Integrative morphological, phytochemical, and molecular identification of three invasive and medicinal Reynoutria species

The three highly invasive Reynoutria species - R. japonica Houtt., Reynoutria sachalinensis (F.Schmidt) Nakai, and a hybrid of them - R. x bohemica J. Chrtek & A. Chrtková are rich sources of biologically active compounds. However, only R. japonica is accepted as an herbal drug source in the pharmacopeias as Polygoni cuspidati rhizoma et radix, while R. sachalinensis is used in folk phytotherapy. The hybrid species can be misidentified as R. japonica due to its morphological resemblance. This study intended to explore genetic identity, diversity, and population structure using sequence-related amplified polymorphisms (SRAP) and start codon targeted (SCoT) markers correlated with morphological characteristics and metabolic profiles in fifteen accessions of Reynoutria species from invasive populations in Central Europe (Wroclaw, south-west Poland). The results suggest that morphological identification based on leaf morphology is clear for R. sachalinensis but fails to distinguish between R. japonica and R. × bohemica unambiguously. The qualitative HPLC/DAD/ESI-HR-QTOF-MS analysis determined the chemical composition of collected samples, revealing 117 compounds belonging to carbohydrates, stilbenes, flavan-3-ols, procyanidins, anthraquinones, organic acids, and naphthalenes. Out of these, the six compounds were quantitated. Based on LC-MS data for 117 compounds, the cluster analysis categorized the fifteen accessions into two distinctive clusters. All SCoT and SRAP markers showed medium (0.1 to 0.25) to high (0.30 to 0.40) levels of PIC, high levels of polymorphic bands (85 and 89%, respectively) and relatively high mean values for Rp (6.42 and 7.0) and were considered informative and differentiating. Partitioning the genetic diversity with AMOVA showed that variation within the populations was higher than among them, as seen in the combined data from SCoT and SRAP (53.04% vs. 46.96%, respectively) and the analyses of individual markers. Genetic diversity indices revealed that chosen genetic markers efficiently assess the variability within the genus Reynoutria. Na varied from 1.57 in R. sachalinensis to 3.34 in R. x bohemica, and Ne was low and almost identical for all species (~ 1.6). Nei's diversity (H) indicated low genetic diversity (< 0.2) in all species, with the highest value for hybrid (0.184). Despite low H values, Shannon's index (I) remains high and similar (R. japonica - 5.21, R. x bohemica - 5.28, and R. sachalinensis - 5.45). The number (NPL) and percentage of polymorphic loci (PPL) for the hybrid (160 and 52.98%) were almost twice as high as for the parent species. The populations exhibited moderate G_ST values (0.159) and a gene flow (Nm = 1.317). Genetic structure analysis with the Evanno test (ΔK = 2) reveals two genetic groups with the highest admixture in the hybrid population. The cluster analysis dendrograms based on LC-MS data, genetic distance, and leaf morphology categorized the fifteen accessions into three clusters, revealing unequivocal separation of the R. sachalinensis from both other taxa. According to Baker's methodology, the highest correlation between dendrograms was observed between genetic and morphological data, with values of 0.85. Lower correlations were noted between molecular markers and phytochemical profile, as well as between morphology and phytochemical profile. The study suggests the effectiveness of combined morphological, phytochemical, and SCoT and SRAP molecular marker systems in assessing the identity of plant material. This approach can also be used in further studies on the population structure, spread, diversity, and evolution of invasive species such as Reynoutria spp.