Impact of Sodium Thiosulfate on Prevention of Nephrotoxicities in HIPEC: An Ancillary Evaluation of Cisplatin-Induced Toxicities in Ovarian Cancer.

PURPOSE: Hyperthermic intraperitoneal chemotherapy (HIPEC) with cisplatin confers a survival benefit in epithelial ovarian cancer (EOC) but is associated with renal toxicity. Sodium thiosulfate (ST) is used for nephroprotection for HIPEC with cisplatin, but standard HIPEC practices vary. METHODS: A prospective, nonrandomized, clinical trial evaluated safety outcomes of HIPEC with cisplatin 75 mg/m2 during cytoreductive surgery (CRS) in patients with EOC (n = 34) and endometrial cancer (n = 6). Twenty-one patients received no ST (nST), and 19 received ST. Adverse events (AEs) were reported according to CTCAE v.5.0. Serum creatinine (Cr) was collected preoperatively and postoperatively (Days 5-8). Progression-free survival (PFS) was followed. Normal peritoneum was biopsied before and after HIPEC for whole transcriptomic sequencing to identify RNAseq signatures correlating with AEs. RESULTS: Forty patients had HIPEC at the time of interval or secondary CRS. Renal toxicities in the nST group were 33% any grade AE and 9% grade 3 AEs. The ST group demonstrated no renal AEs. Median postoperative Cr in the nST group was 1.1 mg/dL and 0.5 mg/dL in the ST group (p = 0.0001). Median change in Cr from preoperative to postoperative levels were + 53% (nST) compared with - 9.6% (ST) (p = 0.003). PFS did not differ between the ST and nST groups in primary or recurrent EOC patients. Renal AEs were associated with downregulation of metabolic pathways and upregulation of immune pathways. CONCLUSIONS: ST significantly reduces acute renal toxicity associated with HIPEC with cisplatin in ovarian cancer patients. As nephrotoxicity is high in HIPEC with cisplatin, nephroprotective agents should be considered.

PTEN functions by recruitment to cytoplasmic vesicles.

PTEN is proposed to function at the plasma membrane, where receptor tyrosine kinases are activated. However, the majority of PTEN is located throughout the cytoplasm. Here, we show that cytoplasmic PTEN is distributed along microtubules, tethered to vesicles via phosphatidylinositol 3-phosphate (PI(3)P), the signature lipid of endosomes. We demonstrate that the non-catalytic C2 domain of PTEN specifically binds PI(3)P through the CBR3 loop. Mutations render this loop incapable of PI(3)P binding and abrogate PTEN-mediated inhibition of PI 3-kinase/AKT signaling. This loss of function is rescued by fusion of the loop mutant PTEN to FYVE, the canonical PI(3)P binding domain, demonstrating the functional importance of targeting PTEN to endosomal membranes. Beyond revealing an upstream activation mechanism of PTEN, our data introduce the concept of PI 3-kinase signal activation on the vast plasma membrane that is contrasted by PTEN-mediated signal termination on the small, discrete surfaces of internalized vesicles.

TDRD3 promotes DHX9 chromatin recruitment and R-loop resolution.

R-loops, which consist of a DNA/RNA hybrid and a displaced single-stranded DNA (ssDNA), are increasingly recognized as critical regulators of chromatin biology. R-loops are particularly enriched at gene promoters, where they play important roles in regulating gene expression. However, the molecular mechanisms that control promoter-associated R-loops remain unclear. The epigenetic 'reader' Tudor domain-containing protein 3 (TDRD3), which recognizes methylarginine marks on histones and on the C-terminal domain of RNA polymerase II, was previously shown to recruit DNA topoisomerase 3B (TOP3B) to relax negatively supercoiled DNA and prevent R-loop formation. Here, we further characterize the function of TDRD3 in R-loop metabolism and introduce the DExH-box helicase 9 (DHX9) as a novel interaction partner of the TDRD3/TOP3B complex. TDRD3 directly interacts with DHX9 via its Tudor domain. This interaction is important for recruiting DHX9 to target gene promoters, where it resolves R-loops in a helicase activity-dependent manner to facilitate gene expression. Additionally, TDRD3 also stimulates the helicase activity of DHX9. This stimulation relies on the OB-fold of TDRD3, which likely binds the ssDNA in the R-loop structure. Thus, DHX9 functions together with TOP3B to suppress promoter-associated R-loops. Collectively, these findings reveal new functions of TDRD3 and provide important mechanistic insights into the regulation of R-loop metabolism.

Characteristics of Phenolic Compounds in Peucedanum japonicum According to Various Stem and Seed Colors.

Total polyphenol and total flavonoid assays were performed to characterize the relationships between the color of Peucedanum japonicum (PJ) seed coat and stem and the content of phytochemical compounds. The samples were divided into two groups based on their stem and seed coat color, with each group containing 23 samples. The stem color group was subdivided into green, light red, and red, whereas the seed coat color group was divided into light brown, brown, and dark brown. In the stem color group, the light red stems exhibited the highest content of phytochemical compounds, with levels over 10% higher than those of the stems of the other colors. Moreover, among the top ten samples with the highest total polyphenol content, eight samples were light red, and the light red group also exhibited the highest total flavonoid content among the examined color groups. In terms of the seed coat color, the plants grown from dark brown seeds exhibited the highest contents of both total polyphenols and total flavonoids. In conclusion, PJ plants with dark brown seeds and light red stems contained the highest levels of phytochemical compounds. Collectively, our findings provide a valuable basis for future seed selection of PJ for pharmaceutical purposes.

Wash-Free Amperometric Escherichia coli Detection via Rapid and Specific Proteolytic Cleavage by Its Outer Membrane OmpT.

Various methods have been developed for the detection of Escherichia coli (E. coli); however, they are complex and time-consuming. OmpT─a cell membrane endopeptidase of E. coli─strongly embedded in the outer membrane of only E. coli, exposed to external solutions, with high proteolytic activity, could be a suitable target molecule for the rapid and straightforward detection of E. coli. Herein, a wash-free, sensitive, and selective amperometric method for E. coli detection, based on rapid and specific proteolytic cleavage by OmpT, has been reported. The method involved (i) rapid proteolytic cleavage of consecutive amino acids, after cleavage by OmpT, linked to an electrochemical species (4-aminophenol, AP), by leucine aminopeptidase (LAP, an exopeptidase), (ii) affinity binding of E. coli on an electrode, and (iii) electrochemical-enzymatic (EN) redox cycling. OmpT cleaved the intermediate peptide bond of a peptide substrate containing alanine-arginine-arginine-leucine-AP (-A-R-R-L-AP), forming R-L-AP, followed by the cleavage of two peptide bonds of R-L-AP sequentially by LAP, to liberate an electroactive AP. Affinity binding and EN redox cycling, in addition to rapid proteolytic cleavage by OmpT and LAP, enabled high electrochemical signal amplification. Two-sequential-cleavage was employed for the first time in protease-based detection. The calculated detection limit for E. coli cells in tap water (approximately 103 CFU/mL after 1 h incubation) was lower than those obtained without affinity binding and EN redox cycling. The detection method was highly selective to E. coli as OmpT is present in only E. coli. High sensitivity, selectivity, and the absence of wash steps make the developed detection method practically promising.

Wash-Free, Sandwich-Type Protein Detection Using Direct Electron Transfer and Catalytic Signal Amplification of Multiple Redox Labels.

Direct electron transfer (DET) between a redox label and an electrode has been used for sensitive and selective sandwich-type detection without a wash step. However, applying DET is still highly challenging in protein detection, and a single redox label per probe is insufficient to obtain a high electrochemical signal. Here, we report a wash-free, sandwich-type detection of thrombin using DET and catalytic signal amplification of multiple redox labels. The detection scheme is based on (i) the redox label-catalyzed oxidation of a reductant, (ii) the conjugation of multiple redox labels per probe using a poly-linker, (iii) the low nonspecific adsorption of the conjugated poly-linker due to uncharged, reduced redox labels, and (iv) a facile DET using long, flexible poly-linker and spacer DNA. Amine-reactive phenazine ethosulfate and NADH were used as the redox label and reductant, respectively. N3-terminated polylysine was used as the poly-linker for the conjugation between an aptamer probe and multiple redox labels. Approximately 11 redox labels per probe and rapid catalytic NADH oxidation enable high signal amplification. Thrombin in urine could be detected without a wash step with a detection limit of ∼50 pM, which is practically promising for point-of-care testing of proteins.

Thermodynamic Study of Am(III)-Isosaccharinate Complexation at Various Temperatures Implicating a Stepwise Reduction in Binding Denticity.

Isosaccharinic acid, a major final product of cellulose degradation under highly alkaline cement porewater conditions, is known to increase the mobility of actinides via strong complex formation. In this study, the formation of Am(III) complexes with α-d-isosaccharinate (ISA) was studied in terms of thermodynamics and coordination structures by combining spectrophotometry, time-resolved laser fluorescence spectroscopy (TRLFS), and density functional theory (DFT) calculations. The formation constants of the Am(III)-ISA complexes were determined by absorption spectroscopy at temperatures in the range of 15-70 °C. The measured reaction enthalpy and entropy changes indicate that the formation of a 1:1 Am(III)-ISA complex is driven by an increase in entropy. By contrast, the 1:2 complex formation is exothermic with a much less increase in entropy. DFT calculations predict that C2- and C4-hydroxyl groups, along with the carboxyl group, participate in the tridentate chelate binding of the primary ISA. The thermodynamic, TRLFS, and DFT results collectively suggest the tridentate binding of the primary ISA to Am(III) via a carboxylate and C2- and C4-hydroxyl groups in the protonated state and reduced dentate binding of the secondary ISA, such as bidentate binding, forming a four-membered ring structure via the carboxylate group.

On-target efficacy of a HIF-2α antagonist in preclinical kidney cancer models.

Clear cell renal cell carcinoma, the most common form of kidney cancer, is usually linked to inactivation of the pVHL tumour suppressor protein and consequent accumulation of the HIF-2α transcription factor (also known as EPAS1). Here we show that a small molecule (PT2399) that directly inhibits HIF-2α causes tumour regression in preclinical mouse models of primary and metastatic pVHL-defective clear cell renal cell carcinoma in an on-target fashion. pVHL-defective clear cell renal cell carcinoma cell lines display unexpectedly variable sensitivity to PT2399, however, suggesting the need for predictive biomarkers to be developed to use this approach optimally in the clinic.

Peptidic degron for IMiD-induced degradation of heterologous proteins.

Current systems for modulating the abundance of proteins of interest in living cells are powerful tools for studying protein function but differ in terms of their complexity and ease of use. Moreover, no one system is ideal for all applications, and the best system for a given protein of interest must often be determined empirically. The thalidomide-like molecules (collectively called the IMiDs) bind to the ubiquitously expressed cereblon ubiquitin ligase complex and alter its substrate specificity such that it targets the IKZF1 and IKZF3 lymphocyte transcription factors for destruction. Here, we mapped the minimal IMiD-responsive IKZF3 degron and show that this peptidic degron can be used to target heterologous proteins for destruction with IMiDs in a time- and dose-dependent manner in cultured cells grown ex vivo or in vivo.

Genome analysis of Streptococcus salivarius subsp. thermophilus type strain ATCC 19258 and its comparison to equivalent strain NCTC 12958.

Streptococcus thermophilus is one of the lactic acid bacteria applied as the main starter for dairy foods. A type strain of Streptococcus salivarius subsp. thermophilus ATCC 19258 has been used in the genetic and biochemical characterization of their genes or gene products. While the genome sequence of NCTC 12958 as an equivalent to ATCC 19258 is available, characterization of whether both collections are identical remains to be validated. Here, we report the complete genome sequence of ATCC 19258, which contains one 2.1 Mb chromosome with a 39.0% of G + C content, and includes 2255 protein-coding sequences, 77 RNAs, 4 riboswitches, and 3 CRISPRs. The data were further compared with NCTC 12958 and found that 54 mutations and 4 gaps occurred in NCTC 12958, resulted in both the mutations and insertions of nucleotides in the genome. Unlike ATCC 19258, pre-termination of three genes encoding IS981 transposase B, MltF, and FetB were detected in NCTC 12958. Our study highlights that type strains of Streptococcus thermophilus in two available independent strain collections are possibly different and therefore, the functions of previously identified or hitherto uncharacterized genes of Streptococcus thermophilus should be carefully assigned based on the genomic database of the strain.

Colorimetric determination of phenolic compounds using peroxidase mimics based on biomolecule-free hybrid nanoflowers consisting of graphitic carbon nitride and copper.

Hybrid nanoflowers consisting of graphitic carbon nitride (GCN) and copper were successfully constructed without the involvement of any biomolecule, by simply mixing them at room temperature to induce proper self-assembly to achieve a flower-like morphology. The resulting biomolecule-free GCN-copper hybrid nanoflowers (GCN-Cu NFs) exhibited an apparent peroxidase-mimicking activity, possibly owing to the synergistic effect from the coordination of GCN and copper, as well as their large surface area, which increased the number of catalytic reaction sites. The peroxidase-mimicking GCN-Cu NFs were then employed in the colorimetric determination of selected phenolic compounds hydroquinone (HQ), methylhydroquinone (MHQ), and catechol (CC). For samples without phenolic compounds, GCN-Cu NFs catalyzed the oxidation of the peroxidase substrate 3,3',5,5'-tetramethylbenzidine (TMB) in the presence of H2O2, producing an intense blue color signal. Conversely, in the presence of phenolic compounds, the oxidation of TMB was inhibited, resulting in a significant reduction of the color signal. Using this strategy, HQ, MHQ, and CC were selectively and sensitively determined in a linear range up to 100 µM with detection limits down to 0.82, 0.27, and 0.36 µM, respectively. The practical utility of this assay system was also validated by using it to detect phenolic compounds spiked in tap water, yielding a good recovery of 97.1-108.9% and coefficient of variation below 3.0%, demonstrating the excellent reliability and reproducibility of this strategy. Colorimetric determination of phenolic compounds using peroxidase mimics based on biomolecule-free hybrid nanoflowers consisting of graphitic carbon nitride and copper.

Extracellular vesicle-associated antigens as a new vaccine platform against scrub typhus.

Scrub typhus is an acute vector-borne disease caused by infection with the intracellular gram-negative bacterium Orientia tsutsugamushi (Ot). The rapid production of an efficient vaccine against Ot using novel strategies is required because of the global increase in mortality caused by these infections; however, no commercial vaccine is currently available. Ot induces T-cell-mediated immunogenic responses upon infection; therefore, a new rapidly producible vaccine that maximizes T-cell responses against Ot is required. In this study, we sought to develop a model vaccine platform for T-cell-mediated Ot infection using T-cell-immunity associated Salmonella-derived extracellular vesicles (EVs). For this purpose, we optimized DNA sequences encoding the full-length Ot proteins, TSA56, ScaA, ScaC, ScaD, and ScaE, and their expression in Salmonella. The sequences were incorporated into a new platform vector, pKST, which ectopically and concurrently produces Ot proteins and EVs. Expression analysis using pKST-antigen plasmids showed that TSA56 and ScaC produced antigen-associated EVs and showed strong T-cell immunogenic responses. We found that mice vaccinated with EVs derived from TSA56-expressing cells were protected from Salmonella-induced mortality. Therefore, our findings showed that Salmonella EV-associated antigen is a model platform for T-cell immune response infections. Our system could help prepare EV-antigen vaccines against scrub typhus in an easy and rapid manner.

Rosette-shaped graphitic carbon nitride acts as a peroxidase mimic in a wide pH range for fluorescence-based determination of glucose with glucose oxidase.

Rosette-shaped graphitic carbon nitride (rosette-GCN) is described as a promising alternative to natural peroxidase for its application to fluorescence-based glucose assays. Rosette-GCN was synthesized via a rapid reaction between melamine and cyanuric acid for 10 min at 35 °C, followed by thermal calcination for 4 h. Importantly, rosette-GCN possesses a peroxidase-like activity, producing intense fluorescence from the oxidation of Amplex UltraRed in the presence of H2O2 over a broad pH-range of, including neutral pH; the peroxidase activity of rosette-GCN was ~ 10-fold higher than that of conventional bulk-GCN. This enhancement of peroxidase activity is presumed to occur because rosette-GCN has a significantly larger surface area and higher porosity while preserving its unique graphitic structure. Based on the high peroxidase activity of rosette-GCN along with the catalytic action of glucose oxidase (GOx), glucose was reliably determined down to 1.2 µM with a dynamic linear concentration range of 5.0 to 275.0 µM under neutral pH conditions. Practical utility of this strategy was also successfully demonstrated by determining the glucose levels in serum samples. This work highlights the advantages of GCNs synthesized via rapid methods but with unique structures for the preparation of enzyme-mimicking catalysts, thus extending their applications to the diagnostics field and other biotechnological fields. Graphical abstract.

Biomedical named entity recognition using deep neural networks with contextual information.

BACKGROUND: In biomedical text mining, named entity recognition (NER) is an important task used to extract information from biomedical articles. Previously proposed methods for NER are dictionary- or rule-based methods and machine learning approaches. However, these traditional approaches are heavily reliant on large-scale dictionaries, target-specific rules, or well-constructed corpora. These methods to NER have been superseded by the deep learning-based approach that is independent of hand-crafted features. However, although such methods of NER employ additional conditional random fields (CRF) to capture important correlations between neighboring labels, they often do not incorporate all the contextual information from text into the deep learning layers. RESULTS: We propose herein an NER system for biomedical entities by incorporating n-grams with bi-directional long short-term memory (BiLSTM) and CRF; this system is referred to as a contextual long short-term memory networks with CRF (CLSTM). We assess the CLSTM model on three corpora: the disease corpus of the National Center for Biotechnology Information (NCBI), the BioCreative II Gene Mention corpus (GM), and the BioCreative V Chemical Disease Relation corpus (CDR). Our framework was compared with several deep learning approaches, such as BiLSTM, BiLSTM with CRF, GRAM-CNN, and BERT. On the NCBI corpus, our model recorded an F-score of 85.68% for the NER of diseases, showing an improvement of 1.50% over previous methods. Moreover, although BERT used transfer learning by incorporating more than 2.5 billion words, our system showed similar performance with BERT with an F-scores of 81.44% for gene NER on the GM corpus and a outperformed F-score of 86.44% for the NER of chemicals and diseases on the CDR corpus. We conclude that our method significantly improves performance on biomedical NER tasks. CONCLUSION: The proposed approach is robust in recognizing biological entities in text.

Escherichia coli OxyS RNA triggers cephalothin resistance by modulating the expression of CRP-associated genes.

Antibiotics have been one of the most successful forms of therapy in medicine. However, the efficiency of antibiotics is compromised by the emergence of antibiotic-resistant pathogens. To reduce antibiotic resistance, complete understanding of bacterial tactics to defend themselves against antibiotics is necessary. Small-noncoding RNAs (sRNAs) modulate gene expression by base-pairing with multiple target mRNAs. Cellular levels of Hfq-dependent sRNAs influence antibiotic resistance by modulating expression of specific target genes; therefore, such sRNAs could be a good tool to identify target mRNAs that modulate antibiotic susceptibility and may themselves be used as druggable molecules. Here, we report the identification of genes and pathways associated with OxyS RNA-mediated cephalothin resistance using phenotypic and expression analyses of OxyS-regulated genes identified by RNA-seq, literature mining, or predictions. From our studies we found that the differential expression of 27 OxyS-regulated genes was involved in cephalothin susceptibility. Among them, 17 gene knockouts showed resistance to the drug and nine from them is associated with cAMP receptor protein (CRP), a transcriptional dual regulator in E. coli. Moreover, levels of OxyS and OxyS-modulated genes (cycA and cysH) were also altered in multidrug-resistant (MDR) E. coli strains. Together, our data suggest that OxyS extensively modulates gene expression in multiple pathways to develop cephalothin resistance. In addition, OxyS and its regulated target genes, either individually or in combination, could be used as molecular markers and targets for the identification and eradication of cephalothin-resistant strains.

Role of acid responsive genes in the susceptibility of Escherichia coli to ciclopirox.

Antibiotic resistance poses a huge threat to the effective treatment of bacterial infections. To circumvent the limitations in developing new antibiotics, researchers are attempting to repurpose pre-developed drugs that are known to be safe. Ciclopirox, an off-patent antifungal agent, inhibits the growth of Gram-negative bacteria, and genes involved in galactose metabolism and lipopolysaccharide (LPS) biosynthesis are plausible antibacterial targets for ciclopirox, since their expression levels partially increase susceptibility at restrictive concentrations. In the present study, to identify new target genes involved in the susceptibility of Escherichia coli to ciclopirox, genome-wide mRNA profiling was performed following ciclopirox addition at sublethal concentrations, and glutamate-dependent acid resistance (GDAR) genes were differentially regulated. Additional susceptibility testing, growth analyses and viability assays of GDAR regulatory genes revealed that down-regulation of evgS or hns strongly enhanced susceptibility to ciclopirox. Further microscopy and phenotypic analyses revealed that down-regulation of these genes increased cell size and decreased motility. Our findings could help to maximise the efficacy of ciclopirox against hard-to-treat Gram-negative pathogens.

The IL-10-producing regulatory B cells (B10 cells) and regulatory T cell subsets in neuromyelitis optica spectrum disorder.

B cells contribute to the pathogenesis of neuromyelitis optica (NMO) by producing Aquaporin 4-specific autoantibodies (AQP4-ab); on the other hand, there are certain B cells that suppress immune responses by producing regulatory cytokines, such as IL-10. In this study, we investigated the presence of IL-10-producing Breg cells among lymphocyte subsets. Twenty-two seropositive NMO spectrum disorder (NMOSD) patients (29 samples) and 13 healthy controls (HCs) (14 samples) were enrolled. All NMOSD patients have received one or more immunosuppressive drugs. The phenotype and frequency of B cell and T cell subsets in the peripheral blood were measured by flow cytometry. We defined Breg cells as IL-10-producing B (B10) cells, which are CD19+CD39+CD1d+IL-10+. The potential relations were evaluated between specific lymphocyte subsets and AQP4-ab intensity measured by the cell-based indirect immunofluorescence assay. The frequency of B10 cells was higher in patients with NMOSD regardless of the disease status than that in HCs (attack samples; p = 0.009 and remission samples; p < 0.001, respectively). In addition, the frequency of IL-17+ Treg cells among Treg cells was higher during remission than during an attack (uncorrected p = 0.032). Among the lymphocyte subsets, B10 cells alone showed a positive correlation with the intensity of AQP4-ab positivity (ρ [rho] = 0.402 and p = 0.031). It was suggested that the suppressive subsets including B10 and IL-17+ Treg cells might have important roles in controlling disease status in NMOSD. Further functional studies may help to elucidate the immunological role of B10 and IL-17+ Treg cells in NMOSD.

A method for named entity normalization in biomedical articles: application to diseases and plants.

BACKGROUND: In biomedical articles, a named entity recognition (NER) technique that identifies entity names from texts is an important element for extracting biological knowledge from articles. After NER is applied to articles, the next step is to normalize the identified names into standard concepts (i.e., disease names are mapped to the National Library of Medicine's Medical Subject Headings disease terms). In biomedical articles, many entity normalization methods rely on domain-specific dictionaries for resolving synonyms and abbreviations. However, the dictionaries are not comprehensive except for some entities such as genes. In recent years, biomedical articles have accumulated rapidly, and neural network-based algorithms that incorporate a large amount of unlabeled data have shown considerable success in several natural language processing problems. RESULTS: In this study, we propose an approach for normalizing biological entities, such as disease names and plant names, by using word embeddings to represent semantic spaces. For diseases, training data from the National Center for Biotechnology Information (NCBI) disease corpus and unlabeled data from PubMed abstracts were used to construct word representations. For plants, a training corpus that we manually constructed and unlabeled PubMed abstracts were used to represent word vectors. We showed that the proposed approach performed better than the use of only the training corpus or only the unlabeled data and showed that the normalization accuracy was improved by using our model even when the dictionaries were not comprehensive. We obtained F-scores of 0.808 and 0.690 for normalizing the NCBI disease corpus and manually constructed plant corpus, respectively. We further evaluated our approach using a data set in the disease normalization task of the BioCreative V challenge. When only the disease corpus was used as a dictionary, our approach significantly outperformed the best system of the task. CONCLUSIONS: The proposed approach shows robust performance for normalizing biological entities. The manually constructed plant corpus and the proposed model are available at http://gcancer.org/plant and http://gcancer.org/normalization , respectively.

Clinical characteristics of late-onset neuromyelitis optica spectrum disorder: A multicenter retrospective study in Korea.

BACKGROUND: There are currently few studies regarding late-onset neuromyelitis optica spectrum disorder (LO-NMOSD). OBJECTIVE: We aimed to describe the characteristic features of patients with LO-NMOSD in Korea. METHODS: Anti-aquaporin-4 antibody-positive patients with neuromyelitis optica spectrum disorder (NMOSD) from nine tertiary hospitals were reviewed retrospectively. The patients were divided into two groups based on age of onset: LO-NMOSD (⩾50 years of age at onset) versus early-onset neuromyelitis optica spectrum disorder (EO-NMOSD) (<50 years of age at onset). Clinical, laboratory, and magnetic resonance imaging (MRI) parameters were investigated. RESULTS: Among a total of 147 patients (125 female; age of onset, 39.4 ± 15.2 years), 45 patients (30.6%) had an age of onset of more than 50 years. Compared to patients with EO-NMOSD, patients with LO-NMOSD had more frequent isolated spinal cord involvement at onset (64.4% vs 37.2%, p = 0.002), less frequent involvement of the optic nerve (40.0% vs 67.7%, p = 0.002), and less frequent brain MRI lesions (31.1% vs 50.0%, p = 0.034). Furthermore, there was a significant positive correlation between age of onset and Expanded Disability Status Scale (EDSS) score at last follow-up ( r = 0.246, p = 0.003). CONCLUSION: Age of onset could be an important predictor of lesion location and clinical course of patients with NMOSD.

A corpus for plant-chemical relationships in the biomedical domain.

BACKGROUND: Plants are natural products that humans consume in various ways including food and medicine. They have a long empirical history of treating diseases with relatively few side effects. Based on these strengths, many studies have been performed to verify the effectiveness of plants in treating diseases. It is crucial to understand the chemicals contained in plants because these chemicals can regulate activities of proteins that are key factors in causing diseases. With the accumulation of a large volume of biomedical literature in various databases such as PubMed, it is possible to automatically extract relationships between plants and chemicals in a large-scale way if we apply a text mining approach. A cornerstone of achieving this task is a corpus of relationships between plants and chemicals. RESULTS: In this study, we first constructed a corpus for plant and chemical entities and for the relationships between them. The corpus contains 267 plant entities, 475 chemical entities, and 1,007 plant-chemical relationships (550 and 457 positive and negative relationships, respectively), which are drawn from 377 sentences in 245 PubMed abstracts. Inter-annotator agreement scores for the corpus among three annotators were measured. The simple percent agreement scores for entities and trigger words for the relationships were 99.6 and 94.8 %, respectively, and the overall kappa score for the classification of positive and negative relationships was 79.8 %. We also developed a rule-based model to automatically extract such plant-chemical relationships. When we evaluated the rule-based model using the corpus and randomly selected biomedical articles, overall F-scores of 68.0 and 61.8 % were achieved, respectively. CONCLUSION: We expect that the corpus for plant-chemical relationships will be a useful resource for enhancing plant research. The corpus is available at http://combio.gist.ac.kr/plantchemicalcorpus .