Heegwon Shin

Triangulation of scattered data in 3D space

Abstract Triangular interpolants are widely used to construct smooth surfaces from scattered data in 3D. To apply smooth triangular interpolants, the input 3D points have to be triangulated. This paper presents an algorithmic procedure for 3D triangulation. Relevant algorithms and datastructures are described in detail. Existing methods of 2D triangulation are based on Thiessen polygonization, which is an optimal partitioning of a geometric domain. To obtain an optimal triangular grid on the domain, the max-min angle criterion is frequently used. As there is no domain for 3D points, a new criterion called the smoothness criterion is proposed.

Regulation of BC200 RNA‐mediated translation inhibition by hnRNP E1 and E2

The long noncoding RNA BC200 (brain cytoplasmic RNA, 200 nucleotides) acts as a translational modulator of local protein synthesis at dendrites. BC200 RNA has been shown to inhibit translation in vitro, but it remains unknown how this translation inhibition might be controlled in a cell. Here, we performed yeast three‐hybrid screening and identified hnRNP E1 and hnRNP E2 as BC200 RNA‐interacting proteins. We found that: these hnRNA proteins could restore BC200 RNA‐inhibited translation; BC200 RNA interacts with hnRNP E1 and E2 mainly through its unique 3′ C‐rich domain; and the RNA binding specificities and modes of the two proteins differed somewhat. Our results offer new insights into the regulation of BC200 RNA‐mediated translation inhibition.

Knockdown of BC200 RNA expression reduces cell migration and invasion by destabilizing mRNA for calcium-binding protein S100A11

ABSTRACT Although BC200 RNA is best known as a neuron-specific non-coding RNA, it is overexpressed in various cancer cells. BC200 RNA was recently shown to contribute to metastasis in several cancer cell lines, but the underlying mechanism was not understood in detail. To examine this mechanism, we knocked down BC200 RNA in cancer cells, which overexpress the RNA, and examined cell motility, profiling of ribosome footprints, and the correlation between cell motility changes and genes exhibiting altered ribosome profiles. We found that BC200 RNA knockdown reduced cell migration and invasion, suggesting that BC200 RNA promotes cell motility. Our ribosome profiling analysis identified 29 genes whose ribosomal occupations were altered more than 2-fold by BC200 RNA knockdown. Many (> 30%) of them were directly or indirectly related to cancer progression. Among them, we focused on S100A11 (which showed a reduced ribosome footprint) because its expression was previously shown to increase cellular motility. S100A11 was decreased at both the mRNA and protein levels following knockdown of BC200 RNA. An actinomycin-chase experiment showed that BC200 RNA knockdown significantly decreased the stability of the S100A11 mRNA without changing its transcription rate, suggesting that the downregulation of S100A11 was mainly caused by destabilization of its mRNA. Finally, we showed that the BC200 RNA-knockdown-induced decrease in cell motility was mainly mediated by S100A11. Together, our results show that BC200 RNA promotes cell motility by stabilizing S100A11 transcripts.

Biosynthesis of brain cytoplasmic 200 RNA

Brain cytoplasmic 200 RNA (BC200 RNA), a neuron-specific non-coding RNA, is also highly expressed in a number of tumors of non-neuronal origin. However, the biosynthesis of BC200 RNA remains poorly understood. In this study, we show that the efficient transcription of BC200 RNA requires both internal and upstream promoter elements in cancer cells. The transcription complex seems to interact with a broad range of sequences within the upstream 100-bp region. The cellular levels and half-lives of BC200 RNA were found to differ across various cancer cell types, but there was no significant correlation between these parameters. Exogenously expressed BC200 RNA had a shorter half-life than that observed for the endogenous version in cancer cells, suggesting that BC200 RNA might be protected by some limiting factor(s) in cancer cells. Transient transfection experiments showed that the transcriptional activity of the exogenous BC200 RNA promoter element varied depending on the cancer cell type. However, the promoter activities together with the half-life data could not explain the differences in the levels of BC200 RNA among different cell types, suggesting that there is another level of transcriptional regulation beyond that detected by our transient transfection experiments.

Identifying the cellular location of brain cytoplasmic 200 RNA using an RNA-recognizing antibody

Brain cytoplasmic 200 RNA (BC200 RNA) is a neuron-specific non-coding RNA, implicated in the inhibition of local synaptodendritic protein synthesis, and is highly expressed in some cancer cells. Although BC200 RNA has been shown to inhibit translation in vitro, the cellular location of this inhibition is unknown. In this study, we used a BC200 RNA-recognizing antibody to identify the cellular locations of BC200 RNA in HeLa cervical carcinoma cells. We observed punctate signals in both the cytoplasm and nucleus, and further discovered that BC200 RNA co-localized with the p-body decapping enzyme, DCP1A, and the heterogeneous nuclear ribonucleoprotein E2 (hnRNP E2). The latter is a known BC200 RNA-binding partner protein and a constituent of p-bodies. This suggests that BC200 RNA is localized to p-bodies via hnRNP E2.

Truncated SRA RNA derivatives inhibit estrogen receptor-α-mediated transcription

The steroid receptor RNA activator (SRA) is a long non-coding RNA (lncRNA) that acts as a putative coactivator for steroid receptor-mediated transcription. A recent study showed that SRA RNA can be structurally dissected into four domains comprising various secondary structures, but the contribution of each domain to the coactivation ability of SRA RNA was previously unknown. Here, we assessed the functional contributions of the various domains of SRA. We examined the effects of each domain on the coactivation of estrogen receptor-α (ERα)-mediated transcription of a luciferase reporter gene in HeLa cells. Then the detailed domain analysis was focused on domain III (D3) not only with the reporter gene in HeLa cells, but also with ERα-responsive genes in MCF7 breast cancer cells. Domain deletion analysis showed that the deletion of any domain decreased the luciferase activity, and that deletion of D3 caused the largest decrease. This D3 deletion effect was not recovered by co-expression of D3 alone; moreover, the expression of D3 fragments (particularly helices H15-H18, which are highly conserved across vertebrates) inhibited luciferase expression in HeLa cells. Moreover, a fragment containing helices H15-H18 reduced ERα-responsive gene expression in MCF7 breast cancer cells. Our findings indicate that D3 inhibited ERα-mediated transcription of a reporter gene in HeLa cells and that helices H15-H18, as a core element responsible for the D3-driven inhibition, reduced expression of ERα-responsive genes in breast cancer cells.