The chromatin architectural protein DEK : assessing its chromatin binding properties

Chromatin is a highly-ordered, yet dynamic structure that is essential in the regulation of all DNA-dependent processes, with the coordinated interplay of non-histone chromosomal factors playing important roles in its regulation. One such factor is the unique oncoprotein DEK, which has been implicated in the maintenance of heterochromatin via protein-protein-mediated mechanisms. However, given its distinct DNA binding and folding activities, it is highly possible that DEK is directly involved in the regulation of chromatin organization. Our results presented here demonstrate that DEK indeed binds to the entry/exit site of nucleosomal DNA in vitro, and that knocking down DEK expression in cells leads to general alterations in the nucleosomal repeat length of in cellular chromatin. As this aspect has not been studied before, multiple hypothesis-driven strategies were chosen to create “DNA-binding-dead” DEK mutants by site-directed mutagenesis. Though these initial approaches proved unsuccessful we found that expression of DEK in bacteria resulted in massive compaction of the bacterial genome due to DEK’s DNA-folding activity which resulted in a complete halt in bacterial growth. This fact in combination with random mutagenesis led to the development of a novel screening procedure, now termed bacterial growth inhibition screen (BGIS) that allowed for the selection of loss-of-function mutants, which were subsequently verified by biochemical analyses. By re-expression of one DNA-binding-dead mutant in newly created DEK knockout cells, first evidence was provided that the DNA binding and folding activities of DEK indeed are involved in the regulation of global chromatin structure.By subjecting other DEK domains with previously unidentified functions to the BGIS, we were able to identify and verify a hitherto unrecognized RNA-binding domain in DEK. Prompted by these findings, we next tested whether artificial interference with the biology of bacteria as a tool can also be exploited for investigation of other protein classes. Through these initial studies, we found that DNA- and RNA-binding/folding activities as well as enzymatic activities from several proteins indeed resulted in complete growth inhibition, suggesting a rather broad applicability of the BGIS. Despite some limitations to this screen, we believe it could aid in the discovery of functional domains in proteins with yet unknown functions. Together with the finding that enzyme-dead mutants did not affect bacterial growth, this opens up possibilities for establishing simple, quick and economical ways for high-throughput approaches for the purpose of drug screening using compound libraries.In conclusion, we established and validated a unique and surprisingly simple screen based on undirected interference with bacterial growth, which allowed us to gain deeper insights into the biology of DEK’s DNA binding activities in the cellular environment. Additionally, due to the seemingly wide applicability and easy implementation in the molecular biology laboratory, we believe that the BGIS may be a valuable addition to the toolbox of methods for the study of protein function