Discussion
AR with cinnabar-like red color secretory cavities have been traditionally used in Chinese medicine for thousand years. However, little is known about the chemical basis of the red secretory cavities, not only between A. lancea natural accessions but also within different tissue types. Although previous studies have shown that AR is rich of certain types of sesquiterpenes and polyacetylenes (Zhang et al., 2021), the spatial distribution of those secondary/specialized metabolites is less understood. According to the geographical variation in the components of AR, here we systematically analysed three representative A. lancea natural accessions (Guo et al., 2008; Takeda et al., 1996), each of which has distinct secretory cavities (red, non-red & mixed SCs). This allowed rational comparative metabolomics with multivariate statistical analysis to identify the potential metabolites that underly the cinnabar-like red color of SC in AR.
Metabolites in plants are often distributed in complex matrices that stock arrays of different classes of chemicals. Some highly valuable compounds are accumulated only in specialized structures, such as the glandular secretory trichomes on Artemisia annua leaves have the capacity to secrete and store artemisinin (Shi et al., 2018), monoterpenoids in pericarp secretory cavities of citrus plants (Voo et al., 2012), and morphine in the laticifers of the aerial organs of opium poppy (Onoyovwe et al., 2013). Accessing the patchy mixtures of specialized metabolites thus are challenging and requires reliable spatial dissection and visualization of metabolomics techniques. We first applied untargeted metabolite analysis using LCM-GC-MS, enabling quantitative comparisons in a diversity of samples. Although the spatial resolution of LCM-GC-MS is limited, it allowed robust comparisons not only between SC and non-SC regions but also SCs that differ among natural accessions. We also applied various of DESI approaches and found that the DESI/PI-MSI (which has a relatively soft ionization step) was able to detect the biomarkers (polyacetylenes) identified by previous comparative metabolomics (Figure S6).