Correlation Network Analysis Visually Identifies Interactions of Antioxidant Compounds with Plant Growth, Leaf Photosynthetic Performance, and Agronomic Quality in Strawberry

Abstract

Strawberry (Fragaria ananassa Duch.) is a rich source of bioactive compounds, including ascorbic acid (ASA), and polyphenols, such as anthocyanins and phenolic acids, most of which exhibit high antioxidant activities both in vitro and in vivo (Giampieri et al., 2012). As antioxidant compounds are associated with health benefits, the dietary intake of strawberry is highly recommended (Hannum, 2004; Cervantes et al., 2019). However, in the strawberry fruit, the content of antioxidant compounds varies because of genetics, environmental factors, and cultural practices. Previously, several studies on strawberry focused on the effects of environmental factors, such as harvest season (Choi et al., 2016), air temperature (Balasooriya et al., 2019; Shin et al., 2007; Sun et al., 2012), and light-related factors, such as light intensity, light duration, and ultraviolet (UV) exposure (Palmieri et al., 2017; Cervantes et al., 2019), on antioxidant compounds. In general, environmental conditions such as air temperature and irradiation have great impacts on plant growth because of changes in the photosynthetic apparatus, such as the photoinhibition of photosystem II (PSII) (Xu et al., 2020). Changes in the photosynthetic apparatus affect antioxidant metabolism through the generation of reactive oxygen species in photosynthesizing tissues (Hajiboland, 2014), which affect plant growth and/or photosynthetic performance, leading to changes in the level of antioxidant compounds. However, the effects of plant growth and photosynthetic performance on fruit antioxidant compounds have not yet been investigated. Therefore, to enhance the level of antioxidant compounds in the strawberry fruit, it is necessary to understand the interactions of antioxidant compounds with plant growth and photosynthetic performance. A correlation network analysis can be used widely to visualize interactions among various factors (Newman, 2003). From the agricultural viewpoint, the correlation network analysis of plant metabolism can provide key insights into biochemical processes and their regulation (Toubiana et al., 2013). Previously, correlation network analysis has been used to perform metabolic data analysis of several horticultural crops, such as tomato (DiLeo et al., 2011; Zushi and Matsuzoe, 2011, 2015), pepper (Silva et al., 2016), and strawberry (Fait et al., 2008). In these studies, the correlation network and its structure have been characterized extensively in efforts to elucidate the design principles of metabolic interactions. For example, in strawberry fruit, the correlation network suggested that metabolism is