Search

Background information. The RNA-binding protein S1-1, also called RBM10 (RNA-binding motif 10), is a paralogue of putative tumour suppressor RBM5 and has been correlated with cancer proliferation and apoptosis. In the present study, we have investigated the cell biology of S1-1.Results. In the extranucleolar nucleoplasm, S1-1 occurred in hundreds of punctate and irregular domains. Some 10–40 of these domains were larger than 0.5 μm and prominent for S1-1 immunostaining. These domains (S1-1 nuclear bodies) were commonly present in tissue cells and in cultured cells. When cellular transcription was globally reduced by heat shock, serum starvation, culture at high cell densities or inhibition with RNA polymerase II inhibitors, small S1-1 domains (S1-1 granules), with weak immunostaining signals, reduced in number, whereas S1-1 nuclear bodies became prominent and increased in size. These altered S1-1 domains were returned to initial states when the cells were placed under normal conditions. Similar to paraspeckles, S1-1 nuclear bodies occurred closely adjacent to nuclear speckles or IGCs (interchromatin granule clusters), as determined by immunoelectron microscopy. However, the S1-1 nuclear bodies did not correspond to paraspeckles or IGAZs (interchromatin-granule-associated zones), but coincided with TIDRs (transcription-inactivation-dependent RNA domains), which we had characterized previously at the RNA level. The enlarged S1-1 nuclear bodies/TIDRs accumulated the S1-1 protein and microinjected primary and spliced mRNAs, presumably for later elevation of gene expression. In addition, electron microscopy revealed that S1-1 was also present on perichromatin fibrils, suggesting the structure of S1-1 granules seen at higher resolution.Conclusions. S1-1 constitutes hundreds of nuclear domains, which dynamically change their structures in a reversible manner. Upon globally reducing RNA polymerase II transcription, S1-1 nuclear bodies enlarge and decrease in number. They are novel domains different from paraspeckles or IGAZs, despite their similar occurrence adjacent to nuclear speckles. We discuss S1-1 granules in terms of their association with gene expression. In addition, this is the first report of a TIDR-localized protein. [ABSTRACT FROM AUTHOR]

Aberrant carbohydrate expression frequently occurs in breast cancer and may endow cells with metastatic potential. Here we first studied the relationship between expression of Vicia villosa agglutinin (lectin) (VVA)-binding carbohydrates and aggressive breast cancer. We then investigated the molecular characteristics of these glycoproteins and compared them with those of glycoproteins recognized by the mouse anti-Tn monoclonal antibody (MAb) HB-Tn1. Histochemical studies of samples from 322 cases of invasive ductal carcinoma demonstrated that VVA-binding carbohydrate expression correlated with tumor stage, lymphatic invasion, and lymph node metastasis (p=0.0385, p=0.0019, and p=0.0430. respectively). Western blotting analysis of frozen materials from 39 cases, under denaturing and reducing conditions, revealed that the major cancer cell-specific VVA-binding proteins were molecules of about 30, 33, and >200 kDa. Cases expressing the ∼33 kDa molecule had significant lymphatic invasion more frequently than did cases not expressing this molecule (p=0.0076). Binding of VVA to the ∼30 and ∼33 kDa molecules was completely lost by preincubation of VVA with 1 mM Tn antigen (N-acetylgalactosamine α1-O-serine). The VVA-binding molecules appeared to react with VU-3C6 anti-MUC1 MAb. Expression of HB-Tn1 in breast cancer cells showed significant correlation with expression of VVA-binding carbohydrate(s) (p<0.0001) but HB-Tn1 reactivity was not clearly related to breast cancer aggressiveness. Because anti-Tn MAbs bound to Tn antigen clusters, we concluded that atypical MUC1 bearing the noncluster form of Tn antigen is implicated in aggressive growth of primary breast cancer cells, particularly in lymphatic metastasis. [ABSTRACT FROM AUTHOR]

Summary MUC1 is a transmembrane molecule characterized by a repeated sequence of 20 amino acid (TAP PAHGVTSAPDTRPAPGS). Abnormal overexpression of MUC1 in cancer cells is thought to contribute to their aggressive growth, but molecular mechanisms associated with this effect are still unclear. Our current study aimed to clarify whether MUC1 expression as recognized by VU-3C6 anti-MUC1 mouse IgG monoclonal antibody (MAb) with a dominant epitope of 12 amino acids: GVTSAPDTRPAP, correlated with aggressive properties of human breast cancer. Immunohistochemical studies of 309 samples of formalin-fixed and paraffin-embedded materials showed no statistical correlation between MUC1 expression and clinicopathological parameters, as well as several breast cancer aggressiveness-related markers. Expression or nonexpression of MUC1 in 50 frozen samples, as determined by Western blotting, demonstrated no correlation with aggressive properties of breast cancer. However, samples with one MUC1-positive band more often had lymphatic vessel invasion and lymph node metastasis than those with more than two or three MUC1-positive bands (p<0.014 and p<0.043, respectively). Because VU-3C6 MAb recognizes MUC1 with short branches of O-glycosylated core carbohydrates, we used immunohistological methods to examine Tn antigen (precursor antigen: GalNAcα-O-Ser/Thr), Thomsen–Friedenreich (T) antigen, and sialyl-Tn antigen (STn) antigen. We found a strong correlation between expression of MUC1 and Tn antigen (p<0.0006), and samples with Tn antigen expression had more lymphatic metastasis than those with no such expression (p<0.08). We concluded that the lack of polymorphic MUC1 expression with Tn antigen is one characteristic related to aggressive breast cancer. [ABSTRACT FROM AUTHOR]

The purpose of this study was to develop a computer-aided diagnosis (CAD) system for the classification of malignant and benign masses in the breast using ultrasonography based on a convolutional neural network (CNN), a state-of-the-art deep learning technique. We explored the regions for the correct classification by generating a heat map that presented the important regions used by the CNN for human malignancy/benign classification. Clinical data was obtained from a large-scale clinical trial previously conducted by the Japan Association of Breast and Thyroid Sonology. Images of 1536 breast masses (897 malignant and 639 benign) confirmed by pathological examinations were collected, with each breast mass captured from various angles using an ultrasound (US) imaging probe. We constructed an ensemble network by combining two CNN models (VGG19 and ResNet152) fine-tuned on balanced training data with augmentation and used the mass-level classification method to enable the CNN to classify a given mass using all views. For an independent test set consisting of 154 masses (77 malignant and 77 benign), our network showed outstanding classification performance with a sensitivity of 90.9% (95% confidence interval 84.5–97.3), a specificity of 87.0% (79.5–94.5), and area under the curve (AUC) of 0.951 (0.916–0.987) compared to that of the two CNN models. In addition, our study indicated that the breast masses themselves were not detected by the CNN as important regions for correct mass classification. Collectively, this CNN-based CAD system is expected to assist doctors by improving the diagnosis of breast cancer in clinical practice. [ABSTRACT FROM AUTHOR]