. Periodic boundary conditions were implemented, and a real space cut-off distance of 1.2 nm was established. The Particle Mesh Ewald (PME) summation employed a grid spacing of 0.16 nm, coupled with fourth-order cubic interpolation to calculate the forces and potential between grid points. The van der Waals cut-off was at 1.2 nm. Energy minimization of the systems was conducted utilizing the steepest descent approach, with temperature and pressure regulated at 310 K and 1 bar through the Nose-Hoover thermostat and Parrinello-Rahman barostat, respectively. A time step of 2 femtoseconds was employed for the numerical integration of the equations of motion. The coordinates were recorded every 20 picoseconds. Three replicates of each cytosolic, apo, and LIR-bound LC3B system were simulated for 1 ps whereas the allosteric site LC3B mutant SEQ ID 3 and allosteric site LC3B mutant SEQ ID 2 were simulated for 1 ps each.

[0089] EXAMPLE 2: Analysis of the trajectories

[0090] The periodic boundary conditions were removed prior to analyzing the trajectories. The protein dynamics was computed using the GROMACS module gmx rms, which utilizes root mean square deviation to compare the structures. A reference structure was used to evaluate each structure within a trajectory. All the distances were calculated using the gmx distance module of gromacs where the center of mass of the selected region is taken as the point for distance calculations. The gromos algorithm of gmx cluster module was utilized to identify the key structural conformation of LC3B in different states. It counts the number of neighbors using the cut-off of 0.2 nm and generates the cluster of a structure having the largest number of neighbors. The gmx chi module was utilized to calculate the time evolution , , and %1 ofP_IN105408

[0091] each residue in the three states. All the analysis was performed on DT1000 resolution. The molecular snapshots were generated with UCSF ChimeraX and VMD. The graphs and plots were produced with the Matplotlib library in Python.

[0092] Pocket Volume calculations

[0093] The Epock tool was employed to obtain the pocket center coordinates for volumetric calculations (Laurent, Benoist et al. “Epock: rapid analysis of protein pocket dynamics.” Bioinformatics (Oxford, England) vol. 31,9 (2015): 1478-80). The binding site volume in each trajectory was computed over time using POVME 3.0 (Wagner, Jeffrey R et al. “POVME 3.0: Software for Mapping Binding Pocket Flexibility.” Journal of chemical theory and computation vol. 13,9 (2017): 4584-4592). The probe size was taken as 6 A whereas the grip spacing was kept as low as 0.5 A for improving the accuracy of the volume calculation.

[0094] EXAMPLE 3: Defining the rotameric states and orderness for each residue

[0095] The rotameric states were defined as explained in Singh et al., 2017 (Singh, Sukrit, and Gregory R Bowman. “Quantifying Allosteric Communication via Both Concerted Structural Changes and Conformational Disorder with CARDS.” Journal of chemical theory and computation vol.

[0096] 13,4 (2017): 1509-1517). The orderness was obtained from the time evolution of rotameric states of the relevant dihedral according to the k-neighborhood scheme. Consequently, the three parameters raw dihedral angles, rotameric states, and orderness states for each residuewere analyzed to quantify residue-residue correlations.

[0097] Calculating the residue-residue correlations

[0098] The residue-residue correlation was defined as the sum of mutual information calculated form the backbone (BC) and sidechain (SC) dihedrals between Ri and Rj.

[0099] Net MI= MIBB dihedral + MIsC dihedral

[0100] The MIBB / SC dihedral is the sum of three parameters.

[0101] MI (BB / SC dihedral)=Ml Dihedral M 1 RotamericStates Ml Orderness

[0102] The resultant net MI matrices were normalized and subsequently utilized for calculating net communication. The net communication between residues Ri and Rj is the aggregate of mutual information between the neighbors of Ri and Rj. The neighborhood is defined as 3 A surrounding the residue selection. The net communication matrix was subsequently normalized using the max-scaling method. The communication strength of each residue was calculatedP_IN105408

[0103] from the normalized net communication matrix. The communication strength of residue Ri is defined as the aggregate of net communication between residue Ri and all other residues.

[0104]

[0105] Information flow calculations and analysis

[0106] To enhance the understanding of the key essential residues for facilitating communication between functional regions, the information flow analysis introduced by Kang, Westerlund et al., 2020was performed. The analysis categorizes a collection of residues as belonging to source and sink space. The examination of information flow is grounded in the current flow methodology within graph theory. Then computed two factors to evaluate the significance of residues in distal communication: a) information flow betweenness and b) efficiency, also referred to as informational flow closeness centrality.

[0107] Initially, an adjacency matrix was created by multiplying the net communication with a contact matrix (Cij). A residue pair was assigned as a contact if the contact probability is greater than 0.3 such that both transient and stable contacts were included.

[0108]

[0109] The Laplacian of the adjacency matrix is calculated by subtracting the diagonal degree matrix from the adjacency matrix, expressed as L = D - A. The degree of residue i is calculated by accumulating its adjacency with the other residues.

[0110]

[0111] Subsequently, sink nodes were eliminated from the Laplacian matrix, and the inverse was computed. Zeros were reintroduced to the rows and columns that had previously been eliminated, leading to a diminished inverse Laplacian L-1. The supply vector transmits the unit current required for application to source residues and currents departing from the sink nodes (bsink= l / n(sink residues)). Potentials are subsequently computed as follows:P_IN105408

[0112] p source) = L1b

[0113] The information flow betweenness centrality of residue i, for the flow originating from the source residue group to the absorbing sink residue group, is computed as follows:

[0114]

[0115] f_i (s) is designated as zero if i belongs to the source or sink group. Otherwise, it is computed as follows:

[0116] / ■(s) = 0.5

[0117]

[0118] Information flow closeness centrality, sometimes referred to as efficiency, is the reciprocal of the potential difference calculated between source and sink pairs.

[0119] 1

[0120]

[0121] Group IPsC5) Pt(sI

[0122] EXAMPLE 4: Protein purification

[0123] Constructs for mouse wild-type (WT) LC3B, allosteric site LC3B mutant SEQ ID 3 (I64K-V89D-V91F) and allosteric site LC3B mutant SEQ ID 2 (I64D-V89P-V91D) were cloned from an eGFP pcDNA3.1 plasmid into the pet28a vector (Pet28a Novagen 69864). The constructs were confirmed via Sanger sequencing. Overexpression was performed in Rosetta cells (Rosetta de3 competent cells Novagen 70954) and Induction was carried out at an OD600 of 0.5-0.7 by adding 0.2 mM IPTG. The cells were kept at 20°C for 22-24 hours in a shaker incubator. The cell pellet was resuspended in lysis buffer (20 mM Na2HPO4, pH 7.5, 1.2 M NaCl, and 10 mM imidazole). Cells were lysed using a sonicator, and the lysate was applied to a His Trap Chelating HP column charged with NiSO4 to purify (His)6-tagged LC3B WT, allosteric site LC3B mutant SEQ ID 3, and allosteric site LC3B mutant SEQ ID 2. The column was washed with the wash buffer (20 mM imidazole) and proteins were eluted using elution buffer (250 mM imidazole). Further purification was performed via gel filtration chromatography (FPLC) using a Superdex 75pg column (Cytiva) into the binding buffer (20 mM HEPES, pH 7.5, 200 mMNaCl, and 0.5 mM Tris(2-carboxy ethyl) phosphine (TCEP). TheP_IN105408

[0124] purified proteins were kept at 4°C until further use. The peptide p62 was synthesized by Genescript (IGIB16122024GAP02638) and resuspended in a binding buffer and subsequently used for further experiments.

[0125] EXAMPLE 5: Isothermal Titration Calorimetry (ITC)

[0126] Calorimetric titrations were performed at 25°C using a Microcal PEAQ-ITC (MicroCai, LLC). The proteins were exchanged in a binding buffer through gel filtration. The titrations involved 19 injections of 2 pL each, where p62 peptide at concentrations of 480-500 pM was titrated into 300pL of 20 pM LC3B WT and mutant constructs. Each experiment was conducted in triplicate. The integrated heat data from the titrations were analyzed using the MicroCai PEAQ-ITC analysis software, employing the fitted offset option to automatically subtract the control heat. The resulting isotherm was fitted using a one-site binding model. We further evaluated LC3 mutants previously reported in the literature to enhance LC3-LIR interactions. Specifically, in these studies the LC3B8AAmutant comprising substitutions F7L, R1 IL, T12N, Q15L, E18D, D19N, K49I, and K51G, as well as the LC3B K49A mutant, were assessed for their binding interactions with the LIR of optineurin (OPTN) and ATG13, respectively, and were observed to exhibit enhanced binding efficiency toward their reported targets. These mutants are reported in literature to enhance the binding of LC3 with LIR peptides other than p62. The calorimetric titration for p62 titrated into LC3B8AA mutant and LC3BK49A mutant resulted in complete loss of binding and consistent weak interactions respectively. Isothermal titration calorimetry (ITC) experiments were performed to comparatively evaluate the binding efficacy of the mutants disclosed herein relative to the previously known LC3 mutants. Notably, the LC3B8AAmutant demonstrated an absence of detectable binding to p62, while the LC3BK49Amutant exhibited weak interaction with p62. (Fig. 3)

[0127] ADVANTAGES OF THE INVENTION

[0128] 1. The invention provides novel LC3B mutants useful in regulating autophagy.

[0129] 2. The mutants will play a significant role in developing therapies for diseases such as cancer, skin or aging disorders.P_IN105408

[0130] 3. The invention throws insights into understanding the lipidated LC3B protein in the membrane environment which provides molecular details of the protein-lipid interplay during autophagosome formation.

[0131] 4. Simulating the physiologically relevant model of all-atom lipidated structure captures the dynamic interactions between the protein and lipid environment, which is not explored in the cytosolic WT structure due to the current limitations of experimental techniques.

[0132] 5. The invention identifies an activation site that can regulate the protein's activity, hence provides a potential target for modulating protein function. This is a significant advancement over WT studies that may focus only on canonical binding or active sites.

[0133] 6. The identification of an activation site opens avenues for the development of small molecules or peptides to enhance or regulate protein activity, providing a new dimension for therapeutic interventions. Therefore, by linking the structural insights of lipidated forms to functional outcomes, the invention advances knowledge on LC3 activity regulation, which could further be utilized to modulate the autophagy pathways.

Claims

P_IN105408We Claim1. A mutant LC3B allosteric switch useful in autophagy regulation comprising a mutant polypeptide wherein the sequence is selected from the group consisting of SEQ ID No. 2 and SEQ ID No.3.

2. The mutant polypeptide as claimed in claim 1 , wherein the mutant polypeptide having SEQ ID No. 2 comprises an amino acid substitution at positions selected from the group consisting of I64D, V89P and V91D.

3. The mutant polypeptide as claimed in claim 1 , wherein the mutant polypeptide having SEQ ID No. 3 comprises an amino acid substitution at positions selected from the group consisting of I64K, V89D and V91F.

4. The mutant polypeptide as claimed in claim 1, wherein the nucleic acid sequence encoding the corresponding polypeptide is selected from the group consisting of SEQ ID No. 4 and SEQ ID No. 5.

5. The mutant polypeptide as claimed in claim 1, wherein the mutations are located at a3-Loop-6 allosteric interface distal from the canonical hydrophobic LIR-binding pockets (HP1 and HP2).

6. The mutant polypeptide as claimed in claim 1, for use in autophagy regulation.

7. An in vitro method for regulating autophagy receptor binding comprising the steps of:a) providing cDNA of the mutant polypeptide having SEQ ID No. 2-3, b) cloning the cDNA obtained in step (a) in a suitable vector;c) delivering the cloned vector obtained in step (b) into a host cell line to assess receptor binding for autophagy regulation.P INl 054088. The method as claimed in claim 7, wherein the mutant polypeptide having SEQ ID No. 2 exhibits reduced autophagy receptor binding.

9. The method as claimed in claim 7, wherein the mutant polypeptide having SEQ ID No. 3 exhibits enhanced autophagy receptor binding.

Images