digestion -

Fig. 9.5 Folded domains can be selected for in vitro by phage display, which establishes a link between a protein and the corresponding gene (thick line): the protein is displayed on the surface of bacteriophage particle (Inovirus) by fusion with a phage coat protein such as protein p3. The phage tips are represented. (A) A folded domain (rectangle) or an unfolded polypeptide (thin line) is inserted between domain 3 and domains 1 and 2 of the protein p3. Digestion by a proteolytic enzyme cleaves the fusion protein if it is unfolded. Phage particles displaying only domain D3 of p3 are not infectious and are counter-selected. Phage particles with the three domains of p3 are infectious, thereby allowing the corresponding gene to be amplified in E. coli. (B) A folded domain (rectangle) or an unfolded polypeptide (thin line) is inserted between protein p3 and an affinity tag 'a'. Digestion by a proteolytic enzyme cleaves the unfolded polypeptide and releases the affinity tag in the solution. Affinity chromatography for the tag 'a' allows the isolation of folded domains and of their genes.

so as to establish a link between the phenotype (the folded or unfolded domain) and the genotype (the corresponding gene which can be amplified) [40, 41]. Pro-teolytic digestion of unfolded domains is coupled to loss of infectivity of the phage particle, thereby preventing its recovery after infection of E. coli. In analogous methods, proteolytic digestion of unfolded domains is coupled to release of an affinity tag, preventing the recovery of the phage particle by affinity chromatography for the tag [42]. Identifying New Folds and New Topologies

The in vitro selection using phage display described in the previous section was applied to a library of about 3 x 107 random DNA fragments of the E. coli genome. 124 unique fragments were found to be resistant to trypsin which cleaves the polypeptide chain at arginine and lysine residues. About two-thirds of the fragments also displayed resistance to the protease thermolysin, which cleaves at aromatic and aliphatic amino acids [42]. This method for identification of structural domains should be useful for structural genomics studies either by NMR which requires small folded domains or by crystallography as crystallization can be improved by removal of the flexible regions in proteins.

A new fold was also identified using the same in vitro selection strategy. A library of random fragments of about 40 amino acids in length from E. coli were fused to the N-terminal half of the cold shock protein A from E. coli and yielded after selection a fusion with a C-terminal fragment of the ribosomal protein S1 [43]. The protein's architecture was found to be a unique six-stranded beta-barrel that assembles to form a tetramer [44].

Enzyme topology was also altered: in the case of Methanococcus jannaschii cho-rismate mutase, the quaternary structure of the protein was changed from a dimer into a monomer. The catalytic activity as measured by the kinetic parameters was found to be similar for both the dimeric and the monomeric enzymes. This method relies on the introduction of dimer destabilizing mutations and on in vivo selection of active enzymes [45]. Combining Domains

Given two domains A and B, domain fusions found in nature occur mainly in a single orientation AB or BA. It was estimated that both orientations (AB and BA) domains appear in only about 2% of the cases [46].

Fusion of domains can be used to endow catalysts with new functions. For example, thermostable DNA-dependent DNA-polymerases such as family A T. aqua-ticus DNA polymerase I or family B Pyrococcus furiosus DNA polymerase were fused to a thermostable protein binding double-stranded DNA, the Sulfolobus sol-fataricus Sso7D protein. These fusion proteins were found to incorporate more nucleotides per DNA binding event, i.e. to have an improved processivity [47]. This has important applications in polymerase chain reactions (PCR) for efficient amplification of long templates of up to 15 kilobases.

0 0

Post a comment