A Chemometric Approach to Distribution of Selenium in Medicinal Plants Cultivated in Poland

Abstract

The aim of this study was to evaluate the level of selenium (Se) in common raw plant materials (herbs, leaves, flowers, fruits, and roots) obtained from medicinal plants cultivated in Poland. Furthermore, the relationship between the morphological part of a plant and its species was examined, and the concentration of Se in it was measured. Spectrofluorimetric determination of Se in 148 samples of 44 plant species revealed that the majority of the plants contained Se at a level from several to several tens of μg/kg dry weight (d.w.). A relatively high Se concentration, in the order of 50 μg/kg d.w., was found in Majoranae herba, Crataegi fructus, and Lini semen. An especially high Se level, >100 μg/kg d.w., was found in only three plants—Equiseti herba, Farfarae folium, and Cichorii radix. Application of a nonparametric Kruskal–Wallis analysis of variance test indicates that the morphological variable influences the level of Se in the studied groups of raw plant materials. The mean concentration of Se in fruits differs significantly from that found in flowers and roots. However, there was no statistically significant difference between the Se content in herbs and leaves and its levels in fruits, flowers, and roots. Cluster analysis and principal component analysis calculations show that it is possible to relate the Se concentration in plant material to the plant species and botanical family of the medicinal plant in only a few cases. Analysis of average Se levels in the plant species showed that the plants belonging to the Apiaceae and Labiatae botanical families are more Se-rich than those belonging to Rosaceae.

Introduction

S elenium (Se) is an essential microelement in human nutrition, and its deficiency or excess has been implicated in several diseases. Because this element behaves both as an antioxidant and anti-inflammatory agent, any condition associated with increased oxidative stress or inflammation might be expected to be influenced by Se level. This is thought to be the case in rheumatoid arthritis and asthma, which are now the most prevalent chronic inflammations.^1,2

As selenocysteine, Se is a component of more than 35 selenoproteins, such as glutathione peroxidases, iodothyronine deiodinases, thioredoxin reductase, selenoprotein P, and selenoprotein W, some of which have important enzymatic functions.^2
–4 This microelement, like iodine, is required for appropriate thyroid hormone synthesis, activation, and metabolism.^5
–7 Low Se status is also associated with a significantly increased incidence of depression, anxiety, confusion, and hostility.^2,5 Furthermore, this bioelement is associated with reduction in total cancer incidence, including lung, colorectal, colon, prostate, and gastric cancers.³

In the industrialized world, dietary Se deficiency is thought to be associated with cardiovascular disease and a weakened immune system.^1,2 Se deficiency can result in accumulation of fatty acid peroxides in the heart and lead to formation of substances that enhance blood clot formation.³ Human Se deficiency diseases have been recognized in the Keshan province of China, where soils have extremely low Se content (Keshan disease is an endemic cardiomyopathy; Kashin-Beck disease is a deforming arthritis).^7

–10 Se deficiency has also been reported in patients with Friedreich's ataxia, and there are histological similarities between Friedreich's cardiomyopathy and Keshan disease.¹¹

These data are evidence of the crucial impact of Se on health conditions of the human population. Due to this fact, numerous investigations have targeted determination of the Se level and forms (chemical speciation) in foods of plant and animal origin. Determination of Se in these samples is challenging, owing to its low level and volatility, especially when the sample is mineralized. Therefore, several methods have been developed for the quantification of Se in plant samples.^12,13 Spectrofluorimetry is one of the simplest, least expensive, and most versatile methods for the determination of Se in food.^14,15 Recently, the hydride generation technique coupled with spectroscopic methods, such as atomic absorption spectrometry^13,16,17 and atomic fluorescence spectrometry,^12,13,18 have been used as an excellent method for the separation of Se from a wide range of matrices through conversion to its volatile hydride. These methods, as well as inductively coupled plasma–mass spectrometry,^19,20 are particularly useful for the trace analysis of Se in plant samples.

The content of Se in agricultural crops is usually <1 mg/kg.²¹ Generally, fruits and vegetables have a low Se concentration. According to Smrkolj et al.,^9,13 the average concentration of Se in lettuce, carrots, and onions varied from 0.3 to 20.0 μg/kg dry weight (d.w.), but in cabbage and beans it ranged 1.1–76.7 and 30.4–81.4 μg/kg, respectively. Similar results were also reported by Pappa et al. ²² and Kadrabova et al.²³ In lettuce, carrots, onions, apples, pears, and strawberries, the average Se level ranged between 0.9 and 8.5 μg/kg. These studies also show that garlic and beans accumulate higher contents of Se, namely, 12.7–14.6 and 20.8–28.1 μg/kg, respectively.²² The highest concentration of Se was determined in beans, 152.1–269.7 μg/kg.²⁴

Because of the lack of data on Se content in plants used in medicine, the aim of this work was to assess the Se level in the most popular medicinal plants growing in Poland, and to identify plants especially rich in this element using multivariate statistical methods, cluster analysis (CA), and principal component analysis (PCA).²⁵ The importance of this study results from the fact that the ability of plants to accumulate and transform inorganic forms of Se into bioactive organic compounds has important implications for human nutrition and health.

For the study, selected species of medicinal plants were chosen, most of which belong to four families: Apiaceae, Asteraceae, Labiatae, and Rosaceae.²⁶ The studied plants contain important substances from the biological point of view, such as essential oils, terpene compounds, and phenolic compounds including phenolic acids, flavonoids, lignans, and tannins. They are commonly administered in Poland in different formulations, such as extracts, infusions, and herbal medicines, to treat many health problems. For example, medicines obtained from these plants are applied to treat colds, flu, insomnia, and digestive ailments; are used as sedative agents, especially to combat muscle pain; and are used to aid circulation, kidney function, and liver function.

Materials and Methods

Plant material

The content of Se was determined in 148 plant samples (herbs, leaves, flowers, fruits, and roots) originating from 44 species of medicinal plants. The samples were collected by Medicinal Plants Works at various factories in Poland: Boguccy (Krakow), Elanda (Rozprza), Herbalux (Warsaw), Herbalux-Bis (Warsaw), Herbapol (Bydgoszcz, Krakow, Lublin, Lodz), Kawon (Gostyn), and Flos (Mokrsko). Before the analysis, the samples were ground using a Knifetec sample mill (Foss-Tecator, Hillerod, Denmark) and kept in plastic containers.

Digestion procedure

For digestion of the samples, a high-pressure microwave system (UniClewer™ BM 1z; Plazmatronika, Wroclaw, Poland) was used. About 2 g of a material was accurately weighed, transferred into a teflon tube, and digested by using a mixture of 7 mL of 65% HNO₃ (Selectipur™; Merck KGaA, Darmstadt, Germany) and 7 mL of double-distilled water obtained from a quartz-glass system (Heraeus Materials, Yverdon, Switzerland). The temperature and pressure of the digestion were maintained within the range 100°C–200°C and 17–45 atm (1722.5–4559.6 kPa), respectively. After digestion, the sample was transferred into a volumetric flask and diluted with the double-distilled water.

Se determination

Se in plant samples was determined by spectrofluorimetry (luminescence spectrometer LS 50B; Perkin Elmer, Waltham, MA, USA). To the analyzed solution, 3 mL of 36% HCl (Instra-Analyzed; Avantor, Deventer, The Netherlands) was added, and to reduce Se(VI) to Se(IV), the solution was heated at 60°C for 20 min. Next, 5 mL of a 50% HCOOH (POCh, Gliwice, Poland) and 5 mL of a stabilizing solution containing EDTA and NH₂OH·HCl (both from PChO, Lublin, Poland) were added. The pH of the mixture was adjusted to ∼2 with a 25% ammonia solution (POCh). Two milliliters of a 0.1% 2,3-diaminonaphthalene solution (Sigma-Aldrich, Poznan, Poland) in 0.1 M HCl extracted twice with cyclohexane was added to the sample and heated on the water bath at 60°C for 20 min. After heating, 5 mL of cyclohexane (Sigma) was added and the mixture was shaken. The organic phase was separated and the cyclohexane layer was measured at 377 nm (excitation wavelength) and 522 nm (emission wavelength). A Se stock standard solution contained 1 mg Se/mL (Merck). Working standard solutions of Se were prepared daily by dilution of the stock standard solution.

Based on linear regression analysis, the linearity for concentrations between 25 and 200 ng Se/25 mL of standard solutions was established and the calibration equation was y=(0.8477±0.0353)x+(40.28±3.81) (n=54, r=0.9993, R ²=0.9986), RSD of the slope was ≤6.79% (n=54), the limits of detection and quantification were 0.00969 and 0.029 ng Se/25 mL of solution, respectively. Besides, by using a Virginia Tobacco Leaves (CTA-VTL-2) certified reference material, the accuracy and precision of Se determination were assessed as follows, the recovery of 75.4%, RSD 5.2% (n=6).

Software and calculations

All calculations were performed by a one-way analysis of variance (ANOVA) test, CA and PCA by using the Statistica 7.1 (Statsoft^®, Krakow) software. A starting point for CA and PCA calculations was a matrix of the data with n×p dimensions, where n is the number of rows, that is, the species of medicinal plants, and p is the number of variables, that is, the average Se concentrations (minimum, maximum, arithmetic mean, and median) in these species.

Results and Discussion

The results of Se determination in 148 plant samples representing the different morphological parts of plants—herbs, leaves, flowers, fruits, and roots—and originating from 44 species of medicinal plants show that three plants, Equiseti herba, Farfarae folium, and Cichorii radix, contained extremely high amounts of Se, >100 μg/kg. A high Se level, ∼50 μg/kg, was also determined in Majoranae herba, Crataegi fructus, and Lini semen. In the remaining plants, the Se level extended from several to several tens of μg/kg.

Equiseti herba, Farfarae folium, and Cichorii radix were represented by nine samples originating from different plant cultivators. Because an extremely high Se content in these samples vitiate the data with the mean concentration of this element in the other plants, the nine samples of Equiseti herba, Farfarae folium, and Cichorii radix were excluded from the total number of 148 samples, to obtain in this way a new database comprising 139 plant samples. The mean Se concentrations in this subgroup of plants are collected in Table 1 and presented graphically in Figure 1.

FIG. 1.

Graphical presentation of the selenium (Se) content in five different morphological parts of the medicinal plants, after rejection of nine samples of Equiseti herba, Farfarae folium, and Cichorii radix.

Table 1.

General Description of the Selenium Content in the Analyzed Plants, After Rejection of Nine Samples of Equiseti Herba, Farfarae Folium, and Cichorii Radix

Plant samples	Herbs	Leaves	Flowers	Fruits	Roots
Number of samples	35	25	32	27	20
Arithmetic mean (μg/kg d.w.)	22.96	21.71	14.76	31.54	14.03
Median (μg/kg d.w.)	19.07	18.18	14.59	27.98	12.21
Minimum (μg/kg d.w.)	2.10	3.93	1.24	6.58	2.76
Maximum (μg/kg d.w.)	91.92	44.33	42.61	89.85	36.93
Range (μg/kg d.w.)	89.82	40.40	41.37	83.27	34.17

d.w., dry weight.

The Se content in particular groups of the samples under study were found to be comparable. The highest average concentration of this element was found in fruits and seeds and the lowest in the roots and flowers. To verify whether these five morphological parts of the plants differ significantly in their Se levels, the ANOVA test was used. The results obtained for 139 plant samples at a P<.05 significance level have shown that application of the nonparametric Kruskal–Wallis ANOVA test indicates that the morphological variable significantly influences the concentration of Se in the studied groups of plants. The mean Se concentration in fruits differs significantly from that determined in flowers and roots. However, there was no statistically significant difference between the Se concentration in herbs and leaves compared with its level found in fruits, flowers, and roots.

Among the plant samples investigated, the highest average Se level, 31.54 μg/kg, was found in the fruits and seeds. In this group, the highest concentration of this element was found in Crataegi fructus, 76.88 μg/kg (Herbapol, Bydgoszcz). This level of Se is higher than those determined in three other samples of the same plant material—45.79, 49.95, and 25.79 μg/kg, collected from different herbal cultivators: Herbapol (Lublin), Kawon, and Herbalux, respectively. Higher Se concentrations, 68.32 and 50.00 μg/kg, respectively, were found in two samples of Lini semen, while the lowest Se content of 9.91 μg/kg was in Rosae fructus.

Rich sources of Se are also herbs with the average concentration of 22.96 μg/kg. In this group of samples, the highest level of the element was found in Equiseti herba, 133.7 μg/kg (Kawon), and this value is 6.5 times higher than the Se level in another sample of the same plant material, 19.90 μg/kg (Boguccy). A high level of Se was also found in two samples of Majoranae herba, 63.04 and 86.16 μg/kg (Herbapol, Bydgoszcz), whereas the third sample of the same plant material contained only 3.82 μg Se/kg (Kawon). Artemisii herba is one of two plants with the lowest Se level, 2.77μg/kg. An important source of Se is also Thymi herba, because three samples of this plant material contained 44.91, 39.40, and 47.85 μg/kg.

A closer inspection of the concentration of Se in leaves showed that the majority of samples contained the element in the range from 15 to 25 μg/kg, with the average level of 21.71 μg Se/kg. In this group of plant materials, the sample with the highest Se concentration among all the studied plants was Farfarae folium (Kawon), with an Se level of 623.8 μg/kg. The remaining three samples of this plant contain lower amounts of Se, 201.8, 108.4, and 142.5 μg/kg. A significantly high level of this element was also found in Menthae folium, 41.99 μg/kg, and in Majoranae folium, 40.31 μg/kg, and the lowest Se concentration was in Melissae folium, 8.45 μg/kg.

Flowers and roots are characterized by the lowest Se level among all of the studied plants, and the average Se content in flowers and roots is 14.76 and 14.03 μg/kg, respectively. Flowers generally contain Se within the range from 10 to 20 μg/kg. The highest amount of the element was determined in Calendulae flos, 35.47 μg/kg (Herbapol, Lublin), and this is a higher value than that in the other samples of the same plant material obtained from Herbapol (Bydgoszcz), 19.57 and 25.88 μg/kg, and from Kawon, 22.79 and 17.55 μg/kg.

Among the roots, Cichorii radix (Kawon) is rich in Se; two samples had Se concentrations of 152.92 and 130.97 μg/kg. The lowest level of the element was in Bardanae radix (Kawon), 4.92 μg/kg; this was much lower in comparison with two other samples of this plant material—15.02 and 22.83 μg/kg.

When comparing the concentration of Se in samples originating from the same plant species, we found that the Se levels differ depending on the geographical region of Poland from which they were collected. For example, the fruits of cumin obtained from the Wielkopolska region (southwestern Poland) were characterized by similar Se levels of 31.10, 31.84, and 33.33 μg/kg, whereas the sample originating from central Poland (the Mazowsze region) had an Se level that was twice lower, 14.42 μg/kg. A higher concentration of Se was found in a sample of hawthorn coming from the Pomorze region (northern Poland), 76.88 μg/kg, as compared with samples obtained from the Wielkopolska and Lubelszczyzna (southeastern Poland) regions, in which the Se levels of 49.95 and 45.79 μg/kg were determined. It should be emphasized that in some cases, the samples originating from the same geographical region of Poland differed significantly in their Se level. For example, in the samples of Cnici benedicti herba from the Wielkopolska region, different amounts of Se were found: 23.57, 33.15, and 11.19 μg/kg. A similar situation in the case of cornflower samples originating from the Lubelszczyzna region was noticed—3.95, 18.19, and 3.77 μg Se/kg.

Average Se levels, expressed as median, arithmetical mean, minimum, and maximum, in the species of medicinal plants from which the analyzed samples originated were statistically evaluated to compare the mean concentration values of the element in the analyzed material. To show whether or not particular samples differed widely in Se levels, two pattern recognition methods, CA and PCA, were applied.²⁵

A graphical illustration of the results concerning CA calculations is shown in Figure 2A. Taking into consideration the Sneath's index, in the dendrogram, three important clusters were designed at 33% of the maximum distance. Cluster I groups three species of medicinal plants—Linum usitatissimum, Crataegus monogyna, and Majorana hortensis. Inspection of the dataset in Table 2 indicates that these differ from the other plant species, with high Se content in the order of 50 μg/kg, and particular morphological parts of two of these species differ in their Se level. The species C. monogyna is represented by three samples of flowers and four samples of fruits, but in cluster I, there are only fruits because of an almost three times higher Se content in comparison with that of the flowers. A similar situation has been noticed in the case of plants of the M. hortensis species. A high average Se concentration in three analyzed samples of the herbs is the reason for their location in cluster I. A leaf sample from the same plant species with a lower Se content creates cluster IIa with four herbs from the Thymus serpyllum species in the order of 40 μg/kg. Both medicinal plants belong to the Labiatae family.

FIG. 2.

Dendrogram of cluster analysis (A) and PC1 vs. PC2 score plot (B) for average Se levels in the species of medicinal plants. Calculations were done based on the data compiled in Table 2, after rejection of nine samples of Equiseti herba, Farfarae folium, and Cichorii radix. The dendrogram clusters (I, IIa and b, and IIIa and b) are described in the text. The two ovals in the principal component analysis indicate clusters IIb and IIIa.

Table 2.

Average Selenium Concentration in the Species of Medicinal Plants

Species of medicinal plants	Plant samples ^a	Range	Arithmetic mean ^b [μg/kg d.w.]	Median
Apiaceae family
Archangelica officinalis	Roots (3)	5.44–22.76	11.27 (9)	9.33
Carum carvi	Fruits (4)	13.02–39.29	27.67 (12)	30.84
Coriandrum sativum	Fruits (3)	19.80–35.60	26.01 (9)	26.78
Foeniculum vulgare	Fruits (4)	21.75–37.45	30.52 (8)	30.39
Levisticum officinale	Roots (3)	2.76–20.51	12.61 (9)	13.86
Pimpinella anisum	Fruits (3)	22.73–29.24	27.13 (7)	27.98
Asteraceae family
Achillea millefolium	Herbs (3)	8.94–26.27	17.05 (7)	17.05
	Flowers (4)	3.68–31.67	11.83 (12)	9.67
Arctium lappa	Roots (3)	5.21–28.57	14.26 (9)	15.02
Arnica montana	Flowers (2)	12.87–19.19	15.90 (6)	15.35
Artemisia absinthium	Herbs (3)	2.10–21.51	11.75 (9)	12.92
Artemisia vulgaris	Herbs (2)	9.04–12.99	10.88 (6)	10.55
Calendula officinalis	Flowers (5)	14.33–42.61	24.25 (15)	24.5
Centaurea cyanus	Flowers (4)	3.10–23.29	11.24 (12)	9.05
Chamomilla recutita	Flowers (3)	10.37–21.70	15.40 (9)	15.30
Cichorium intybus	Roots (3)	126.8–159.6	141.9 (7)	141.9
Cnicus benedictus	Herbs (4)	6.07–38.46	18.81 (12)	16.55
Helichrysum arenarium	Flowers (2)	1.76–20.05	10.91 (7)	10.90
Inula helenium	Roots (3)	10.87–15.07	12.45 (5)	12.47
Mentha piperita	Leaves (4)	3.93–44.33	20.49 (12)	16.61
Solidago virgaurea	Herbs (3)	5.61–27.35	13.42 (9)	10.58
Taraxacum officinale	Roots (3)	3.62–11.96	7.94 (9)	8.11
Tussilago farfara	Leaves (4)	100.6–624.2	296.1 (12)	170.5
Labiatae family
Lamiom album	Flowers (3)	1.24–18.38	10.47 (9)	11.67
Lavandula officinalis	Flowers (4)	7.58–18.49	12.48 (12)	12.74
Majorana hortensis	Herbs (3)	2.91–81.28	51.00 (9)	61.32
	Leaves (1)	34.30–43.79	40.31 (3)	42.85
Melisa officinalis	Herbs (2)	16.37–29.85	22.98 (6)	23.23
	Leaves (3)	5.26–26.71	15.60 (9)	16.40
Organum vulgare	Herbs (2)	15.74–24.46	20.22 (6)	20.04
Salvia officinalis	Leaves (4)	10.26–36.60	29.06 (12)	32.53
Thymus serpyllum	Herbs (4)	17.90–51.87	37.90 (12)	41.58
Thymus vulgaris	Herbs (4)	23.27–33.07	26.85 (12)	27.23
Rosaceae family
Crataegus monogyna	Flowers (3)	8.31–22.53	17.88 (9)	17.18
	Fruits (4)	22.89–89.85	46.60 (12)	49.12
Fragaria vesca	Leaves (3)	12.13–21.31	15.45 (7)	15.71
Potentilla anserina	Herbs (3)	9.25–18.17	14.27 (9)	14.86
Potentilla tormentilla	Roots (3)	9.83–22.5	14.39 (9)	12.89
Rosa canina	Fruits (4)	7.74–27.79	16.30 (10)	15.85
Rubus fruticosus	Leaves (4)	10.93–20.32	15.33 (10)	14.79
Rubus idaeus	Leaves (4)	14.43–31.93	20.78 (12)	17.96
Sorbus aucuparia	Fruits (3)	6.56–26.74	18.00 (7)	18.14
Other families
Aesculus hippocastanum	Flowers (1)	8.17–14.83	8.61 (3)	9.06
Arctostaphylos uva ursi	Leaves (2)	17.77–38.34	29.10 (6)	30.55
Equisetum arvense	Herbs (2)	18.04–144.9	76.75 (6)	71.68
Glycyrrhiza gabra	Roots (2)	22.87–31.18	29.86 (6)	30.70
Linum usitatissimum	Seeds (2)	47.32–79.97	59.16 (6)	57.38
Melilotus officinalis	Herbs (2)	9.40–22.64	17.43 (6)	19.03

Number of plant samples under study is given in parentheses.

Number of independent spectrofluorimetric determinations is given in parentheses.

Cluster IIb encompasses 31 samples (fruits, leaves, flowers, herbs, and roots), originating from 9 medicinal plants. The largest group in this cluster is the one containing plant species from the Apiaceae family—Carum carvi, Coriandrum sativum, Foeniculum vulgare, Pimpinella anisum—and is represented by 14 fruit samples. The other plant species placed in cluster IIb are Salvia officinalis—Thymus vulgaris (Labiatae family), Calendula officinalis (Asteraceae family), Arctostaphylos uva ursi (Ericaceae family), and Glycyrrhiza gabra (Fabaceae family). All the specified plant species have in common the fact that the average Se content in particular samples of fruits, leaves, flowers, herbs and roots is ∼30 μg/kg.

In cluster III, the 94 remaining samples represent species of medicinal plants shown in Table 2. Particular attention should be paid to a small cluster on the left of cluster IIIa, with the following plant species: Cnicus benedictus and Mentha piperita (Asteraceae family), Melisa officinalis and Origanum vulgare (Labiatae family), and Rubus idaeus (Rosaceae family). These species have a common feature, which is an average Se concentration in the order of 20 μg/kg in eight samples of herbs and eight samples of leaves. The remaining plant samples located on the right of cluster III can be characterized by a close similarity in their Se level, ∼12 μg/kg. This is confirmed by small distances between adjacent clusters, showing that no relation exists between the Se content of particular samples and plant species from which the studied materials originated.

Interpretation of PCA calculations shown graphically in Figure 2B fully confirms the results obtained by the use of CA. Three medicinal plant species with a high Se level (∼50 μg/kg)—L. usitatissimum (44), C. monogyna (33), and M. hortensis (24)—are located in the left part of the PC1 versus PC2 plot. A similar group is made up of the leaves of M. hortensis (25) and the herbs of T. serpyllum (30), which corresponds to cluster IIa in the CA dendrogram (Fig. 2A). Two other clusters, IIb and IIIa—located in the central area of the PC1 versus PC2 plot, and encompassed by two ellipses—involve plants with a similar Se concentration of ∼30 and ∼20 μg/kg, respectively. These results indicate that only in few cases, it is possible to find a relation between the Se content and the analyzed plant species. For the majority of the crude herbal drugs, it is not possible to derive such a relation.

Summary

The results of this study show that the majority of samples originating from medicinal plants cultivated in Poland contain Se at a level from several to several tens of μg/kg. A relatively high Se concentration, around 50 μg/kg, was determined in Majoranae herba, Crataegi fructus, and Lini semen. An especially high Se level, >100 μg/kg, was found only in three plant materials—Equiseti herba, Farfarae folium, and Cichorii radix.

It has also been noted that the morphological variable has a crucial impact on the Se level in the herbs, leaves, flowers, fruits, and roots. Statistically significant differences were found between the Se concentration in fruits and its levels in flowers and roots. No significant difference was noticed in the Se concentration between herbs and leaves, nor between fruits, flowers, and roots.

The results of CA and PCA calculations show that in only a few cases, it is possible to relate the Se concentration in a plant material to the plant species and the botanical family of the medicinal plant. A chemometric analysis also shows that more rich in Se are the plants belonging to the Apiaceae and Labiatae than to the Rosaceae botanical families. In the case of the Asteraceae family, the majority of the plant species are characterized by a relatively low Se content, with the exception of two species—Cichorium intybus and Tussilago farfara.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Brown

, Arthur

. Selenium, selenoproteins and human health: a review. Public Health Nutr, 2001; 4:593–599.

Rayman

. The importance of selenium to human health. Lancet, 2000; 356:233–241.

Holben

, Smith

. The diverse role of selenium within selenoproteins: a review. J Am Diet Assoc, 1999; 99:836–843.

Rasmussen

, Hollenbach

, Laurberg

, Carlé

, Hög

, Jørgensen

, Vejbjerg

, Ovensen

, Schomburg

. Serum selenium and selenoprotein P status in adult Danes—8-year followup. J Trace Elem Med Biol, 2009; 23:265–271.

Sher

. Role of thyroid hormones in the effects of selenium on mood, behavior, and cognitive function. Med Hypothes, 2001; 57:480–483.

Beckett

, Arthur

. Selenium and endocrine systems. J Endocrinol, 2005; 184:455–465.

Yao

, Pei

, Kang

. Selenium, iodine, and the relation with Kashin-Beck disease. Nutrition, 2011; 27:1095–1100.

Baoyan

, Xuelin

. Regional ecological risk assessment of selenium in Jilin province, China. Sci Total Environ, 2000; 262:103–110.

Gao

, Liu

, Huang

, Lin

, Bañuelos

, Hon-Wah Lam

. Daily selenium intake in a moderate selenium deficiency area of Suzhu, China. Food Chem, 2011; 126:1008–1093.

10.

Mistry

, Broughton Pipkin

, Redman

CWG

, Poston

. Selenium in reproductive health. Am J Obstet Gynecol, 2012; 206:21–30.

11.

Fryer

. Rationale for clinical trials of selenium as an antioxidant for the treatment of the cardiomyopathy of Friedreich's ataxia. Med Hypotheses, 2002; 58:127–132.

12.

Ventura

, Stibilj

, do Carmo Freitas

, Pacheco

AMG

. Determination of ultratrace levels of selenium in fruit and vegetables samples grown and consumed in Portugal. Food Chem, 2009; 115:200–206.

13.

Smrkolj

, Pograjc

, Hlastan-Ribić

, Stibilj

. Selenium content in selected Slovenian foodstuffs and estimated daily intakes of selenium. Food Chem, 2005; 90:691–697.

14.

Costa-Silva

, Marques

, Matos

, Barros

, Nunes

. Selenium contents of Portuguese commercial and wild edible mushrooms. Food Chem, 2011; 126:91–96.

15.

Gűler

, Maden

, Bekirdere

, Ataman

, Volkan

. Speciation of selenium in vitamins tablets using spectrofluorimetry following cloud point extraction. Food Chem, 2011; 129:1793–1799.

16.

Sigrit

, Brusa

, Campagnoli

, Beldomenico

. Determination of selenium in selected food samples from Argentina and estimation of their contribution to the Se dietary intake. Food Chem, 2012; 134:1932–1937.

17.

Tuzen

, Soylak

. Evaluation of trace element contents in canned food marketed from Turkey. Food Chem, 2007; 102:1089–1095.

18.

Smrkolj

, Stibilj

. Determination of selenium in vegetables by hydride generation atomic fluorescence spectrometry. Anal Chim Acta, 2004; 512:11–17.

19.

Molan

, Flanagan

, Wei

, Moughan

. Selenium-containing green tea higher antioxidant and prebiotic activities than regular green tea. Food Chem, 2009; 114:829–835.

20.

Younju

Choi

, Jiyung

Kim

, Haeng-Shin

Lee

, Cho-Il

Kim

, Kyeong

Hwang

, Hye Kyung

Park

, Chang-Hwan

. Selenium content in representative Korean foods. J Food Comp Anal, 2009; 22:117–122.

21.

Lauchli

. Selenium in plants: uptake, functions, and environmental toxicity. Bot Acta, 1993; 106:455–468.

22.

Pappa

, Pappas

, Surai

. Selenium content in selected foods from the Greek market and estimation of the daily intake. Sci Total Environ, 2006; 372:100–108.

23.

Kadrabova

, Madaric

, Ginter

. The selenium content of selected food from the Slovak Republic. Food Chem, 1997; 58:29–32.

24.

Diaz-Alarcon

, Navarro-Alarcon

, Lopez-Garcia

, Lopez-Martinez

. Determination of selenium in cereals, legumes and dry fruits from southeastern Spain for calculation of daily dietary intake. Sci Total Environ, 1996; 184:183–189.

25.

Hill

, Lewicki

. Statistics, Methods, Applications. A Comprehensive Reference for Science, Industry, and Data Mining. StatSoft, Inc.: Tulsa, OK, USA, 2006.

26.

Kohlműnzer

. Pharmacognosy. PZWL: Warszawa, Poland, 2004.