On the evolutionary relationship of the 4-alpha-helical heme proteins. The comparison of cytochrome b562 and cytochrome c'

Secondary structure analyses of protein films on gold surfaces by circular dichroism

In order to analyze the secondary structures of protein molecules adsorbed on gold surfaces, circular dichroism (CD) spectra were measured and the secondary structure contents of protein ultra-thin films were estimated quantitatively. A disulfide group was introduced to cytochrome b(562) (cyt.b562), which is a water-soluble b-type heme protein. The cyt.b562 molecules self-assembled to form an ultra-thin protein film both on a gold substrate modified with 2,2(')-dithiodiacetic acid and on a bare gold surface. CD measurements were carried out both in solution and in air, and these results were compared. The protein denaturation was partially prevented, not only in solution but also in air, by both the modification of the substrate and the introduction of the anchor group to the protein molecule. The secondary structure contents of ultra-thin protein films on flat gold surfaces were observed for the first time both in solution and in air by CD spectra.

Sequence conservation in families whose members have little or no sequence similarity: the four-helical cytokines and cytochromes

Proteins for which there are good structural, functional and genetic similarities that imply a common evolutionary origin, can have sequences whose similarities are low or undetectable by conventional sequence comparison procedures. Do these proteins have sequence conservation beyond the simple conservation of hydrophobic and hydrophilic character at specific sites and if they do what is its nature? To answer these questions we have analysed the structures and sequences of two superfamilies: the four-helical cytokines and cytochromes c'-b(562). Members of these superfamilies have sequence similarities that are either very low or not detectable. The cytokine superfamily has within it a long chain family and a short chain family. The sequences of known representative structures of the two families were aligned using structural information. From these alignments we identified the regions that conserve the same main-chain conformation: the common core (CC). For members of the same family, the CC comprises some 50% of the individual structures; for the combination of both families it is 30%. We added homologous sequences to the structural alignment. Analysis of the residues occurring at sites within the CCs showed that 30% have little or no conservation, whereas about 40% conserve the polar/neutral or hydrophobic/neutral character of their residues. The remaining 30% conserve hydrophobic residues with strong or medium limitations on their volume variations. Almost all of these residues are found at sites that form the "buried spine" of each helix (at sites i, i+3, i+7, i+10, etc., or i, i+4, i+7, i+11, etc.) and they pack together at the centre of each structure to give a pattern of residue-residue contacts that is almost absolutely conserved. These CC conserved hydrophobic residues form only 10-15% of all the residues in the individual structures.A similar analysis of the cytochromes c'-b(562), which bind haem and have a very different function to that of the cytokines, gave very similar results. Again some 30% of the CC residues have hydrophobic residues with strong or medium conservation. Most of these form the buried spine of each helix and play the same role as those in the cytokines. The others, and some spine residues bind the haem co-factor.

Alpha-helical assembly of biologically active peptides and designed helix bundle protein

The formation of alpha-helical assembly by complexing biologically active peptides with de novo designed protein is described. The de novo designed protein described here is a cystine-linked 4-helix bundle protein constructed with 80 amino acid residues and forms a hydrophobic core region surrounded by 4 helices in an aqueous solution. The biologically active peptides, such as melittin and human growth hormone releasing factor, contain the sequences that are able to form amphiphilic helices. These peptides alone do not form the alpha-helix structure in a diluted solution with low ion strength. But on mixing with the designed helix bundle protein, the peptides are strongly bound to the protein with the induction of alpha-helical structure in the biologically active peptides. The content of induced alpha-helix is in accord with that estimated from the amphiphilic sequence. The results mean that a novel architecture composed of alpha-helices is formed. Fluorescent and temperature-scanning measurement revealed that the alpha-helical assembly is constructed with hydrophobic interaction. Also, it is shown by means of fluorescence depolarization that the assembly has a compact globular form corresponding to 1:1 complex.

Refolding of cytochrome b562 and its structural stabilization by introducing a disulfide bond

The packing mechanism of the secondary structures (4-alpha-helices and 3(10)-helix) of cytochrome b562 is simulated by the "island model," where the formation of protein structure is accomplished by the growth-type mechanism with the driving force of packing of the long-range and specific hydrophobic interactions. Packing proceeds through the formation of the structure at the nonhelical part, where a lot of hydrophobic pairs are distributed. Consequently, conformation, nearly similar to the native one, is successfully obtained. With the help of this result, the theoretical prediction of the possibility of forming this disulfide mutant (N22C/G82C) of b562 can be performed prior to the experiments by our geometrical criterion ("lampshade"). This criterion is expected to be a significant principle for introducing possible disulfide bonds into a protein to be engineered.

Cloning and expression of the gene encoding the soluble cytochrome b562 of Escherichia coli

The gene for the soluble cytochrome b562 from Escherichia coli B has been cloned on a SalI fragment. The analysis of the gene reveals the presence of a leader sequence in front of the sequence encoding the mature protein. Expression of cytochrome b562 using the lac-promoter produced the protein to a level of 3-5% of total protein. This over-production enables employment of a simple, high-yield purification protocol to obtain homogeneous cytochrome b562. Spectroscopic and N-terminal sequence analyses of the purified protein demonstrate that it is identical to the chromosomally expressed cytochrome b562 purified and characterized from E. coli B [Itagaki, E. & Hager, L.P. (1966) J. Biol. Chem. 241, 3687-3695]. It is demonstrated that the genomic sequence codes for a classic N-terminal signal sequence and that mature cytochrome b562 is translocated to the periplasmic space.

Why do c-type cytochromes exist? Reprise

An earlier paper (Wood, P.M. (1983) FEBS Lett. 164, 223-226) proposed that the covalent links that characterize c-type haem originated in order to prevent the haem being lost into the external medium. This is discussed in relation to Gram-negative and Gram-positive bacteria, cyanobacterial thylakoids and eukaryotes. The practical application is for unravelling complex electron transfer chains: c-type cytochromes may be assumed to be absent from the cytoplasm. Sulphate reducing bacteria provide the only confirmed exception to this rule. Examples of non-covalent haem at risk of being lost are considered. Analogies are drawn with flavin and pyrroloquinoline quinone as prosthetic groups.

Bacterial 4-alpha-helical bundle cytochromes

The biological functions of cytochrome c' and bacterioferritin, both haemoproteins with a common 4-alpha-helical bundle structure, are discussed and an example given of one of Kamen's laws, namely: comparative studies of prokaryotic cytochromes and their eukaryotic counterparts are useful. In the present case, the comparison is between bacterioferritin and its animal counterpart, haemoferritin.

1H NMR study of the solution molecular and electronic structure of Escherichia coli ferricytochrome b562: evidence for S = 1/2 in equilibrium S = 5/2 spin equilibrium for intact His/Met ligation

The solution 500-MHz 1H NMR spectral parameters for ferricytochrome b562, a soluble 12-kDa electron carrier from Escherichia coli with axial His/Met coordination, are shown to be strongly influenced by protein concentration and ionic strength at low pH and 25 degrees C in a manner consistent with significant aggregation at low ionic strength. At high ionic strength a well-resolved 1H NMR spectrum reveals over 40 hyperfine-shifted resonances which arise from two isomeric species in the ratio 2:1. 2D COSY and NOESY maps at 25 degrees C for the hyperfine-shifted resonances allow the assignment of a number of axial His resonances and all heme peripheral substituent peaks. The resulting asymmetric heme contact shift patterns, together with the halving of the number of lines when reconstituting with 2-fold symmetric hemin, demonstrate the molecular basis of the solution heterogeneity to be heme orientational disorder. The strongly upfield-shifted axial Met-7 resonances, characteristic of low-spin ferricytochromes c with His/Met ligation, appear upfield only at very low temperatures. At elevated temperatures, all resonances, in particular those of the axial Met, move strongly downfield. Detailed analysis of the deviation from Curie behavior for different functional groups demonstrates the presence of a low spin in equilibrium high spin equilibrium with an intact His-Fe-Met coordination. The weaker axial field in ferricytochrome b562, relative to the purely low-spin ferricytochromes c, is attributed to a perturbed iron-Met bond. The contact shifts for a coordinated Met in the high-spin state are estimated. A link between equatorial hemin and axial ligand interactions is indicated by a differential population of the high-spin form for the two hemin orientations.

A sensitive procedure to compare amino acid sequences

Methods are discussed that provide sensitive criteria for detection of weak sequence homologies. They are based on the Dayhoff relatedness odds amino acid exchange matrix and certain residue physical characteristics. The search procedure uses several residue probe lengths in comparing all possible segments of two protein sequences, and search plots are shown with peak values displayed over the entire search length. Alignments are automatically effected using the highest search matrix values and without the necessity of gap penalties. Tests for significance are derived from actual protein sequences rather than a random shuffling procedure.

Identification of heme axial ligands of cytochrome c' from Alcaligenes sp. N.C.I.B. 11015

The spectral properties of both ferric and ferrous cytochromes c' from Alcaligenes sp. N.C.I.B. 11015 are reported. The EPR spectra at 77 K and the electronic, resonance Raman, CD and MCD spectra at room temperature have been compared with those of the other cytochromes c' and various hemoproteins. In the ferrous form, all the spectral results at physiological pH strongly indicated that the heme iron(II) is in a high-spin state. In the ferric form, the EPR and electronic absorption spectra were markedly dependent upon pH. EPR and electronic spectral results suggested that the ground state of heme iron(III) at physiological pH consists of a quantum mechanical admixture of an intermediate-spin and a high-spin state. Under highly alkaline conditions, identification of the axial ligands of heme iron(III) was attempted by crystal field analysis of the low-spin EPR g values. Upon the addition of sodium dodecyl sulfate to ferric and ferrous cytochrome c', the low-spin type spectra were induced. The heme environment of this low-spin species is also discussed.

A spectroscopic investigation of the structure and redox properties of Escherichia coli cytochrome b-562

The six-coordinate monohaem ferricytochrome b-562 from Escherichia coli exhibits two haem-linked pH-dependent transitions detected by NMR and optical spectroscopy. Only one of these transitions, that of the Fe(III)-coordinated His-102, is detected by EPR and MCD; the ionisation of a haem propionate is not. Both ionisations are redox-state-dependent and the midpoint redox potential of the protein is markedly pH-dependent. Over the pH range 5.0 to 8.5 the potential drops from 260 mV to 110 mV and at least five single proton ionisations are responsible for this. In addition to the two spectroscopically identified ferricytochrome ionisations, there are at least three unidentified ionisations, two of which occur in the ferrous protein. From a consideration of the X-ray structure, together with NMR data, it seems probable that at least one of these ionisations involves an amino acid carboxylate. The X-ray structure also suggests that the relatively low pKa of His-102 is a result of its proximity to Arg-98. However, an appreciable interaction between these groups requires that the solution conformation differs slightly from the X-ray structure. The fast rate of electron self-exchange, over 4 X 10(6) M-1 X s-1 at 315 K and pH* 7, may be a reflection of the fact that, as shown by the X-ray structure, a large amount of the haem and axial histidine ligand are exposed at the molecular surface with an asymmetric distribution of charged groups surrounding them.

The complete amino-acid sequence of the low-spin class II cytochrome c-556 from Agrobacterium tumefaciens strain B2a

The amino acid sequence of the soluble monohaem cytochrome c-556 from Agrobacterium tumefaciens, strain B2a, has been determined. The sequence was derived from peptides obtained by digestion of the apoprotein with trypsin and chymotrypsin, and by subdigestion of some of the peptides with Staphylococcus aureus protease and thermolysin. Sequencing of the various peptides was achieved by a combination of manual dansyl-Edman degradation and automatic liquid-phase sequence analysis. The main characteristic of this cytochrome is that the haem-binding sequence Cys-Xaa-Yaa-Cys-His occurs in the C-terminal region of the polypeptide chain, the first cysteine being located 11 residues ahead of the C-terminal lysine-122. As such, the protein belongs to cytochrome c sequence class II (sensu Ambler). The cytochrome c-556 is the first example known of a class II cytochrome of the low-spin type isolated from an obligate aerobic organism.