Search

Single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI) plays a crucial role in the optimal treatment strategy for patients with coronary heart disease. We tested the feasibility of feature extraction from MPI using a deep convolutional autoencoder (CAE) model.Eight hundred and forty-three pairs of stress and rest myocardial perfusion images were collected from consecutive patients who underwent cardiac scintigraphy in our hospital between December 2019 and February 2022. We trained a CAE model to reproduce the input paired image data, so as the encoder to output a 256-dimensional feature vector. The extracted feature vectors were further dimensionally reduced via principal component analysis (PCA) for data visualization. Content-based image retrieval (CBIR) was performed based on the cosine similarity of the feature vectors between the query and reference images. The agreement of the radiologist's finding between the query and retrieved MPI was evaluated using binary accuracy, precision, recall, and F1-score.A three-dimensional scatter plot with PCA revealed that feature vectors retained clinical information such as percent summed difference score, presence of ischemia, and the location of scar reported by radiologists. When CBIR was used as a similarity-based diagnostic tool, the binary accuracy was 81.0%.The results indicated the utility of unsupervised feature learning for CBIR in MPI.

During the course of our studies towards the identification of promising chemotherapeutic candidates from plants against two human T-cell lymphotropic virus type I-infected T-cell lines (MT-1 and MT-2), we screened 17 extracts from 9 rutaceous plants against MT-1 and MT-2 cells. The extracts from the aerial parts and roots of Boenninghausenia japonica, as well as the leaves and roots of Ruta graveolens showed potent antiproliferative effects. After activity-guided fractionation, we isolated 44 compounds from two rutaceous plants, including three new compounds (1-3), which were classified into 26 coumarin analogs (13 coumarins, 8 furanocoumarins, 4 dihydrofuranocoumarins and one dihydropyranocoumarin), 15 alkaloid analogs (7 quinolone alkaloids, 4 acridone alkaloids, 3 furanoquinoline alkaloids and one tetrahydroacridone alkaloid) and 3 flavonoid glycosides. Structure-activity relationship studies were also evaluated. The coumarin compounds (2, 3 and 7-9) bearing a 3-dimethylallyl moiety showed potent activity. Similarly, of all the furanocoumarins evaluated in the current study, compound 17 bearing a 3-dimethylallyl group also showed potent activity. A dihydrofuranocoumarin (27) bearing a 3-dimethylallyl moiety showed the most potent activity. Following 27, compound 28 showed potent activity. These results therefore suggested that the presence of a 3-dimethylallyl moiety was important to the antiproliferative activity of these coumarin analogs.

Adult T-cell leukemia/lymphoma (ATL) is a malignancy of mature peripheral T lymphocytes caused by human T-cell lymphotropic virus type 1 (HTLV-1). There are an estimated 5 million to 20 million HTLV-1-infected individuals worldwide; their lifetime risk of developing ATL is 3-5 %, and high HTLV-1 proviral loads have been shown to be an independent risk factor. Although conventional chemotherapeutic regimens used against other malignant lymphomas have been administered to ATL patients, the prognosis is often poor. In previous studies, we screened 459 extracts from 344 plants to isolate components exhibiting antiproliferative activity against HTLV-1-infected T-cell lines (MT-1 and MT-2). In our continuing search for potential anti-HTLV-1 natural products, 15 extracts of Asclepiadaceae plants were further tested against MT-1 and MT-2 cells. The MeOH extract of aerial parts of Tylophora tanakae showed antiproliferative activity. Activity-guided fractionation resulted in the isolation of 6 phenanthroindolizidine alkaloids (including a new compound), and we examined their antiproliferative activity against MT-1 and MT-2 cells. The EC50 value of some of the alkaloids was in the low nanomolar range, comparable to that of the clinically used antineoplastic drug doxorubicin. Structure-activity relationship analyses suggested that a 14β-hydroxy moiety is essential for activity against HTLV-1-infected T cells. In contrast, the presence of a 2-methoxy moiety, a 7-methoxy moiety, or an N-oxide moiety appears to reduce the potency of the antiproliferative activity against HTLV-1-infected T cells.

Adult T-cell leukemia/lymphoma (ATL) is a peripheral T-cell malignancy caused by human T-cell lymphotropic virus type I (HTLV-1). Clinical manifestations of ATL range from smoldering to chronic, lymphoma and acute subtypes. Patients with acute and lymphoma-type ATL require therapeutic intervention. Conventional chemotherapeutic regimens used against other malignant lymphoma have been administered to ATL patients, but the therapeutic outcomes of acute and lymphoma-type ATL remain very poor. In this study, 214 extracts from 162 plants belonging to 65 families were screened for the purpose of elucidating the anti-proliferative effect against HTLV-1-infected T-cell lines. Extracts from aerial parts of Physalis pruinosa showed potent inhibitory effect. We isolated five withanolides from the extracts by activity-guided fractionation and examined the structure-activity relationships. The presence of a 5β,6β-epoxy function is suggested to be essential for the activity, and the most active principle showed selective toxicity to HTLV-1-infected T-cell lines.

In order to clarify the efficacy of natural-occurring hepatoprotective flavonoids and related miscellaneous compounds, we have investigated the hepatoprotective activity of 41 phenolic compounds on human liver-derived HepG2 cells and evaluated their structure-activity relationships. The hepatoprotective activity of catechin (cyanidanol), which has been used to treat chronic hepatitis, was almost equal to that of glycyrrhizin at 200 μM. The structure-activity relationships of flavanol derivatives as revealed by this investigation suggest that (1) the configuration of the hydroxy group at C-3 does not affect the hepatoprotective activity; (2) the pyrogallol and catechol groups in the B-ring are almost comparable in terms of hepatoprotection; (3) the galloyl group is a key factor for enhancing hepatoprotective activity; (4) the free carboxyl group of gallic acid reduces hepatoprotective activity. The common structural features of the most potent hepatoprotective phenolic compounds are summarized.

Larvae of the diamondback moth, Plutella xylostella, a crucifer specialist, refuse to feed on a crucifer, Barbarea vulgaris, because of the presence of a feeding deterrent, which is extractable with chloroform. We isolated a feeding deterrent from B. vulgaris leaves, by successive fractionations with silica-gel, ODS, i.e., C18 reversed phase, and Sephadex LH-20 column chromatographies, and ODS-HPLC, guided by a bioassay for feeding deterrent activity. The structure of the compound was determined to be a monodesmosidic triterpenoid saponin, 3-O-[O-β-D-glucopyranosyl-(1→4)-β-D-glucopyranosyl]-hederagenin, based on FAB-MS, 1H- and 13C-NMR spectra, and hydrolysis experiments. When the compound was applied to cabbage leaf disks at greater than 0.18 μg/mm2, consumption of the disks by third instars was less than 11% of control disks treated with the solvent alone. Furthermore, all first instars died on the disks treated with the same concentrations. Because the concentration of the compound in the fresh leaves of B. vulgaris was comparable to the effective dose in the cabbage leaf disk tested, we conclude that the unacceptability of B. vulgaris to P. xylostella larvae is primarily due to this saponin.