Rationale of Protein (enzyme) Engineering
Although thousands of proteins have been characterized in prokaryotes and eukaryotes, only few became commercially important. This is due to the high cost of isolating and purifying enzymes in sufficient quantities. Although the cost aspect has been overcome by producing an enzyme in large quantities in bacteria, for its industrial application, an enzyme (outside the cell) should also have some characteristics in addition to those of enzymes in the cells. These characteristics may include the following: (i) enzyme should be robust with a long life; (ii) enzyme should be able to use the substrate supplied in the industry even if it differs slightly from that in the cell; (iii) enzyme should be able to work under conditions (e.g. extremes of pH, temperature and concentration) of the industry even if these conditions differ from those in the cell.

For a particular class of enzymes, variation in nature may occur for each of the above properties so that one may like to combine the optimum properties to get the most efficient form of enzyme. Sometimes, however it may not be possible to get a combination of optimum properties. For instance, an enzyme with highest activity may not be the most stable. Therefore, a compromise in properties may have to be made, if we have to select an enzyme from the available variability or even if we create variability by mutagenesis. However, if structure and function relationship of an enzyme is known, the structural features for desirable function may be combined and protein engineering techniques may then be used to create a novel enzyme exhibiting a combination of all desirable functional properties. This aspect of protein engineering will be illustrated using the example of glucose isomerases, which convert glucose into other isomers like fructose and are used to make high fructose corn syrup that is so vital (or soft drink industry. It exhibits wide variation in its properties.

Glucose isomerase belongs to a TIM barrel family of enzymes which resemble each other in having a highly characteristic domain called TIM barrel, with active, site for catalytic action at one end. This TIM barrel may be found in enzymes that may differ in sequences and may catalyze different reactions. As earlier discussed, since similarities in sturcture of protein meant similarities in function, TIM barrel presents a challenge to this concept. However, it is curious that some enzymes in this family appear in pairs in their metabolic pathways so that they catalyse two consecutive steps thus showing coupling of their functions. As an example of two enzymes of TIM barrel family, while triose phosphate isomerase is one of the most efficient catalysts, glucose isomerase is relatively very inefficient. Therefore, if glucose isomerase enzyme is redesigned to use the highly efficient domain of TIM barrel family, it will be a remarkable achievement for soft drink industry. Efforts in this direction are being made (see later for methods of protein engineering).