PPI Network Of Transcription Factors Regulating Liver Regeneration

Liver is one of the most active regeneration organs of our body, and liver regeneration has been known by human for thousands of years, however still little is understood about the phenomenon. Generally, the process can be separated to three phases, namely initiation, regeneration and termination. In fact, the mechanisms lay behind the three phases are all mystery. The evolution of ideas pertaining to the mechanisms of liver regeneration may also be summarized into three phases: (1) the original view that a single hormonal agent could function as a key element, capable of unlocking all the events required for liver regeneration; (2) the idea that the activation of one pathway involving multiple components could be responsible for regeneration; and (3) the more recent idea that the activation of multiple pathways is required for liver regeneration to finish. Now, it has been known that liver regeneration is more complicated than originally expected, and it is a result of several signal pathways impacting on and interacting with each other. Guided by this view, new research focus had been shifted from traditional individual cell factor or pathway study to more attention on the combinatory effect of multi factors and pathways. A series publication on liver regeneration using gene-chip analysis, with attempt to learn the phenomenon at genomic level, had appeared in recent years, however, the systematic study of liver regeneration has a long way to go. This dissertation endeavored to shed light on the proteomic study of the field, in addition to the gene-chip technique, yeast two-hybrid method was also applied in this work, with a hope that we can start to draw a blue print for the regulation of liver regeneration through complicated and well orchestrated networks of protein-protein interactions.The result can be divided into following parts:1. Screen for liver regeneration associated mRNAs. Previously, a large-scale screening was performed for transcriptional factors that can either stimulate or repress liver cell growth in cell culture system. Totally, 129 ORFs were isolated from screening a human cDNA library (containing 29,910 genes), through transfecting individual ORF into liver cancer cell line SMMC7721 and scoring for colony formation. At the same time, a gene chip assay on liver regeneration was also carried out to find genes with their mRNA levels fluctuated during regeneration of mouse liver. From our study, the expression of protein correlates well to mRNA synthesis, it was thus considered that these factors with mRNA level changed during liver regeneration may play important roles in regulating the process. So, the common genes isolated from both screens may very well be associated with liver regeneration. Finally, 27 transcription factors and 5 kinases fulfill were selected for the following Y2H assay.2. Protein interaction network of transcription factors established: All the 32 ORFs were subcloned into both plasmids pGADT7 and pGBKT7, and then transformed into yeast AH109 and Y187 for use in Y2H system as prey and bait respectively. Self-activation phenotype was tested on appropriate amino acids-lacking medium plates and bait BD-HBP1, BD-STAT3, BD-FOXM1 and prey AD-FOXM1,AD-REA were excluded for the following yeast mating since they could activate report gene expression alone. Totally, more than 870 mating assays were performed to achieve one-to-one matrix hybrid screen. All four Y2H system reporter genes, namely HIS3,ADE2,and MEL(1α,β-gal), were applied to screen all the diploid phenotype, 64 protein pairs were demonstrated to be capable of activating at least one reporter gene expression and that all screen assays were repeated once. All the protein-protein interaction (PPI) were categorized intoⅠ,Ⅱ,ⅢandⅣgrades according to their degree of reporter gene activation. The four grades represented approximately the four levels of interaction strength and gradeⅠstand for relatively the weakest PPI strength and gradeⅣfor the strongest PPI. Based on those interactions, a PPI network of transcriptional factors for liver regeneration was initially established.3. Verification of the PPI network: In order to confirm the quality of this network, all 32 ORFs were subcloned into prokaryotic plasmids, pGST and pET, and eukaryotic plasmids, pCMV02 and pCMV03 from pCMVscript. All the expression plasmids were subjected to expression in prokaryotic and eukaryotic cells, and analyzed by SDS-PAGE and subsequent western blots. Co-immunoprecipitation and GST Pull-down were also used to verify some of the Y2H derived protein-protein interactions, as results, 11 out of 13 pairs were verified using alternative methods.4. Functional analysis of the PPI network during liver regeneration: The ultimate goal of building a protein-protein interaction network for liver regeneration associated transcriptional factors, is to have further understanding of the process at proteome level. From the identification of all the possible interacting partners of STAT3 and ATF3, and what we know about the functions of these two proteins in regulating liver regeneration, it is predicted that ATF3 and STAT3 might play focal roles in connecting the network, and the net effects of the FHL2 to ATF3 and the FHL2 to STAT3 interactions are inhibitory to hepatic growth. In fact, almost all 32 transcripts that we used to build the PPI network exert growth inhibitory functions, which make this network an important mechanism to terminate hepatic proliferation at the later stages of liver regeneration. For the first time, we identified the interaction between FHL2 and ATF3, and believed its function is to repress cell growth, and the in vivo study of this interaction is underway to reveal its physiological impact on liver regeneration.5. Additional studies based on the PPI network: As the mapping of interaction domains is necessary to dissect and eventually to understand the mechanisms of protein-protein interaction, we performed YTH method to define PPI responsive domains for the network proteins. Various deletion mutants of FHL2, ATF3 and ZNF408, together with ZNF212, ID3 were constructed as both bait and prey. By using the same screening method as previously used, the half Lim domain of FHL2 and the transcriptional activation and repression domains of ATF3 were shown to be necessary for mediating the interaction between FHL2 and ATF3. For ZNF408 self-interaction, the middle zinc finger repeating domain was proven to be the interaction-responsive region.In summary, a PPI network of transcription factors associated with liver regeneration and hepatic growth was established in this dissertation. The quality of the network was confirmed by GST Pull-down and Co-IP assays and which was proven to be quite reliable. A possible mechanism for network controlled liver regeneration and further studies were proposed. This research sets foundation for PPI networking and new dimension of liver proteomic studies, with an ultimate goal of providing more scientific insights to the understanding of liver regeneration.