Protein Function PDF

Topic 06: Protein function Protein function Proteins have an enormous range of functions in nature While complex, protein functions arise from some combination of: Recognizing and binding other molecules – be they proteins, nucleic acids or small molecules Catalyzing chemic...

Topic 06: Protein function Protein function Proteins have an enormous range of functions in nature While complex, protein functions arise from some combination of: Recognizing and binding other molecules – be they proteins, nucleic acids or small molecules Catalyzing chemical reactions (enzymes) Serving as a substrate for modifications When combined with a protein’s intrinsic structure and dynamics, these elementary functions can give rise to very complex roles E.g. hemoglobin binds heme, reacts with CO2, binds bis-phosphoglycerate, and binds O2; the net effect is the ability to effectively bind O2 in the lungs, and release it at different sites in the body. Binding Cells contain an enormous variety of molecules – ions, proteins, nucleic acids, lipids, and metabolites These molecules diffuse and collide with one another continuously (in bulk solution or within a membrane) An individual protein will “recognize” specific molecules from the complex mix Recognition implies that the protein and its partner form a sufficiently stable complex that they stay associated long enough to have a biologically meaningful effect Such complexes are the basis of signaling events, catalysis, and protein localization  Binding energetics Binding requires that the complex formed between the two molecules have an overall lower energy than the two individual molecules in solution Binding is driven by hydrogen bonds, van der Waals interactions, and the hydrophobic effect etc. This is offset by the energy required to displace water, plus loss of conformational heterogeneity Generally, interactions maximize the Peptidoglycan binding to lysozyme contact surface to maximize 1HEW interaction energy Stable vs. transient interactions Interactions that persist over a long period of time generally show low micromolar (10-6 M) or better affinity (below picomolar (10-12 M) in some cases) For other biological processes, interactions are required to be relatively weak and fleeting These interaction affinities are typically in the micromolar to millimolar range For weak interactions, a large fraction of the protein may be unbound at any given moment, even in the presence of ligand Small molecule binding Cells contain an enormous variety of small (non- polymer) molecules These include metabolic intermediates, cellular building blocks, enzyme co-factors and ions The binding site of a protein that is meant to recognize a specific small molecule is generally “pre-organized” to bind that ligand in a specific conformation The protein binding site is generally complementary in properties to the ligand being recognized  e.g. Streptavidin biotin Streptavidin is a 159 a.a. protein isolated from the bacterium Streptomyces avidini Streptavidin functions as a biotin scavenging protein Recall that biotin is an essential cofactor for most enzymes catalyzing carboxylation reactions (e.g. acetyl- CoA carboxylase in FA synthesis) Streptavidin has very high affinity for biotin (~10-14 mol/L), among the highest affinity interactions known to occur in nature Streptavidin is used extensively in biotechnological applications, allowing biotin tagged molecules to be reliably captured even if they are very dilute  Streptavidin structure Streptavidin is organized into eight b-strands These form a single b-sheet, with a simple up-down topology The first strand pairs with the last to form a b-barrel The functional tetramer is organized from four streptavidin protomers in D2 symmetry  Streptavidin binding biotin Biotin (yellow) binds each protomer in the center of the barrel Note that a short helix in the apo structure (purple) reorganizes to from a loop that covers the bound biotin Similar structural transitions are commonly observed during binding  Details of biotin binding Buried polar groups are stabilized by side chains that make hydrogen bonds mostly non-polar side chains around Biotin is otherwise non-polar atoms non-polar, and packs against non-polar side chains The hydrophobic effect contributes significantly to biotin hydrogen bonds to binding polar urea group  Biotin shape complementarity The streptavidin binding pocket closely complements the shape of biotin Note that the molecular surface (at van der Waals radius) just touches the van der Waals spheres of non-polar atoms This means that these atoms are in van der Waals contact Polar atoms are closer than this, as hydrogen bonds are shorter Protein-protein binding In addition to cases where proteins form stable, permanent associations (see quaternary structure), proteins can also form short lived complexes Such complexes can be used to localize a protein to a specific place in the cell, may have a regulatory role, or may allow one protein to act as a substrate or allosteric modulator for the other Proteins bind other proteins through extended interfaces Generally, at least 700 Å2 of surface needs to be buried for a reasonably strong interface The interactions are generally similar to oligomeric interfaces, though less hydrophobic if the proteins form only temporary complexes  E.g. Ran and RanGAP form a short-lived signaling complex Ran (purple) and RanGAP (grey) form a temporary complex that promotes exchange of GDP for GTP in Ran’s active site Both proteins fold independently, and recognize each other in a transient binding event The individual proteins do not shift much, but the Ran active site opens slightly - promoting GDP exchange Note that these proteins bury an extended surface between them 1TX4  E.g. Ankyrin complex in erythrocytes membrane Valesse et al NSMB 2022 Ankyrin (red) is a large helical repeat protein It connects the cytoskeleton (via spectrin) to the membrane In erythrocytes, it binds the regulatory domains of three copies of the bicarbonate/chloride antiporter in the cytosol (orange, lime, mauve) as well as the Rh protein (blue) and aquaporin This complex anchors the cytoskeleton to the membrane  Protein - Peptide binding Peptide binding differs from protein binding in that short peptides have no intrinsic structure This means that the peptide can adapt itself structurally to a target protein surface The peptide has the flexibility to maximize interactions However, the peptide pays a cost in terms of loss of conformational SH3 domain – peptide complex entropy upon binding  Nucleic acid binding Nucleic acid binding proteins generally use multiple basic residues (especially arginine) to interact with the polyphosphate backbone Hydrogen bonds with exposed polar groups add specificity For RNA binding, stacking on the exposed edges of bases can be an important RNA contributor to stability Disordered, basic residue rich regions that become ordered upon NA binding P22 peptide are a very common theme P22 N-peptide binding boxB RNA (1A4T) Enzyme catalysis Many proteins can catalyze chemical reactions Broadly, most reactions catalyzed can be thought of as being either: acid-base chemistry (where protons are added and/or removed) or redox chemistry (where electrons or hydride ions are added or removed Proteins have multiple groups suited to acid base chemistry, but often rely on external co-factors for redox chemistry Enzyme active sites Enzymes bind their substrates in a conformation that pre-organizes them for the reaction Catalytic residues then initiate the reaction by polarizing bonds, and/or attacking reaction centers nucleophilically or electrophilically, and/or removing or adding protons or electrons chymotrypsin (1dlk) Catalysis mostly involves charged or titratable amino acids Catalysis is dominated by a handful of residues His is the most common catalytic residue, despite being fairly rare overall Catalytic residues generally enric hed are charged, or have a pKa near neutral ted deple Polar side chains partake more rarely Non-polar aliphatic sidechains (G,A,V,L,I,P) very rarely play catalytic roles Ribeiro et al JBC 2020  Protein cofactors SAM pyridoxal phosphate biotin Enzymes (and other proteins) can use a variety of co-factors to accomplish chemically challenging tasks that their side chains can not Some co-factors are tightly bound or covalently linked and remain permanently associated with a given protein They can allow the protein to capture light energy (e.g. chlorophyll, retinal) or use chemically challenging groups (heme: O2; biotin: CO2-; cobalamin CH3; FAD: H-; pyridoxal phosphate: NH3) Other co-factors can carry activated chemical groups between enzymes e.g. hydride (NADPH), methyl groups (SAM), carboxylic acids (CoA), phosphate (ATP) These bind and release every reaction cycle, and are recharged by enzymes in dedicated metabolic pathways (e.g. pentose phosphate pathway)  E.g.: D-lactate dehydrogenase D-lactate dehydrogenase reduces pyruvate to D-lactic acid (redox reaction) This enzyme occurs in bacteria, but a human homolog plays an essential role in serine metabolism D-LDH is a tetramer, with the active site found in a cleft between two Rossmann domains DLDH catalytic groups Proteins have no good hydride donors, so the hydride ion is donated by the bound co-factor NADH (cyan) A proton is transferred from histidine to the pyruvyl keto group; glutamate helps shift the His pKa Arginine is required to polarize the keto group and make it more reactive These elements are all essential catalytic elements in this family, and are conserved by all family members that catalyze similar reactions D-LDH domain closure D-LDH’s small domain rotates to enclose the substrates This ensures that no extra room is available that might allow bulkier, non-specific substrates to bind It also produces a non-polar environment to promote hydride transfer Such motions are an important aspect of the function of many proteins  D-lactate dehydrogenase binding D-LDH binds a small substrate (pyruvate) and makes favourable contacts with every atom to ensure specificity Pyruvate stacks on the nicotinamide ring of NAD (cyan) Note: catalytic residues also contribute to binding Here, the catalytic Arginine and Histidine form hydrogen bonds with the keto group D-lactate dehydrogenase lactic acid + NAD ternary complex (3KB6) D-LDH – carboxylate binding The carboxylate group of the substrate is bound by a set of hydrogen bonds, mediated by both backbone and ligands Two backbone amide nitrogen atoms bind the substrate carboxylate Tyrosine makes a third hydrogen bond D-LDH substrate specificity In D-LDH, the methyl group sits in a hydrophobic pocket contributed by a Phe and Tyr side chain These provide favourable van der Waals interactions to help drive binding Importantly, there are various other common metabolites that resemble pyruvate but differ in having bulkier groups in place of the methyl The Phe and Tyr residues therefore are important for the specificity of D-LDH  Zn2+ Metal ions can be recruited by select amino acids Many proteins bind specific metal ions Metal ions can be catalytic (e.g. metallo- Zinc finger proteases), structural (e.g. zinc finger proteins) or regulatory (e.g. calmodulin) Fe2+ S cluster Metal ions are commonly coordinated mostly by Cys, His, Asp or Glu, H2O or S For transition metal ions (e.g. Fe, Cu, Zn, Co) the bonds are almost covalent in strength Transition metals are commonly used in redox reactions E.g. electrons can jump 10+ Å between the Ferredoxin iron-sulfur clusters of proteins E.g. Src kinase binding ATP Src kinase binds ATP in a pre-organized pocket Hydrophobic residues stack on the top and bottom of the adenine base + ribose Hydrogen bonds to the edge of the base are contributed by protein Mg2+ backbone atoms The phosphate group is coordinated by a magnesium ion that itself mediates interactions with the enzyme A Lys + Arg residue also help neutralize the PO4 charge 3DQW Note that many interactions are made to ATP by bound water molecules  E.g. Calmodulin C-terminal Calmodulin is a Calcium EF hand modulated protein Calmodulin responds to elevated calcium levels in the cell linker (secondary signal) by binding a wide variety of target proteins It is built as two distinct 4 helix motifs known as EF-hands N-terminal The EF hands each form a four- EF hand helix bundle, connected by a linker (yellow) Calmodulin EF hands The EF hands are each built as a pair of helix-turn-helix motifs The turns connecting the helices (white) are relatively long and include two Asp residues, with an additional Glu at the end of the second helix These residues are key to calcium binding Note – similar calcium binding motifs are found in dozens of other proteins Calcium binding to calmodulin One calcium ion binds to each loop region of the helix-loop- helix motifs The loop provides the flexibility to position each residue so that it points towards Ca Calcium prefers oxygen ligands, especially carboxylates The calcium is coordinated by 2x Asp, Glu, Asn plus a backbone carbonyl oxygen  Binding calcium forces a structural change + Ca2+ In order to bind calcium, the helices need to repack from being almost parallel to packing at right angles The Ca-Calmodulin complex exposes a hydrophobic patch The structural rearrangement opens up the hydrophobic core of the EF hand This exposes an extended hydrophobic surface Note that the energetic cost of exposing hydrophobic residues to water is compensated by the favourable energy of calcium binding Ca-Calmodulin binds diverse proteins, regulating them Ca-Calmodulin can bind exposed non- polar regions on a variety of targets The EF hands can move relative to one another, allowing the shape of the cavity to be adjusted The methionines add side chain flexibility to allow optimized packing Calmodulin can bind and regulate over 300 different proteins Here it is bound to a fragment of calmodulin-dependent kinase (orange) 1iq5 Forming this complex activates the kinase  Many proteins are covalently modified glycosylation Over 200 distinct covalent modifications of proteins have been characterized Glycosylation can direct proteins and act as intracellular signals Lipidation can help localize proteins to the membrane Ubiquitination tags proteins for degradation palmitoyl group wnt-fizzled complex  Protein glycosylation N-glycans on Thy-1 Erythrocyte surface – David Goodsell In eukaryotes, the extracellular domains of proteins are typically heavily modified by oligosaccharides Oligosaccharides can contribute ~20 % of total protein mass These affect cellular trafficking, and are important targets for interactions with the immune system, neighbouring cells, and pathogens (bacteria and viruses) Saccharides are added to Asn (N-linked) or Ser/Thr (O-linked) residues as they mature through the ER & Golgi Regulatory modifications Eukaryotic proteins are often modified by small molecular groups These modulate the activity of a protein or the ability of the protein to interact with other proteins E.g. phosphate groups are added by kinases to Tyr, Ser or Thr residues Acetyl groups can be added to Lys e.g. on histones Modifications can also use whole proteins – e.g. ubiquitin or Sumo Addition of lipid groups can temporarily or permanently target a protein to the membrane Function and structure constrains sequences Proteins are subject to mutations Protein coding sequences are subject to mutations caused by mutagens or DNA replication errors These are rare, but do slowly accumulate in populations Even within a species, small differences can be found in the sequence of a given protein These are generally neutral, but can impact function When comparing equivalent proteins in different species, more differences are apparent In general, the more distantly related the species, the more changes you find These changes are not randomly distributed though; some sites are much more prone to mutations than others Active sites residues are functionally critical and very conserved An enzyme’s active site should be complementary to the ligand, and be capable of catalyzing the reaction Most residues will be important, and mutations here will decrease enzyme efficiently Active site residues tend to be very conserved Even small changes can result in alterations in substrate specificity Lysozyme; 1HEW Buried residues are conserved Buried residues mediate the packing required to stabilize the protein, and correctly orient key elements These residues tend to be quite conserved They do commonly allow substitution of similar residues (E.g. Val->Ile) Lysozyme; 1HEW Surface residues are more variable Surface residues that are not part of the active site or any binding site interact primarily with neighbours and solvent They tend to be quite mobile Substitutions of these residues are generally very well tolerated The general nature of the surface (e.g. polar for soluble proteins) is the main requirement Lysozyme; 1HEW Different sequences can result in a protein with function X Most residues are not critical for protein function, and are subject to gradual drift over many generations Once consequence of this is that any given protein has a (at least slightly) different sequence in every species E.g. chicken (white) vs turkey (orange) lysozyme differs by 5 amino acids These substitutions are on the surface, and minimally impact function Lysozyme; 1HEW vs 1TEW Protein divergence and function change As protein sequences slowly accumulate small changes over evolutionary time periods, they can potentially retain the same function Alternatively, they could also potentially evolve a new function It can be difficult to predict how much sequence change might indicate a change in the specificity of a protein Many human and bacterial enzymes have the same function, but have less than a third identical residues On the other hand, some proteins can switch function with very few amino acid changes Substrate specificity in D-LDH homologs D-2-hydroxy isocaproate DH is related to D-LDH but makes the precursor to Leucine Changes in D-LDH’s methyl binding pocket, including residue shifts and Phe->Leu helps accommodate the larger substrate Other family members use different groups to bind different substrates D-LDH homologs recognize different substrates by altering the methyl binding pocket The residues lining the “methyl” pocket clearly differentiate D- LDH orthologs from other related enzymes with differing substrates different enzyme specificities HicDH D-LDH  b-carbonic anhydrases Carbonic anhydrases hydrolyse CO2 This is an unusual reaction that uses a zinc-activated water ion The resulting OH- ion is used to attack CO2 The first structure determined showed the active site is formed by a dimer The catalytic zinc ion is bound by a Cys and a His-X-X-Cys motif An Asp/Arg motif following the Cys is essential for catalysis A hydrophobic pocket helps bind CO2 Kimber and Pai, EMBO J, 2000  b-carbonic anhydrase divergence This is a bacterial CA originally assigned to a new class (epsilon) This protein has 11 % sequence identity (near noise levels) to the pea enzyme However, the structure has clear resemblances The protein has two CA domains, only one is active The active site is formed by a single pseudo-symmetric chain Only 5 essential residues in the active site are conserved Function is nevertheless identical between the two enzymes Human A/B/O glycosyltransferases GTA and GTB Human blood groups are primarily classified on the basis of presence of A/B/O antigens on cell surfaces The difference lies in whether a GlcNAC, Glc or no saccharide is added to a ceramide pentasaccharide The specific version(s) of a specific glycosyltransferase we inherit determines our blood type (an inactive allele of this enzyme gives rise to the O antigen) Lehninger 6th Fig. 10-15 GTA vs GTB glycosyltransferases The GT comes in two allelic variants GTA transfers N-acetyl glucosamine (GlcNAc) from UDP-GlcNAC to the antigen GTB transfers Glucose from UDP-Glc to the antigen These substrates differ only on the presence of the amide at C2 5nrb Patenaude et al, 2002 GTA vs GTB active site GTA and GTB differ in 4 a.a. Only two changes are important Leu->Met closes the space that the amide group needs to bind These differences are enough to produce a distinct biochemical function Patenaude et al, 2002 Protein divergence and function Sequence divergence in a protein functional site can lead to the evolution of new functions Changes in a local area that distinguish between similar substrates are often key (e.g. DLDH) In some cases, only a couple of changes are needed to change the function (e.g. GTA v GTB) In some rarer cases, proteins will retail a function despite having deeply different sequences with only a few key residues conserved (b-CA) Shared protein function can vary unpredictably with sequence similarity Proteins can evolve new functions most easily when they have access to structurally/ chemical similar partners/substrates E.g. if two substrates are available in the cell that only differ in minor ways, switching between them may only require changing a couple of interacting side chains E.g. GTA vs GTB (two similar sugars), peptide binding proteins If all alternative substrates are highly dissimilar, then switching requires many changes, and is much less likely E.g. carbonic anhydrases; there is little else in nature like CO2 Bacterial enzymes encompass enormous diversity Eukaryotes are well-studied and generally share a core metabolism that does not vary very much Most eukaryotic enzymes (esp. in central metabolism) can be confidently assigned functions Bacterial metabolism is very diverse - they can synthesize and break down almost any molecule found in nature Bacteria repeatedly evolve new functions within existing enzyme families This can make assigning functions to bacterial enzymes on the basis of sequence challenging