Protein Splicing Mechanisms And Applications

An intein is defined as a self-splicing element that exists as an in-frame protein fusion with two flanking protein fragments, called exteins. An intein is able to excise itself out of the precursor protein with the concomitant ligation of its extein sequences through a native peptide bond by a process termed protein splicing. In nature, three categories of inteins have been classified depending on their structure:canonical inteins, mini-inteins, and split-inteins. The sizes of currently identified inteins range from 134 to 1650 amino acid residues and show low sequence similarity. Protein splicing reaction depends on four highly conserved residues. These conserved residues are Ser or Cys at the first residue of the N-terminus of the intein (the '1'aa); the TXXH in the block B; the dipeptide His-Asn at the intein C-terminus; and Ser, Thr or Cys (the+1 aa) at the first residue of the C-extein. Thr has not been observed naturally at an intein N-terminus. The standard protein splicing mechanism consists of four nucleophilic displacement reactions, step1:N-O or N-S acyl rearrangement form a linear ester intermediate converting the peptide bond at the N-terminal splice junction to a (thio)ester bond involving the nucleophilic amino acid residue Cysl or Serl; step 2:formation of a branched ester intermediate by the attack of the nucleophilic residue Cys+1 or Thr+1, Ser+1 at the C-terminal splice junction; step 3:cyclization of the asparagine residue adjacent to the C-terminal splice junction resolves the branched intermediate yielding free intein and (thio)ester linked exteins; Step 4:a spontaneous (O/S)-N acyl shift restores the natural amide bond between the exteins.The protein splicing domain of inteins typically consists of 12β-strands and is shaped like a flattened disk, which brings the N- and C-terminal sequences of the intein into proximity in a catalytic pocket located near the center of the disk-like structure. The catalytic pocket of intein typically has two nucleophilic catalytic residues:one at the beginning of the intein (named position 1) and another at the beginning of the C-extein (named position+1). The mechanism of protein splicing requires the nucleophilic amino acid residue at both position 1 and position +1, however it was not clear whether or how the three different nucleophilic residues (Cys, Ser, and Thr) would work differently at these two positions. To use intein in a target protein of interest, one needs to choose an intein insertion site to have a nucleophilic residue at position +1, therefore it is desirable to know what nucleophilic residue(s) are preferred by different inteins. In order to answer these questions, in chapter 3 we began with a statistical analysis of known inteins, which showed an unequal distribution of the three nucleophilic residues at positions 1 and+1, and then subjected six different mini-inteins to site-directed mutagenesis to systematically test the functionality of the three nucleophilic residues at the two positions. At position 1, most natural inteins had Cys and none had Thr. When the Cys at position 1 of the six inteins was mutated to Ser and Thr, the splicing activity was abolished in all except one case. At position+1, Cys and Ser were nearly equally abundant in natural inteins, and they were found to be functionally interchangeable in the six inteins of this study. When the two positions were studied as 1/+1 combination, the Cys/Ser combination was abundant in natural inteins, whereas the Ser/Cys combination was conspicuously absent. Similarly, all of the six inteins of this study spliced with the Cys/Ser combination, whereas none spliced with the Ser/Cys combination. These findings have interesting implications on the mechanism of splicing and the selection of intein insertion sites, and they also produced two rare mini-inteins that could splice with Thr at position+1.Studies showed that these four steps in the splicing pathway occur independently of each other. Intein mutations or intein expression in non-native host conditions often disrupt the coordination of these steps, and instead of splicing yield side reactions that are incapable of forming the mature host protein. For example, without the coupling to transesterification, the hydrolysis of the ester bonds between the linear or branched ester intermediate releases the N-extein. However, C-cleavage product is produced when the C-terminal Asn cyclizes in the absence of the transesterification that transfers the N-extein to the C-extein. These cleavage reactions have been exploited in protein engineering and applied widely for other purposes. However, existing methods employing conventional, contiguous inteins often result in spontaneous cleavage and thus lower yields of protein product. In chapter 4 we report controllable cleavages using three engineered S1 split-inteins, Ter ThyX, Ssp GryB and Rma DnaB and four engineered S1 1 split-inteins, Ssp DnaX, Ter ThyX, Ssp GyrB and Rma DnaB. In controllable C-cleavage design, the C terminus of an S1-Ic sequence was fused with thioredoxin to form a precursor protein, and a synthetic S1-In peptide from the Ssp DnaB S1 split-intein was used together with DTT to trigger C-cleavage at the C-terminus of the Ic, where this peptide can trigger C-cleavage in all three inteins. In controllable N-cleavage design, the N terminus of an S11-IN sequence was fused with maltose binding protein to form a precursor protein, and a synthetic S1 1-IC peptide from the Ssp GyrB S1 1 split-intein was used together with DTT to trigger N-cleavage at the N-terminus of the IN, which both can trigger N-cleavage except for Ter ThyX and they have synergistic effect. These findings provide more options for protein purification using controllable intein N-cleavage and C-cleavage without spontaneous cleavage and give some clues about intein structure-function relationship.Previously reported showed that al44-aa C-terminal fragment of the Ssp DnaB SI split-intein, which was termed C-intein, was completely inactive when fused at its C-terminus to a target protein of interest. After expression and purification of the fusion protein, the missing 11-aa N-terminal fragment of the intein, which was termed N-intein, was added as a synthetic peptide in trans to trigger the C-cleavage reaction, resulting a separation of the C-intein and the target protein. It was concluded that the N-intein was required for the C-cleavage reaction (Asn cyclization at the C-terminus of C-intein). It was thought that the N-intein was needed to fill a structural void in the catalytic pocket formed by C-intein alone, although the N-intein was not known to participate directly in catalyzing the Asn cyclization. In our study of chapter 5, we unexpectedly discovered that a similarly constructed C-intein derived from a different intein, namely the Ssp DnaX intein, could undergo spontaneous C-cleavage without addition of its complementary N-intein. The missing N-intein sequence could be up to 15 aa long and contained the conserved intein motif A. We also show that the spontaneous C-cleavage activity of the C-intein could be decreased significantly by a mutation of the conserved Thr residue in the conserved intein motif B, although this Thr residue was not known to participate directly in catalyzing the Asn cyclization. These findings suggest a robust intein structure without the motif A and a larger role of the motif B in the third step of the protein splicing mechanism.Inteins and their protein splicing function have found many useful applications in protein research and biotechnology. Split-inteins capable of protein trans-splicing can be used to join separately produced polypeptides, including chemically synthesized peptides to form a mature protein of interest; segmental isotope labeling of proteins for NMR studies; in production of cyclic proteins or peptides, in some protein two-hybrid methods for detecting protein-protein interactions and sub-cellular protein localization. In chapter 6, we developed a protein-based tandem protein trans-splicing system to splice recombinant spider silk protein. We chose three different artificially SO split-inteins Rma DnaB, Ssp GyrB and Ter DnaE-3 and their N- and C-terminus halves as the protein splicing elements fused with spider silk protein. Though a denaturation/renaturation step before splicing occurs in vitro, we report production of recombinant Argiope trifasciata aciniform silk up to 256kDa, and we spun an 180kDa recombinant Argiope trifasciata aciniform silk into fibers, and characterization of the mechanical properties.We also used improved splicing efficiency intein Ssp GryB by directed envolution to splice RicinA protein. Ricin is a heterodimeric protein produced in the seeds of the castor oil plant. It consists of an A-Chain bound by a disulfide bond to a B-Chain. The B-subunit binds to glycoproteins on the surface of epithelial cells, enabling the A-subunit to enter the cell via receptor-mediated endocytosis. A-subunit inactivates ribosomal RNA by removing an adenine residue, thereby inhibiting protein synthesis. Some researchers have speculated about using ricins in the treatment of cancer, as a so-called "magic bullet" to destroy targeted cells. The major problem with ricin is that its native internalization sequences are distributed throughout the protein. If any of these native internalization sequences are present in a therapeutic, then the drug will be internalized by, and kill, untargeted epithelial cells as well as targeted cancer cells. In order to overcome this problem we seperated the whole RicinA chain to two inactive protein fragments and then fused with N and C terminals of split-intein individualy. Then cloned the fused genes to AAV vector and be genetically linked to a monoclonal antibody to target malignant cells recognized by the antibody. Now we have selected three mini-inteins having high splicing efficiency in vivo and one S11 split-intein having～50% splicing activity in vitro. The next step we hope to enhance the intein splicing efficiency and use this method to target the cancer cells and killed them.In this thesis we focus on the intein mechanism and application study. We studied the two nucleophilic catalytic residues at 1 and+1 sites of mini-inteins, the controllable cleavage of S1 and S11 split-inteins, the spontaneous C-cleavage of Ssp DnaX intein and successfully using split-intein trans-splicing system to splice spider silk protein and RicinA protein. These findings not only afford new insight of protein splicing also expand the useful of split-inteins.