An important issue in modern protein biophysics is whether structurally homologous proteins share common stability and/or folding features. Flavodoxin is an archetypal alpha/beta protein organized in three layers: a central beta-sheet (strand order 21345) flanked by helices 1 and 5 on one side and helices 2, 3, and 4 on the opposite side. The backbone internal dynamics of the apoflavodoxin from Anabaena is analyzed here by the hydrogen exchange method. The hydrogen exchange rates indicate that 46 amide protons, distributed throughout the structure of apoflavodoxin, exchange relatively slowly at pH 7.0 (k(ex) < 10(-1) min(-1)). According to their distribution in the structure, protein stability is highest on the beta-sheet, helix 4, and on the layer formed by helices 1 and 5. The exchange kinetics of Anabaena apoflavodoxin was compared with those of the apoflavodoxin from Azotobacter, with which it shares a 48% sequence identity, and with Che Y and cutinase, two other alpha/beta (21345) proteins with no significant sequence homology with flavodoxins. Both similarities and differences are observed in the cores of these proteins. It is of interest that a cluster of a few structurally equivalent residues in the central beta-strands and in helix 5 is common to the cores.

We have destabilized apoflavodoxin by site-specific excision of its C-terminal helix. The resulting flavodoxin fragment (Fld1-149) is compact and monomeric at pH 7.0, with spectroscopic properties of a molten globule and a low conformational stability. To study if Fld1-149 is cooperatively stabilized, we have measured the equilibrium urea unfolding by fluorescence, circular dichroism, and size-exclusion chromatography. The three techniques produced coincident unfolding curves. Furthermore, the thermal unfolding seems also to be cooperative as the same temperature of half-denaturation is obtained using fluorescence and circular dichroism. Fld1-149 displays cold denaturation. The equilibrium properties of Fld1-149 demonstrate that molten globules lacking well-defined tertiary interactions can still be cooperatively stabilized and that cooperatively may appear in protein conformations of very low stability. This suggests that protein folding intermediates, can, in principle, be cooperatively stabilized.

Knowledge of protein stability principles provides a means to increase protein stability in a rational way. Here we explore the feasibility of stabilizing proteins by replacing solvent-exposed hydrogen-bonded charged Asp or Glu residues by the neutral isosteric Asn or GLN: The rationale behind this is a previous observation that, in some cases, neutral hydrogen bonds may be more stable that charged ones. We identified, in the apoflavodoxin from Anabaena PCC 7119, three surface-exposed aspartate or glutamate residues involved in hydrogen bonding with a single partner and we mutated them to asparagine or glutamine, respectively. The effect of the mutations on apoflavodoxin stability was measured by both urea and temperature denaturation. We observed that the three mutant proteins are more stable than wild-type (on average 0.43 kcal/mol from urea denaturation and 2.8 degrees C from a two-state analysis of fluorescence thermal unfolding data). At high ionic strength, where potential electrostatic repulsions in the acidic apoflavodoxin should be masked, the three mutants are similarly more stable (on average 0.46 kcal/mol). To rule out further that the stabilization observed is due to removal of electrostatic repulsions in apoflavodoxin upon mutation, we analysed three control mutants and showed that, when the charged residue mutated to a neutral one is not hydrogen bonded, there is no general stabilizing effect. Replacing hydrogen-bonded charged Asp or Glu residues by Asn or Gln, respectively, could be a straightforward strategy to increase protein stability.

Electrostatic contributions to the conformational stability of apoflavodoxin were studied by measurement of the proton and salt-linked stability of this highly acidic protein with urea and temperature denaturation. Structure-based calculations of electrostatic Gibbs free energy were performed in parallel over a range of pH values and salt concentrations with an empirical continuum method. The stability of apoflavodoxin was higher near the isoelectric point (pH 4) than at neutral pH. This behavior was captured quantitatively by the structure-based calculations. In addition, the calculations showed that increasing salt concentration in the range of 0 to 500 mM stabilized the protein, which was confirmed experimentally. The effects of salts on stability were strongly dependent on cationic species: K + , Na + , Ca 2+ , and Mg 2+ exerted similar effects, much different from the effect measured in the presence of the bulky choline cation. Thus cations bind weakly to the negatively charged surface of apoflavodoxin. The similar magnitude of the effects exerted by different cations indicates that their hydration shells are not disrupted significantly by interactions with the protein.Site-directed mutagenesis of selected residues and the analysis of truncation variants indicate that cation binding is not site-specific and that the cation-binding regions are located in the central region of the protein sequence. Three-state analysis of the thermal denaturation indicates that the equilibrium intermediate populated during thermal unfolding is competent to bind cations. The unusual increase in the stability of apoflavodoxin at neutral pH affected by salts is likely to be a common property among highly acidic proteins.

Flavodoxins are well known one-domain ␣/␤ electrontransfer proteins that, according to the presence or absence of a ϳ20-residue loop splitting the fifth ␤-strand of the central ␤-sheet, have been classified in two groups: long and short-chain flavodoxins, respectively. Although the flavodoxins have been extensively used as models to study electron transfer, ligand binding, protein stability and folding issues, the role of the loop has not been investigated. We have constructed two shortened versions of the long-chain Anabaena flavodoxin in which the split ␤-strand has been spliced to remove the original loop. The two variants have been carefully analyzed using various spectroscopic and hydrodynamic criteria, and one of them is clearly well folded, indicating that the long loop is a peripheral element of the structure of long flavodoxins. However, the removal of the loop (which is not in contact with the cofactor in the native structure) markedly decreases the affinity of the apoflavodoxin-FMN complex. This seems related to the fact that, in long flavodoxins, the adjacent tyrosine-bearing FMN binding loop (which is longer and thus more flexible than in short flavodoxins) is stabilized in its competent conformation by interactions with the excised loop. The modest role played by the long loop of long flavodoxins in the structure of these proteins (and in its conformational stability, see Ló pezLlano, J., Maldonado, S., Jain, S., Lostao, A., Godoy-Ruiz, R., Sanchez-Ruiz, Cortijo, M., Ferná ndez-Recio, J., and Sancho, J. (2004) J. Biol. Chem. 279, 47184 -47191) opens the possibility that its conservation in so many species is related to a functional role yet to be discovered. In this respect, we discuss the possibility that the long loop is involved in the recognition of some flavodoxin partners. In addition, we report on a structural feature of flavodoxins that could indicate that the short flavodoxins derive from the long ones.The flavodoxins are electron transfer proteins involved in both photosynthetic and non-photosynthetic reactions, which carry a molecule of non-covalently bound FMN as their only redox center (1, 2). Soon after their discovery, it was realized that they could be isolated in two sizes and were accordingly divided in two classes: 1) the short-chain flavodoxins (i.e. Clostridium beijerincki and Desulfovibrio vulgaris flavodoxins) and 2) the long-chain ones (i.e. Synechococcus sp. (strain PCC 7942) and Anabaena sp. (strain PCC 7119) flavodoxins). Once the x-ray structures of representatives of the two groups became available (3-6), the structural difference was seen to be due to the presence in long flavodoxins of an extra loop that splits the fifth strand of the central ␤-sheet (Fig. 1). Because many of the functional and thermodynamic properties of short and long flavodoxins are similar (redox potentials, affinity for the FMN redox cofactor, and so forth), it is not clear yet what role the extra loop of the long flavodoxins may play. In our laboratory, we have used the holoform (6) and apoform ...

No abstract

Flavodoxins are classified in two groups according to the presence or absence of a ϳ20-residue loop of unknown function. In the accompanying paper (36), we have shown that the differentiating loop from the longchain Anabaena PCC 7119 flavodoxin is a peripheral structural element that can be removed without preventing the proper folding of the apoprotein. Here we investigate the role played by the loop in the stability and folding mechanism of flavodoxin by comparing the equilibrium and kinetic behavior of the full-length protein with that of loop-lacking, shortened variants. We show that, when the loop is removed, the three-state equilibrium thermal unfolding of apoflavodoxin becomes two-state. Thus, the loop is responsible for the complexity shown by long-chain apoflavodoxins toward thermal denaturation. As for the folding reaction, both shortened and wild type apoflavodoxins display threestate behavior but their folding mechanisms clearly differ. Whereas the full-length protein populates an essentially off-pathway transient intermediate, the additional state observed in the folding of the shortened variant analyzed seems to be simply an alternative native conformation. This finding suggests that the long loop may also be responsible for the accumulation of the kinetic intermediate observed in the full-length protein. Most revealing, however, is that the influence of the loop on the overall conformational stability of apoflavodoxin is quite low and the natively folded shortened variant ⌬(120 -139) is almost as stable as the wild type protein.The fact that the loop, which is not required for a proper folding of the polypeptide, does not even play a significant role in increasing the conformational stability of the protein supports our proposal (36) that the differentiating loop of long-chain flavodoxins may be related to a recognition function, rather than serving a structural purpose.The flavodoxins are well known electron transfer proteins involved in both photosynthetic and non-photosynthetic reactions that carry a non-covalently bound FMN molecule as a redox center (1, 2). Given their key biological function and a series of practical facts (i.e. they were among the first proteins for which x-ray structures became available (3, 4), their purification in the pre-recombinant era was relatively easy, and they are reasonably stable to handle), flavodoxins were soon found to be convenient models to investigate electron transfer and molecular recognition (1, 2) and, more recently, protein stability (5-20) and folding (18 -21). Based on molecular weight and sequence comparisons, they were divided in two families: short-chain and long-chain flavodoxins (the latter containing an extra ϳ20-residue segment, subsequently shown in Refs. 22 and 23 to form a loop in the folded protein, as highlighted in Fig. 1 of our accompanying article (36)). Despite the wealth of structural and functional information available for several flavodoxins of either family, it is still unclear whether the differentiating loop plays a structura...