One-Pot Determination Of Transcriptional Complexes By CDNA Display And Proximity Ligation

All biological functions are dependent on transcriptional regulation, which establishes complex relationships between genotype and phenotype. Therefore, investigation of the mechanisms underlying transcriptional regulation is at the core of the life sciences. Transcriptional complexes play a key role in the transcriptional regulation process. To clarify the mechanisms of transcription, the complicated protein-protein and protein-DNA interactions(PPI and PDI) involved in transcriptional complexes must be determined. Various methods, including high-throughput-sequencing-based techniques, have been developed in recent decades to determine PPI or PDI. However, most of these methods measure the congeneric physical interactions between interaction pairs, which can not be used to determine PPI and PDI synchronously. Because the conformation and composition of transcription factors(TFs) are dependent on DNA binding, a precise elucidation of transcriptional regulation mechanism needs the determination of PPI/PDI in transcriptional complexes. Together, it is still a great challenge to parallel analyse the composition of transcriptional complexes in a single experiment. This difficulty basically arises from the different principles and pipelines for protein and DNA sequencing.In this study, using a combination of cDNA display and proximity ligation, we established a method to determine the PPI/PDI of transcriptional complexes in one pot, which we have called the One-Pot-seq method. The pipeline for this method is as follows. First, the target proteins are covalently linked with their corresponding DNA oligonucleotides using cDNA display, forming barcoded proteins. Barcoded proteins are then incubated with the bait DNA and enriched using an iterative selection. The screened protein-protein and protein-DNA complexes are further enriched by solid-phase purification and joined via in situ blunt end ligation. The captured ligation products are amplified by in situ nested PCR and analysed by DNA sequencing to decode the interacting PPI/PDI complexes. In order to expand application scope, solid phase and solution phase two different selection methods were designed according to pull down principle and subtractive differential electrophoretic mobility shift assay, respectively.In this study, first to examine the feasibility of the One-Pot-seq method, a proof-of-concept experiment was performed to determine the cognate binding DNA fragments of the c-Fos/c-Jun complex. c-Fos/c-Jun are the chief constituents of the activator protein-1(AP-1) complex(belonging to the bZIP transcription factor family). After one round of selection, the PCR products were cloned and randomly chosen for sequencing, the results were used to define the PPI/PDI involved in the transcriptional complex. The results showed that all of the ligated sequences corresponded to c-Fos-c-Jun, c-Fos-DNA and c-Jun-DNA and lacked GST sequences, bait DNA mutants or c-Jun-c-Jun ligation. The absence of c-Jun-c-Jun ligation products is consistent with a previous report that the c-Fos-c-Jun heterodimer, rather than the c-Jun-c-Jun homodimer, binds DNA. Therefore, the combined use of cDNA display and proximity ligation provides an in vitro selection platform to determine transcriptional complexes in one pot.Second, to evaluate the feasibility of the present methodology used to as a platform for library to library selection. The c-Fos/c-Jun transcriptional system bearing both cis and trans mutations was studied. The final captured ligation products were amplified and cloned. Of the ligated sequences, the wild-type PPI/PDI interactions are predominant, with 23% corresponded to c-Fos-c-Jun, 31% corresponded to c-Fos-DNA and 33% corresponded to c-Jun-DNA, indicating that the wide-type c-Fos/c-Jun-DNA complex is most stable. This result strongly suggests that we achieve the effective selection with One-Pot-seq in the(2 × 1012 trans mutations × 2 × 1013 cis mutations) library vs. library format. This technique holds the promise of direct molecular counting of all-by-all pairwise or even higher-order interactions in a complex mixture.Last, we applied the One-Pot-seq selection system to the selection of TRE-binding proteins from a mouse liver barcode-protein library. After 4 rounds solid selection, the resulting librariy and control library were cloned, and then 100 and 20 randomly chosen clones were sequenced, respectively. The results showed that we capture ten interaction pairs, 9 of which were confirmed with reference. This implies that the One-Pot-seq selection system affords a low false-positive rate and this selection system can be used to detect the “biological” interaction.