Inactivation of a human kinetochore by specific targeting of chromatin modifiers
Here I describe the construction and characterization of a new generation of human artificial chromosome that contains an array of DNA sequences that can be used to manipulate the chromosome in vivo and possibly in vitro. This HAC was originated in human fibrosarcoma HT1080 cells from a synthetic alphoid DNA containing an array of TetOperator sequences, cloned in a BAC-based vector. This synthetic ά-satellite DNA formed HACs that were stably maintained throughtout replication and segregation in HT1080 cells. However, I succeeded to also transfer and manipulate the alphoidtetO HAC into a HeLa-based hybrid cell line. The synthetic alphoidtetO HAC chromatin was similar to the chromatin at endogenous centromeric alphoid DNA. Importantly, the DNA sequences embedded in the synthetic HAC were accessible to targeting TetR-fused constructs in vivo. The alphoidtetO HAC could be successfully targeted with a number of TetR:fusion proteins without affecting its chromatin structure, kinetochore assembly and mitotic behaviour. However, the targeting of a transcriptional activator (tTA) inactivated the HAC synthetic alphoidtetO DNA in a fraction of transfected cells. Surprisingly, the targeting of the transcriptional repressor tTS, co-repressor KAP1 or the heterochromatin-associated protein HPIά severely inactivated the synthetic alphoidtetO kinetochore . In fact, upon targeting several inner and outer kinetochore proteins were delocalized from the alphoidtetO sequences. The dissociation of kinetochore proteins CENP-H and CENP-C appeared to precede that of CENP-A. The alphoidtetO HAC lacking inner kinetochore protein complexes showed mitotic defects including misalignment at the metaphase plate and defective anaphase segregation, ultimately being included in tiny DAPI-positive nano-nuclei in the cytoplasm. The transcriptional repressor tTS repressed the low levels of transcription from the alphoidtetO sequences. In addition, targeting of transcriptional repressors altered the HAC chromatin towards a more “closed”, heterochromatic conformation, as seen from the changes in histone tail modifications. Interestingly, the targeting of the histone methyltransferase EZH2 to the alphoidteto HAC showed a much milder inactivating activity compared to KAP1. Based on these results, I propose that the formation of HPI-type of heterochromatin or accumulation of HPIά to the centromeric regions could disrupt the association of constitutive kinetochore proteins to the underlying sequences. Centromeric alphoid sequences lacking a functional kinetochore structure then also loose the centromere-specific histone H3 variant CENP-A becoming definitively inactive. Alternatively, a basal transcriptional activity from centromeric sequences might be required for centromere functionality.