Heterodimer formation between superoxide dismutase and its copper chaperone

Copper, zinc superoxide dismutase (SOD1) is activated in vivo by the copper chaperone for superoxide dismutase (CCS). The molecular mechanisms by which CCS recognizes and docks with SOD1 for metal ion insertion are not well understood. Two models for the oligomerization state during copper transfer have been proposed: a heterodimer comprising one monomer of CCS and one monomer of SOD1 and a dimer of dimers involving interactions between the two homodimers. We have investigated protein-protein complex formation between copper-loaded and apo yeast CCS (yCCS) and yeast SOD1 for both wild-type SOD1 (wtSOD1) and a mutant SOD1 in which copper ligand His 48 has been replaced with phenylalanine (H48F-SOD1). According to gel filtration chromatography, dynamic light scattering, analytical ultracentrifugation, and chemical cross-linking experiments, yCCS and this mutant SOD1 form a complex with the correct molecular mass for a heterodimer. No higher order oligomers were detected. Heterodimer formation is facilitated by the presence of zinc but does not depend on copper loading of yCCS. The complex formed with H48F-SOD1 is more stable than that formed with wtSOD1, suggesting that the latter is a more transient species. Notably, heterodimer formation between copper-loaded yCCS and wtSOD1 is accompanied by SOD1 activation only in the presence of zinc. These findings, taken together with structural, biochemical, and genetic studies, strongly suggest that in vivo copper loading of yeast SOD1 occurs via a heterodimeric intermediate.     

Coexpression of yeast copper chaperone (yCCS) and CuZn-superoxide dismutases in Escherichia coli yields protein with high copper contents

To fully understand the function of the Cu- and Zn-containing superoxide dismutases in normal and disordered cells, it is essential to study protein variants with full metal contents. We describe the use of an Escherichia coli-based expression system for the overproduction of human intracellular wild type CuZn-superoxide dismutase (SOD), the CuZnSOD variant F50E/G51E (monomeric), two amyotrophic lateral sclerosis-related mutant CuZnSOD variants (D90A and G93A), and PseudoEC-SOD, all with high Cu contents. This system is based on coexpression of the SOD variants with the yeast copper chaperone yCCS during growth in a medium supplemented with Cu(2+) and Zn(2+). The recombinant SOD enzymes were all found in the cytosol and represented 30-50% of the total bacterial protein. The enzymes were purified to homogeneity and active enzymes were obtained in high yield. The resulting proteins were characterized through immunochemical reactivity and specific activity analyses, in conjunction with mass-, photo-, and atomic absorption-spectroscopy.     

Heterodimeric structure of superoxide dismutase in complex with its metallochaperone

The copper chaperone for superoxide dismutase (CCS) activates the eukaryotic antioxidant enzyme copper, zinc superoxide dismutase (SOD1). The 2.9 A resolution structure of yeast SOD1 complexed with yeast CCS (yCCS) reveals that SOD1 interacts with its metallochaperone to form a complex comprising one monomer of each protein. The heterodimer interface is remarkably similar to the SOD1 and yCCS homodimer interfaces. Striking conformational rearrangements are observed in both the chaperone and target enzyme upon complex formation, and the functionally essential C-terminal domain of yCCS is well positioned to play a key role in the metal ion transfer mechanism. This domain is linked to SOD1 by an intermolecular disulfide bond that may facilitate or regulate copper delivery.     

Metal ion reconstitution studies of yeast copper-zinc superoxide dismutase: the "phantom" subunit and the possible role of Lys7p

 Using a corrected molar extinction coefficient for yeast apo copper-zinc superoxide dismutase (CuZnSOD), we have confirmed that the metal binding properties of this protein in vitro differ greatly from those of the bovine and human CuZnSOD enzymes. Thus yeast apo CuZnSOD was found to bind only one Co²⁺ per protein dimer under the conditions in which the bovine and human CuZnSOD apoenzymes readily bind two per dimer. The spectroscopic properties characteristic of the two Cu²⁺ plus two Co²⁺ per dimer or four Cu²⁺ per dimer metal-substituted bovine apo CuZnSOD derivatives were obtained for the yeast apoprotein but by the addition of only half of the appropriate metals, i.e., one Cu²⁺ plus one Co²⁺ per dimer or two Cu²⁺ per dimer. This half-metallated yeast CuZnSOD has been characterized by UV-visible and EPR spectroscopy as well as by native polyacrylamide gel electrophoresis. We conclude that yeast apo CuZnSOD, unlike the bovine and human apoproteins, cannot be reconstituted fully with metal ions under the same conditions. Instead, only one subunit of the homodimer, the "normal" subunit, can be remetalled in a fashion reminiscent of the well-characterized bovine protein. The other "phantom" subunit is not competent to bind metals in this fashion. Furthermore, we have shown that CuZnSOD protein isolated from Saccharomyces cerevisiae that lacks the gene coding for the copper chaperone, Lys7p, contains only one metal ion, Zn²⁺, per protein dimer. The possibility that yeast CuZnSOD can exist in multiple conformational states may represent an increased propensity of the yeast protein to undergo changes that can occur in all CuZnSODs, and may have implications for amyotrophic lateral sclerosis. Received: 8 June 1998 / Accepted: 9 September 1998     

Crystal structure of the second domain of the human copper chaperone for superoxide dismutase

The human copper chaperone for superoxide dismutase (hCCS) delivers the essential copper ion cofactor to copper,zinc superoxide dismutase (SOD1), a key enzyme in antioxidant defense. Mutations in SOD1 are linked to familial amyotrophic lateral sclerosis (FALS), a fatal neurodegenerative disorder. The molecular mechanisms by which SOD1 is recognized and activated by hCCS are not understood. To better understand this biochemical pathway, we have determined the X-ray structure of the largest domain of hCCS (hCCS Domain II) to 2. 75 A resolution. The overall structure is closely related to that of its target enzyme SOD1, consisting of an eight-stranded beta-barrel and a zinc-binding site formed by two extended loops. The first of these loops provides the ligands to a bound zinc ion, and is analogous to the zinc subloop in SOD1. The second structurally resembles the SOD1 electrostatic channel loop, but lacks many of the residues important for catalysis. Like SOD1 and yCCS, hCCS forms a dimer using a highly conserved interface. In contrast to SOD1, however, the hCCS structure does not contain a copper ion bound in the catalytic site. Notably, the structure reveals a single loop proximal to the dimer interface which is unique to the CCS chaperones.     

Loss of SOD1 and LYS7 sensitizes Saccharomyces cerevisiae to hydroxyurea and DNA damage agents and downregulates MEC1 pathway effectors

Aerobic metabolism produces reactive oxygen species, including superoxide anions, which cause DNA damage unless removed by scavengers such as superoxide dismutases. We show that loss of the Cu,Zn-dependent superoxide dismutase, SOD1, or its copper chaperone, LYS7, confers oxygen-dependent sensitivity to replication arrest and DNA damage in Saccharomyces cerevisiae. We also find that sod1Delta strains, and to a lesser extent lys7Delta strains, when arrested with hydroxyurea (HU) show reduced induction of the MEC1 pathway effector Rnr3p and of Hug1p. The HU sensitivity of sod1Delta and lys7Delta strains is suppressed by overexpression of TKL1, a transketolase that generates NADPH, which balances redox in the cell and is required for ribonucleotide reductase activity. Our results suggest that the MEC1 pathway in sod1Delta mutant strains is sensitive to the altered cellular redox state due to increased superoxide anions and establish a new relationship between SOD1, LYS7, and the MEC1-mediated checkpoint response to replication arrest and DNA damage in S. cerevisiae.     

Copper activation of superoxide dismutase 1 (SOD1) in vivo. Role for protein-protein interactions with the copper chaperone for SOD1

Insertion of copper into superoxide dismutase 1 (SOD1) in vivo requires the copper chaperone for SOD1 (CCS). CCS encompasses three protein domains: copper binding Domains I and III at the amino and carboxyl termini, and a central Domain II homologous to SOD1. Using a yeast interaction mating system, yeast CCS was seen to physically interact with SOD1, and this interaction required sequences at the predicted dimer interface of CCS Domain II. Interactions with SOD1 also required sequences of Domain III, but not Domain I. Mutations were introduced at the dimer interface of yeast SOD1, and the corresponding mutant failed to interact with CCS. When loaded with copper independent of CCS, this mutant SOD1 exhibited superoxide scavenging activity, but was normally inactive in vivo because CCS failed to recognize the enzyme. Activation of SOD1 by CCS was also examined using an in vivo assay for copper incorporation into SOD1. Yeast CCS was observed to insert copper into a pre-existing pool of apoSOD1 without the need for new SOD1 synthesis or for protein unfolding by the major SSA cytosolic heat shock proteins. Our data are consistent with a model in which prefolded dimers of apoSOD1 serve as substrate for the CCS copper chaperone.     

The structural plasticity of the human copper chaperone for SOD1: insights from combined size-exclusion chromatographic and solution X-ray scattering studies

The incorporation of copper into biological macromolecules such as SOD1 (Cu,Zn superoxide dismutase) is essential for the viability of most organisms. However, copper is toxic and therefore the intracellular free copper concentration is kept to an absolute minimum. Several proteins, termed metallochaperones, are charged with the responsibility of delivering copper from membrane transporters to its intracellular destination. The CCS (copper chaperone for SOD1) is the major pathway for SOD1 copper loading. We have determined the first solution structure of hCCS (human CCS) by SAXS (small-angle X-ray scattering) in conjunction with SEC (size-exclusion chromatography). The findings of the present study highlight the importance of this combined on-line chromatographic technology with SAXS, which has allowed us to unambiguously separate the hCCS dimer from other oligomeric and non-physiological aggregated states that would otherwise adversely effect measurements performed on bulk solutions. The present study exposes the dynamic molecular conformation of this multi-domain chaperone in solution. The metal-binding domains known to be responsible for the conveyance of copper to SOD1 can be found in positions that would expedite this movement. Domains I and III of a single hCCS monomer are able to interact and can also move into positions that would facilitate initial copper binding and ultimately transfer to SOD1. Conversely, the interpretation of our solution studies is not compatible with an interaction between these domains and their counterparts in an hCCS dimer. Overall, the results of the present study reveal the plasticity of this multi-domain chaperone in solution and are consistent with an indispensable flexibility necessary for executing its dual functions of metal binding and transfer.     

Crystal structure of the copper chaperone for superoxide dismutase.

Cellular systems for handling transition metal ions have been identified, but little is known about the structure and function of the specific trafficking proteins. The 1.8 A resolution structure of the yeast copper chaperone for superoxide dismutase (yCCS) reveals a protein composed of two domains. The N-terminal domain is very similar to the metallochaperone protein Atx1 and is likely to play a role in copper delivery and/or uptake. The second domain resembles the physiological target of yCCS, superoxide dismutase I (SOD1), in overall fold, but lacks all of the structural elements involved in catalysis. In the crystal, two SOD1-like domains interact to form a dimer. The subunit interface is remarkably similar to that in SOD1, suggesting a structural basis for target recognition by this metallochaperone.     

The crystal structure of the C-terminal fragment of yeast Hsp40 Ydj1 reveals novel dimerization motif for Hsp40

The molecular chaperone Hsp40 functions as a dimer. The dimer formation is critical for Hsp40 molecular chaperone activity to facilitate Hsp70 to refold non-native polypeptides. We have determined the crystal structure of the C-terminal fragment of yeast Hsp40 Ydj1 that is responsible for Ydj1 dimerization by MAD method. The C-terminal fragment of Ydj1 comprises of the domain III of Ydj1 and the Ydj1 C-terminal dimerization motif. The crystal structure indicates that the dimerization motif of type I Hsp40 Ydj1 differs significantly from that of yeast type II Hsp40. The C terminus of type I Hsp40 Ydj1 from one monomer forms beta-strands with the domain III from the other monomer in the homo-dimer. The L372 from Ydj1 C terminus inserts its side-chain into a hydrophobic pocket on domain III. The modeled full-length Ydj1 dimer structure reveals that a large cleft is formed between the two monomers. The domain IIs of Ydj1 monomers that contain the zinc-finger motifs points directly against each other.     

Copper stabilizes a heterodimer of the yCCS metallochaperone and its target superoxide dismutase

The copper chaperone for superoxide dismutase (CCS) activates the antioxidant enzyme Cu,Zn-SOD (SOD1) by directly inserting the copper cofactor into the apo form of SOD1. Neither the mechanism of protein-protein recognition nor of metal transfer is clear. The metal transfer step has been proposed to occur within a transient copper donor/acceptor complex that is either a heterodimer or heterotetramer (i.e. a dimer of dimers). To determine the nature of this intermediate, we generated a mutant form of SOD1 by replacing a copper binding residue His-48 with phenylalanine. This protein cannot accept copper from CCS but does form a stable complex with apo- and Cu-CCS, as observed by immunoprecipitation and native gel electrophoresis. Fluorescence anisotropy measurements corroborate the formation of this species and further indicate that copper enhances the stability of the dimer by an order of magnitude. The copper form of the heterodimer was isolated by gel filtration chromatography and contains one copper and one zinc atom per heterodimer. These results support a mechanism for copper transfer in which CCS and SOD1 dock via their highly conserved dimer interfaces in a manner that precisely orients the Cys-rich copper donor sites of CCS and the His-rich acceptor sites of SOD1 to form a copper-bridged intermediate.     

A multinuclear copper(I) cluster forms the dimerization interface in copper-loaded human copper chaperone for superoxide dismutase

Copper binding and X-ray aborption spectroscopy studies are reported on untagged human CCS (hCCS; CCS = copper chaperone for superoxide dismutase) isolated using an intein self-cleaving vector and on single and double Cys to Ala mutants of the hCCS MTCQSC and CSC motifs of domains 1 (D1) and 3 (D3), respectively. The results on the wild-type protein confirmed earlier findings on the CCS-MBP (maltose binding protein) constructs, namely, that Cu(I) coordinates to the CXC motif, forming a cluster at the interface of two D3 polypeptides. In contrast to the single Cys to Ser mutations of the CCS-MBP protein (Stasser, J. P., Eisses, J. F., Barry, A. N., Kaplan, J. H., and Blackburn, N. J. (2005) Biochemistry 44, 3143-3152), single Cys to Ala mutations in D3 were sufficient to eliminate cluster formation and significantly reduce CCS activity. Analysis of the intensity of the Cu-Cu cluster interaction in C244A, C246A, and C244/246A variants suggested that the nuclearity of the cluster was greater than 2 and was most consistent with a Cu4S6 adamantane-type species. The relationship among cluster formation, oligomerization, and metal loading was evaluated. The results support a model in which Cu(I) binding converts the apo dimer with a D2-D2 interface to a new dimer connected by cluster formation at two D3 CSC motifs. The predominance of dimer over tetramer in the cluster-containing species strongly suggests that the D2 dimer interface remains open and available for sequestering an SOD1 monomer. This work implicates the copper cluster in the reactive form and adds detail to the cluster nuclearity and how copper loading affects the oligomerization states and reactivity of CCS for its partner SOD1.     

Structure and dimerization of translation initiation factor aIF5B in solution

Translation initiation factor 5B (IF5B) is required for initiation of protein synthesis. The solution structure of archaeal IF5B (aIF5B) was analysed by small-angle X-ray scattering (SAXS) and dynamic light scattering (DLS) and was indicated to be in both monomeric and dimeric form. Sedimentation equilibrium (SE) analytical ultracentrifugation (AUC) of aIF5B indicated that aIF5B forms irreversible dimers in solution but only to a maximum of 5.0-6.8% dimer. Sedimentation velocity (SV) AUC at higher speed also indicated the presence of two species, and the sedimentation coefficients s(20,w)(0) were determined to be 3.64 and 5.51±0.29 S for monomer and dimer, respectively. The atomic resolution (crystallographic) structure of aIF5B (Roll-Mecak et al. [6]) was used to model monomer and dimer, and theoretical sedimentation coefficients for these models were computed (3.89 and 5.63 S, respectively) in good agreement with the sedimentation coefficients obtained from SV analysis. Thus, the structure of aIF5B in solution must be very similar to the atomic resolution structure of aIF5B. SAXS data were acquired in the same buffer with the addition of 2% glycerol to inhibit dimerization, and the resultant monomeric aIF5B in solution did indeed adopt a structure very similar to the one reported earlier for the protein in crystalline form. The p(r) function indicated an elongated conformation supported by a radius of gyration of 37.5±0.2 ? and a maximum dimension of ~130 ?. The effects of glycerol on the formation of dimers are discussed. This new model of aIF5B in solution shows that there are universal structural differences between aIF5B and the homologous protein IF2 from Escherichia coli.     

A copper chaperone for superoxide dismutase that confers three types of copper/zinc superoxide dismutase activity in Arabidopsis

The copper chaperone for superoxide dismutase (CCS) has been identified as a key factor integrating copper into copper/zinc superoxide dismutase (CuZnSOD) in yeast (Saccharomyces cerevisiae) and mammals. In Arabidopsis (Arabidopsis thaliana), only one putative CCS gene (AtCCS, At1g12520) has been identified. The predicted AtCCS polypeptide contains three distinct domains: a central domain, flanked by an ATX1-like domain, and a C-terminal domain. The ATX1-like and C-terminal domains contain putative copper-binding motifs. We have investigated the function of this putative AtCCS gene and shown that a cDNA encoding the open reading frame predicted by The Arabidopsis Information Resource complemented only the cytosolic and peroxisomal CuZnSOD activities in the Atccs knockout mutant, which has lost all CuZnSOD activities. However, a longer AtCCS cDNA, as predicted by the Munich Information Centre for Protein Sequences and encoding an extra 66 amino acids at the N terminus, could restore all three, including the chloroplastic CuZnSOD activities in the Atccs mutant. The extra 66 amino acids were shown to direct the import of AtCCS into chloroplasts. Our results indicated that one AtCCS gene was responsible for the activation of all three types of CuZnSOD activity. In addition, a truncated AtCCS, containing only the central and C-terminal domains without the ATX1-like domain failed to restore any CuZnSOD activity in the Atccs mutant. This result indicates that the ATX1-like domain is essential for the copper chaperone function of AtCCS in planta.     

AtCCS is a functional homolog of the yeast copper chaperone Ccs1/Lys7

In plant chloroplasts two superoxide dismutase (SOD) activities occur, FeSOD and Cu/ZnSOD, with reciprocal regulation in response to copper availability. This system presents a unique model to study the regulation of metal-cofactor delivery to an organelle. The Arabidopsis thaliana gene AtCCS encodes a functional homolog to yeast Ccs1p/Lys7p, a copper chaperone for SOD. The AtCCS protein was localized to chloroplasts where it may supply copper to the stromal Cu/ZnSOD. AtCCS mRNA expression levels are upregulated in response to Cu-feeding and senescence. We propose that AtCCS expression is regulated to allow the most optimal use of Cu for photosynthesis.     

A molecular dynamics study of the C-terminal fragment of the L7/L12 ribosomal protein

The crystallographic dimer of the C-terminal fragment (CTF) of the L7/L12 ribosomal protein has been subjected to molecular dynamics (MD) simulations. A 90 picosecond (ps) trajectory for the protein dimer, 19 water molecules and two counter ions has been calculated at constant temperature. Effects of intermolecular interactions on the structure and dynamics have been studied. The exact crystallographic symmetry is lost and the atomic fluctuations differ from one monomer to the other. The average MD structure is more stable than the X-ray one, as judged by accessible surface area and energy calculations. Crystal (non-dimeric) interactions have been simulated in another 40 ps trajectory by using harmonic restraints to represent intermolecular hydrogen bonds. The conformational changes with respect ot the X-ray structure are then virtually suppressed.The unrestrained dimer trajectory has been scanned for cooperative motions involving secondary structure elements. The intrinsic collective motions of the monomer are transmitted via intermolecular contacts to the dimer structure.The existence of a stable dimeric form of CTF, resembling the crystallographic one, has been documented. At the cost of fairly small energy expenditure the dimer has considerable conformational flexibility. This flexibility may endow the dimer with some functional potential as an energy transducer.     

Metal-binding stoichiometry and selectivity of the copper chaperone CopZ from Enterococcus hirae.

We studied the interaction of several metal ions with the copper chaperone from Enterococcus hirae (EhCopZ). We show that the stoichiometry of the protein-metal complex varies with the experimental conditions used. At high concentration of the protein in a noncoordinating buffer, a dimer, (EhCopZ)2-metal, was formed. The presence of a potentially coordinating molecule L in the solution leads to the formation of a monomeric ternary complex, EhCopZ-Cu-L, where L can be a buffer or a coordinating molecule (glutathione, tris(2-carboxyethyl)phosphine). This was demonstrated in the presence of glutathione by electrospray ionization MS. The presence of a tyrosine close to the metal-binding site allowed us to follow the binding of cadmium to EhCopZ by fluorescence spectroscopy and to determine the corresponding dissociation constant (Kd = 30 nm). Competition experiments were performed with mercury, copper and cobalt, and the corresponding dissociation constants were calculated. A high preference for copper was found, with an upper limit for the dissociation constant of 10-12 m. These results confirm the capacity of EhCopZ to bind copper at very low concentrations in living cells and may provide new clues in the determination of the mechanism of the uptake and transport of copper by the chaperone EhCopZ.     

Oligomeric state of lipocalin-1 (LCN1) by multiangle laser light scattering and fluorescence anisotropy decay

Multiangle laser light scattering and fluorescence anisotropy decay measurements clarified the oligomeric states of native and recombinant tear lipocalin (lipocalin-1, TL). Native TL is monomeric. Recombinant TL (5-68 microM) with or without the histidine tag shows less than 7% dimer formation that is not in equilibrium with the monomeric form. Fluorescence anisotropy decay showed a correlation time of 9-10 ns for TL (10 microM-1 mM). Hydrodynamic calculations based on the crystallographic structure of a monomeric TL mutant closely concur with the observed correlation time. The solution properties calculated with HYDROPRO and SOLPRO programs from the available crystallographic structure of a monomeric TL mutant concur closely with the observed fluorescence anisotropy decay. The resulting model shows that protein topology is the major determinant of rotational correlation time and accounts for deviation from the Stokes-Einstein relation. The data challenge previous gel filtration studies to show that native TL exists predominantly as a monomer in solution rather than as a dimer. Delipidation of TL results in a formation of a complex oligomeric state (up to 25%). These findings are important as the dynamic processes in the tear film are limited by diffusional, translational as well as rotational, properties of the protein.     

Molecular cloning and characterization of a copper chaperone for copper/zinc superoxide dismutase from the rat

Copper chaperone is an essential cytosolic factor that maintains copper homeostasis in living cells. Cytosolic metallochaperones have been recently identified in plant, yeast, rodents, and human cells. During our investigation, we found a new member of the copper chaperone family for copper/zinc superoxide dismutase, which was cloned from rats. The new copper chaperone was named rCCS (rat Copper Chaperone for Superoxide dismutase). The cDNA of rCCS was found to have a length of 1094 bp, and the protein analyzed from the cDNA was deduced to contain 274 amino acids. The amino acid sequence of rCCS consists of three domains: A metal binding domain, which has a MXCXXC motif in domain I, a homolog of the Cu/Zn SOD in domain II, and a CXC motif in domain III. The binding of rCCS to Cu/Zn SOD was analyzed by GST column binding assay, and the domain II of rCCS was found to be essential for binding to Cu/Zn SOD, which in turn activates Cu/Zn SOD.     

Dissociation of human copper-zinc superoxide dismutase dimers using chaotrope and reductant. Insights into the molecular basis for dimer stability

The dissociation of apo- and metal-bound human copper-zinc superoxide dismutase (SOD1) dimers induced by the chaotrope guanidine hydrochloride (GdnHCl) or the reductant Tris(2-carboxyethyl)phosphine (TCEP) has been analyzed using analytical ultracentrifugation. Global fitting of sedimentation equilibrium data under native solution conditions (without GdnHCl or TCEP) demonstrate that both the apo- and metal-bound forms of SOD1 are stable dimers. Sedimentation velocity experiments show that apo-SOD1 dimers dissociate cooperatively over the range 0.5-1.0 M GdnHCl. In contrast, metal-bound SOD1 dimers possess a more compact shape and dissociate at significantly higher GdnHCl concentrations (2.0-3.0 M). Reduction of the intrasubunit disulfide bond within each SOD1 subunit by 5-10 mM TCEP promotes dissociation of apo-SOD1 dimers, whereas the metal-bound enzyme remains a stable dimer under these conditions. The Cys-57 --> Ser mutant of SOD1, a protein incapable of forming the intrasubunit disulfide bond, sediments as a monomer in the absence of metal ions and as a dimer when metals are bound. Taken together, these data indicate that the stability imparted to the human SOD1 dimer by metal binding and the formation of the intrasubunit disulfide bond are mediated by independent molecular mechanisms. By combining the sedimentation data with previous crystallographic results, a molecular explanation is provided for the existence of different SOD1 macromolecular shapes and multiple SOD1 dimeric species with different stabilities.