In recent years, rare sugars have gained attention as a result of their numerous applications. According to the International Society of Rare Sugars (ISRS), they are both monosaccharides and monosaccharide derivatives [1]. Their content is scarce in nature; they are found primarily in fresh milk, corn, wheat, forest trees, etc. Obtaining them by extraction is difficult, which limits their development and application [2]. Psicose and allose are two examples of rare sugars.

D-Allose is an aldose isomer of D-psicose, which is structurally a distinct isomer from D-glucose. It is particularly abundant in some natural plant extracts and bacterial metabolites [3] and has also received increased attention as a result of its numerous physiological functions. D-Allose, for example, could be employed as a low-calorie carbohydrate sweetener and bulking agent [4], an anticancer agent for a variety of cancer cells [5], [6], [7], [8], an antioxidant [9], an anti-aging agent [10], an antihypertensive [11], a cryoprotectant, and an immunosuppressant [12], [13].

The chemical synthesis method uses redox to change the configuration of various sugars to synthesize allose, but the procedure is laborious and results in many byproducts [14], [15]. The biotransformation method refers to the catalytic synthesis of allose by isomerization of isomerase, which includes many advantages over chemical methods. Examples include simple processing, improved activity, higher specificity under mild conditions, and reduced byproduct conversion [16].

The Izumoring strategy provides almost all efficient strategies for the production of ketohexose, aldohexose, and hexositol. Among them, D-psicose could be isomerized to produce both D-allose and D-altrose [17]. L-rhamnose isomerase (L-Rhi), which was derived from Pseudomonas stutzeri was discovered in the laboratory in 1997 [18]. Another class of enzymes that potentially catalyze the isomerization of D-psicose to D-allose comes from the ribose/galactose isomerase family of enzymes, which include ribose-5-phosphate isomerase (RPI) [19], galactose-6-phosphate isomerase (GPI) [20], etc. These are examples of phosphoglucoisomerase enzymes, which generally do not require the addition of metal ions when catalyzing isomerization reactions [21].

L-Rhi is an aldo-keto isomerase that catalyzes the conversion of L-rhamnose and L-rhamnulose; it is widely present in a variety of microorganisms and originates from an array of microbial sources. The catalytic production of D-allose is more efficient but might be accompanied by the production of the byproduct altrose [22]. Studies have demonstrated that the optimum temperature of L-Rhi derived from different microorganisms was maintained above 60 °C, and the optimum pH value was between neutral and alkaline. Mn2+ and cobalt (Co2+) are two common dependent ions, and a few of them have been demonstrated to possess reliable thermal stability and extensive substrate profiles [23], [24], [25], [26].

Currently, the three-dimensional (3D) structures of L-Rhi from Escherichia coli, Pseudomonas stutzeri, and Bacillus halodurans have been reported [27], [28], [29]. Most of them are tetramer structures, which are dependent on metal ions. It has been observed that the activity of L-Rhi was higher in comparison to other enzymes; however, its substrate specificity during the process of L-Rhi isomerization of D-psicose into D-allose was not high. Therefore, its conversion rate could not meet the production demands of D-allose.

To increase the affinity of L-Rhi for D-psicose and the conversion rate of D-allose, L-Rhi from B. subtilis was overexpressed in E. coli BL21. Initially, SWISS-MODEL and PDB database screening were used to construct a 3D structure of L-Rhi from B. subtilis. Following that, alanine scanning, directed evolution of saturated mutations, and structural analysis were employed to design a library of mutants that was conducive to increasing the conversion rate of allose; subsequently, enzyme mutants that significantly improved the characteristics of enzymes and the conversion rate of allose were further screened. Finally, the MROMAS software was utilized for mutant enzyme kinetic analysis.