Dandelion (
At present, dandelion quality evaluation in China is reliant exclusively on China Pharmacopoeia. China Pharmacopoeia (Ed. 2015) stipulated that caffeic acid content in dandelion should be at least 0.02%; the newly released version stipulated that cichoric acid content shall be >0.45%. However, most researchers pointed out that very complicated effective compounds have been found in dandelion, including phenolic acids, flavonoids and terpenes,
According to the theory of traditional Chinese medicine, dandelion functions mainly include heat-clearing and detoxifying effects, and these are attributed to the rich phenolic acids in dandelion, such as caffeic acid, cichoric acid and chlorogenic acid (González-Castejón et al., 2012). These phenolic acids possess similar chemical molecular structure and chemical properties (Grauso et al., 2019; Tajner-Czopek et al., 2020). Therefore, we may infer that the most reasonable way to evaluate the quality of dandelion intended for therapeutic usage would be to determine the concentration of a few main phenolic acid compounds as key indices.
The usage of multiple indices or statistical methods to evaluate plant traits finds broad application in the literature. Li et al. (2021) evaluated dandelion (
Since from 2017, we have collected 96 dandelion resources from different regions of China and together these have been transplanted to the demonstration base of the Institute of Coastal Agriculture, Hebei Academy of Agriculture and Forestry Sciences (North China, 39.23°N, 118.57°E). These resources were cultured in different sites involving greenhouse, saline-alkali land and common field, and harvested in different seasons (spring and autumn). Finally, we obtained a total of 578 dandelion samples. The aboveground parts were harvested after cultivation for 1 month. For related field management, the standard, “Technical specification for cultivation of dandelion in saline-alkali soil” was referred to DB13/T 2986-2019. Dandelion samples were washed and dried at 60 °C. Dry samples were ground and passed by 100-mesh sieve. Water content in the resulting fine powders was controlled <13% according to the stipulation of China Pharmacopoeia. Parts of dandelion samples with some indices were assigned to testing companies, and related details of dandelion samples and examination results are shown in Table S1 in Supplementary Materials.
Phenolic acids were extracted based on our previously published method (Wu et al., 2020a). About 0.5 g dry powder was mixed with 15 mL cellulase solution (cellulase dosage, 0.1%; enzyme activity, 3,000 U · g−1) and placed in water bath at 60 °C for 30 min. Then, 15 mL of methanol was added, and ultrasonic treatment was continued for 30 min (ultrasonic power 400 W). The extract was cooled to room temperature, filtered with a 0.45 μm membrane and then examined using high performance liquid chromatography (HPLC).
The extract was checked by HPLC (HPLC 1200 series, Agilent Technologies Inc., USA) equipped with chromatographic column (Mars ODS-AQ (4.6 mm × 250 mm, 5 μm), Hming Technologies Co. Ltd, China) under our optimised HPLC conditions based on the technical specifications laid down under China Pharmacopoeia (Eds 2015 and 2020). The optimal determination conditions based on our equipment were the following: injection volume, 0.5 μL; flow rate, 1 mL · min−1; methanol:0.2% phosphoric acid ratio, 40:60; detection wavelength, 327 nm; and detection time, 15 min. Meanwhile, the contents of four kinds of phenolic acids were determined by the following linear regression equations (Table 1) based on quantitative analysis with standard chemicals of caftaric acid, chlorogenic acid, caffeic acid and cichoric acid (Li et al., 2021).
Linear regression equations for four compounds in dandelion.
Compounds | Equations | Valid scope (mg · mL−1) | |
---|---|---|---|
Caftaric acid | Y = 0.6342x + 1.0263 | 0.9965 | 10–70 |
Chlorogenic acid | Y = 0.4946x − 0.8089 | 0.9669 | 5–30 |
Y = 0.6148x + 0.9279 | 0.9997 | 30–105 | |
Caffeic acid | Y = 0.4605x − 0.8895 | 0.9905 | 5–30 |
Y = 0.39x + 0.814 | 0.9994 | 30–105 | |
Cichoric acid | Y = 0.4564x + 3.0071 | 0.9995 | 45–220 |
Y = 0.569x − 60.97 | 0.9502 | 240–512 |
Y represents the component content (μg · mL−1), and × the HPLC peak area.
HPLC, high performance liquid chromatography.
Fifteen random HPLC detection spectrums from 578 dandelion samples were imported into the software of Similarity Evaluation System for Chromatographic Fingerprint of TCM published by Chinese Pharmacopoeia Commission (version 2012, Beijing, China) using median method and an adjustment retention time (RT) of 0.1 min. According to the combination analysis of similarity assessment and matching numbers for each component shown in HPLC spectrum, the common and dominant phenolic compounds were chosen as the main index for quality evaluation (Wu et al., 2020).
Data pertaining to the weights of the main compounds contained in the dandelion samples, together with data from quality evaluation related references, were analysed by SPSS 19.0 software (IBM SPSS Statistics, USA) using descriptive statistics involving mean, mode, median and frequency. Then, based on statistics and general level division methods with a certain percentage used in the literature cited in the present article or in the industrial standards in force in China, each compound content was divided into five levels from low to high, of which Level 1 represented the highest content.
Using the equal weight average calculation method, the content level of each component was converted into a quality index (QI) according to the following formula:
According to the QI formula, totally 17 QIs were obtained and then the probability or proportion of each QI was calculated based on the completely random combination of all levels for four components. Considering that setting industry standards in China is possible only by satisfying the requirements of high standard and universality, here we stipulated that the probability of QI <1% be classified into Grade 1, indicating the super high content, while most samples were classified into Grades 2–5, and the overall hierarchical structure was spindle-shaped.
We tested HPLC conditions following the method from China Pharmacopoeia and found that some peaks, except for caffeic acid or cichoric acid, were obviously shown in the HPLC spectrum, but the separation of these peaks was unclear (Figure 1A). Thus, we adjusted HPLC conditions and obtained an ideal HPLC spectrum (Figure 1B). Through HPLC fingerprint comparison analysis, totally 17 compounds were found, of which four compounds (corresponding retention times, ~4.2 min, ~4.7 min, ~6.1 min and ~10.0 min, respectively) showed 100% matching numbers with the reference fingerprint, and they were identified as caftaric acid, chlorogenic acid, caffeic acid and cichoric acid by standard chemicals (Figure 1C and Table S2 in Supplementary Materials). The four compounds possessed similar molecular structures and belonged to phenolic acids (González-Castejón et al., 2012; Tajner-Czopek et al., 2020). Thus, the four compounds were preliminarily considered as candidates for main quality evaluation indices.
Dandelion is theorised to contain complex components and the presence of all of these components has not been proved as yet, and neither have the therapeutic effectiveness or other useful functions of a majority of these components been demonstrated as yet (Fatima et al., 2018). We referred to the traditional Chinese medicine theory that attributes heat-clearing and detoxifying effects to dandelion, and additionally considered the inference that these effects derive mainly from the phenolic acids that are present in this plant. Therefore, we selected some main phenolic acids as the key evaluation indices that could reflect dandelion quality with some extent (Fatima et al., 2018; Lis and Olas, 2019). However, choosing too many or few indices was deemed unsuitable owing to redundancy or insufficiency, respectively. From Figure 1, we see that some unknown compounds (e.g. retention time at ~5.5 min, ~7.1 min and ~10.8 min) were also found in parts of the dandelion samples; however, the four phenolic acids were found in all of the 15 dandelion samples and belonged to caffeic acid derivatives (Tajner-Czopek et al., 2020).
The content (by weight) of each compound in dandelion was different, and the effectiveness of each of these compounds with regard to a particular function (e.g., with regard to a therapeutic function, studies have identified chlorogenic acid as exerting a significant hypoglycaemic action) was also different. Compounds such as caffeic acid, chlorogenic acid and chicoric acid were the most abundant phenolic compounds in dandelion, and these have been recognised to exhibit significant antioxidant, anti-inflammatory, anti-bacterial and anti-carcinogenic effects (Didier et al., 2011). Dandelion has also been reported to contain significant amounts of phytosterols and derivatives such as plant sterols, stanols and sitosterol (Ovesnaâ et al., 2004), and also to exhibit anti-inflammatory activity, anti-inflammatory effect and anti-cancer properties (Aldini et al., 2014). However, the effectiveness of each component in dandelion is difficult to compare. In order to better evaluate the overall quality, some researchers used membership function, principal component analysis and fingerprint similarity analysis to determine the weight of each compound in dandelion. However, the actual operation of these methods is complex and inconvenient, especially for the purposes of popularisation and application for people lacking a specialised knowledge in these statistical techniques. Hence, the current method of dandelion quality evaluation in China mainly follows China Pharmacopoeia, although the single evaluation index used in the Pharmacopoeia has been subject to some dispute. To ensure a comprehensive approach, we adopted the equal weight average calculation method, and referring to local standards presently applicable in China, converted the above four phenolic acids into a QI to evaluate the quality of dandelion, as follows.
We conducted statistical analyses on dandelion samples and collected data from some references meanwhile (shown in Table 2 and 3; and Table S1 in Supplementary Materials). The statistical data in Table 2 and 3 are the reference sources used to classify the content level of each component.
Descriptive statistics of four components (%) in dandelion.
Compounds | Numbers | Min–Max | Mean | Mode | Median |
---|---|---|---|---|---|
Cichoric acid | 470 | 0.0175–2.1732 | 0.3834 | 0.0714 | 0.0369 |
Caffeic acid | 437 | 0.0010–0.0600 | 0.0155 | 0.0138 | 0.0025 |
Chlorogenic acid | 437 | 0.0011–0.5337 | 0.0828 | 0.0294 | 0.0023 |
Caftaric acid | 465 | 0.0234–0.6419 | 0.1705 | 0.0858 | 0.0576 |
Data of components contents (%) from relevant references.
Literature sources | Numbers | Statistics | Cichoric acid | Caffeic acid | Chlorogenic acid | Caftaric acid |
---|---|---|---|---|---|---|
Literature 1 (Hao, 2010) | 25 | Min | – | 0.008 | 0.0050 | – |
Max | 0.0455 | 0.0470 | ||||
Mean | 0.0189 | 0.0173 | ||||
Literature 2 (Lang et al., 1999) | 29 | Min | – | 0.0089 | 0.0363 | – |
Max | 0.0559 | 3.7819 | ||||
Mean | 0.0248 | 0.5154 | ||||
Literature 3 (Liu et al., 2017) | 11 | Min | 0.104 | 0.013 | 0.015 | 0.103 |
Max | 0.599 | 0.051 | 0.072 | 0.441 | ||
Mean | 0.272 | 0.033 | 0.035 | 0.307 | ||
Literature 4 (Chen et al., 2018) | 15 | Min | 0.266 | 0.022 | 0.030 | 0.180 |
Max | 0.909 | 0.060 | 0.072 | 0.359 | ||
Mean | 0.608 | 0.036 | 0.046 | 0.286 | ||
Literature 5 (Ning et al., 2012) | 11 | Min | – | 0.0281 | 0.0338 | – |
Max | 0.0184 | 0.0802 | ||||
Mean | 0.0281 | 0.0576 |
– indicates not checked.
The four compound contents showed different extents of positive correlation according to Pearson correlation coefficient analysis (Table 4). China Pharmacopoeia (Ed. 2020) stipulated the content of cichoric acid as the one and only quality evaluation index. However, further linear regression analysis showed that the coefficient of determination (
Pearson correlation analysis for four components in dandelion.
Components | Cichoric acid | Caffeic acid | Chlorogenic acid | Caftaric acid |
---|---|---|---|---|
Cichoric acid | 1 | 0.410 | 0.934 | 0.889 |
Caffeic acid | 1 | 0.451 | 0.674 | |
Chlorogenic acid | 1 | 0.774 | ||
Caftaric acid | 1 |
China Pharmacopoeia (Ed. 2020) stipulated that the content of cichoric acid in dandelion shall exceed 0.45%, and that in dandelion herbal pieces it shall not be <0.3%. The cichoric acid content in Table 2 ranges from 0.0175% to 2.1732%, with an average of 0.3834%. Based on referring the division method used in Hao's dissertation (2010), content <80% of the average was defined as low (literature 1; Table 3), and thus that value was just in line with the provisions of Pharmacopoeia. Therefore, the content <0.3% was defined as low (Level 5).
Figure 3A showed that the contents of most samples were distributed within 1.2%, which was almost consistent with studies from the literature cited in Table 3. Thus, the valid division range was considered from 0% to 1.2%, and the division of other content levels was comprehensively determined as follows: the reasonable division was reckoned at 0.3%–0.5% as Level 4 referenced from China Pharmacopoeia, 0.5%–0.65% as Level 3 referenced from Chen et al. (2018) (literature 4; Table 3), 0.65%–0.75% as Level 2 and >0.75% as Level 1 referenced from histograms (Figure 3A).
China Pharmacopoeia (Ed. 2015) stipulated that the content of caffeic acid in dandelion shall be >0.02%. The caffeic acid content in Table 2 ranges from 0.001% to 0.06%, with an average of 0.0155%. Similarly, content <80% of the average, and for convenience of application, here <0.015%, was defined as low (Level 5). Histograms of caffeic acid contents showed a relatively uniform distribution (Figure 3B), and thus the percentage distance of 0.01 was reasonable for the division of the rest of content levels as follows: 0.015%–0.025% was determined as Level 4, 0.025%–0.035% as Level 3, 0.035%–0.045% as Level 2 and >0.045% as Level 1.
China Pharmacopoeia did not indicate chlorogenic acid content as QI, and therefore we had to divide its content level based on statistical analysis and related references. The chlorogenic acid content in Table 2 ranges from 0.001% to 0.5337%, with an average of 0.0828%, mode of 0.0294% and median of 0.0023%. The overall contents’ distribution was relatively uniform, except for the two obvious parts divided at 0.1% (Figure 3C). Ling et al. (1999) (literature 2; Table 3) showed the highest content with an average of 0.5154%, which may be the outcome of either actually high content in dandelion or errors of determination results. However, data from the remaining studies cited in this paper were consistent with the statistical analysis. In summary, referring to the mode in Table 2, <0.03% was defined as low (Level 5); <0.085% according to the mean was Level 4; <0.15% (around twice of mean) was Level 3; <0.25% was Level 2 and >0.25% was Level 1 based on histograms (Figure 3C).
Similarly, Caftaric acid was not mentioned in China Pharmacopoeia for dandelion. The caftaric acid content in Table 2 ranged from 0.0234% to 0.6419%, with an average of 0.1705%, mode of 0.0858% and median of 0.0576%. The overall contents’ distribution was relatively uniform (Figure 3D). The data distribution evident in the studies cited in Table 3 was similar to that discerned from the statistical analysis. Thus, <0.1% was defined as low (Level 5) according to the mean and mode; and then the remaining content levels were divided sequentially based on the interval distance of 0.1, as follows: 0.1%–0.2% was reckoned as Level 4, 0.2%–0.3% as Level 3, 0.3%–0.4% as Level 2 and >0.4% as Level 1. Finally, we summarise the four kinds of phenolic acids’ content levels in Table 5.
Contents (%) level division for dandelion.
Index/level | 1 | 2 | 3 | 4 | 5 |
---|---|---|---|---|---|
Cichoric acid | ≥0.750 | ≥0.650 | ≥0.500 | ≥0.300 | <0.300 |
Caffeic acid | ≥0.045 | ≥0.035 | ≥0.025 | ≥0.015 | <0.015 |
Chlorogenic acid | ≥0.250 | ≥0.150 | ≥0.085 | ≥0.030 | <0.030 |
Caftaric acid | ≥0.400 | ≥0.300 | ≥0.200 | ≥0.100 | <0.100 |
The equal weight average calculation method was used here. Actually, this method is often adopted in domestic or industrial standard setting in China, and some examples are “Rules for characterisation and evaluation of cotton salt tolerance” (DB13/T 1339-2010), “Technical code of practice for identification of salt tolerance in rice” (NY/T 3692-2020) and “Evaluation guidance for water security” (DB37/T 4499-2022). In the present study, the content levels of four components were converted into a dandelion QI. From Table 5 and using the QI formula, totally 625 complete combinations and 17 QIs were obtained. According to the design requirements of quality standards, the probability within 1% was classified into Grade 1, and the rest were divided according to spindle structure. The above results are shown in Table 6.
Dandelion QI and grade division.
QI | Grade | Numbers | Total | Result | Probability (%) |
---|---|---|---|---|---|
20 | 1 | 1 | 5 | Super high | 0.8 |
25 | 1 | 4 | |||
30 | 2 | 10 | 117 | High | 18.72 |
35 | 2 | 20 | |||
40 | 2 | 35 | |||
45 | 2 | 52 | |||
50 | 3 | 68 | 233 | Medium | 37.28 |
55 | 3 | 80 | |||
60 | 3 | 85 | |||
65 | 4 | 80 | 200 | Qualified | 32 |
70 | 4 | 68 | |||
75 | 4 | 52 | |||
80 | 5 | 35 | 70 | Low | 11.2 |
85 | 5 | 20 | |||
90 | 5 | 10 | |||
95 | 5 | 4 | |||
100 | 5 | 1 |
QI, quality index.
A total of 578 dandelion samples were checked, of which the quality levels of 324 samples with four phenolic acids were evaluated, as shown in Table 7 and Table S1 in Supplementary Materials. Of that quantity, Grade 1 samples amounted to 0.62%, and Grade 5 to 58.95%, indicating that the overall quality of this batch of dandelion samples was low, and that the overall samples collected from the greenhouse presented a lower compounds’ content compared with samples cultured in saline land (Table S1 in Supplementary Materials).
Quality evaluation result for 578 dandelion samples.
Grade | Super high (Grade 1) | High (Grade 2) | Medium (Grade 3) | Qualified (Grade 4) | Low (Grade 5) |
---|---|---|---|---|---|
Quantity | 2 | 54 | 32 | 45 | 191 |
Percentage (%) | 0.62 | 16.67 | 9.88 | 13.89 | 58.95 |
According to the 2020 report from National Institutes for Food and Drug Control (Zhang et al., 2021), the overall dandelion quality in Hebei Province, China was generally substandard, going by the method of assessment recommended by China Pharmacopoeia (Ed. 2015). It has been suggested that the low concentration of phenolic acid compounds in dandelion plants cultivated in this province is attributable to the cultivation method employed, since producers focussed on maximising dandelion yield and accordingly input a large amount of fertiliser and water, leading to the dandelions’ rapid growth, and thus insufficient accumulation of the needed components (Wu et al., 2019; Zhang et al., 2021). Considering this situation, the dandelion evaluation standard could be appropriately decreased; however, improving dandelion quality with regard to the aspects of cultivation technology or breeding technology has remained the fundamental strategy. In addition, we proposed an idea for dandelion quality evaluation; however, the determination of evaluation indices may be adjusted following the deepening understanding of functional components of dandelion in future.
A dandelion quality evaluation method that is relatively comprehensive and capable of widespread application was developed based on phenolic acids’ analysis from 578 samples and related references. Four phenolic acids, namely cichoric acid, caffeic acid, chlorogenic acid and caftaric acid, were chosen as the main evaluation indices. Contents of the four phenolic acids were divided into five levels and then converted into a dandelion QI using the equal weight average calculation method. Finally, five grades of dandelion quality were identified according to the QI, namely Grade 1 (super high, 0.8%), Grade 2 (high, 18.72%), Grade 3 (medium, 37.28%), Grade 4 (qualified, 32%) and Grade 5 (low, 11.2%). Advantageously, this method enabled avoiding conflict with that of the single evaluation index used in China Pharmacopoeia, possesses the characteristics of scientific nature and widespread applicability or reproducibility and can be more conveniently adopted by dandelion industries or research facilities even in the absence of specialised knowledge of statistical techniques.