ALBERT

Notes: Abstract The acquisition of the native three-dimensional structure of proteins consists of sequential folding reactions with well-populated and well-defined structural intermediates. For small proteins successive stages in the folding have been resolved kinetically; these suggest that H-bonded elements of secondary structure are formed first, followed by folding steps to generate the complete tertiary structure. The rate determining step in the folding of a number of small proteins has been shown to be proline cis ⇌ tram isomerization. As indicated by experiments using fast kinetics the overall folding mechanism, even in a small single-domain molecule like ribonuclease, involves more than one intermediate. Large protein molecules contain domains which may fold independently. For multi-domain proteins, the pathway of folding therefore involves “folding by parts”, followed by merging of folded domains. In the case of assembly systems (e.g., oligomeric or multimeric enzymes) folding and association have to be subtly interconnected with respect to the time scale, since the correct assembly of subunits requires their proper folding. In this sense the initial function of oligomeric proteins is their own self-assembly. The corresponding mechanism underlying the spontaneous formation of the native quaternary structure of oligomeric proteins must be the consecutive folding and association of the constituent polypeptide chains. Equilibrium and kinetic studies have been concerned with a number of dimeric, tetrameric and multimeric enzymes, using enzymatic activity to measure structure formation: alcohol dehydrogenase, aldolase, glyceraldehyde-3-phosphate dehydrogenase, lactic dehydrogenase, malic dehydrogenase, pyruvate dehydrogenase, triose phosphate isomerase, tryptophan synthase. These experiments make use of the reversibility of protein denaturation, focusing on refolding and reassociation rather than folding and association, because there is no direct approach to structural investigations of the nascent polypeptide chain in vivo. Optimum conditions of reconstitution yield up to 100% reactivation. After separation of “irreversibly denatured protein”, reconstituted and native protein turn out to be indistinguishable. The major side reaction leading to “wrong aggregation” is due to competition between folding and association. Due to the high specificity of the association reaction “chimeric” species are not observed, and multimeric systems containing different component enzymes show specific assembly. The kinetics of reconstitution generally obey an irreversible sequential first- order/second-order mechanism involving inactive monomers; only in the case of aldolase is subunit activity suggested. For a number of oligomeric enzymes renaturation from various denaturants, in the absence or presence of coenzyme is characterized by identical kinetics. For glyceraldehyde-3-phosphate dehydrogenase, however, free NAD as well as a covalently bound NAD-analog are found to enhance the reconstitution. In the case of assembly structures exceeding the dimer, the observed consecutive folding/association mechanism does not allow us to decide whether the observed second order processes belong to the formation of the dimer or tetramer. Chemical cross-linking and hybridization techniques allow the equilibrium state and the assembly kinetics of oligomeric systems to be analyzed quantitatively. Using this method, e.g., for lactic dehydrogenase, it is obvious that dissociation leads to the homogeneous monomer, while tetramer formation is found to parallel reactivation. In general, equilibrium and kinetic experiments prove that full enzymatic activity requires association. In the case of multisubunit enzymes (multienzyme complexes) heterologous interactions of the component enzymes seem to be involved in the rate determining (first order) “reshuffling” processes which generate catalytic activity in the overall enzymatic reaction.