Amino acid sequences alternative conformations

Many virus coats have 180 subunits or a number that is some other multiple of 60. However, in these coats the subunits cannot all be in identical environments. Two cases may be distinguished. If all of the subunits have identical amino acid sequences they probably exist in more than one distinct conformation that permit them to pack efficiently. (Next section) Alternatively, two or more subunits of differing sequence and structure may associate to form 60 larger subunits that do pack with icosahedral symmetry. For example, the polioviruses (diameter 25 nm) contain three major coat proteins (a, P, and y or VP1, VP2, and VP3). These are formed by cleavage of a large precursor protein into at least four pieces.76 77 Tire three largest pieces of 33-, 30-, and 25-kDa mass (306,272, and 238 residues, respectively) aggregate as (aPy)60. Sixty copies of a fourth subunit of 60 residues are found within the shell. 

A single polypeptide chain can in theory exist in an infinite number of different conformations. However, one specific conformation generally appears to be the most stable for any given sequence of amino acids, and this conformation is assumed by the chain as it is synthesized within the cell. Thus, the primary structure of the polypeptide chain also determines its three-dimensional secondary and tertiary structures. It is conceivable that in some cases there may be several alternative conformations ("conforraers ) of a single chain that are of nearly equal stabilities and therefore these alternative forms may coexist. This possibility was first suggested to account for the heterogeneity noted in preparations of the cytoplasmic and mitochondrial isoenzymes of malate dehydrogenase and has also been proposed as an explanation of the multiple electrophoretic zones of erythrocyte acid phosphatase. However, no multiple enzyme forms have been shown unequivocally to be due to conformational isomerism. 

Proline has long been known to help a peptide adopt a reverse-turn conformation (71). For example, the classic type VI turn was defined as having a cm-amide bond between residues i + 1 and i + 2, which proline facilitates in the i + 2 position due to its disubstituted amide nitrogen. The sequence d-Pro-l-Pro in particular has been found to adopt a reverse-turn conformation (84-86) and alternating d/l amino acid sequences facilitate cycli-zation of small peptides as discussed in a review of cyclization of peptides and depsipeptides by Davies (87). Particular difficulties in cyclization of linear tetrapeptides containing residues of the same chirality have been found (88). Durani has discussed designing small folded proteins based on an alphabet of d- and L-amino acids (89). 

The largest group of facial amphiphilic peptides consists of the alpha-helical peptides. Facial amphiphilic alpha helices, often referred to as amphipathic alpha helices, are not amphiphilic in their random coil conformation and their amphiphilicity is not directly obvious from then-sequence. However, folding of the peptide into its preferred secondary structure, leads to the formation of an alpha helix, of which the hydrophilic amino acids occupy one face and the hydrophobic amino acids are located at the other face. Alpha-helical peptides have a periodicity of 3.6 amino acid residues per turn, and because of this, for two turns, roughly every third and seventh amino acids are on the same face of the alpha helix. In order to make a helix amphiphilic, the sequence of amino acids should alternate between hydrophobic and hydrophilic every three to four residues, which becomes more clear in a helical wheel representation (Figure 3). An example of such a facial amphiphilic alpha helix is magainin 2, a 23 amino acid antibiotic peptide. Studies have shown that magainin... 

The amino acid compositions and sequences of the /3-strands in porin proteins are novel. Polar and nonpolar residues alternate along the /3-strands, with polar residues facing the central pore or cavity of the barrel and nonpolar residues facing out from the barrel where they can interact with the hydrophobic lipid milieu of the membrane. The smallest diameter of the porin channel is about 5 A. Thus, a maltodextrin polymer (composed of two or more glucose units) must pass through the porin in an extended conformation (like a spaghetti strand). 

The conformational preferences of mixed /9-peptides containing both /9 - and /9 -amino acid residues in their sequence differ markedly from that of the corresponding homopolymers consisting exclusively of /9 - or /9 -amino acid residues. Several types of mixed /9-peptides have been investigated including block peptides constructed with triads of /9 -amino acid residues and triads of /9 -amino acid residues (e.g. 93) [104,161], as well as alternating peptides of jf lff type (e.g., 72,... 

Bioactive sequences of up to six amino acid residues known to assume (1- or "/-turns in the bioactive conformation are suitable for such libraries. If the sequence is short, residues have to be added in a manner to retain the desired physicochemical properties of the peptide (e.g., to short polar active sequences hydrophobic residues are preferentially added and vice versa). The choice of the scaffold depends on the number of structure-inducing amino acids such as Gly or Pro present in the native sequence. In absence of such residues scaffolds (1) or (4) (Scheme 24) are selected, whereas if Gly or Pro is present alternative scaffolds can be considered. Then the components of the four stereoisomeric sublibraries of Scheme 26 (or their equivalents if other scaffolds are chosen) are synthesized according to procedures described in the preceding sections. 

---