Predictive Model for Bioactive Compounds Based On Dry Matter Content, Color Indicators, and Soluble Solids of Leafy Greens Planted In Greenhouse, High Tunnel, and Open Field

Associations between botanical food and human health have been the subject of many research investigations in recent years. Low in fat, sodium, and calories, leafy vegetables are a source of fiber, nutrients, and bioactive compounds in a healthful diet. Among various phytochemicals, phenolic-based compounds have been of particular interest due to their contribution to human health and potential impact on sensory perception of fruits and vegetables. Radical scavenging capacity of leafy greens has also been a focus of academic and industry studies due to potential health benefits. The purpose of this study was to use predictive models for total phenolic compounds (TPC), and radical scavenging capacity (RSC) based on color co-ordinates, total soluble solids (TSS), and dry matter (DW) of arugula, komatsuna, mizuna, pac choi, huazontel, amaranth, tai sai, and red iceberg lettuce. Three spectrocolorimeter coordinates (L*, a*, b*), two calculated color attributes (chroma [=(a*2+b*2)1/2] and hue angle [= tan-1(b*/a*)], TSS, and DW were used as predictors of TPC and RSC (dry- and fresh-weight basis) using a Stepwise Selection Method to select statistically significant (α=0.05) predictor variables and their parameter estimates (PE). The analysis was conducted for the eight selected leafy greens as well as three subgroups of (1) cultivars with visible red/purple pigmentations (huauzontle, red iceberg lettuce, and amaranth); (2) cultivars without visible red/purple pigmentations i.e. green in color (arugula, komatsuna, tai sai, mizuna, pac choi); (3) leafy greens from Brassica genus (komatsuna, tai sai, mizuna, pac choi) for plants grown in greenhouse, high tunnel, and open field environments. For the main analysis of the eight taxa, DW and/or TSS were the only significant (P<0.05) predictors of TPC (fresh-weight basis) across the three production methods (PE=5.22, 48.6, and 51.07 for greenhouse, high tunnel, and open field, respectively). Although not significant in the main analysis, color co-ordinates were a significant predictor of the selected secondary metabolites for some of the subgroups across the production methods. Color co-ordinate L*, as an example, was a significant predictor of RSC (dry-weight basis) in cultivars with and without visible red/purple pigmentation subgroups in high tunnel-grown samples (PE=0.76, 0.52, respectively). Results of this study indicate dry matter content, color, and soluble solids show potential for prediction of TP and RSC of leafy vegetables.