Construction of molecular assemblies via docking: modeling of tetramers with D2 symmetry

Comparative modeling methods are commonly used to construct models of homologous proteins or oligomers. However, comparative modeling may be inapplicable when the number of subunits in a modeled oligomer is different than in the modeling template. Thus, a dimer cannot be a template for a tetramer because a new monomer-monomer interface must be predicted. We present in this study a new prediction approach, which combines protein-protein docking with either of two tetramer-forming algorithms designed to predict the structures of tetramers with D2 symmetry. Both algorithms impose symmetry constraints. However, one of them requires identification of two of the C2 dimers within the tetramer in the docking step, whereas the other, less demanding algorithm, requires identification of only one such dimer. Starting from the structure of one subunit, the procedures successfully reconstructed 16 known D2 tetramers, which crystallize with either a monomer, a dimer or a tetramer in the asymmetric unit. In some cases we obtained clusters of native-like tetramers that differ in the relative rotation of the two identical dimers within the tetramer. The predicted structural pliability for concanavalin-A, phosphofructokinase, and fructose-1,6-bisphosphatase agrees semiquantitatively with the observed differences between the several experimental structures of these tetramers. Hence, our procedure identifies a structural soft-mode that allows regulation via relative rigid-body movements of the dimers within these tetramers. The algorithm also predicted three nearly correct tetramers from model structures of single subunits, which were constructed by comparative modeling from subunits of homologous tetrameric, dimeric, or hexameric systems.     

Yeast glyceraldehyde-3-phosphate dehydrogenase. Evidence that subunit cooperativity in catalysis can be controlled by the formation of a complex with phosphoglycerate kinase

Sepharose-bound tetrameric, dimeric and monomeric forms of yeast glyceraldehyde-3-phosphate dehydrogenase were prepared, as well as immobilized hybrid species containing (by selective oxidation of an active center cysteine residue with H2O2) one inactivated subunit per tetramer or dimer. The catalytic properties of these enzyme forms were compared in the forward reaction (glyceraldehyde-3-phosphate oxidation) and reverse reaction (1,3-bisphosphoglycerate reductive dephosphorylation) under steady-state conditions. In the reaction of glyceraldehyde-3-phosphate oxidation, immobilized monomeric and tetrameric forms exhibited similar specific activities. The hybrid-modified dimer contributed on half of the total activity of a native dimer. The tetramer containing one modified subunit possessed 75% of the activity of an unmodified tetramer. In the reaction of 1,3-bisphosphoglycerate reductive dephosphorylation, the specific activity of the monomeric enzyme species was nearly twice as high as that of the tetramer, suggesting that only one-half of the active centers of the oligomer were acting simultaneously. Subunit cooperativity in catalysis persisted in an isolated dimeric species. The specific activity of a monomer associated with a peroxide-inactivated monomer in a dimer was equal to that of an isolated monomeric species and twice as high as that of a native immobilized dimer. The specific activity of subunits associated with a peroxide-inactivated subunit in a tetramer did not differ from that of a native immobilized tetramer; this indicates that interdimeric interactions are involved in catalytic subunit cooperativity. A complex was formed between the immobilized glyceraldehyde-3-phosphate dehydrogenase and soluble phosphoglycerate kinase. Three monomers of phosphoglycerate kinase were bound per tetramer of the dehydrogenase and one per dimer. Evidence is presented that if the reductive dephosphorylation of 1,3-bisphosphoglycerate proceeds in the phosphoglycerate kinase - glyceraldehyde-3-phosphate dehydrogenase complex, all active sites of the latter enzyme act independently, i.e. subunit cooperativity is abolished.     

Anabaena sp. DyP‐type peroxidase is a tetramer consisting of two asymmetric dimers

DyP‐type peroxidases are a newly discovered family of heme peroxidases distributed from prokaryotes to eukaryotes. Recently, using a structure‐based sequence alignment, we proposed the new classes, P, I and V, as substitutes for classes A, B, C, and D [Arch Biochem Biophys 2015;574:49–55]. Although many class V enzymes from eukaryotes have been characterized, only two from prokaryotes have been reported. Here, we show the crystal structure of one of these two enzymes, Anabaena sp. DyP‐type peroxidase (AnaPX). AnaPX is tetramer formed from Cys224‐Cys224 disulfide‐linked dimers. The tetramer of wild‐type AnaPX was stable at all salt concentrations tested. In contrast, the C224A mutant showed salt concentration‐dependent oligomeric states: in 600 mM NaCl, it maintained a tetrameric structure, whereas in the absence of salt, it dissociated into monomers, leading to a reduction in thermostability. Although the tetramer exhibits non‐crystallographic, 2‐fold symmetry in the asymmetric unit, two subunits forming the Cys224‐Cys224 disulfide‐linked dimer are related by 165° rotation. This asymmetry creates an opening to cavities facing the inside of the tetramer, providing a pathway for hydrogen peroxide access. Finally, a phylogenetic analysis using structure‐based sequence alignments showed that class V enzymes from prokaryotes, including AnaPX, are phylogenetically closely related to class V enzymes from eukaryotes. Proteins 2016; 84:31–42. © 2015 Wiley Periodicals, Inc.     

New insights into the half‐of‐the‐sites reactivity of human aldehyde dehydrogenase 1A1

Aldehyde dehydrogenases (ALDHs) couple the oxidation of aldehydes to the reduction of NAD(P)⁺. These enzymes have gained importance as they have been related to the detoxification of aldehydes generated in several diseases involving oxidative stress. It has been determined that tetrameric ALDHs work only with two of their four active sites (half‐of‐the‐sites reactivity), but the mechanistic reason for this feature remains unknown. In this study, tetrameric human aldehyde dehydrogenase class 1A1 (ALDH1A1) was dimerized to study the correlation of the oligomeric structure with the presence of half‐of‐the‐sites reactivity. Stable dimers from ALDH1A1 were generated by combining the mutation of two residues of the dimer–dimer interface in the tetramer (previously shown to render a low‐active and unstable enzyme) and the fusion of green fluorescent protein (GFP) in the C‐terminus of the mutant. Some kinetic properties of the GFP‐fusion mutant resembled those of human aldehyde dehydrogenase class 3A1, a native dimer, in that the fusion dimer did not show burst in the generation of nicotinamide adenine dinucleotide (NADH) and was less sensitive to the action of specific modulators. The presence of primary isotope effect indicated that the rate‐limiting step changed from NADH release to hydride transfer. The mutant showed higher activity with malondialdehyde and acrolein and was more resistant to inactivation by acrolein compared with the wild type. The mutant kinetic profile showed two hyperbolic components when the substrates were varied, suggesting the presence of two active sites with different affinities and catalytic capacities. In conclusion, the ALDH1A1–GFP dimeric mutant exhibits full site reactivity, suggesting that only the tetrameric structure induces the half‐of‐the‐sites reactivity. Proteins 2013; 81:1330–1339. © 2013 Wiley Periodicals, Inc.     

The endoplasmic reticulum HSP40 co‐chaperone ERdj3/DNAJB11 assembles and functions as a tetramer

ERdj3/DNAJB11 is an endoplasmic reticulum (ER)‐targeted HSP40 co‐chaperone that performs multifaceted functions involved in coordinating ER and extracellular proteostasis. Here, we show that ERdj3 assembles into a native tetramer that is distinct from the dimeric structure observed for other HSP40 co‐chaperones. An electron microscopy structural model of full‐length ERdj3 shows that these tetramers are arranged as a dimer of dimers formed by distinct inter‐subunit interactions involving ERdj3 domain II and domain III. Targeted deletion of residues 175‐190 within domain II renders ERdj3 a stable dimer that is folded and efficiently secreted from mammalian cells. This dimeric ERdj3 shows impaired substrate binding both in the ER and extracellular environments and reduced interactions with the ER HSP70 chaperone BiP. Furthermore, we show that overexpression of dimeric ERdj3 exacerbates ER stress‐dependent reductions in the secretion of a destabilized, aggregation‐prone protein and increases its accumulation as soluble oligomers in extracellular environments. These results reveal ERdj3 tetramerization as an important structural framework for ERdj3 functions involved in coordinating ER and extracellular proteostasis in the presence and absence of ER stress.     

Direct demonstration that homotetrameric chaperone SecB undergoes a dynamic dimer-tetramer equilibrium

We have shown here that the cytosolic bacterial chaperone SecB is a structural dimer of dimers that undergoes a dynamic equilibrium between dimer and tetramer in the native state. We demonstrated this equilibrium by mixing two tetrameric species of SecB that can be distinguished by size. We showed that the homotetrameric species exchanged dimers, because when the mixture was analyzed both by size exclusion chromatography and native polyacrylamide gel electrophoresis a third hybrid tetrameric species was detected. Furthermore, treatment of SecB with 5,5'-dithiobis-(2-nitrobenzoic acid), which modifies the sulfhydryl group on cysteines, caused irreversible dissociation to a dimer indicating that cysteine must be involved in the stabilizing interactions at the dimer interface. It is clear that the two dimer-dimer interfaces of the SecB tetramer are differentially stable. Dissociation at one interface allows for a dynamic dimer-tetramer equilibrium. Because only dimers were exchanged it is clear that the other interface between dimers is significantly more stable, otherwise oligomers should have formed with a random distribution of monomers.     

Crystal structures of carbohydrate recognition domain of blood dendritic cell antigen‐2 (BDCA2) reveal a common domain‐swapped dimer

We report on crystal structures of a carbohydrate recognition domain (CRD) of human C‐type lectin receptor blood dendritic cell antigen‐2 (BDCA2). Three different crystal forms were obtained at 1.8–2.3 Å resolution. In all three, the CRD has a basic C‐type lectin fold, but a long loop extends away from the core domain to form a domain‐swapped dimer. The structures of the dimers from the three different crystal forms superimpose well, indicating that domain swapping and dimer formation are energetically stable. The structure of the dimer is compared with other domain‐swapped proteins, and a possible regulation mechanism of BDCA2 is discussed. Proteins 2014; 82:1512–1518. © 2013 Wiley Periodicals, Inc.     

Domain swapping in allosteric modulation of DNA specificity

SgrAI is a type IIF restriction endonuclease that cuts an unusually long recognition sequence and exhibits allosteric self-modulation of cleavage activity and sequence specificity. Previous studies have shown that DNA bound dimers of SgrAI oligomerize into an activated form with higher DNA cleavage rates, although previously determined crystal structures of SgrAI bound to DNA show only the DNA bound dimer. A new crystal structure of the type II restriction endonuclease SgrAI bound to DNA and Ca(2+) is now presented, which shows the close association of two DNA bound SgrAI dimers. This tetrameric form is unlike those of the homologous enzymes Cfr10I and NgoMIV and is formed by the swapping of the amino-terminal 24 amino acid residues. Two mutations predicted to destabilize the swapped form of SgrAI, P27W and P27G, have been made and shown to eliminate both the oligomerization of the DNA bound SgrAI dimers as well as the allosteric stimulation of DNA cleavage by SgrAI. A mechanism involving domain swapping is proposed to explain the unusual allosteric properties of SgrAI via association of the domain swapped tetramer of SgrAI bound to DNA into higher order oligomers.     

Crystal structures reveal transient PERK luminal domain tetramerization in endoplasmic reticulum stress signaling

Stress caused by accumulation of misfolded proteins within the endoplasmic reticulum (ER) elicits a cellular unfolded protein response (UPR) aimed at maintaining protein‐folding capacity. PERK, a key upstream component, recognizes ER stress via its luminal sensor/transducer domain, but the molecular events that lead to UPR activation remain unclear. Here, we describe the crystal structures of mammalian PERK luminal domains captured in dimeric state as well as in a novel tetrameric state. Small angle X‐ray scattering analysis (SAXS) supports the existence of both crystal structures also in solution. The salient feature of the tetramer interface, a helix swapped between dimers, implies transient association. Moreover, interface mutations that disrupt tetramer formation in vitro reduce phosphorylation of PERK and its target eIF2α in cells. These results suggest that transient conversion from dimeric to tetrameric state may be a key regulatory step in UPR activation.     

Variation in assembly stoichiometry in non‐metazoan homologs of the hub domain of Ca2+/calmodulin‐dependent protein kinase II

The multi‐subunit Ca²⁺/calmodulin‐dependent protein kinase II (CaMKII) holoenzyme plays a critical role in animal learning and memory. The kinase domain of CaMKII is connected by a flexible linker to a C‐terminal hub domain that assembles into a 12‐ or 14‐subunit scaffold that displays the kinase domains around it. Studies on CaMKII suggest that the stoichiometry and dynamic assembly/disassembly of hub oligomers may be important for CaMKII regulation. Although CaMKII is a metazoan protein, genes encoding predicted CaMKII‐like hub domains, without associated kinase domains, are found in the genomes of some green plants and bacteria. We show that the hub domains encoded by three related green algae, Chlamydomonas reinhardtii, Volvox carteri f. nagarensis, and Gonium pectoral, assemble into 16‐, 18‐, and 20‐subunit oligomers, as assayed by native protein mass spectrometry. These are the largest known CaMKII hub domain assemblies. A crystal structure of the hub domain from C. reinhardtii reveals an 18‐subunit organization. We identified four intra‐subunit hydrogen bonds in the core of the fold that are present in the Chlamydomonas hub domain, but not in metazoan hubs. When six point mutations designed to recapitulate these hydrogen bonds were introduced into the human CaMKII‐α hub domain, the mutant protein formed assemblies with 14 and 16 subunits, instead of the normal 12‐ and 14‐subunit assemblies. Our results show that the stoichiometric balance of CaMKII hub assemblies can be shifted readily by small changes in sequence.     

Quaternary interactions in eye lens beta-crystallins: basic and acidic subunits of beta-crystallins favor heterologous association

beta-Crystallins are complex eye lens proteins made up of several related basic and acidic subunits that combine to form differently sized oligomers each displaying extensive polydispersity. As the sequences are homologous to the X-ray-determined bilobal structure of gamma-crystallin, beta-subunits are visualized as having a similar structure with additional N- and C-terminal extensions. Two basic (beta B2 and beta B3) and two acidic (beta A3 and beta A4) subunits have been isolated in deaggregating media, refolded, and reassociated in various combinations to determine which components favor dimers or higher oligomers. Homopolymers were compared with beta B2 homodimer in terms of charge, using Mono Q fast protein liquid chromatography, and size, using Superose 12 chromatography. Heterooligomeric formations were monitored by their intermediate charge properties compared with homooligomers. beta B2 associates with either beta B3- or beta A4-forming heterodimers whereas a larger oligomer is formed with beta A3. Naturally occurring beta-crystallin oligomers were analyzed by Mono Q chromatography and PhastGel electrophoresis. Whereas beta B2, beta B3, and beta A4 can each be reassociated to homodimers, beta A4 dimers are not found in native beta-crystallins. beta B2-beta A3 is a major component of intermediate-sized beta L1-crystallin and is absent from dimeric beta L2-crystallin. It is suggested that the pH dependence of the size of beta L1-crystallin is due to a dimer to tetramer equilibrium. By following dimer interactions using Superose 12 chromatography, beta B2-beta A4 was shown to interact with beta B2-beta A3. A model of beta-crystallin structure is proposed based on beta-subunits forming dimers with the next level of organization requiring an acidic subunit, beta A3, with a long N-terminal extension.     

High‐order oligomerization is required for the function of the H‐NS family member MvaT in Pseudomonas aeruginosa

H‐NS is an abundant DNA‐binding protein that has been implicated in the silencing of foreign DNA in several different bacteria. The ability of H‐NS dimers to form higher‐order oligomers is thought to aid the polymerization of the protein across AT‐rich stretches of DNA and facilitate gene silencing. Although the oligomerization of H‐NS from enteric bacteria has been the subject of intense investigation, little is known regarding the oligomerization of H‐NS family members from bacteria outside of the enterobacteriaceae, many of which share little sequence similarity with their enteric counterparts. Here we show that MvaT, a member of the H‐NS family of proteins from Pseudomonas aeruginosa, can form both dimers and higher‐order oligomers, and we identify a region within MvaT that mediates higher‐order oligomer formation. Using genetic assays we identify mutants of MvaT that are defective for higher‐order oligomer formation. We present evidence that these mutants are functionally impaired and exhibit DNA‐binding defects because of their inability to form higher‐order oligomers. Our findings support a model in which the ability of MvaT to bind efficiently to the DNA depends upon protein–protein interactions between MvaT dimers and suggest that the ability to form higher‐order oligomers is a conserved and essential feature of H‐NS family members.     

3D domain swapping: a mechanism for oligomer assembly.

3D domain swapping is a mechanism for forming oligomeric proteins from their monomers. In 3D domain swapping, one domain of a monomeric protein is replaced by the same domain from an identical protein chain. The result is an intertwined dimer or higher oligomer, with one domain of each subunit replaced by the identical domain from another subunit. The swapped "domain" can be as large as an entire tertiary globular domain, or as small as an alpha-helix or a strand of a beta-sheet. Examples of 3D domain swapping are reviewed that suggest domain swapping can serve as a mechanism for functional interconversion between monomers and oligomers, and that domain swapping may serve as a mechanism for evolution of some oligomeric proteins. Domain-swapped proteins present examples of a single protein chain folding into two distinct structures.     

Biophysical Characterization of the Dimer and Tetramer Interface Interactions of the Human Cytosolic Malic Enzyme

The cytosolic NADP⁺-dependent malic enzyme (c-NADP-ME) has a dimer-dimer quaternary structure in which the dimer interface associates more tightly than the tetramer interface. In this study, the urea-induced unfolding process of the c-NADP-ME interface mutants was monitored using fluorescence and circular dichroism spectroscopy, analytical ultracentrifugation and enzyme activities. Here, we demonstrate the differential protein stability between dimer and tetramer interface interactions of human c-NADP-ME. Our data clearly demonstrate that the protein stability of c-NADP-ME is affected predominantly by disruptions at the dimer interface rather than at the tetramer interface. First, during thermal stability experiments, the melting temperatures of the wild-type and tetramer interface mutants are 8–10°C higher than those of the dimer interface mutants. Second, during urea denaturation experiments, the thermodynamic parameters of the wild-type and tetramer interface mutants are almost identical. However, for the dimer interface mutants, the first transition of the urea unfolding curves shift towards a lower urea concentration, and the unfolding intermediate exist at a lower urea concentration. Third, for tetrameric WT c-NADP-ME, the enzyme is first dissociated from a tetramer to dimers before the 2 M urea treatment, and the dimers then dissociated into monomers before the 2.5 M urea treatment. With a dimeric tetramer interface mutant (H142A/D568A), the dimer completely dissociated into monomers after a 2.5 M urea treatment, while for a dimeric dimer interface mutant (H51A/D90A), the dimer completely dissociated into monomers after a 1.5 M urea treatment, indicating that the interactions of c-NADP-ME at the dimer interface are truly stronger than at the tetramer interface. Thus, this study provides a reasonable explanation for why malic enzymes need to assemble as a dimer of dimers.     

Self-assembly of apoferritin from horse spleen after reversible chemical modification with 2,3-dimethylmaleic anhydride

Apoferritin from horse spleen is composed of 24 subunits that undergo partial dissociation after chemical modification with 2,3-dimethylmaleic anhydride (DMMA), yielding dimeric, trimeric, and tetrameric intermediates, stable at pH 8.5 and 0 degrees C. Deacylation at neutral pH and elevated temperature provides a means to initiate reassembly by appropriate shifts of the solvent conditions. In order to monitor the pathway of self-assembly, starting from different intermediates of dissociation, dimers, trimers, and tetramers were isolated and investigated with respect to their capacity to accomplish reassociation. Intrinsic protein fluorescence, gel permeation chromatography, and analytical ultracentrifugation were applied to characterize the intermediate and final stages of association. The assembly of both the dimer and trimer yields greater than 85% of the native tetracosamer; the overall rate, starting from the dimer, exceeds the one starting from the trimer. Under comparable conditions, the tetramer exhibits only partial reassociation via the dimer and monomer; the corresponding dissociation reaction determines the observed slower rate. Significant assembly intermediates are "structured monomers", dimers, trimers, and dodecamers. Polymerization of the dimer via the tetramer, octamer, etc., does not occur on the pathway of assembly. The results confirm the assembly scheme proposed previously on the basis of cross-linking and spectroscopic experiments [Gerl, M., & Jaenicke, R. (1987) Eur. Biophys. J. 15, 103-109]. Comparison of structural models involving the different subunit interactions responsible for the sequential association supports the monomer----dimer----trimer----hexamer----dodecamer----tetracosamer mechanism of apoferritin self-assembly.     

Conversion of bovine cardiac adenosine cyclic 3',5'-phosphate dependent protein kinase to a heterodimer by removal of 45 residues at the N-terminus of the regulatory subunit

The type II adenosine cyclic 3',5'-phosphate (cAMP) dependent protein kinase from bovine heart, consisting of a dimeric regulatory subunit and two catalytic subunits, was converted to a heterodimer by limited tryptic digestion. Loss of the tetrameric structure was accompanied by proteolysis of the regulatory subunit to a form with an apparent molecular weight of 45 000 vs. 52 000 for the native subunit. The proteolyzed subunit behaved as a monomer, in contrast to the dimeric native subunit. Amino acid sequence analysis established that proteolysis removed 45 residues at the N-terminus, indicating that these 45 residues constitute the dimerizing domain of this protein. The kinetic properties of this heterodimer were indistinguishable from those of the native tetramer: half-maximal kinase activation occurred at 48 nM cAMP with a Hill coefficient of 1.45, the regulatory subunit bound 1.5 equiv of cAMP with half-maximal binding occurring at 33 nM, and kinetics for dissociation of bound cAMP were biphasic, indicating the presence of two different binding sites. These observations suggest that residues 1-45 function only in the formation of dimers and that dimerization has little influence on other functional properties of the regulatory subunit. More extensive proteolysis cleaved the monomeric fragment at Lys-311. The fragments resulting from this second cleavage did not dissociate, and the complex inhibited the catalytic subunit in a cAMP-dependent manner.     

Regulation of the aggregation state of maize phosphoenolpyruvate carboxylase: evidence from dynamic light-scattering measurements

The molecular weights of different aggregational states of phosphoenolpyruvate carboxylase purified from the leaves of Zea mays have been determined by measurement of the molecular diameter using a Malvern dynamic light scattering spectrometer. Using these data to identify the monomer, dimer, tetramer, and larger aggregate(s) the effect of pH and various ligands on the aggregational equilibria of this enzyme have been determined. At neutral pH the enzyme favored the tetrameric form. At both low and high pH the tetramer dissociated, followed by aggregation to a "large" inactive form. The order of dissociation at least at low pH appeared to be two-step: from tetramer to dimers followed by dimer to monomers. The monomers then aggregate to a large aggregate, which is inactive. The presence of EDTA at pH 8 protected the enzyme against both inactivation and large aggregate formation. Dilution of the enzyme at pH 7 at room temperature results in driving the equilibrium from tetramer to dimer. The presence of malate with EDTA stabilizes the dimer as the predominant form at low protein concentrations. The presence of the substrate phosphoenolpyruvate alone and with magnesium and bicarbonate induced formation of the tetramer, and decreased the dissociation constant (Kd) of the tetrameric form. The inhibitor malate, however, induced dissociation of the tetramer as evidenced by an increase in the Kd of the tetramer.     

Distinguishing between functional monomeric and oligomeric complexes of the Ca,Mg-ATPase in sarcoplasmic reticulum

Techniques are described for using blocking agents to distinguish between enzymes which are functional monomers and oligomers. To achieve this distinction the blocking agent must react exclusively at the active site with a stoichiometry of one mole of site per mole enzyme. The effect of the blocking agent on enzymatic activity in oligomers of n = 2 and 4 are described and the optimal degree of blocking is considered for tests of enzyme activity at saturating and less than saturating substrate concentrations. For saturating concentrations and a dimer the distinction between dimer and monomer is best observed with 50 per cent of sites blocked. For a tetramer the distinction is best made at higher degrees of blockade. The use of saturating substrate concentrations is thus limited to small oligomers. If nonsaturating substrate concentrations are used and normalized double reciprocal plots of the dependence of enzyme activity on substrate concentrations are made then the distinction between monomer and oligomer can readily be made for dimers, tetramers, and higher n-mers. The principles developed to distinguished monomeric from oligomeric enzymes are applied to published data obtained with the Ca Mg-ATPase of sarcoplasmic reticulum. Fluorescein isothiocyanate is the blocking agent. Plots of the published data support both the monomeric and tetrameric models for allosteric regulation with the preponderance of the data supporting the monomeric model.     

The reconstitution of denatured phosphoglycerate mutase

The reconstitution of the tetrameric enzyme yeast phosphoglycerate mutase after denaturation in guanidine hydrochloride has been studied. Denaturation is almost completely reversible at enzyme concentrations greater than 10 micrograms/ml. Cross-linking by glutaraldehyde has been used to monitor the reassociation of the subunits; the kinetics of this process has been analyzed in terms of a model involving an equilibrium between monomer and dimer followed by a bimolecular association of two dimers to give a tetramer. Reactivation is found to parallel the appearance of tetramer. Structural changes during reconstitution have been measured by circular dichroism and fluorescence. Both methods reveal complex kinetics indicating the rapid formation of structured monomers (half-time less than 10 s), followed by slow subunit association. For comparison, preliminary reconstitution experiments were performed on the dimeric phosphoglycerate mutase from rabbit muscle.