Amino acid sequences invariant positions

The electron transport protein, cytochrome c, found in the mitochondria of all eukaryotic organisms, provides the best-studied example of homology. The polypeptide chain of cytochrome c from most species contains slightly more than 100 amino acids and has a molecular weight of about 12.5 kD. Amino acid sequencing of cytochrome c from more than 40 different species has revealed that there are 28 positions in the polypeptide chain where the same amino acid residues are always found (Figure 5.27). These invariant residues apparently serve roles crucial to the biological function of this protein, and thus substitutions of other amino acids at these positions cannot be tolerated. 

Much of the amino acid sequence in the first half of the ferredoxin chain is repeated in the second half, suggesting that the chain may have originated as a result of gene duplication. Many invariant positions are present in the sequence, including those of the cysteine residues forming the Fe-S cluster. Ferredoxins with single Fe4S4 clusters are also known 270... 

This interaction pattern is also reflected in the amino acid sequences of proteins such as tropomyosin, which can be shown to be composed of a basic seven-residue sequence that is repeated 40 times without interruption (Hodges et al., 1972 McLachlan and Stewart, 1975). Hydrophobic residues almost invariably occupy the second and fifth positions of the heptad and are presumably directed toward the major axis of the superhelix, where they serve to stabilize the structure by hydrophobic effects (Fig. 15). The charged side chains are also nonrandomly distributed and are believed to form interhelical ion pairs that further stabilize the structure (Talbot and Hodges, 1982). 

Fig. 3. Periplasmic pilus chaperone consensus sequence. Amino acid sequence of the PapD chaperone (top line) and consensus sequence derived from the comparison of twelve chaperones (second line). Amino acids are indicated using the one-letter code. In the consensus sequence, a letter shows a residue that is present in at least eight out of twelve sequences, an asterisk designates an invariant residue, and a box shows a position with a hydrophobic residue in all twelve periplasmic chaperones. The arrows underneath the sequence represent the /3 strands found in the PapD structure.

With multiple crystal structures and amino acid sequences of a large number of iLBPs, the invariant and highly conserved amino acid sites can be analyzed to determine if they are needed for correct folding, for function, or in some instances for both. However, even in a small protein of known structure, individual roles of conserved residues can be difficult to determine. This is easiest if the comparisons are limited to a subfamily in which the level of invariant positions is relatively high. 

In summary, although the number of invariant residues is limited, a large number of highly conserved positions occur in the amino acid sequence, as is apparent in Table V. Most of the conserved positions are associated with the backbone of hydrophobic residues and adjacent polar solvent-exposed amino acids, which help stabilize the backbone. Some of the same amino acids that belong to the backbone also form a portion of the cavity wall, providing some of its hydrophobic nature. 

The structural genes of the Fe proteins of the V nitrogenase of A. chroococcum and A. vinelandii and of the third nitrogenase of A. vine-landii have been cloned and sequenced 13a, 14,22). Comparison of the derived amino acid sequences shows the Fe proteins of Mo and V nitro-genases to be very similar (91% identical) and that of the third nitrogenase to be 61% identical. All sequences show the spacing of the five invariant Cys residues, a conserved Arg residue near position 100, and a consensus nucleotide-binding sequence. 

Fig, 1. Schematic representation of the a and /3 polypeptides of the MoFe protein, the VFe protein, and the third nitrogenase. The vertical lines represent the position of the five invariant Cys residues of the a subunit and the three invariant Cys residues of the /3 subunit. The flanking amino acid sequences are shown to indicate the degree of sequence homology around these Cys residues. The wide horizontal lines represent the polypeptide chains of the subunits of the three nitrogenases (see Ref 25). 

The glycosyltransferase which glycosylates the hydroxyl group of 5-hydroxylysine in collagens and similar proteins appears to recognize the amino acid sequence -G1 y-X—HyZ-Gly-Y-Arg- Whereas the amino acids X and Y are quite variable, the positions of Gly, Hyl, and Arg are invariant in the glycopeptides of ten vertebrate and invertebrate collagens (169). 

Many positions in the V, regions are neither invariant nor hypervariable but, within a subgroup, show a moderate frequency of variation. The nature of this variability can best be understood by referring to amino acid sequences tabulated in Chapter 4. Usually, differences outside hypervariable regions, within a subgroup, are associated with a single base change in the DNA sequence. 

Fig. 7. The crystal structure of L7/L12 CTF. The spirals are a -helices and the arrows represent the B-strands. The positions of the amino acid residues are indicated the invariant ones in the five known amino acid sequences are encircled. The vertical line is a crystallographic two-fold axis that might correspond to a molecular two-fold axis relating one CTF domain to another. The residues 77-79 form a fourth strand as a continuation of the B -structure across the dimer surface.

In different catalytic active CAs, Trp is conserved at position 97, while any one of the hydrophobic amino acids (Phe, Leu, lie, Met) is retained at positions 93 and 95. It has been proposed that these hydrophobic shell residues enhance protein-metal affinity by establishing a metal site environment with a reduced dielectric constant (123). The direct and indirect ligands are invariant in all sequenced and catalytically active a-CAs (103a). 

The synthetic [2Fe-2S] model complex of the 20-peptide complex exhibits two LMCT absorption maxima at 423 and 461 nm in DMF, maxima which are near to those of the native plant-type ferredoxin (423 and 466 nm) (69). Two redox couples for — 3/—2 were observed at — 0.64 V versus SCE and at —0.96 V versus SCE in DMF. One of them is very close to the value (—0.64 V versus SCE) of native ferredoxin. The 20-peptide complex containing invariant sequences Cys-A-B-C-D-Cys-X-Y-Cys and Leu-Thr-Cys-Val possesses all essential factors for a model of the active site except for the peptide conformation. The positive-shifted redox potential of the 20-peptide complex in DMF is undoubtedly due to the interactions between the Fe2S22+ core and adjacent amino-acid residues, giving rise to NH--S hydrogen bonding. 

---