This is not an obligatory material, it is for students more interested in proteins. Protein composition and structure - supplement Attila Ambrus.

This is not an obligatory material, it is for students more interested in proteins.

Protein composition and structure - supplement

Attila Ambrus

branching at the -carbon tends to destabilize an -helix (Val, Thr, Ile) due to steric clashes, these aa are better suited to -sheets where they stand out of the chain

Ser, Asp, Asn disrupt -helices due to H-bonding donor/acceptors sites near the main chain where they compete for N-H or C=O

Analysis of amino acid composition of secondary structural elements

Pro tends to disrupt both -helices and -strands due to its ring structure

Gly readily fits into every sorts of structural motifs because of its small size

predictions of secondary structure of 6 or fewer residues, taking the above (and other) considerations into account, proved to be ~60-70% accurate; reasons include the not abrupt change of preference one aa hasfor one structural motif or another

tertiary interactions may push the same peptide to adopt another secon-dary structure in a different local environment

many sequences can adopt alternative conformations in different proteins

the VDLLKN sequence shown in purple assumes a helical or a -strand confor-mation in two different proteins (3WRP.pdb and 2HLA.pdb)

Folds

protein tertiary structures are divided into five main classes according to the secondary structure content of their domains:

all- domains, all- domains (-barrels, e.g. Greek key motif), + domains (irregular fashion of arrangement), / domains (-- motifs) and “others”

each class contains many different folds further classified into families

no necessary functional connection is in this type of classification: a certain type of function is often, but not always, restricted to a certain type of fold (convergent evolution) – fold of a protein is “only” a scaffold to which functions (active sites and different binding sites) are “added”

a protein of 100 aa may have 20100 possible sequences/conformations, but anestimate says that there are only ~1000 folds in nature (we know a few hund-reds of them so far); if we consider only the # of human genes (even without splice variants) there should be more than ~30,000 individual conformations if all protein sequences would adopt a new structure

conformation is more conserved than sequence

modules: genetically mobile units manifested as separate protein domains, functional units that are shared by proteins of similar functions; there are domain (fold) families throughout the phylogenetic tree to deliver similar functions (small differences in sequence with evolutionary conserved regionsfeatured in multiple sequence alignment of protein primary structures)

some folds are more favored than others as they represent a more stablestructure and some proteins may converge towards these folds over the course of evolution; a number of folds though are found in only one group of proteins

the CATH domain database classifies domains into approximately 800 fold families, ten of these folds are highly populated and are referred to as 'super-folds‘; super-folds are defined as folds for which there are at least three structures without significant sequence similarity (the most populatedis the α/β-barrel super-fold)

similar folds may have very low sequence identity and that is true vica versa, nevertheless, ~30% sequence identity generally means similar structure (may have diverged from a common ancestor; homology modeling of proteins is based on this premise)

large sheet that twists and coils to form a closed structure

first and last -strands are H-bonded

typical antiparallel arrangement of strands

found in proteins spanning membranes (e.g. porins) and in proteins that bind hydrophobic ligands inside the barrel (e.g. in a lipocalin fold)

-barrels

CATH (Class, Architecture, Topology, Homologous) database

Similar database: SCOP (Structural Classification Of Proteins)

all known protein structures are sorted according to their folds

Superhelices

-keratin (main component of wool and hair) consists of two right-handed -helices intertwined to form a left-handed superhelix called coiled coil (superfamily of coiled-coil proteins, ~60 proteins in humans)

2 or more helices can entwine and form a stable, even 1000 Å (0.1 m) or longer, structure found in cytoskeleton, filaments, muscle proteins

3.5 residues/turn, heptad repeats, every 7th residue is Leu on each strandand these two Leu interact (hydrophobic interaction), 2 Cys can also interact (S-S) stabilizing fiber

wool can be stretched (some interactions among helices brake, S-S does not and pulls back after release)

hair and wool have fewer cross-links, horn, claw, hoof are hard

Collagen

most abundant protein in mammals, main fibrous component of skin, bone, teeth, cartilage and tendon

extracellular protein, rod shape, ~3000 Å long/15 Å in diameter, 3 helical protein chains (~1000 residues each, every 3rd residue is Gly, Gly-Pro-(Pro-OH) triad is frequent, Pro-OH (4-hydroxyproline) is a naturalamino acid derivative)

no H-bonds inside the helical strands, stabilization occurs via steric repulsion between Pro and Pro-OH

~3 residues/turn, 3 helices wind in a superhelical cable that is stabilizedby H-bond in between strands (Pro-OH participates in H-bonding networkand lack of –OH on Pro in collagen lead to the disease scurvy (Vitamin C deficiency, ascorbate reduces Fe3+ to Fe2+ in prolyl hydroxylase for its continuous activity)

Pro rings are on the outside, Gly in every 3rd position is needed because the superhelix is very crowded inside and there is no place for any otherbigger amino acid

Denaturation of proteins

denaturating agents have chaotropic properties and disrupt the 3D struc-ture of proteins (or DNA/RNA)

chaotropic agents interfere with intramolecular H-bonding and van der Waals forces (hydrophobic interactions) and denature biomacromolecules

chaotropes also break down the H-bonded network of H2O allowing proteinsmore structural freedom and encouraging extension and denaturation

Examples of chaotropic agents: 6-8 M urea, 2 M thiourea, 6 M guanidiniumchloride, 4.5 M LiClO4

and in general high generic salt concentrations can also exhibit chaotropic effects: they shield electronic charges preventing stabilization of salt bridges and also weaken H-bridges which are more stable in less polar media (not being completely solvated at high concentration, ions interact with dipoles of H-binding partners, which is more favorable than H-bridging itself); they also perturb solubility of proteins taking H2O out of the hydration sphere of proteins – (reversible) precipi-tation/fractionation of proteins

opposite of chaotropes (disorder-maker, destabilizer) are kosmotropes (order-maker, stabilizer): they stabilize proteins in solution, increase structuring of water molecules

SO42-, HPO4

2-, Mg2+ , Ca2+ , Li+, Na+, H+, OH- and HPO42- (small ions with high

charge density) are good kosmotropes exhibiting stronger interactions with H2O than H2O with itself and therefore capable of breaking H2O-H2O

H-bonds; non-ionic kosmotropes: trehalose, glucose, proline, terc-butanol

SCN-, H2PO4-, HSO4

-, HCO3-, I-, Cl-, NO3

-, NH4+, Cs+, K+, (NH2)3C+ (guani-

dinium) and (CH3)4N+ (tetramethylammonium) ions are rather chaotropes

proteins are most stable in solution when surrounded by fully H-bonded H2Oas H2O with spare H-bonding capacity has higher entropy and is more “agg-ressive”; such reactive H2O behaves in a similar way to raising T that dena-tures proteins

optimum stabilization of biological macromolecule by salt requires a mixtureof a kosmotropic anion with a chaotropic cation and the chaotropic ions (with their weak aqueous interactions) should be the direct counterions to the protein and the kosmotropic ions (with their strong aqueous interactions)in the bulk; (NH4)2SO4 is a good salt for stabilizing protein structure/activity

when the anion and cation have similar affinities for H2O they are able to remove H2O from each other most easily, to become ion-paired. A small ion of high charge density plus a large counter-ion of low charge density forms a highly soluble, solvent-separated hydrated but clustered ion pair asthe large ion cannot break through its counter-ion's hydration shell (for example, CaI2, AgF and LiI versus CaF2, AgI)

Hofmeister series of ions precipitating proteins:(Franz Hofmeister was also the one who proposed first in 1902 that amino acids build up proteins via peptide bonds (even before Emil Fischer))

true when proteins are of net negative charge, pH>pI, may reverse if pH<pI, different counterion or pH is present

in the original experiment they used a mixture of egg white proteins, did not control pH and ovalbumin was of negative charge and they got the following series:

anions: citrate3- > SO42- = tartrate2- > HPO4

2- > CrO42- > acetate- >

HCO3- > Cl- > NO3

- > ClO3-

cations: Mg2+ > Li+ > Na+ = K+ > NH4+

(reversible) Salting out/precipitation of proteins

based on smaller solubility of proteins at high salt concentration

critical concentration varies for different proteins (fractionation/purifi-cation of proteins, e.g. albumins vs. globulins)

used also to concentrate proteins from dilute solutions (e.g. after gel filtration [size-exclusion chromatography])

done generally by (solid) (NH4)2SO4 (final concentration expressed as the

% of the saturated (NH4)2SO4 solution) followed by filtration or centrifu-gation

dissolution in appropriate buffer and dialysis is used to remove high salt concentrations afterwards and get the protein dissolved back again

Caution: some ions first increase the solubility of a protein (salting in) while others may permanently denature/precipitate/poison certain proteinsor enzymes (e.g. heavy metal poisoning – irreversible complexation occurs)

protein “salting out” results from interfacial effects of strongly hydrated anions near the protein surface so removing water molecules from the protein solvation sphere and dehydrating the surface

protein “salting in” results from protein-counter ion binding and the con-sequently higher net protein charge and solvation; it occurs where the pro-tein has little net charge near its pI primarily by weakly hydrated anions.

protein solubility is minimal at the pI (net charge is zero), below or abovecharged protein molecules repel each other resulting in better solubility

precipitation is not necessarily accompanied by denaturation and vica versa

strong acids and bases can permanently destroy the H-bonding/salt-brid-ging network of proteins, denature and/or precipitate them; this is usedin the lab to test for protein content (TCA, sulfosalicylic acid)

ethanol or acetone can also precipitate proteins by shifting the dielectricconstant of solvent water that results in lower solubility of solute protein

heat denaturation/precipitation is of pathological relevance (high fever)

Folding/refolding of proteins

intriguing field of research for folding pathways

refolding techniques are used and optimized to increase protein yield in heterologous protein expression and purification experiments (over-expressed excess protein may precipitate in the form of inclusion bodies that contain protein in a (partially) denatured insoluble form)

refolding is not always spontaneous after dialysis of denaturant, helper materials are used to facilitate/initiate the folding process (native pros-thetic groups/cofactors/substrates/ligands and e.g. PEG, arginine, CHAPS, lauril maltoside, glycerol, Triton X-100, BSA, etc. are good helper materials)

a redox-shuffling system (Cys-cystine, GSH-GSSG, -SH-EtOH, DTT) helpsresolve wrongly made S-S bonds and find the thermodynamically most favorable conformation

sharp transition from the foldedto the unfolded state (“all or none”

cooperative process; same trend when the protein is refolded)

half-folded ??

if one part of the protein structure is deteriorated (getting thermodyna-mically unstable under the given con-ditions), the whole structurewill brake down (cooperatively) since the interactions that stabilized the rest of the protein are lost with this (unfolded) part of the enzyme

there are transient intermediates of folding at the atomic level (progressive stabilization of intermediates); proteins may also get transiently stabilized in a molten globule form that contains native-like secondary structural elements but a rather dynamic tertiary structure somewhere inbetween the denatured and the native states)

What is the pathway to fold up?

unfolded proteinunique conformation

in folded state?

the protein should try out all the possible conformations to find the energe-tically most favorable one?

this would take for a 100 aa protein that samples 3 conformations/aa, eachin 100 fs, ~1027 years (Levinthal`s paradox)……not a good option!

Richard Dawkins in “The blind watchmaker” asked how long it would take for a monkey to spell out accidentally on a typewriter Hamlet`s remark to Polonius “Methinks it is like a weasel“…cal-culated…it would happen (probably) in about 1040 random keystrokes

however, if we preserve the correct keystrokes and let the monkey retype only the wrong punches, the whole process would only take couple of thousands of trials! (cumulative selection, partly correct intermediates are retained)

it is the way for a protein to correctly fold in a reasonable time frame to follow an at least partly defined folding pathway with intermediates on the road to the folded form (nucleation-condensation model, energy surface funnel model with multiple possible pathways to the same final stable structure at the bottom of the funnel, deepening in the energy-funnel means fewer and fewer conformations accessible to be adopted)

intracellular proteins assist in folding/preventing misfolding or aggregationof biomacromolecules and help assemble complex macromolecular structures,these proteins are called chaperones or chaperonins and some of them are even called foldases or unfoldases

some chaperones assist in correctly folding newly synthesized protein chains as a minority of protein structures would not be able to correctly fold all by themselves

they also assist in disassembling/unfolding of macromolecular structures

they help assemble already folded structures to higher level structures (e.g. oligomers)

they sometimes need co-chaperons to fully exhibit the chaperon action

they do not convey “steric information” to fold a protein per se, they ratherprevent transformation to non-functional structures

they use sometimes ATP as an energy source for doing their folding action

Chaperones

cellular shock (e.g. heat shock) leads to higher propensity of protein aggre-gation and specialized proteins, so-called “heat-shock proteins (HSP)”, help avoid this aggregation; not all chaperones are HSPs

HSPs express as a response to higher T or other cellular shocks

important chaperons (found especially in the ER): calnexin, calreticulin,different HSPs, protein disulfide isomerase, peptidyl prolyl cis/trans isome-rase

Hsp60 (GroEL/GroES complex in E. coli, Group I chaperonin, GroES is a co-chaperonin) is the best characterized large (~ 1 MDa) chaperone complex, also found in the mitochondrial matrix

other HSPs: HSP70 (prevent apoptosis), HSP90, HSP100, etc. (the number means MW)

the mechanism of action generally requires ATP hydrolysis and major con-formational changes from the chaperon`s side to be able to encapsulate theunfolded protein to the chaperon`s “lumen” where it will start folding

Protein misfolding and aggregation – pathological relevance

some infectious neurological diseases were recently revealed to be trans-mitted by virus-sized protein particles

such examples are bovine spongiform encephalopathy (mad cow disease) and the analogous disease in humans, the Creutzfeldt-Jakob disease (CJD)

the agents causing these diseases are called prions

for proving the hypothesis that diseases can be transmitted purely by proteins, Stanley Prusiner in 1997 was awarded the Nobel Prize in Physiologyor Medicine

such proteins are massive, resistant to most regular treatments, aggregatedproteins formed from a regular cellular, mostly helical, protein in the brain, PrP (prion protein); PrPSC is insoluble and of heterogenous state

evidences say that helical and -turn protein content gets converted to -strand conformations that link to other -strands of similar nature and form extended -sheets and eventually protein aggregates (amyloids)

the infectious agent in prion diseases is an aggregated form of a protein

amyloids are insoluble fibrous protein aggregates sharing specific struc-tural traits; abnormal accumulation of amyloids in organs may lead to amyloidosis that plays role in various (neurodegenerative) diseases (CJD, Alzheimer`s, Parkinson`s, Huntington`s diseases, Atherosclerosis, Dia-betes mellitus type II, etc.)

amyloid plaques in the small intestine

(tau protein)

-protein

normal PrP pool

PrPSC nucleus

aggregation

the protein-only model for prion disease transmission

the disease can be transferred from one organism to another by trans-ferring the nucleus (mad cow disease outbrake in the 1990s in the UK, animals were fed with feed of infected animal origin)

A is derived from the cellular amyloid precursor protein (APP) through specific proteases; it is prone to form insoluble aggregates and its struc-ture by solid-state NMR spectroscopy showed extended parallel -sheet arrangements