Targeting plant UBX proteins: AI-enhanced lessons from distant cousins
Junrui Zhang, Alexandra I. Vancea, Stefan T. Arold
Abstract
Ubiquitin regulatory X (UBX) domain-containing adaptor proteins guide and control the AAA-ATPase cell division control protein 48 (CDC48) to mediate an astonishing variety of fast-response mechanisms.Recent structural and functional studies have clarified the biological roles of many UBX proteins from animals and fungi, but most plant UBX domain-containing (PUX) proteins remain poorly understood.Despite their evolutionary distance, many domains and their combinations are conserved across eukaryotic UBX proteins, suggesting that they form functionally coherent 'work packages'.The artificial intelligence (AI)-based program AlphaFold is now able to accurately predict protein 3D structures and connections between structural units.The combination of a cross-kingdom comparison and AlphaFold analysis reveals conserved motifs and higher-order features for PUX proteins that collectively provide testable hypotheses to efficiently guide future experimental research. Across all eukaryotic kingdoms, ubiquitin regulatory X (UBX) domain-containing adaptor proteins control the segregase cell division control protein 48 (CDC48), and thereby also control cellular proteostasis and adaptation. The structures and biological roles of UBX proteins in animals and fungi have garnered considerable attention. However, their counterparts in plants remain markedly understudied. Since 2021, the artificial intelligence (AI)-based algorithm AlphaFold has provided predictions of protein structural features that can be highly accurate. Predictions of the proteomes of all major model organisms are now freely accessible to the entire research community through user-friendly web interfaces. We propose that the combination of cross-kingdom comparison with AF analysis produces a wealth of testable hypotheses to inspire and guide experimental research on plant UBX domain-containing (PUX) proteins. Across all eukaryotic kingdoms, ubiquitin regulatory X (UBX) domain-containing adaptor proteins control the segregase cell division control protein 48 (CDC48), and thereby also control cellular proteostasis and adaptation. The structures and biological roles of UBX proteins in animals and fungi have garnered considerable attention. However, their counterparts in plants remain markedly understudied. Since 2021, the artificial intelligence (AI)-based algorithm AlphaFold has provided predictions of protein structural features that can be highly accurate. Predictions of the proteomes of all major model organisms are now freely accessible to the entire research community through user-friendly web interfaces. We propose that the combination of cross-kingdom comparison with AF analysis produces a wealth of testable hypotheses to inspire and guide experimental research on plant UBX domain-containing (PUX) proteins. By extracting and unfolding proteins from various locations in the cell, CDC48 assures a balanced proteome (see Glossary) and serves as a fast-response mechanism to changing conditions [1.Stolz A. et al.Cdc48: a power machine in protein degradation.Trends Biochem. Sci. 2011; 36: 515-523Abstract Full Text Full Text PDF PubMed Scopus (193) Google Scholar]. CDC48 is essential for all eukaryotes, and CDC48A is the major isoform in Arabidopsis thaliana (Arabidopsis). CDC48 is also called p97 or valosin-containing protein (VCP) in mammals; we use CDC48/p97 to refer to all orthologs. CDC48/p97 is a hexameric AAA-ATPase. Its sequence and function as a segregase and unfoldase are highly conserved across the plant, fungal, and animal kingdoms. The combination of CDC48/p97 with specific adaptor proteins allows it to participate in diverse biological functions, such as promoting cancer cell growth in humans or reshaping lipid droplets in tobacco pollen tubes [2.Haines D.S. p97-containing complexes in proliferation control and cancer: emerging culprits or guilt by association?.Genes Cancer. 2010; 1: 753-763Crossref PubMed Scopus (36) Google Scholar,3.Kretzschmar F.K. et al.PUX10 is a lipid droplet-localized scaffold protein that interacts with CELL DIVISION CYCLE48 and is involved in the degradation of lipid droplet proteins.Plant Cell. 2018; 30: 2137-2160Crossref PubMed Scopus (62) Google Scholar]. In addition, post-translational modifications can also contribute to the regulation of CDC48/p97 [4.Hanzelmann P. Schindelin H. The interplay of cofactor interactions and post-translational modifications in the regulation of the AAA+ ATPase p97.Front. Mol. Biosci. 2017; 4: 21Crossref PubMed Scopus (97) Google Scholar]. The UBX domain-containing proteins constitute the largest and most heterogeneous group of adaptor proteins for CDC48/p97 [5.Schuberth C. Buchberger A. UBX domain proteins: major regulators of the AAA ATPase Cdc48/p97.Cell. Mol. Life Sci. 2008; 65: 2360-2371Crossref PubMed Scopus (215) Google Scholar]. UBX proteins have a modular architecture where the combination of building blocks (i.e., domains and motifs) can produce various functions. Remarkably, despite almost a billion years of distinct evolution, UBX proteins across all eukaryotic kingdoms still use the same building blocks and combinations thereof (Box 1 and Figure 1). These building blocks are connected through extended and poorly conserved regions that have a high probability of being intrinsically disordered. Unfortunately, the resulting dynamic 'beads-on-a-string' architecture makes it challenging to experimentally determine how the 3D structural features of full-length UBX proteins synergize to control the catalytic activity and ligand engagement of CDC49/p97. However, the AI-driven structure prediction algorithm AlphaFold may now help to close this knowledge gap.Box 1Background information on CDC48/p97 and UBX domain adaptorsThe multifunctionality of CDC48/p97 arises from its interactions with different adaptor proteins. Although most adaptors interact with the N-terminal domain of CDC48/p97, some bind to its C-terminal tail (Figure I). In addition to the domains and motifs that interact with CDC48/p97, other domains or motifs of UBX domain proteins may interact with substrates, anchor to a target site, or sense environmental changes (Table I).Table ICommon additional domains and motifs of UBX proteinsNameaAbbreviations: aa, amino acids; CC, coiled-coil; HL, helical lariat; HP, hydrophobic patch; PUB, peptide N-glycanase/UBA or UBX-containing proteins; SEP: shp1, eyc, and p47; SHP box, linear motif derived from shp1p; shp1, eyes closed and p47; UAS, ubiquitin-associating; UBA, ubiquitin-associated; UBL, ubiquitin-like; UBX, ubiquitin regulatory X; UIM, ubiquitin-interacting motif; VIM, VCP-interacting motif; ZnF, zinc finger.Length (aa)FeaturesbIn the consensus sequences, x indicates any residue, h indicates a hydrophobic residue, and a represents an acidic residue.Possible functionsRefsUBA40Three-helix bundleUbiquitin binding[32.Hofmann K. Bucher P. The UBA domain: a sequence motif present in multiple enzyme classes of the ubiquitination pathway.Trends Biochem. Sci. 1996; 21: 172-173Abstract Full Text PDF PubMed Scopus (356) Google Scholar]SEP65Conserved loop residuesUbiquitin-independent binding[33.Soukenik M. et al.The SEP domain of p47 acts as a reversible competitive inhibitor of cathepsin L.FEBS Lett. 2004; 576: 358-362Crossref PubMed Scopus (17) Google Scholar]UAS110Positive surface patchMay interact with fatty acids[28.Kim H. et al.UAS domain of Ubxd8 and FAF1 polymerizes upon interaction with long-chain unsaturated fatty acids.J. Lipid Res. 2013; 54: 2144-2152Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar]UBL70Ubiquitin foldUbiquitin-like modifier[34.Bogan J.S. et al.Endoproteolytic cleavage of TUG protein regulates GLUT4 glucose transporter translocation.J. Biol. Chem. 2012; 287: 23932-23947Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar]HP40Hydrophobic residuesMembrane anchor[25.Deruyffelaere C. et al.PUX10 is a CDC48A adaptor protein that regulates the extraction of ubiquitinated oleosins from seed lipid droplets in Arabidopsis.Plant Cell. 2018; 30: 2116-2136Crossref PubMed Scopus (51) Google Scholar]UIM20aaxxhxxAhxxS consensusBinds to the autophagy adaptor Atg8[35.Hofmann K. Falquet L. A ubiquitin-interacting motif conserved in components of the proteasomal and lysosomal protein degradation systems.Trends Biochem. Sci. 2001; 26: 347-350Abstract Full Text Full Text PDF PubMed Scopus (383) Google Scholar]ZnF30C2HC typeUbiquitin binding[16.Zhang J. et al.Versatile control of the CDC48 segregase by the plant UBX-containing (PUX) proteins.Comput. Struct. Biotechnol. J. 2021; 19: 3125-3132Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar]a Abbreviations: aa, amino acids; CC, coiled-coil; HL, helical lariat; HP, hydrophobic patch; PUB, peptide N-glycanase/UBA or UBX-containing proteins; SEP: shp1, eyc, and p47; SHP box, linear motif derived from shp1p; shp1, eyes closed and p47; UAS, ubiquitin-associating; UBA, ubiquitin-associated; UBL, ubiquitin-like; UBX, ubiquitin regulatory X; UIM, ubiquitin-interacting motif; VIM, VCP-interacting motif; ZnF, zinc finger.b In the consensus sequences, x indicates any residue, h indicates a hydrophobic residue, and a represents an acidic residue. Open table in a new tab Figure 1Domain architecture and proposed classification of ubiquitin regulatory X (UBX) domain-containing proteins.Show full captionFolded protein domains are shown boxed with domain names. The first two letters of the protein name indicate the species: At, Arabidopsis thaliana; Ce, Caenorhabditis elegans; Hs, protein are as p47 and aa, amino The to the of The full and of domains and motifs are in domains or motifs but experimentally are by The new resulting from this are in the and in the (see for full The UBX domain of has a loop Figure The multifunctionality of CDC48/p97 arises from its interactions with different adaptor proteins. Although most adaptors interact with the N-terminal domain of CDC48/p97, some bind to its C-terminal tail (Figure I). In addition to the domains and motifs that interact with CDC48/p97, other domains or motifs of UBX domain proteins may interact with substrates, anchor to a target site, or sense environmental changes (Table I).Table ICommon additional domains and motifs of UBX proteinsNameaAbbreviations: aa, amino acids; CC, coiled-coil; HL, helical lariat; HP, hydrophobic patch; PUB, peptide N-glycanase/UBA or UBX-containing proteins; SEP: shp1, eyc, and p47; SHP box, linear motif derived from shp1p; shp1, eyes closed and p47; UAS, ubiquitin-associating; UBA, ubiquitin-associated; UBL, ubiquitin-like; UBX, ubiquitin regulatory X; UIM, ubiquitin-interacting motif; VIM, VCP-interacting motif; ZnF, zinc finger.Length (aa)FeaturesbIn the consensus sequences, x indicates any residue, h indicates a hydrophobic residue, and a represents an acidic residue.Possible functionsRefsUBA40Three-helix bundleUbiquitin binding[32.Hofmann K. Bucher P. The UBA domain: a sequence motif present in multiple enzyme classes of the ubiquitination pathway.Trends Biochem. Sci. 1996; 21: 172-173Abstract Full Text PDF PubMed Scopus (356) Google Scholar]SEP65Conserved loop residuesUbiquitin-independent binding[33.Soukenik M. et al.The SEP domain of p47 acts as a reversible competitive inhibitor of cathepsin L.FEBS Lett. 2004; 576: 358-362Crossref PubMed Scopus (17) Google Scholar]UAS110Positive surface patchMay interact with fatty acids[28.Kim H. et al.UAS domain of Ubxd8 and FAF1 polymerizes upon interaction with long-chain unsaturated fatty acids.J. Lipid Res. 2013; 54: 2144-2152Abstract Full Text Full Text PDF PubMed Scopus (30) Google Scholar]UBL70Ubiquitin foldUbiquitin-like modifier[34.Bogan J.S. et al.Endoproteolytic cleavage of TUG protein regulates GLUT4 glucose transporter translocation.J. Biol. Chem. 2012; 287: 23932-23947Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar]HP40Hydrophobic residuesMembrane anchor[25.Deruyffelaere C. et al.PUX10 is a CDC48A adaptor protein that regulates the extraction of ubiquitinated oleosins from seed lipid droplets in Arabidopsis.Plant Cell. 2018; 30: 2116-2136Crossref PubMed Scopus (51) Google Scholar]UIM20aaxxhxxAhxxS consensusBinds to the autophagy adaptor Atg8[35.Hofmann K. Falquet L. A ubiquitin-interacting motif conserved in components of the proteasomal and lysosomal protein degradation systems.Trends Biochem. Sci. 2001; 26: 347-350Abstract Full Text Full Text PDF PubMed Scopus (383) Google Scholar]ZnF30C2HC typeUbiquitin binding[16.Zhang J. et al.Versatile control of the CDC48 segregase by the plant UBX-containing (PUX) proteins.Comput. Struct. Biotechnol. J. 2021; 19: 3125-3132Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar]a Abbreviations: aa, amino acids; CC, coiled-coil; HL, helical lariat; HP, hydrophobic patch; PUB, peptide N-glycanase/UBA or UBX-containing proteins; SEP: shp1, eyc, and p47; SHP box, linear motif derived from shp1p; shp1, eyes closed and p47; UAS, ubiquitin-associating; UBA, ubiquitin-associated; UBL, ubiquitin-like; UBX, ubiquitin regulatory X; UIM, ubiquitin-interacting motif; VIM, VCP-interacting motif; ZnF, zinc finger.b In the consensus sequences, x indicates any residue, h indicates a hydrophobic residue, and a represents an acidic residue. Open table in a new tab protein domains are shown boxed with domain names. The first two letters of the protein name indicate the species: At, Arabidopsis thaliana; Ce, Caenorhabditis elegans; Hs, protein are as p47 and aa, amino The to the of The full and of domains and motifs are in domains or motifs but experimentally are by The new resulting from this are in the and in the (see for full The UBX domain of has a loop most protein sequences, AF can predict the structure of domains and the of regions with high the can between in the sequence J. et protein structure prediction with 2021; PubMed Scopus Google (Box AlphaFold can be to connections between protein building blocks and structure in protein connections and structural are conserved across they may indicate functionally coherent and interaction The AlphaFold predictions for the proteomes of most model and and to all in the K. et protein structure prediction for the 2021; PubMed Scopus Google Scholar]. These predictions are freely and where web their analysis knowledge of structural to for knowledge of structural the by AF can sequence motifs and structural connections between protein building In Figure we the use of on two from group and from group for knowledge of structural the by AF can sequence motifs and structural connections between protein building In Figure we the use of on two from group and from group We propose that UBX proteins across kingdoms to their domain and AlphaFold features information between and can testable hypotheses for the of UBX proteins. We this for the PUX proteins are poorly with their animal we is for that this knowledge may be to other We also AlphaFold features that are conserved in and propose functional hypotheses resulting from to the of UBX to features different from we is and provide additional We proteins that a UBX domain to group (Figure 1). from this group p97 to the by to the the M. et of the adaptor a for in Biol. PubMed Scopus Google Scholar]. The UBX domain to suggesting that the motifs are in the N-terminal of Although the N-terminal has sequence to may functionally in this to CDC48 to specific protein complexes in a The AlphaFold 3D structural and that a is of the UBX domains of and to the UBX domain through a motif h is a hydrophobic and x is any we the (Box with an motif are to the UBX domains of other plant and animal UBX proteins (Box and suggesting functional The is from the UBX that to and may provide structural or additional motifs and by AF analysis to PUX proteins conserved sequence motifs of as shown in AF analysis also structural by building blocks as shown in Figure the structure and sequence of motifs and the of the the UBX of the Open table in a new tab functional Open table in a new tab AF analysis to PUX proteins conserved sequence motifs of as shown in AF analysis also structural by building blocks as shown in Figure the structure and sequence of motifs and the of the the UBX of the Open table in a new tab functional Open table in a new tab The of group proteins is a N-terminal UBX domain that to this domain to form a helical or in Figure 1). the UBX domain of and to the the CDC48/p97 thereby the et interactions and for the regulation of Arabidopsis CDC48/p97 by the plant ubiquitin regulatory X (UBX) domain-containing Biol. Chem. Full Text Full Text PDF PubMed Scopus Google A. et interaction reveals an extended UBX domain in that functional p97 PubMed Scopus Google et of AAA by their Full Text Full Text PDF PubMed Scopus Google Scholar]. AlphaFold also a for suggesting that all group CDC48/p97 through to domain-containing UBX proteins also a UBX We proteins to group UBX is in and amino to group proteins, suggesting a However, from group to p97 M. et is a VCP-interacting protein that is involved in Res. PubMed Scopus Google P. Schindelin H. The structural and functional of the protein motif of to the N-terminal domain of Biol. Chem. 2011; Full Text Full Text PDF PubMed Scopus Google A dynamic for in of PubMed Scopus Google Scholar]. The to p97 may be the UBX domain of a motif in its it from to et is a for and 2018; PubMed Scopus Google Scholar]. its domain to the C-terminal motif of and it is the motif that to (Box 1). also has an loop in its UBX suggesting that it bind to J. et al.Versatile control of the CDC48 segregase by the plant UBX-containing (PUX) proteins.Comput. Struct. Biotechnol. J. 2021; 19: 3125-3132Abstract Full Text Full Text PDF PubMed Scopus (6) Google and a as a interaction for J. et al.Versatile control of the CDC48 segregase by the plant UBX-containing (PUX) proteins.Comput. Struct. Biotechnol. J. 2021; 19: 3125-3132Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar]. However, of a VIM, a SHP in its we that all domain-containing UBX proteins from group form an with CDC48/p97, thereby the UBX domain for other for their UBX may be We the SEP domain-containing and to group with different The in group and group bind to CDC48/p97 their C-terminal SHP and UBX domain M. et by a p97 AAA-ATPase Cell. 2018; Full Text Full Text PDF PubMed Scopus (42) Google suggesting CDC48/p97 also for the other of The biological of SEP domains remain J. et al.Versatile control of the CDC48 segregase by the plant UBX-containing (PUX) proteins.Comput. Struct. Biotechnol. J. 2021; 19: 3125-3132Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar]. and a and a SHP motif of their SEP domain J. et al.Versatile control of the CDC48 segregase by the plant UBX-containing (PUX) proteins.Comput. Struct. Biotechnol. J. 2021; 19: 3125-3132Abstract Full Text Full Text PDF PubMed Scopus (6) Google M. et by p97 is through of catalytic and regulatory by the p97 Mol. Biol. PubMed Scopus Google Scholar]. AlphaFold that this the SEP domain and that the in in a structural (Box However, functional may between of and the a UBA In the UBA domain to ubiquitinated in a et al.The of two proteins, p47 and in the degradation of a Cell. Biochem. 2017; PubMed Scopus Google and the in the of CDC48/p97 et and interactions of a major adaptor of the AAA J. 2004; PubMed Scopus Google of functions. group a the of serves as a that allows the degradation of by p97 M. et by a p97 AAA-ATPase Cell. 2018; Full Text Full Text PDF PubMed Scopus (42) Google M. et by p97 is through of catalytic and regulatory by the p97 Mol. Biol. PubMed Scopus Google Scholar]. group and for also have a of the has conserved a a and hydrophobic and suggesting that it may have a conserved ligand We refer to this as the (Box The roles of the SEP remain to be to it to group AlphaFold that the SEP domains form a structural However, this AlphaFold prediction is of group proteins with or two and it shown that the target proteins to where they mediate the of et proteins a new of autophagy adaptors and Full Text Full Text PDF PubMed Scopus Google Scholar]. The UBX domains of and bind to CDC48/p97 et of the plant ubiquitin regulatory X (UBX) domain-containing protein in Arabidopsis 2013; PubMed Scopus Google et structures of the UBX domain of and its with p97 Res. 2017; PubMed Scopus Google suggesting that of this group have UBX The mechanism through CDC48/p97 is but the of a UBA domain the of AlphaFold that group PUX proteins an additional acidic that in the sequence motif and in and (Box is between the most C-terminal motif and the UBX A the in group from other but in other suggesting that it is specific for the function of group group proteins a domain of biological and also a from a in and The of with a suggesting they mediate also a motif that is of the but where the motif is by a sequence to and in We and UBA, UAS, and UBX domains in group of has functionally and we to with the same domain in fungi or AlphaFold that the UBA, UAS, and UBX domains are and that all UBX domains a on we that and are involved in protein UBX proteins with a domain and a hydrophobic form group and an N-terminal and their is in a C-terminal The to the for its function in degradation J. et of a new protein that protein Sci. PubMed Scopus Google Scholar]. an additional N-terminal UBA The of this group is between the UBA and and serves as an anchor for lipid droplets of and C. et al.PUX10 is a CDC48A adaptor protein that regulates the extraction of ubiquitinated oleosins from seed lipid droplets in Arabidopsis.Plant Cell. 2018; 30: 2116-2136Crossref PubMed Scopus (51) Google the C. Buchberger A. to ubiquitin and their to protein Biol. PubMed Scopus Google Scholar]. The UBX domains of and group bind to J. et of a new protein that protein Sci. PubMed Scopus Google A. et of to a UBX domain from proteins involved in and 2004; 21: PubMed Scopus Google Scholar]. group domains have but their function The domain of may sense fatty H. et al.UAS domain of Ubxd8 and FAF1 polymerizes upon interaction with long-chain unsaturated fatty acids.J. Lipid Res. 2013; 54: 2144-2152Abstract Full Text Full Text PDF PubMed Scopus (30) Google the domain is the amino that to the protein that the domain to ubiquitinated the UBA domain of et and bind a involved in Biol. PubMed Scopus Google Scholar]. group proteins are to use their UBA domain for group proteins may use their domain to ubiquitinated AlphaFold that group a helical of the C-terminal group a of their UBX These structural are to have functional and for in the that and A. et of to a UBX domain from proteins involved in and 2004; 21: PubMed Scopus Google Scholar]. group AlphaFold that the helical is with the CC, the and UBX domains structural (Box to the of the group with an an and in how UBX proteins control the CDC48/p97 segregase is for how fungi, and animals to changes and a We propose that UBX proteins can be on their building combinations and AlphaFold constitute conserved 'work By a for knowledge across eukaryotic such a classification may future experimental research on UBX proteins to domain or higher-order We this to the PUX protein for experimental information We features in biological roles experimental (see also where knowledge be for of UBX proteins in is the function of the domain and of the group UBX or of PUX proteins in is the function of group (see UBX proteins across kingdoms reveals that may functionally coherent UBX proteins can PUB, or but a combination of the two linear interaction motifs and SHP in the same protein and are or in combination with the SEP the of or the UBA the of and UBX proteins with an also a the biological of features help in the the of PUX proteins, post-translational modifications and the of or AlphaFold that PUX proteins multiple interaction connected through this is a of proteins involved in through PUX proteins have to is involved in the of CDC48 by in plants et the of PubMed Scopus Google and animal of the UBA domain can produce in p97-containing et and of the PubMed Scopus Google Scholar]. is that PUX proteins the of the AlphaFold may also help in protein and provide structural for in ligand AlphaFold or AlphaFold and of predictions and be to inspire to In we propose that the combination of AlphaFold features and a cross-kingdom comparison a wealth of testable hypotheses to efficiently guide future experimental research. PUX proteins, the information can help to the that proteins use to control The resulting of the of the UBX proteins in eukaryotic and may have in various from to is to how the different features of PUX proteins control the biological of and contribute to plant and We propose that the are the most this are the biological of the conserved structural features the and the the and synergize to biological is the of the UBX domain from group is it present the UBX domains of this group bind to CDC48/p97 its are the and biological roles of the domain is in PUX proteins and from other kingdoms, but its function is the biological of group PUX proteins (i.e., or their they have in other kingdoms, and their function in plants is are PUX proteins for through or post-translational or and PUX proteins of functional and structural the of PUX proteins in However, in has The is to how the different features of PUX proteins control the biological of and contribute to plant and We propose that the are the most this are the biological of the conserved structural features the and the the and synergize to biological is the of the UBX domain from group is it present the UBX domains of this group bind to CDC48/p97 its are the and biological roles of the domain is in PUX proteins and from other kingdoms, but its function is the biological of group PUX proteins (i.e., or their they have in other kingdoms, and their function in plants is are PUX proteins for through or post-translational or and PUX proteins of functional and structural the of PUX proteins in However, in has research by the of and through the and from the of We for to the of this and we H. and A. for their on the are with diverse cellular A and functionally diverse protein that for The is for extracting or protein a protein that two or different proteins with other through interactions with of this artificial intelligence (AI)-based algorithm produces highly predictions of protein structural features and that its and AlphaFold predictions for the proteomes of Arabidopsis and can be or a protein domain is an amino sequence that a 3D by specific can produce are or specific interactions to a biological a linear sequence with a distinct amino the to a conserved sequence with a biological are with the same function in different the AlphaFold is a of the in the between two In the of a it can indicate between a by AlphaFold for 3D protein structural The from to with of and are by AlphaFold to be protein regions with a may be intrinsically a by protein complexes that proteins by the of proteins in a cell or a protein that protein from its In this activity is with an unfolding of the target a protein in can be as a post-translational for other proteins. can proteins in proteasomal a protein that the 3D structure of a target