Abstract
In practical applications of protein splicing, intein domain can be splitted into two building blocks which respectively conjugate with their exteins and are defined as an amino-terminal fragment (ExtN-IntN) and a carboxyl-terminal one (IntC-ExtC). The two fragments reconstitute spontaneously via inter-fragment complementation of split intein. If the splicing activity of intein can be recovered, the flanking exteins will be spliced in trans, which is referred to protein transsplicing (PTS). Various native and engineered split inteins have been exploited in different fields, especially in protein semi-synthesis and segmental isotope labeling in NMR. Nonetheless, how to efficiently design the split inteins with great trans-splicing activity, this problem remains hitherto unsolved. In terms of this perspective, we developed a strategy to systematically identify novel split sites on Nostoc punctiforme (Npu) dnaE intein, a sophisticated intein with extraordinary splicing and trans-splicing activity. Because few split sites identified in Npu dnaE intein, protein circular permutation (CP) prediction is employed as part of strategy to screen the new split sites. A newly introduced disconnection based on CP prediction provided intein CP variants with highly structural identity as to native intein and great structural stability. These CP sites implied potential split sites in split intein design, which preserve protein trans-splicing function. We incorporated in silico intein CP prediction with experimental verification to facilitate the search of proper intein split sites. Comparing the characterizations between the native intein and its distinct CPs, we offer a rational strategy about how to design a viable split intein to increase the versatility in protein trans-splicing application.