Int&in: A machine learning-based web server for active split site identification in inteins

Protein trans-splicing: optimization of intein-mediated GFP assembly as a model for the development of gene therapy

Adeno-associated virus (AAV)-based gene therapy has become one of the key directions of modern translational medicine geared towards treatment of hereditary disorders by means of gene replacement. At the moment, about 5,000 different syndromes are associated with mutations in large genes, which presents a great problem due to the AAV packaging capacity of 5 kilobases. The main strategies for overcoming this obstacle were the creation of truncated gene versions, overloading the viral vector, and separate delivery of partial genetic material to restore the whole gene at the level of DNA, RNA, or protein. At present, genome editing via prime editors, most effectively delivered by AAV, relies on the intein pair used to restore the protein complex. The amazing integration speed of intein-based protein trans splicing technology makes it a versatile tool for a variety of applications, albeit not always successful on the first attempt. This study discusses the key points of working with Ssp, Npu, and Ava inteins of the DnaE group, known as the most effective for assembly of large proteins. Using green fluorescent protein (GFP) as a model, we demonstrate that the successful protein assembly requires not only cysteine at position C+1 but also certain aminoacid residues on either side in its immediate environment. Furthermore, the conformation of extein-intein composition, difficult to predict by computer modeling, has an additional effect, as demonstrated by experimental tests of the three split sites optimal in amino acid composition. The NpuDnaE variant demonstrated the highest kinetics of interaction between the N and C parts in the DnaE group of inteins. Optimization of conditions using NpuDnaE intein led to GFP assembly in 80% of transfected HEK293 cells and in 55% of AAV5-transduced cells, as demonstrated by flow cytometry. The efficiency of GFP assembly post-plasmid DNA transfection or AAV transduction of the HEK293 cell line was 15% higher than that of the ARPE19 cell line. We hope that the obtained data will facilitate the development of gene therapies for the treatment of hereditary disorders caused by mutations in large genes.

Int&in: A machine learning-based web server for active split site identification in inteins

Inteins are proteins that excise themselves out of host proteins and ligate the flanking polypeptides in an auto-catalytic process called protein splicing. In nature, inteins are either contiguous or split. In the case of split inteins, the two fragments must first form a complex for the splicing to occur. Contiguous inteins have previously been artificially split in two fragments because split inteins allow for distinct applications than contiguous ones. Even naturally split inteins have been split at unnatural split sites to obtain fragments with reduced affinity for one another, which are useful to create conditional inteins or to study protein-protein interactions. So far, split sites in inteins have been heuristically identified. We developed Int&in, a web server freely available for academic research (https://intein.biologie.uni-freiburg.de) that runs a machine learning model using logistic regression to predict active and inactive split sites in inteins with high accuracy. The model was trained on a dataset of 126 split sites generated using the gp41-1, Npu DnaE and CL inteins and validated using 97 split sites extracted from the literature. Despite the limited data size, the model, which uses various protein structural features, as well as sequence conservation information, achieves an accuracy of 0.79 and 0.78 for the training and testing sets, respectively. We envision Int&in will facilitate the engineering of novel split inteins for applications in synthetic and cell biology.