Loss of Dok-3 in Non-tumor Cells Induces Malignant Transformation of Benign Epithelial Tumor Cells of the Intestine.

The fundamental difference between benign and malignant tumors lies in their invasive ability. It is believed that malignant conversion of benign tumor cells is induced by a tumor cell-intrinsic accumulation of driver gene mutations. Here, we found that disruption of the Dok-3 tumor suppressor gene led to malignant progression in the intestinal benign tumor model ApcMin/+ mice. However, Dok-3 gene expression was undetectable in epithelial tumor cells and the transplantation of bone marrow cells lacking the Dok-3 gene-induced malignant conversion of epithelial tumor cells in ApcMin/+ mice, indicating a previously unrecognized tumor cell-extrinsic mechanism. Moreover, the Dok-3 loss-induced tumor invasion in ApcMin/+ mice required CD4+ and CD8+ T lymphocytes, but not B lymphocytes. Finally, whole-genome sequencing showed an indistinguishable pattern and level of somatic mutations in tumors irrespective of the Dok-3 gene mutation in ApcMin/+ mice. Together, these data indicate that Dok-3 deficiency is a tumor-extrinsic driving force of malignant progression in ApcMin/+ mice, providing a novel insight into microenvironments in tumor invasion. Significance: This study uncovers tumor cell-extrinsic cues that can induce malignant conversion of benign tumors without intensifying mutagenesis in tumors, a novel concept potentially providing a new therapeutic target in malignancy.

Loss of C/EBPδ enhances apoptosis of intestinal epithelial cells and exacerbates experimental colitis in mice.

Inflammatory bowel diseases (IBDs) are characterized by chronic inflammation involving intestinal tissue damage, which include ulcerative colitis and Crohn's disease as major entities. Accumulating evidence suggests that excessive apoptosis of intestinal epithelial cells (IECs) contributes to the development of IBD. It was recently reported that the transcription factor CCAAT/enhancer-binding protein delta (C/EBPδ) is involved in inflammation; however, its role in colitis remains unclear. Here, we found that C/EBPδ knockout mice showed enhanced susceptibility to dextran sodium sulfate (DSS)-induced colitis, a mouse model of IBD, which was associated with severe colonic inflammation and mucosal damage with increased IEC apoptosis. Additionally, DSS stimulation induced increased expression of pro-apoptotic BH3-only protein Bim in the colon of C/EBPδ knockout mice. Collectively, our findings demonstrate that C/EBPδ plays an essential role in suppressing DSS-induced colitis, likely by attenuating IEC apoptosis.

Amorphous protein aggregation monitored using fluorescence self-quenching.

Biophysical understanding of amorphous protein aggregation can significantly impact diverse area of biotechnology. Here, we report the time dependent salt-induced formation of amorphous aggregation as monitored by fluorescence self-quenching and compare the results with conventional methods for detecting protein aggregation [static light scattering (LS) and dynamic light scattering (DLS)]. As a model protein, we used a bovine pancreatic trypsin inhibitor (BPTI) variant extended by two glycines (C2G) at its C terminus, and three variants where three types of Solubility Controlling Peptide tags (SCP tags) made of five serines (C5S), alanines (C5A) or aspartic acids (C5D) were added to the C terminus of C2G. All variants have a native-like BPTI structure and trypsin inhibitory activity, but different solubilities controlled by the SCP tags. The BPTIs were labeled using NHS-Fluorescein (FAM) conjugated to BPTI's lysines, and we measured the changes in fluorescence intensity occurring upon the addition of NaCl. The fluorescence of all FAM-BPTIs decreased almost immediately, albeit to a different extent, upon addition of salt and became constant after 10 min for 24 h or more. On the other hand, LS and DLS signal changes were dependent on the type of tags. Namely, C2G's LS and DLS signals changed immediately, the signals of C5S and C5A tagged FAM-BPTIs increased slowly from 10 min to 24 h, and those of C5D remained constant. These observations indicated the presence of at least one intermediate step, with increased protein-protein interaction yielding a 'molecular condensation' phase. According to this model, C2G would rapidly turn from 'condensates' to aggregates, whereas C5S and C5A tagged FAM-BPTIs would do so slowly, and the soluble C5D tagged variant would remain in the molecular condensation state.