Phenolic compounds in taxonomy of Myricaria longifolia and Myricaria bracteata (Tamaricaceae)

For the first time, the phenolic profile of poorly studied Siberian species of the genus Myricaria Desv. Myricaria longifolia (Willd.) Ehrenb. was investigated by HPLC and LC-MS/MS in comparison with that of wide-ranging species M. bracteata Royle. 65 quantitative parameters of the phenolic profiles were processed by ANOVA and princ ipal component analysis (PCA). The results suggest that the distinc tio between the species is mainly determined by the variance in total flavo noids and free quercetin in the leaves. Free gallic and ferulic acids, hyper oside and total phenolics in aqueous ethanol extract, as well as kaempferol a nd rhamnazin in the hydrolyzed extract contributed to the difference be tween the species. The significant differences justify the positions of th ese species in two individual series of the genus Myricaria established before. The statistical analysis of the biochemical data allowed us to iden tify both the characters that determined the distinction between species, an d indicators of heterogeneity of the species that varies abnormally (the concentrations of isorhamnetin and rhamnazin, and their ratio).


Introduction
Myricaria Desv. (Tamaricaceae) is a genus of shrubs growing on mountain river banks in Asia and Eastern Europe. Himalayan region was considered to be the origin of the distribution of this genus [1]. By now, 11 to 13 Myricaria species were identified, and their taxonomy is very complicated [2]. E.G. Bobrov (1967) attributed the genus Myricaria to the ancient genera of the desert flora of the Eastern Hemisphere whose origin and development were associated with Central Asia. He regard M. bracteata Royle as a species the closest to the original form. It is widely distributed in extensive territories in the West Hymalayas, Central and Nothern Asia. Meanwhile, M. longifolia (Willd.) Ehrenb. occurs mainly in the mountains of southern Siberia and represents a line of further development of the genus [3]. The genus Myricaria has a long history of taxonomic study. So far, there is no consensus on taxonomy of Myricaria species in botanical literature. For instance, various researchers counted 2 to 5 Myricaria species in the Siberian flora [3,4]. According to recent data in our investigation, we consider two species: M. bracteata and M. longifolia. In their identification, shape and size of leaves were accepted as the main diagnostic characters. M. bracteata was shown to belong to small-leaved species assigned to series Germanicae Gorschk., whereas M. longifolia is large-leaved species, which was placed in another series Dahuricae Gorschk. or series Elegantes Bobr. [3]. Another diagnostic character, flower racemes arrangement, was revealed to vary during plant development, and its diagnostic value is questionable [4]. The difficulties of species delimitation associated with high polymorphism and a substantial spatial and temporal variation of many morphological characters induce us to increase the reliability of taxonomic analysis due to biochemical features. In this regard, secondary metabolite profiling creates additional opportunities for the identification of the phylogenetic relations between plant taxa. Along with morphological parameters, phenolic compounds are indicators of the biological diversity of plants. These compounds were regularly used as taxonomic markers and facilitated the solution of taxonomic problems at different levels [5]. Their considerable structural diversity is the basis of taxon-specificity of phenolic profiles. Phenolic compounds of Myricaria species are poorly studied. The investigation of phenolic aglycon composition in M. bracteata leaves by LC-MS/MS have revealed the diversity of methylated flavonols and the presence of flavones in the leaves of this species [6]. The aim of this study is to determine phenolic profile of M. longifolia in comparison to the profile of M. bracteata, and to identify their species-specificity. For LC-MS/MS and HPLC analyses, precisely weighed samples of air-dried plant material (0.3 g) were exhaustively extracted with an ethanol:water mixture (70:30, v/v) in a water bath at 60-70°C. Dry-weight concentration in the samples was calculated by the gravimetric method. A hydrolyzed extract was obtained by hydrolysis of the aqueous ethanol extract with 2N HCl for 2 h in a boiling water bath, followed by purification by means of a C16 Diapack cartridge and redissolution in ethanol. It was used for identification and quantification of the aglycones. The identification was performed by liquid chromatography with tandem mass spectrometry. MS analysis was carried out at the Core Facility of Mass Spectrometric Analysis (ICBFM SB RAS) as described before [6]. The quantification of phenolic compounds was conducted by the external-standard method. Chromatographic analysis for absolute quantification of phenolics was carried out with an Agilent 1200 chromatograph with a diode matrix detector and the system for processing the chromatographic data ChemStаtion (Agilent Technologies, USA). The separation was performed with a Zorbаx SB-C18 column 4.6 × 150 mm in size, with particle diameter 5 mm. The mobile phase consisted of MeOH (solvent A) and 0.1% orthophosphoric acid in water (solvent B). Chromatographic separations of aglycones in hydrolyzed extracts with BIO Web of Conferences 38, 00051 (2021) Northern Asia Plant Diversity 2021 https://doi.org/10.1051/bioconf/20213800051 gradient 1 and native compounds (phenolic acids, flavonoid glycosides and free aglycones) with gradient 2 was described previously [6,7]. To construct calibration curves, chemical reference standards of gallic and ferulic acids from Serva (Heidelberg, Germany), ellagic acid, ascorbic acid, quercetin, kaempferol, rhamnetin, isorhamnetin, luteolin, apigenin, naringenin, astragalin, isoquercitrin, and isorhamnetin 3-O-rutinoside from Sigma (St. Louis, MO, USA), and hyperoside from Fluka (Sigma-Aldrich Chemie GmbH, Munich, Germany) were used. The sum of organic acids was calculated as ascorbic acid equivalents at 250 nm, and flavonol glycosides 18, 23, 25-27 were calculated as hyperoside equivalents at 350 nm. All the data were processed in the Statistica 10.0 software (Statsoft Inc., Tulsa, OK, USA). The significance of difference between datasets was determined by one-way analysis of variance (ANOVA). Data from the quantification of phenolic compounds were expressed as mean ± SE of 9 (3 biological x 3 technical) replicates and were compared by Duncan's multiple range test. Differences between the means of any parameters were considered statistically significant at the 5% level (p < 0.05). To evaluate the variations among the Myricaria species and to identify major compounds contributed to delimiting species, principal components analysis (PCA) was performed. PCA was applied to 64 biochemical parameters, including the concentrations, ratios, and number of the compounds.
Quantification of these compounds in hydrolyzed extracts of the leaves by HPLC showed the substantial variation in levels of most of the compounds. Organic acids and ellagic acid were permanent major organic constituents of both species, and gallic acid, isorhamnetin, and rhamnazin levels varied between populations from high to negligible. Apigenin and rhamnetin occurs only in traces in both species.
M. longifolia exceeded M. bracteata in the concentrations of ferulic acid and quercetin, but the concentration of kaempferol in the leaves of M. longifolia was lower ( Table 1). The differences in the average concentrations of isorhamnetin and rhamnazin were not statistically significant due to substantial variation in their concentrations. Here and in Table 2, means in rows followed by the same letter do not differ significantly according to Duncan's test (p < 0.05).
Among intact compounds in the aqueous ethanol extracts of the leaves, organic acids, free gallic and ferulic acids, free ellagic acid, flavonol glycosides (including hyperoside, isoquercitrin, astragalin, and isorhamnetin 3-O-rutinoside), and free aglycones (including quercetin, naringenin, luteolin and apigenin) were detected ( Table 2). Organic acids, free gallic acid and hyperoside were major components in both species. M. longifolia surpassed M. bracteata in the concentrations of most of components, as well as in total phenolic compounds and total flavonoids. On the other hand, M. bracteata had an increased level of ferulic acid. The results of PCA confirmed this finding (Table 3). According to PCA, the main parameters that determine the difference between the phenolic profiles of the species were total flavonoids in aqueous ethanol extract and free quercetin, strongly correlated with PC1. Free gallic and ferulic acids, hyperoside, total phenolics in aqueous ethanol extract, as well as kaempferol and rhamnazin in the hydrolyzed extract showed a high correlation with PC1 too.
Thus, the statistical analysis of the biochemical data allowed us to identify both the characters that determine the distinction between species, and abnormally varying characters indicating the variability within the species. Proportions of phenolic compounds are common in plant taxonomy [8]. Previously, we dealt with a high ratio of kaempferol to quercetin as taxonomic character of the series Elegantes Pojark. in the genus Spiraea (Rosaceae) [9]. In this case, the ratio of rhamnazin to isorhamnetin seems to reflect a differentiation within the studied species. Considerable variation in this ratio between two significantly distant of each other populations of M. bracteata we have discovered before [6].
For the first time, the phenolic profile of poorly studied Siberian species Myricaria longifolia was investigated by HPLC and LC/MS/MS in comparison with that of wideranging species M. bracteata. The results suggest that the distinction between the species is mainly determined by the variance of total flavonoids and free quercetin in the leaves. The significant differences justify the positions of these species in two individual series of the genus Myricaria in the classifications of S.G. Gorshkova (1949) and E.G. Bobrov (1967). High variation in the concentrations of major flavonoid aglycones between the populations within the species confirms a possible subdivision of M. longifolia and M. bracteata into species or subspecies. Our study fills the gap in knowledge about differentiation and relationships in the genus Myricaria.
The work was financially supported by the draft State assignments of the Central Siberian Botanical Garden of the Siberian Branch of the Russian Academy of Sciences # АААА-А21-121011290025-2 and # АААА-А21-121011290027-6 within the framework of a government contract and with material of CSBG representing USFs (Unique Scientific Facilities) "Collections of living plants indoors and outdoors" USU 440534.