• Korean Journal of Food Science and Technology
• /
• v.42 no.4
• /
• pp.481-487
• /
• 2010

The nutritional components, antioxidant, and neuroprotective effects of water and a 50% methanol extract from litchi fruit pericarp were investigated. The most abundant mineral, amino acid, and fatty acid were K, proline, and palmitic acid, respectively. In addition, the total water phenolics and 50% methanol extracts were 8.02 and 12.28 mg/g, respectively. The DPPH, ABTS radical scavenging activities and ferric reducing antioxidant power of the water and 50% methanol extracts showed dose-dependent antioxidant activity. In a cell viability assay using MTT, almost all extracts showed a protective effect against $H_2O_2$-induced neurotoxicity, and lactate dehydrogenase leakage was also inhibited by the pericarp extracts. In particular, the 50% methanol extract showed a higher cell membrane protective effect than the water extract at the highest concentration. Consequently, these data suggest that litchi fruit pericarp can be utilized as an effective and safe functional food substances for natural antioxidants and may reduce the risk of neurodegenerative disorders.

• Korean Journal of Weed Science
• /
• v.14 no.3
• /
• pp.163-170
• /
• 1994

The experiment was conducted in 1993 to find out a simple prediction method of weeds and to make the prediction models of weeds in paddy fields. The annuals producing fine seeds were apt to emerge at sampling soil only, on the contrary the perennials and the annuals producing large seeds tended not to emerge at sampling soil due to the miss of seeds at sampling. There was no appropriate regression between a total number of weeds emerged at sampling soil and that of weeds occurred in fields. The important annual weeds occurring in fields were able to predict by the number of weeds emerged at sampling soil, but it was difficult to predict the important perennial weeds. In case of Bidens tripartita producing large seeds and Eleocharis kuroguwai producing large tubers, the prediction coefficients were high as above 1.0, and that of Echinochloa crus-galli and Sagittaria pygmaea were comparatively high as 0.175 and 0.172, respectively. However the coefficients of the other weeds were much low as below 0.08. The prediction models for 9 species were made. The model of six species including E. crus-galli, M. vaginalis, R. indica, B. tripartita, E. triandra and S. pygmaea were linear regression with high significance, however that of 3 species including C. difformis, S. juncoides and E. kuroguwai were curve regression with high significance.