For PLAC1, to determine the basis for its extraordinarily selective tissue-specific expression, we have shown that the gene is expressed from two promoters, P1 and P2, spaced 105 Kilobases apart and is alternatively spliced. By cloning both promoters from mouse and human, defined the minimal promoter regions. The minimal promoter region binds nuclear receptors Retinoic Acid X Receptor alpha (RXR-), LXR-beta, and Steroidogenic factor 1 (SF1)/ Estrogen related receptor beta (ERR-) at specific sites and their binding has a positive effect stimulating transcription >10 fold, in the presence of their respective agonists. In a follow up publication, in Oncogenesis (2013), Plac1 expression in cancer cells was evaluated by a classical approach establishing cancer cell lines; SV40 mediated transformation of primary cells WI38 and IMR90 cells. We found that following SV40 mediated transformation the primary cells induced PLAC1 and a series of steps are catalyzed by Large T antigen encoded by SV40 early regions that modify Tp53 repressor properties normally bound to the promoter region such that it loses its repressive ability, bring about changes in chromatin from closed to open status facilitating Plac1 transcription. The transcription is then further stimulated in the presence of nuclear receptors and if an additional coactivator NCOA2 (nuclear receptor co-activator2) is present, it recruits RB, leading to additional up-regulation of the gene. Thus, we have defined a major way in which the gene is activated in cancer cells, which thereby provides a route to repress the gene activity. Currently, we have shown at the biochemical level that Plac1interacts with desmosomes. Desmosome function is itself complex, and we have continued to study the association with specific components of the membrane organelle. In related work on the regulation of FOXL2 gene, we have shown that FOXL2 mediates Col1a2 gene regulation. For rDNA structure analysis the cloning and analysis problems were resolved with collaborations with 2 other NIH groups. J.H. Kim and Vladimir Larionov at NCI created an advanced approach to cloning based on transformation associated recombination (TAR), that provided stable clones with up to 2 repeat units of rDNA; Adam Phillippy and Alex Dilthey at NHGRI adapted advanced long-read sequencing techniques (PacBio and Nanopore) to facilitate sequence recovery and assembly; and we supplied annotation and context for the analyses. The major findings thus far are that ribosomal DNA, and transcribed regions that included 5 and 3external transcribed sequences, and internal transcribed sequences, all of which are eliminated during assembly but are essential for formation of mature ribosomes harbor many variants, with a fraction of them deeply seated in human evolution. Thirteen clones, about 0.32-fold coverage (0.82 Mb) of the chromosome 21 rDNA complement, revealed a previously missed 2 kb tract, several palindromic structures, and over 300 variants; 85 variants fall in mature 18S/28S rRNA sequences. Palindromic breakpoints and >80% of 45S variant alleles were also found in independent whole-genome or RNA-Seq data, indicating that many variants are long established in human populations. We have developed an updated 44,838 bp rDNA reference sequence annotated with detected variants, suggesting a possible route to complete analysis of the rDNA component of the human genome. The large number of variants reveal more - and more universal - heterogeneity in human ribosomal DNA than previously considered, opening the possibility of corresponding variations in ribosome dynamics. Further, we have extended the study to rDNA units in chromosome 22 from mouse-human hybrid cell-line containing human chromosome 22, collected additional clones from chromosome 22, sequenced, assembled, annotated and submitted to Genbank. These isolates include the telomeric and centromeric borders flanking the rDNA repeats from chromosome 22.
