Akira Tanigami

Automated extraction of information on protein–protein interactions from the biological literature

MOTIVATION To understand biological process, we must clarify how proteins interact with each other. However, since information about protein-protein interactions still exists primarily in the scientific literature, it is not accessible in a computer-readable format. Efficient processing of large amounts of interactions therefore needs an intelligent information extraction method. Our aim is to develop an efficient method for extracting information on protein-protein interaction from scientific literature. RESULTS We present a method for extracting information on protein-protein interactions from the scientific literature. This method, which employs only a protein name dictionary, surface clues on word patterns and simple part-of-speech rules, achieved high recall and precision rates for yeast (recall = 86.8% and precision = 94.3%) and Escherichia coli (recall = 82.5% and precision = 93.5%). The result of extraction suggests that our method should be applicable to any species for which a protein name dictionary is constructed. AVAILABILITY The program is available on request from the authors.

Assessment of prediction accuracy of protein function from protein--protein interaction data.

Functional prediction of open reading frames coded in the genome is one of the most important tasks in yeast genomics. Among a number of large‐scale experiments for assigning certain functional classes to proteins, experiments determining protein–protein interaction are especially important because interacting proteins usually have the same function. Thus, it seems possible to predict the function of a protein when the function of its interacting partner is known. However, in vitro experiments often suffer from artifacts and a protein can often have multiple binding partners with different functions. We developed an objective prediction method that can systematically include the information of indirect interaction. Our method can predict the subcellular localization, the cellular role and the biochemical function of yeast proteins with accuracies of 72.7%, 63.6% and 52.7%, respectively. The prediction accuracy rises for proteins with more than three binding partners and thus we present the open prediction results for 16 such proteins. Copyright

A radiation hybrid map of the rat genome containing 5,255 markers

A whole-genome radiation hybrid (RH) panel was used to construct a high-resolution map of the rat genome based on microsatellite and gene markers. These include 3,019 new microsatellite markers described here for the first time and 1,714 microsatellite markers with known genetic locations, allowing comparison and integration of maps from different sources. A robust RH framework map containing 1,030 positions ordered with odds of at least 1,000:1 has been defined as a tool for mapping these markers, and for future RH mapping in the rat. More than 500 genes which have been mapped in mouse and/or human were localized with respect to the rat RH framework, allowing the construction of detailed rat-mouse and rat-human comparative maps and illustrating the power of the RH approach for comparative mapping.

Macular corneal dystrophy type I and type II are caused by distinct mutations in a new sulphotransferase gene

Macular corneal dystrophy (MCD; MIM 217800) is an autosomal recessive hereditary disease in which progressive punctate opacities in the cornea result in bilateral loss of vision, eventually necessitating corneal transplantation. MCD is classified into two subtypes, type I and type II, defined by the respective absence and presence of sulphated keratan sulphate in the patient serum, although both types have clinically indistinguishable phenotypes. The gene responsible for MCD type I has been mapped to chromosome 16q22, and that responsible for MCD type II may involve the same locus. Here we identify a new carbohydrate sulphotransferase gene (CHST6), encoding an enzyme designated corneal N-acetylglucosamine-6-sulphotransferase (C-GlcNAc6ST), within the critical region of MCD type I. In MCD type I, we identified several mutations that may lead to inactivation of C-GlcNAc6ST within the coding region of CHST6. In MCD type II, we found large deletions and/or replacements caused by homologous recombination in the upstream region of CHST6. In situ hybridization analysis did not detect CHST6 transcripts in corneal epithelium in an MCD type II patient, suggesting that the mutations found in type II lead to loss of cornea-specific expression of CHST6.

Rox, a novel bHLHZip protein expressed in quiescent cells that heterodimerizes with Max, binds a non‐canonical E box and acts as a transcriptional repressor

Proteins of the Myc and Mad family are involved in transcriptional regulation and mediate cell differentiation and proliferation. These molecules share a basic‐helix–loop–helix leucine zipper domain (bHLHZip) and bind DNA at the E box (CANNTG) consensus by forming heterodimers with Max. We report the isolation, characterization and mapping of a human gene and its mouse homolog encoding a new member of this family of proteins, named Rox. Through interaction mating and immunoprecipitation techniques, we demonstrate that Rox heterodimerizes with Max and weakly homodimerizes. Interestingly, bandshift assays demonstrate that the Rox–Max heterodimer shows a novel DNA binding specificity, having a higher affinity for the CACGCG site compared with the canonical E box CACGTG site. Transcriptional studies indicate that Rox represses transcription in both human HEK293 cells and yeast. We demonstrate that repression in yeast is through interaction between the N–terminus of the protein and the Sin3 co‐repressor, as previously shown for the other Mad family members. ROX is highly expressed in quiescent fibroblasts and expression markedly decreases when cells enter the cell cycle. Moreover, ROX expression appears to be induced in U937 myeloid leukemia cells stimulated to differentiate with 12‐O‐tetradecanoylphorbol‐13‐acetate. The identification of a novel Max‐interacting protein adds an important piece to the puzzle of Myc/Max/Mad coordinated action and function in normal and pathological situations. Furthermore, mapping of the human gene to chromosome 17p13.3 in a region that frequently undergoes loss of heterozygosity in a number of malignancies, together with the biochemical and expression features, suggest involvement of ROX in human neoplasia.

Isolation of a novel human lung‐specific gene, LUNX, a potential molecular marker for detection of micrometastasis in non‐small‐cell lung cancer

We have isolated a novel human lung‐specific gene, LUNX (lung‐specific X protein), by differential‐display mRNA analysis. The full‐length cDNA contained 1,015 nucleotides including an open reading frame of 768 nucleotides encoding 256 amino acids. We localized the gene to chromosomal region 20p11.1‐q12 by radiation hybrid mapping. Using an RT‐PCR assay specific for LUNX mRNA, 35 non‐small‐cell lung‐cancer (NSCLC) tumors and 0 of 16 normal lymph nodes were positive. Furthermore, LUNX mRNA expression was enhanced in 26 (84%) of 31 NSCLC tumors vs. corresponding cancer‐free lung tissues by semi‐quantitative analyses with multiplex RT‐PCR. We assessed the possibility of LUNX mRNA as a molecular marker for detection of micrometastasis in dissected lymph nodes obtained from 20 patients with NSCLC tumors. LUNX mRNA was detected in 16 (80%) of 20 histologically positive lymph nodes and 21 (25%) of 84 histologically negative lymph nodes. Comparative analyses of the conventional histological examination and the RT‐PCR detection assay for LUNX mRNA showed that the detection rate of metastases in lymph nodes by the RT‐PCR assay was higher in 12 and consistent in 6 of the total 20 NSCLC patients. We demonstrate that the LUNX RT‐PCR assay is a potential diagnostic method for detection of micrometastases in lymph nodes of NSCLC patients.

Embryonic Stem Cells form Articular Cartilage, Not Teratomas, in Osteochondral Defects of Rat Joints

Embryonic stem (ES) cells are considered to be a potential tool for repairing articular cartilage defects, but so far it has been impossible to cause these cells to differentiate into chondrocytes exclusively, either in vivo or in vitro. To explore a potential new cell source of cell transplantation for articular cartilage defects, we transplanted ES cells into articular cartilage defects in immunosuppressed rats. ES cells (AB2.2 or CCE cells) were transplanted into articular cartilage defects in the patellar groove of immunosuppressed rats treated with cyclosporine. The cells were histologically observed until 8 weeks after transplantation. To determine whether the repair tissue in the defect in the AB2.2-transplanted group was derived from the transplanted cells, the neomycin-resistant gene, which had been transfected into AB2.2 cells but does not exist in rat cells, was used for detection. The cells produced cartilage, resulting in repair of the defects from 4 weeks until 8 weeks after the transplantation without forming any teratomas. The neomycin-resistant gene was detected in every sample, demonstrating that the repair tissue in the AB2.2-transplanted group was derived from the transplanted AB2.2 cells. The environment of osteochondral defects is chondrogenic for ES cells. ES cells may thus be a potential tool for repairing articular cartilage defects.

OASIS is a transcriptional activator of CREB/ATF family with a transmembrane domain

Murine OASIS is a putative CREB/ATF family transcription factor that is induced in gliosis, but its molecular role has not been determined. We have isolated the human OASIS gene and investigated the potential of OASIS protein as a transcriptional activator. We found that OASIS can activate transcription through box-B elements but not through the somatostatin CRE. OASIS contains a putative C-terminal hydrophobic transmembrane domain, a typical structural feature for the transcription factors activated by regulated intramembrane proteolysis. Truncation of the OASIS transmembrane domain resulted in a significant increase in transcriptional activity and altered its subcellular localization from the endoplasmic reticulum to the nucleus. Western blot analysis of transfected cells identified OASIS polypeptides of 82 and 66 kDa. These results suggest that the transmembrane domain plays an important role in the regulation of transcriptional activation by OASIS.

Overexpression of LAMP3/TSC403/DC-LAMP Promotes Metastasis in Uterine Cervical Cancer

LAMP3 (DC-LAMP, TSC403, CD208) was originally isolated as a gene specifically expressed in lung tissues. LAMP3 is located on a chromosome 3q segment that is frequently amplified in some human cancers, including uterine cervical cancer. Because two other members of the LAMP family of lysosomal membrane glycoproteins, LAMP1 and LAMP2, were previously implicated in potentially modulating the interaction of vascular endothelial and cancer cells, we hypothesized that LAMP3 might also play an important part in metastasis. To clarify the metastatic potential of LAMP3 in cervical cancers, we transfected a LAMP3 expression vector into a human uterine cervical cancer cell line, TCS. In an in vitro invasion assay, the migration of LAMP3-overexpressing TCS cells was significantly higher than in control TCS cells. In an in vivo metastasis assay, distant metastasis was detected in 9 of 11 LAMP3-overexpressing TCS cell-injected mice and in only 1 of 11 control mice. Histologic study showed that LAMP3-overexpressing cells readily invaded into the lymph-vascular space. In clinical samples, quantitative real-time reverse transcription-PCR (RT-PCR) analyses showed that LAMP3 mRNA was significantly up-regulated in 47 of 47 (100%) cervical cancers and in 2 of 15 (13%) cervical intraepithelial neoplasias, compared with a low level of LAMP3 mRNA expressed in normal uterine cervixes. Interestingly, high LAMP3 expression was significantly correlated with the overall survival of patients with stage I/II cervical cancers. These findings indicate that LAMP3 overexpression is associated with an enhanced metastatic potential and may be a prognostic factor for cervical cancer.

Matrix morphogenesis in cornea is mediated by the modification of keratan sulfate by GlcNAc 6-O-sulfotransferase

Matrix assembly and homeostasis in collagen-rich tissues are mediated by interactions with proteoglycans (PGs) substituted with sulfated glycosaminoglycans (GAGs). The major GAG in cornea is keratan sulfate (KS), which is N-linked to one of three PG core proteins. To ascertain the importance of the carbohydrate chain sulfation step in KS functionality, we generated a strain of mice with a targeted gene deletion in Chst5, which encodes an N-acetylglucosamine-6-O-sulfotransferase that is integral to the sulfation of KS chains. Corneas of homozygous mutants were significantly thinner than those of WT or heterozygous mice. They lacked high-sulfated KS, but contained the core protein of the major corneal KSPG, lumican. Histochemically stained KSPGs coassociated with fibrillar collagen in WT corneas, but were not identified in the Chst5-null tissue. Conversely, abnormally large chondroitin sulfate/dermatan sulfate PG complexes were abundant throughout the Chst5-deficient cornea, indicating an alteration of controlled PG production in the mutant cornea. The corneal stroma of the Chst5-null mouse exhibited widespread structural alterations in collagen fibrillar architecture, including decreased interfibrillar spacing and a more spatially disorganized collagen array. The enzymatic sulfation of KS GAG chains is thus identified as a key requirement for PG biosynthesis and collagen matrix organization.