The NF-kappa B activation pathway: a paradigm in information transfer from membrane to nucleus.

Nuclear factor kappa B (NF-kappaB)/Rel proteins are dimeric, sequence-specific transcription factors involved in the activation of an exceptionally large number of genes in response to inflammation, viral and bacterial infections, and other stressful situations requiring rapid reprogramming of gene expression. In unstimulated cells, NF-kappaB is sequestered in an inactive form in the cytoplasm bound to inhibitory IkappaB proteins. Stimulation leads to the rapid phosphorylation, ubiquitinylation, and ultimately proteolytic degradation of IkappaB, which frees NF-kappaB to translocate to the nucleus and activate the transcription of its target genes. The multisubunit IkappaB kinase (IKK) responsible for the inducible phosphorylation of IkappaB appears to be the initial point of convergence for most stimuli that activate NF-kappaB. IKK contains two catalytic subunits, IKKalpha and IKKbeta, both of which phosphorylate IkappaB at sites phosphorylated in vivo. Gene knockout studies indicate that IKKbeta is primarily responsible for the activation of NF-kappaB in response to proinflammatory stimuli, whereas IKKalpha is essential for keratinocyte differentiation. The activity of IKK is regulated by phosphorylation. IKK contains a regulatory subunit, IKKgamma, which is critical for activation of IKK and is postulated to serve as a recognition site for upstream activators. When phosphorylated, the IKK recognition site on IkappaBalpha serves as a specific recognition site for the kappa-TrCP-like component of a Skp1-Cullin-F-box-type E3 ubiquitin-protein ligase. A variety of other signaling events, including phosphorylation of NF-kappaB, phosphorylation of IKK, new synthesis of IkappaBs, and the processing of NF-kappaB precursors provide mechanisms of modulating the amount and duration of NF-kappaB activity.

Equilibrium and kinetic constants for the thiol-disulfide interchange reaction between glutathione and dithiothreitol.

The equilibrium and rate constants for the reaction between oxidized and reduced glutathione and oxidized and reduced dithiothreitol have been determined at several pH values and temperatures. The measurements involve approach to equilibrium from both directions, quenching of the reaction by lowering the pH or by addition of methyl methanethiosulfonate, separation of reactants and products by reverse-phase HPLC, and determination of their concentrations. Analysis of reaction mixtures was carried out at various times to assure that equilibrium had been reached and to determine kinetic constants prior to the attainment of equilibrium.

Role of non-native aromatic and hydrophobic interactions in the folding of hen egg white lysozyme.

The folding kinetics have been determined for hen egg white lysozyme and two mutants in which Trp-62 and Trp-108 have been individually replaced by tyrosine (Tyr-62-lysozyme and Tyr-108-lysozyme, respectively). An earlier study of wild-type lysozyme [Denton, M. E., Rothwarf, D. M., & Scheraga, H. A. (1994) Biochemistry 33, 11225-11236] had indicated that two transient intermediates were formed during the early stages of refolding. Both intermediates were characterized by substantial quenching of tryptophan fluorescence which suggested that, during the refolding process, Trp-62 and/or Trp-108 was involved in a non-native tertiary interaction(s). Both Tyr-108- and Tyr-62-lysozyme fold significantly faster than wild-type lysozyme (7- and 13-fold, respectively). These results indicate that the rate-limiting step in the folding of lysozyme arises not from any inherent slowness in the formation of the native structure but rather as a consequence of the formation of a highly stable intermediate which contains significant non-native structure which must be disrupted prior to, or in concert with, subsequent folding. The data suggest that aromatic and hydrophobic interactions play a pivotal role in the formation of the non-native intermediate. The general role that non-native interactions play in the folding process is discussed.

Regeneration of bovine pancreatic ribonuclease A. 1. Steady-state distribution.

The regeneration of bovine pancreatic ribonuclease A (RNase A) from the reduced to the native form with mixtures of oxidized and reduced dithiothreitol has been studied at 25 degrees C, pH 8.0, by using a variety of current experimental techniques, including quenching the regeneration reaction with 2-aminoethyl methanethiosulfonate, fractionation of intermediates by HPLC, and analysis by both UV and disulfide-specific detection systems. The disulfide-containing protein intermediates achieve a steady-state distribution after which the native protein regenerates at a rate comparable to the rates observed previously during the regeneration of RNase A with glutathione. Equilibrium constants at 25 degrees C, pH 8.0, for the interconversion of species containing different numbers of disulfide bonds are evaluated from the concentrations in the steady-state distribution. These equilibrium constants are compared with those obtained earlier when native RNase A is regenerated with glutathione. The observed equilibrium constants (with the dithiothreitol system) for the interconversions among all intermediates are very similar once statistical factors arising from the different numbers of disulfide-containing species in each grouping are taken into account. None of the disulfide-containing intermediates has any significant enzymatic activity, in agreement with earlier conclusions that these intermediates are considerably disordered. This is in sharp contrast to disulfide-containing intermediates populated during the regeneration of bovine pancreatic trypsin inhibitor, which have significant nativelike structure.

Regeneration of bovine pancreatic ribonuclease A. 3. Dependence on the nature of the redox reagent.

Monothiol reagents such as oxidized and reduced glutathione (GSSG and GSH, respectively) form stable mixed disulfides with protein thiols while dithiol reagents such as oxidized and reduced dithiothreitol (DTTox and DTTred, respectively) do not. This large difference in the stabilities of the mixed disulfides is reflected in much greater rates of formation and reduction of protein disulfide bonds with GSSG/GSH than with DTTox/DTTred. With dithiothreitol, the concentrations of intermediate species in the steady state depend on the redox potential, i.e., on the [DTTox]/[DTTred] ratio, and not on the absolute concentrations of these reagents. With glutathione, the redox potential and hence the concentrations of intermediate species in the steady state depend on the [GSSG]/[GSH]2 ratio; hence, with glutathione, in contrast to dithiothreitol, the absolute values of the concentrations do affect the steady-state concentrations. Consequently, the regeneration pathways of bovine pancreatic ribonuclease A depend on the nature of the redox reagent as well as the redox potential at which they are used. The use of GSSG/GSH favors multiple regeneration pathways, while the use of DTTox/DTTred favors regeneration through fewer pathways. These concepts are also illustrated by an analysis of literature data for the regeneration pathways of bovine pancreatic trypsin inhibitor.

Regeneration of bovine pancreatic ribonuclease A. 2. Kinetics of regeneration.

Analysis of the experimental data of the previous paper [Rothwarf, D. M., & Scheraga, H. A. (1993) Biochemistry (first of four papers in this issue)], using the method of Konishi et al. [Konishi, Y., Ooi, T., & Scheraga, H. A. (1981) Biochemistry 20, 3945-3955; Konishi, Y., Ooi, T., & Scheraga H. A. (1982) Biochemistry 21, 4734-4740], and a discussion of the validity of the steady-state kinetic treatment of the data analyzed here are presented. The analysis reveals that RNase A regenerates with oxidized and reduced dithiothreitol (DTTox and DTTred, respectively) through a rearrangement pathway involving one or more three-disulfide species; i.e., multiple pathways could be involved. This pathway is different from that observed when RNase A is regenerated with oxidized and reduced glutathione (GSSG and GSH, respectively). These differences result primarily from the very different characteristics of the oxidation of thiols with DTTox and GSSG, respectively. In addition, the concept of multiple pathways, as applied to the regeneration of RNase A, is developed.

Regeneration of bovine pancreatic ribonuclease A. 4. Temperature dependence of the regeneration rate.

The rate of regeneration of bovine pancreatic ribonuclease A with oxidized and reduced dithiothreitol (DTTox and DTTred, respectively) decreases by a factor of 10 when the temperature is increased from 25 to 37 degrees C. The rate of regeneration of RNase A with oxidized and reduced glutathione increases slightly over that same range of temperature. This suggests that the regeneration processes with the two types of redox reagents proceed through different pathways. There is a significant change in the distribution of three-disulfide intermediates populated during regeneration with DTTox/DTTred over the range of temperature 15-37 degrees C that suggests that the three-disulfide species populated at 15 degrees C are directly involved in the major regeneration pathway observed at 25 degrees C.

Kinetic studies of the regeneration of recombinant hirudin variant 1 with oxidized and reduced dithiothreitol.

The regeneration of native recombinant hirudin variant 1 (rHV1) from the reduced unfolded form to the fully oxidized native state has been carried out with mixtures of oxidized and reduced dithiothreitol at pH 8.3 and 12 degrees C. The regeneration reaction was quenched at various times by the addition of 2-aminoethyl methanethiosulfonate to block unreacted sulfhydryl groups. The quenched protein-folding intermediates were fractionated by both capillary electrophoresis and a combination of anion exchange and reverse phase HPLC and characterized by mass spectrometry, amino acid analysis, and disulfide analysis. These intermediates (before quenching) were found to interconvert rapidly so as to achieve a steady-state distribution early in the regeneration process. The experimental data were fitted to a steady-state kinetic scheme. The analysis reveals that the rate-determining step in the regeneration of rHV1 with oxidized and reduced dithiothreitol involves the oxidation of one or more two-disulfide-containing species, most likely those already containing two native disulfide bonds. This regeneration mechanism is different from one that has been proposed by Chatrenet and Chang [(1993) J. Biol. Chem. 268, 20988]. The differences are discussed, and possible explanations for the differences are presented.

Mechanism of reductive protein unfolding.

The reductive unfolding of ribonuclease A with dithiothreitol proceeds through parallel pathways with the formation of two well-populated partially-unfolded three-disulphide intermediates. Two distinct local unfolding events rather than a global one are involved in the rate-limiting steps. These results are contrary to the current view that protein unfolding generally follows an all-or-none mechanism, and that the rate-limiting step is controlled by an extensive rearrangement of the native structure. Sequential breakage of disulphide bonds through local unfolding events is energetically more favourable than disruption of the native structure through global unfolding. The results also indicate that the oxidative refolding of ribonuclease A from the fully-reduced form proceeds through parallel conformationally-distinct transition states.

The nature of the initial step in the conformational folding of disulphide-intact ribonuclease A.

Here we investigate conformational folding reaction of disulphide-intact ribonuclease A in the absence of the complicating effects due to non-native interactions (such as cis/trans proline isomerization) in the unfolded state. The conformational folding process is found to be intrinsically very fast occurring on the milliseconds time scale. The kinetic data indicate that the conformational folding of ribonuclease A proceeds through the formation of a hydrophobically collapsed intermediate with properties similar to those of equilibrium molten-globules. Furthermore, the data suggest that the rate-limiting transition states on the unfolding and refolding pathways are substantially different with the refolding transition state having non-native-like properties.

A very fast phase in the refolding of disulfide-intact ribonuclease A: implications for the refolding and unfolding pathways.

The refolding and unfolding of disulfide-intact ribonuclease A has been studied by using single-jump and double-jump stopped-flow techniques. Absorbance and fluorescence detection methods were used to follow the kinetics. By appropriate choice of solution conditions (1.5 M guanidine hydrochloride, pH 3.0, at temperatures < or = 15 degrees C) to slow the refolding process, a new very fast phase has been observed in addition to the usual fast and slow phases that involve the unfolded species Uf and U(s), respectively. Double-jump experiments consisting of an unfolding step at 4.2 M guanidine hydrochloride and pH 2.0 followed by a refolding step at 1.5 M guanidine hydrochloride and pH 3.0 were carried out to monitor the unfolding process. These experiments demonstrated that the new phase arises from a separate unfolded species, Uvf, which is present to the extent of about 6% in the equilibrium ensemble of unfolded protein at high guanidine hydrochloride concentration and low pH. A new model for the unfolding pathway and interconversion among unfolded species is proposed based on two independent isomerization processes. The equilibrium constants and activation energies obtained for each process suggest that they involve the isomerization of cis prolines. We propose that the isomerizations occur at the X-Pro peptide bonds of Pro 93 and 114. In the model, Uvf is the first species to form without isomerization at any cis X-Pro peptide bonds when the native protein is unfolded; Uf and U(s) then form from Uvf through two independent isomerization processes. Both prolines are in the native (cis) conformation in Uvf. In Uf, Pro 114 is in a nonnative (trans) conformation while, in U(s), Pro 93 is in a nonnative (trans) conformation. The slow folding species, U(s), actually consists of (at least) two species: U(s) alpha with Pro 93 in a nonnative (trans) conformation and U(s) beta with both Pro 93 and 114 in nonnative (trans) conformations. Finally, the kinetic data suggest that the presence of a nonnative trans conformation at the Tyr 92-Pro 93 peptide bond impedes the refolding rate of ribonuclease A much more than the presence of a nonnative trans conformation at the Asn 113-Pro 114 peptide bond.

IKK-gamma is an essential regulatory subunit of the IkappaB kinase complex.

Pro-inflammatory cytokines activate the transcription factor NF-kappaB by stimulating the activity of a protein kinase that phosphorylates IkappaB, an inhibitor of NF-kappaB, at sites that trigger its ubiquitination and degradation. This results in the nuclear translocation of freed NF-kappaB dimers and the activation of transcription of target genes. Many of these target genes code for immunoregulatory proteins. A large, cytokine-responsive IkappaB kinase (IKK) complex has been purified and the genes encoding two of its subunits have been cloned. These subunits, IKK-alpha and IKK-beta, are protein kinases whose function is needed for NF-kappaB activation by pro-inflammatory stimuli. Here, by using a monoclonal antibody against IKK-alpha, we purify the IKK complex to homogeneity from human cell lines. We find that IKK is composed of similar amounts of IKK-alpha, IKK-beta and two other polypeptides, for which we obtained partial sequences. These polypeptides are differentially processed forms of a third subunit, IKK-gamma. Molecular cloning and sequencing indicate that IKK-gamma is composed of several potential coiled-coil motifs. IKK-gamma interacts preferentially with IKK-beta and is required for the activation of the IKK complex. An IKK-gamma carboxy-terminal truncation mutant that still binds IKK-beta blocks the activation of IKK and NF-kappaB.

Kinetics of folding of guanidine-denatured hen egg white lysozyme and carboxymethyl(Cys6,Cys127)-lysozyme: a stopped-flow absorbance and fluorescence study.

The folding kinetics of hen egg white lysozyme and of a three-disulfide derivative of lysozyme [carboxymethyl(Cys6,Cys127)-hen egg white lysozyme] have been studied by absorbance- and fluorescence-detected stopped-flow techniques. A "very-fast" phase with a time constant in the millisecond range has been observed by both absorbance and fluorescence when unfolded lysozyme in 4 M guanidine hydrochloride, 100 mM phosphate buffer, and pH 2.0 is refolded at 0.5 M guanidine hydrochloride, 100 mM phosphate, and pH 6.7. Data obtained from fluorescence-detected refolding studies show that a transient intermediate is formed during the very-fast refolding phase. This intermediate is characterized by substantial quenching of tryptophan fluorescence. In addition, analysis of the fluorescence data indicates the presence of an additional "burst" phase that occurs within the dead time of the instrument, < 3 ms. The very-fast phase is not observed during the refolding of the three-disulfide derivative. In addition, the three-disulfide derivative re-attains the final native folded conformation more rapidly than the unmodified protein over the range of temperatures studied (10-20 degrees C). We conclude that, not only does the presence of the disulfide bond between Cys6 and Cys127 slow down the overall folding process of lysozyme, but it also directs the folding of lysozyme through a pathway characterized by a non-native tertiary interaction(s).

Structural characterization of a three-disulfide intermediate of ribonuclease A involved in both the folding and unfolding pathways.

Earlier studies of the unfolding pathway of native bovine pancreatic ribonuclease A (using dithiothreitol as the reducing agent) revealed that the three-disulfide species lacking the disulfide bond between cysteine 65 and cysteine 72 is the most highly populated intermediate [Rothwarf & Scheraga (1991) J. Am. Chem. Soc. 113, 6293-6294]. This unfolding intermediate is referred to as des-[65-72]-RNase A. In order to determine the role of des-[65-72]-RNase A, i.e. of the 65-72 disulfide bond, in the structural folding/unfolding processes of RNase A, the stability and structure of this unfolding intermediate were determined by examining its thermal transition curve and by using two- and three-dimensional homonuclear 1H NMR spectroscopy. The midpoint of the thermal transition of des-[65-72]-RNase A was found to be 17.8 degrees C lower than that of native RNase A. A set of conformations that are consistent with the NMR-derived constraints was obtained by minimizing, first, a variable-target function and, then, the conformational energy. These conformations exhibit a well-defined structure that is very similar to that of native ribonuclease A in regions where the native protein has a regular backbone structure such as a beta-sheet or a helix. Some of the loop regions of the several computed structures exhibit large deviations from each other as well as from native ribonuclease A. However, these results indicate that des-[65-72]-RNase A has a close structural similarity to RNase A in all regions with the only major differences occurring in a loop region comprising residues 60-72. This led to the conclusion that, in reduction pathways that include des-[65-72]-RNase A (at 25 degrees C, pH 8.0), the rate-determining step corresponds to a partial unfolding event in one region of the protein and not to a global conformational unfolding process. The results further suggest that, in the regeneration pathways involving des-[65-72]-RNase A, the loop region from 60 to 72 is the last to fold.

Regeneration of bovine pancreatic ribonuclease A: identification of two nativelike three-disulfide intermediates involved in separate pathways.

During the regeneration of bovine pancreatic ribonuclease A (RNase A) from the reduced to the native form with mixtures of oxidized and reduced dithiothreitol at 25 degrees C, pH 8.0, the disulfide-containing protein intermediates achieve a steady-state distribution. By manipulating the redox conditions after the attainment of the steady-state condition, it has been possible to kinetically trap and, thereby, isolate and identify the disulfide-bonded species that follow the rate-determining step in the regeneration pathway. Two three-disulfide species have been identified by peptide mapping. Both species contain three native disulfide-bond pairings, one lacks the 65-72 disulfide bond (des-[65-72]), and the other lacks the 40-95 disulfide bond (des-[40-95]). These species are the same as those identified during the reduction of RNase A. By restarting the regeneration process from isolated des-[65-72] and des-[40-95], it is shown that both intermediates lie directly on regeneration pathways.

Regeneration of bovine pancreatic ribonuclease A: detailed kinetic analysis of two independent folding pathways.

The regeneration of bovine pancreatic ribonuclease A (RNase A) from the reduced to the native form with mixtures of oxidized and reduced dithiothreitol at 25 degrees C, pH 8.0, proceeds through two separate pathways in which separate nativelike three-disulfide species are populated. The populations of these two three-disulfide species during the regeneration process have been monitored directly through the use of a reduction pulse. A detailed kinetic analysis of the regeneration process using improved experimental procedures and data analysis has been carried out to obtain rate constants for disulfide interconversion among the various disulfide-bonded intermediates. This analysis indicates that these two pathways can account for essentially 100% of the native protein regenerated and that the relative amount of native protein regenerated through these two pathways is insensitive to the redox conditions used. These results indicate that the rate-determining step in both pathways involves formation of the nativelike three-disulfide species, a step in which most of the conformational folding takes place. The experimentally determined rate constants indicate that these two pathways are sufficient to explain the differences in the temperature dependence of the regeneration rate with different redox reagents. In addition, the population of a fully oxidized species that contains three native disulfide bonded pairs and a dithiothreitol bridging cysteines 65 and 72 has been observed.

Nonrandom distribution of the one-disulfide intermediates in the regeneration of ribonuclease A.

The one-disulfide intermediates formed during the oxidative refolding of ribonuclease A (RNase A) have been characterized. This information is important for understanding the folding pathways of RNase A. The one-disulfide intermediates were blocked with 2-aminoethyl methanethiosulfonate, fractionated using ion-exchange chromatography, and digested with trypsin and chymotrypsin. The resulting peptide fragments were fractionated using reversed phase high-performance liquid chromatography, and identified using mass spectrometry. The relative population of each one-disulfide intermediate was determined from its disulfide bond concentration using a postcolumn disulfide detection system. A total of 24 out of 28 possible one-disulfide intermediates were found to be populated (greater than 0.3%) in the one-disulfide mixture. The population of one-disulfide intermediates displays a nonrandom distribution. All four native disulfide pairings have populations greater than those predicted by loop entropy calculations, suggesting the presence of enthalpic contributions stabilizing these species. The one-disulfide intermediate [65, 72], containing the disulfide bond between cysteines 65 and 72, comprises 40% of the entire one-disulfide population. The interactions that stabilize this intermediate may play an important role in the regeneration pathways of RNase A.

Circular dichroism evidence for the presence of burst-phase intermediates on the conformational folding pathway of ribonuclease A.

Refolding of the very-fast-folding unfolded species (Uvf) of disulfide-intact bovine pancreatic ribonuclease A has been monitored by circular dichroism (CD) at 222 and 275 nm at 0.9 or 2.6 M guanidine hydrochloride, pH 7.0, and 5 degrees C. The refolding of Uvf represents a purely conformational folding process which is not complicated by cis-trans proline isomerization. The data indicate that there are at least two intermediates on the refolding pathway of Uvf and that both intermediates form in the burst phase when the refolding is monitored by CD. At the initiation of folding, Uvf is converted to a largely unfolded intermediate, termed Iu, which then undergoes a hydrophobic collapse to form the molten-globule-like intermediate I phi. The CD values obtained for Iu and I phi indicate that IU has no significant secondary structure and presumably differs from Uvf by a local structural rearrangement, while I phi has a substantial population of secondary and tertiary structures, about 40%-50% of that of native.

The high-molecular-weight proteins of bovine brain plain synaptic vesicles.

High molecular mass polypeptides (Mr greater than 100,000) of plain synaptic vesicles from bovine cerebral cortex were separated using porous polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. Four major bands, of Mr 262,000, 249,000, 216,000, and 173,000, were resolved. Investigations into the membrane association of the Mr 216,000 and 173,000 proteins by means of solubilization experiments and Sepharose 4B chromatography indicate that the former is a peripheral protein and the latter is more firmly attached, possibly an integral protein. Finally, the Mr 216,000 protein was shown to be highly enriched in synaptic vesicles compared to other brain subfractions. It thus appears to be specifically associated with synaptic vesicles and therefore may have an important role specific to synaptic vesicle function or structure.

A cytokine-responsive IkappaB kinase that activates the transcription factor NF-kappaB.

Nuclear transcription factors of the NF-kappaB/Rel family are inhibited by IkappaB proteins, which inactivate NF-kappaB by trapping it in the cell cytoplasm. Phosphorylation of IkappaBs marks them out for destruction, thereby relieving their inhibitory effect on NF-kappaB. A cytokine-activated protein kinase complex, IKK (for IkappaB kinase), has now been purified that phosphorylates IkappaBs on the sites that trigger their degradation. A component of IKK was molecularly cloned and identified as a serine kinase. IKK turns out to be the long-sought-after protein kinase that mediates the critical regulatory step in NF-kappaB activation.