Evaluation of Anti-Inflammatory Effects of Celery Leaf and Stem Extracts in LPS-Induced RAW 264.7 Cells Using Nitric Oxide Assay and LC-MS Based Metabolomics.

The present work demonstrated and compared the anti-inflammatory effects of celery leaf (CLE) and stem (CSE) extracts. LC-MS-based metabolomics were an effective approach to achieve the biomarker identification and pathway elucidation associated with the reduction in inflammatory responses. The celery extracts suppressed LPS-induced NO production in RAW 264.7 cells, and CLE was five times more effective than CSE. Distinct differences were revealed between the control and celery-treated samples among the 24 characteristic metabolites that were identified. In celery-treated LPS cells, reversals of intracellular (citrulline, proline, creatine) and extracellular (citrulline, lysine) metabolites revealed that the therapeutic outcomes were closely linked to arginine metabolism. Reversals of metabolites when treated with CLE (aspartate, proline) indicated targeted effects on the TCA and urea cycles, while, in the case of CSE (histidine, glucose), the glycolysis and the pentose phosphate pathways were implicated. Subsequently, apigenin and bergapten in CLE were identified as potential biomarkers mediating the anti-inflammatory response.

High throughput human plasma N-glycan analysis using DNA analyzer and multivariate analysis for biomarker discovery.

Carbohydrates form the majority of organic compounds found in nature and their presence on proteins influences many important bioactivities. Therefore, glycan profiling shows potential in clinical applications. This work demonstrates the use of a high-throughput GlycanAssure™ sample preparation technology and multi-capillary DNA analyzer for the analysis of the major N-linked glycans (N-glycans) found in human plasma. The application involves two biomarker studies: (1) in profiling patients with chronic kidney disease and (2) in differentiating heart disease patients with normal controls in response to an antiplatelet drug from hypo-responders. Due to complexity of the study data, bio-statistical methods were applied to data processing. 37 N-glycan peaks were observed from separation results, with confirmed structure for most glycans. Principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA) were used to build models to differentiate the patient groups. The percentages of correct classification of the models reached 95.45% for the chronic kidney disease dataset and 85.42% for the anti-platelet drug response dataset. Given that blood N-glycan profiles had been shown to reflect certain disease states, this high-throughput platform could potentially be used for the simultaneous screening of multiple glycan biomarkers, with as little as one drop of blood sample.