Changes in Profiles of Classes and of Individual Polyphenols in Leaves of Spiraea chamaedryfolia and Spiraea media along an Altitudinal Gradient

Plants in high-altitude habitats are exposed to severe environmental stressors, including extreme temperatures and irradiation, which can have wide-ranging effects on changes of secondary-metabolite profiles in higher plants. Altitude-related variation of levels of polyphenols in organs of medicinal and food plant species has not yet been investigated sufficiently. This study was focused on variation in quantitative profiles of classes and of individual biologically active phenolic compounds in leaf extracts of resource species Spiraea chamaedryfolia and Spiraea media from the family Rosaceae in coenopopulations of the Altai Mountains, along an altitudinal gradient. High-performance liquid chromatography revealed 22 polyphenolic compounds in the extracts of S. media leaves, with the main polyphenolic compounds being flavonols. Sixteen compounds were found in S. chamaedryfolia leaf extracts, and the major ones were flavonols and a flavanone. Opposite responses to changes in the altitude-associated growth conditions were documented for levels of some individual polyphenolic compounds. With an increase in altitude, concentrations of chlorogenic acid and of flavanone in the extracts of S. chamaedryfolia leaves significantly increased, while concentrations of cinnamic acid, astragalin, and kaempferol diminished. A statistically significant positive correlation between the altitude of plant habitats and total levels of polyphenols and phenolcarboxylic acids was detected. In leaf extracts from S. media, an altitude increase was significantly positively correlated with astragalin, avicularin, and cinnamic acid levels and negatively correlated with hyperoside concentration.


Introduction
In high-altitude conditions, plants are exposed to several adverse factors, such as low atmospheric pressure, low air temperature, strong winds, a short growing season, and highintensity ultraviolet radiation [1]. An altitude change affects a wide range of environmental conditions of plant habitats, such as hydrothermal and mineral composition of soils, which directly or indirectly influence levels of secondary metabolites [2][3][4]. The survival of plants under extreme growth conditions depends on their physiological adaptation to various adverse factors that cause oxidative stress. In plants, the defensive mechanism that counteracts oxidative stress consists of the synthesis of antioxidant-defense components, including biologically active phenolic compounds (PCs) [5,6]. Furthermore, shifts in the profile of PCs along an altitudinal gradient may be species-specific [7]. In some studies on variation in profiles of classes and of individual PCs in leaf extracts from Lonicera caerulea L. (Caprifoliaceae Juss.) in populations of the Altai Mountains (Seminsky Ridge), it has been found that various PC classes and individual PCs are characterized by different directions of the response to shifts of growth conditions along the altitudinal gradient [8].

Results and Discussion
We performed a comparative examination of retention times of the peaks of the compounds in chromatograms of the analyzed and standard samples; we also carried out a computational comparison of the absorption spectra obtained by chromatography of the substances with a library that we have. As a result, profiles of classes of PCs and of these individual compounds were determined in the aqueous-ethanol extracts from the leaves of the plant species being studied: S. chamaedryfolia and S. media (Tables 1 and 2).
The comparative analysis of the levels of PC classes in the plant leaves showed that the total polyphenol content in S. media varied within 24.6-52.4 mg/g, with major PCs being flavonols. S. chamaedryfolia is distinguished by a significantly lower concentration of polyphenols (11.7-23.6 mg/g), the main of which are flavonols and flavanones. The analyzed species manifested high variation both of the total content of PCs and of their individual classes and representatives, depending on the sampling site (Figures 1-5).
The leaf extracts of S. chamaedryfolia were found to contain 16 biologically active PCs (Table 1). Major PCs of the extracts were an unidentified flavanone (retention time [t R ] = 11.5) and flavonols: rutin, spiraeoside, and an unidentified flavonol (t R = 7.3). In all the studied coenopopulations, chlorogenic and cinnamic acids, an unidentified phenolcarboxylic acid (t R = 44.0), flavones (t R = 4.4 and 23.8), isoquercitrin, and astragalin were detected in minor amounts. In this species, other PCs were found in some coenopopulations only in small amounts. The largest set of polyphenols was detected in leaf extracts of the S. chamaedryfolia collected in CP1 and CP2, respectively, at 1070 and 1222 m above sea level (a.s.l.). The highest total content of polyphenols (23.8 mg/g) was noted in CP7 (1700 m a.s.l.), whereas the lowest content (11.7 mg/g) was registered at the lowest location (CP1 at 1072 m a.s.l.).  being flavonols. S. chamaedryfolia is distinguished by a significantly lower concentration of polyphenols (11.7-23.6 mg/g), the main of which are flavonols and flavanones. The analyzed species manifested high variation both of the total content of PCs and of their individual classes and representatives, depending on the sampling site (Figures 1-5). The leaf extracts of S. chamaedryfolia were found to contain 16 biologically active PCs (Table 1). Major PCs of the extracts were an unidentified flavanone (retention time [tR] = 11.5) and flavonols: rutin, spiraeoside, and an unidentified flavonol (tR = 7.3). In all the studied coenopopulations, chlorogenic and cinnamic acids, an unidentified phenolcarboxylic acid (tR = 44.0), flavones (tR = 4.4 and 23.8), isoquercitrin, and astragalin were detected in minor amounts. In this species, other PCs were found in some coenopopulations only in small amounts. The largest set of polyphenols was detected in leaf extracts of the S. chamaedryfolia collected in CP1 and CP2, respectively, at 1070 and 1222 m above sea level (a.s.l.). The highest total content of polyphenols (23.8 mg/g) was noted in CP7 (1700 m a.s.l.), whereas the lowest content (11.7 mg/g) was registered at the lowest location (CP1 at 1072 m a.s.l.).
The levels of individual polyphenols and of their classes in the leaf extracts of S. chamaedryfolia varied significantly depending on the location of a coenopopulation (Figures 2 and 3). In terms of these concentrations, we noticed some patterns. Along the gradient of increasing altitude, there was a significant (p < 0.01) increase in the level of chlorogenic acid (r = 0.89), and at p < 0.05, an increase of flavanone concentration (r = 0.71) and a decrease in concentrations of astragalin (r = −0.87, p < 0.01), cinnamic acid (r = −0.91, p < 0.01), and kaempferol (r = −0.82, p < 0.05). A direct statistically significant (p < 0.05) correlation was found between the altitude where the plants grew and the total contents of PCs and of phenolcarboxylic acids in the extracts of S. chamaedryfolia leaves (r = 0.84 and 0.85, respectively).  The levels of individual polyphenols and of their classes in the leaf extracts of S. chamaedryfolia varied significantly depending on the location of a coenopopulation (Figures 2 and 3). In terms of these concentrations, we noticed some patterns. Along the gradient of increasing altitude, there was a significant (p < 0.01) increase in the level of chlorogenic acid (r = 0.89), and at p < 0.05, an increase of flavanone concentration (r = 0.71) and a decrease in concentrations of astragalin (r = −0.87, p < 0.01), cinnamic acid (r = −0.91, p < 0.01), and kaempferol (r = −0.82, p < 0.05). A direct statistically significant (p < 0.05) In S. media leaf extracts, 22 PCs were found ( Table 2). The major ones were flavonols: rutin; hyperoside; quercetin; spiraeoside; astragalin; avicularin; kaempferol; unidentified flavonols 4, 5, 14, and 17 (compound ID numbers in Table 2); and an unidentified flavanone (tR = 11.5). Unidentified flavonols 15, 20, and 21 were present only in minor amounts, whereas in some coenopopulations, the same was true for n-coumaric acid and dihydroquercetin. The largest set of PCs was found in the leaves of S. media plants collected in CP2 and CP7, which are the lowest location (for S. media) (1242 m a.s.l.) and the highest location (1701 m a.s.l.), respectively. The highest total contents of PCs were also  With the increasing altitude, the concentration of astragalin (r = 0.98) increased significantly (p < 0.001), and at p < 0.05, the same was true for avicularin (r = 0.83) and cinnamic acid (r = 0.87). Meanwhile, the hyperoside content diminished significantly (p < 0.05; r = 0.82). Various compounds could be grouped according to the nature of their accumulation in the tested coenopopulations ( Figure 5). The levels of astragalin, avicularin, and cinnamic acid went up with altitude, while the level of hyperoside decreased.  Patterns of accumulation of rutin, spiraeoside, chlorogenic acid, flavonol 4, and flavonol 5 resembled one another. In the plant leaf extracts, the lowest concentrations of these PCs were detected in CP5 (1660 m a.s.l.; Figure 5). With the increasing distance from this location up and down the altitudinal gradient, the concentration of this set of components increased. Quercetin, phenolic acid 18, and flavonols 14 and 17 also showed patterns similar to one another, but in CP5, there was an increase in their concentrations as compared to neighboring coenopopulations ( Figure 5). A possible reason is special In S. media leaf extracts, 22 PCs were found ( Table 2). The major ones were flavonols: rutin; hyperoside; quercetin; spiraeoside; astragalin; avicularin; kaempferol; unidentified flavonols 4, 5, 14, and 17 (compound ID numbers in Table 2); and an unidentified flavanone (t R = 11.5). Unidentified flavonols 15, 20, and 21 were present only in minor amounts, whereas in some coenopopulations, the same was true for n-coumaric acid and dihydroquercetin. The largest set of PCs was found in the leaves of S. media plants collected in CP2 and CP7, which are the lowest location (for S. media) (1242 m a.s.l.) and the highest location (1701 m a.s.l.), respectively. The highest total contents of PCs were also observed in the same habitats (Figure 4): 41.1 and 52.4 mg/g, respectively. The lowest total contents of PCs were observed in coenopopulations located at 1409, 1660, and 1690 m a.s.l. (24.6, 30.9, and 28.8 mg/g, respectively).
With the increasing altitude, the concentration of astragalin (r = 0.98) increased significantly (p < 0.001), and at p < 0.05, the same was true for avicularin (r = 0.83) and cinnamic acid (r = 0.87). Meanwhile, the hyperoside content diminished significantly (p < 0.05; r = 0.82). Various compounds could be grouped according to the nature of their accumulation in the tested coenopopulations ( Figure 5). The levels of astragalin, avicularin, and cinnamic acid went up with altitude, while the level of hyperoside decreased.
Patterns of accumulation of rutin, spiraeoside, chlorogenic acid, flavonol 4, and flavonol 5 resembled one another. In the plant leaf extracts, the lowest concentrations of these PCs were detected in CP5 (1660 m a.s.l.; Figure 5). With the increasing distance from this location up and down the altitudinal gradient, the concentration of this set of components increased. Quercetin, phenolic acid 18, and flavonols 14 and 17 also showed patterns similar to one another, but in CP5, there was an increase in their concentrations as compared to neighboring coenopopulations ( Figure 5). A possible reason is special micro conditions at this location.
Our results about the profiles of classes of PCs and of individual PCs in leaf extracts from S. media and S. chamaedryfolia along the altitudinal gradient in the Multa River valley (Altai Mountains) revealed significant variation in concentrations of PC classes and individual PCs. In S. media leaves, the main PCs were found to be affiliated with the class of flavonols, whereas in S. chamaedryfolia leaves, in addition to flavonols, a high level of flavanone was registered. Different directions of the response of individual PCs to changes in growth conditions (associated with altitude) were documented in our work. For instance, an increase in altitude had a significant positive correlation with the concentration of chlorogenic acid and flavanone in the extracts of S. chamaedryfolia leaves and a negative correlation with concentrations of cinnamic acid, astragalin, and kaempferol there. A direct statistically significant correlation was detected between the altitude where the plants grew and the total contents of polyphenols and of phenolcarboxylic acids in this species. In the extracts of S. media leaves, as altitude increased, levels of astragalin, avicularin, and cinnamic acid increased too, while the concentration of hyperoside declined. Regarding the variation of concentrations of the other identified compounds in the extracts, no significant dependence on altitude was found (in both species).
The opposite directions of some responses of PC classes and of individual PCs to the changes in growth conditions (associated with altitude) can apparently be explained by the different functions performed by individual polyphenols in the physiological processes of the plants. Many researchers have reported either a positive correlation between altitude and levels of PCs [28,29] or different directions of this correlation for different PC classes and for individual PCs [30][31][32][33]. Previously, we have also noticed different directions of the response of classes of PCs to shifts of growth conditions along an altitudinal gradient in populations of Lonicera caerulea in the Altai Mountains [8,34].
The patterns of changes in concentrations of PC classes and of individual PCs along the altitudinal gradient turned out to be different between the two Spiraea species analyzed in the current work. These differences between the closely related species are determined by the aforementioned features of the profiles of individual PCs and of their classes and by species-specific optima of growth conditions. It is likely that the different ecological and phytocoenotic affiliations of these species differently affect the synthesis of classes and of individual secondary metabolites in these plants. S. chamaedryfolia grows in the forest belt and subalpine vegetation belt in the mountains of Europe, Southern Siberia, Northern China, and Mongolia, where it occurs at up to 2000 m a.s.l. [26,27], in contrast to S. media, which has a wider geographic range, including plains. The latter species occurs throughout Russia and in Europe, Central Asia, China, Korea, Japan, and Mongolia [26,27]. In the Altai Mountains, the Spiraea species under study differ in growth conditions. S. chamaedryfolia is confined more to mountain forest communities, while S. media is confined more to shrubby meadow communities. It is likely the differences in the patterns of synthesis of polyphenols between the analyzed Spiraea species are also related to the confinement of these species to dissimilar ecological and phytocoenotic conditions.

Plant Material
The study was conducted in 2019 in the Republic of Altai, in a valley of the Multa River (Ust-Koksinsky District) within a geobotanical sub-province called the Central Altai. The study sites are located in Katunsky Nature Reserve and its border area; this choice ensured the absence of technogenic pollution and of its possible influence on the variation of a secondary metabolism.

Preparation of the Extract
To assay PCs, aqueous-ethanol (70% ethyl alcohol) extracts from S. media and S. chamaedryfolia leaves were prepared by means of a water bath. Approximately 1.0 g of the raw material (precise weight was recorded) that passed through a sieve with a pore diameter of 2-3 mm was placed into a 100 mL round-bottom flask with a ground-in stopper. The raw material was covered with 30 mL of 70% ethanol, and the flask was attached to a reflux condenser and placed in a boiling water bath for 30 min incubation. The flask was shaken from time to time to wash particles of the raw material off the walls. After that, the flask with the extract was cooled, and the first portion of the extract was passed through a paper filter into a conical flask (volume 100 mL) with a ground-in stopper. Next, the raw material from the filter was put in a round-bottom flask and again covered with 30 mL of 70% ethanol and extracted for 30 min in the boiling water bath. The second portion of the extract was cooled and filtered into the first portion of the extract in the 100 mL flask. The procedure was repeated one more time. The three portions of the extract were mixed, and the volume of the resulting combined extract was measured. The sampling was carried out simultaneously for the two investigated species at the end of the flowering stage. The plant material was collected for 10 days to ensure that the sampling was performed in the same phase of seasonal plant development. The plant leaves were air-dried completely under natural conditions.
The study sites are located in Katunsky Nature Reserve and its border area; this choice ensured the absence of technogenic pollution and of its possible influence on the variation of a secondary metabolism.

Preparation of the Extract
To assay PCs, aqueous-ethanol (70% ethyl alcohol) extracts from S. media and S. chamaedryfolia leaves were prepared by means of a water bath. Approximately 1.0 g of the raw material (precise weight was recorded) that passed through a sieve with a pore diameter of 2-3 mm was placed into a 100 mL round-bottom flask with a ground-in stopper. The raw material was covered with 30 mL of 70% ethanol, and the flask was attached to a reflux condenser and placed in a boiling water bath for 30 min incubation. The flask was shaken from time to time to wash particles of the raw material off the walls. After that, the flask with the extract was cooled, and the first portion of the extract was passed through a paper filter into a conical flask (volume 100 mL) with a ground-in stopper. Next, the raw material from the filter was put in a round-bottom flask and again covered with 30 mL of 70% ethanol and extracted for 30 min in the boiling water bath. The second portion of the extract was cooled and filtered into the first portion of the extract in the 100 mL flask. The procedure was repeated one more time. The three portions of the extract were mixed, and the volume of the resulting combined extract was measured.

Determination of the Set and Levels of Individual PCs by High-Performance Liquid Chromatography (HPLC)
One milliliter of each aqueous-ethanol extract was diluted with double-distilled water to 5 mL and passed through a Diapak C16 concentrating cartridge (BioChemMac Corporation, Moscow, Russia). Substances were eluted from the cartridge by means of 3 mL of 70% ethanol and then using 2 mL of 96% ethanol. The combined eluate was passed through a membrane filter with 0.45 µm pore size.
PCs in the eluate were studied by means of an analytical HPLC system consisting of an Agilent 1200 liquid chromatograph (USA) equipped with a diode array detector, an autosampler, and a ChemStation system for the collecting and processing of chromatographic data by a method of T.A. van Beek [35] with modifications, on a Zorbax SB-C18 column (4.6 × 150 mm, 5 µm). The chromatographic analysis was first carried out as isocratic elution in a methanol-0.1% orthophosphoric acid (31:69) system for 27 min, and then in a gradient elution mode: in the mobile phase, the proportion of methanol in the mixture with orthophosphoric acid (0.1% in water) was changed from 33% to 46% for 11 min, then from 46% to 56% during the next 12 min, and finally from 56% to 100% for 4 min. The eluent flow rate was 1 mL/min, column temperature was 26 • C, and injection volume was 10 µL. Detection was carried out at wavelengths (λ) 254, 270, 290, 340, 360, and 370 nm.
The quantification of PCs was conducted as reported previously [36]. The quantitative analysis of individual PCs in the plant samples was performed by the external standard method at λ = 360 nm. Standard samples were prepared from cinnamic acid (Serva Heidelberg, Germany), chlorogenic and p-coumaric acids, quercetin, and kaempferol (Sigma-Aldrich Taufkirchen, Germany), isoquercitrin, rutin, avicularin, spiraeoside, and astragalin (Fluka, Chemie AG, Buchs, Switzerland), and dihydroquercetin (Austrowaren, Austria). By means of spectral characteristics, unidentified compounds were assigned to a class of PCs according to ultraviolet spectral characteristics from the literature [37][38][39]. The calculation of phenolic acids' concentrations was carried out on the basis of chlorogenic acid, and the calculation of flavonoids' concentrations on the basis of rutin.
The relative standard deviation (σ r,rel , a metric of repeatability) of the quantification of PCs was 0.011; the relative standard deviation in terms of retention time in the HPLC was 0.0018.

Statistical Analysis
Statistical processing of the experimental data was carried out using the STATISTICA 6.1 software package. Correlation analysis was performed by the Pearson method; critical significance levels are given in ref. [40]. Student's t-test was performed to determine the statistical significance of a difference in substance levels between specimens of Spiraea species. Multiple comparisons were performed by one-way ANOVA followed by Tukey's honestly significant difference (HSD) test to evaluate the significance of differences among the means [40]. The significance level was set to p < 0.01 and 0.05, and the data are presented as mean ± standard deviation.

Conclusions
A comparative study on the variation in quantitative profiles of classes and of individual biologically active PCs in leaf extracts from resource species S. chamaedryfolia and S. media in coenopopulations of the Altai Mountains in the Multa River valley showed a significant variation of secondary-metabolites' concentrations depending on the sampling site along an altitudinal gradient. In S. chamaedryfolia leaf extracts, the total content of polyphenols and phenolcarboxylic acids significantly increased with increasing altitude, mainly owing to the significant upregulation of chlorogenic acid and flavanone. Levels of cinnamic acid, astragalin, and kaempferol significantly diminished with altitude. For S. media, an increase in altitude correlated significantly positively with concentrations of astragalin, avicularin, and cinnamic acid in the leaf extracts and correlated negatively with the hyperoside level. The species-specific patterns of changes in PCs' levels along the altitudinal gradient are probably determined by the confinement of these species to different ecological and phytocoenotic conditions.  Data Availability Statement: Raw data of this study are available upon request from the corresponding author.