Pharmacognosy Magazine

: 2019  |  Volume : 15  |  Issue : 61  |  Page : 177--182

Quality assessment of pollen typhae by high-performance liquid chromatography fingerprint, hierarchical cluster analysis, and principal component analysis

Xin Ma1, Haimiao Zou2, Yibin Pan3, Jing Su1, Yujiao Qiu4, Mingfeng Qiu1,  
1 Research Center of Traditional Dai Medicine, School of Pharmacy, Shanghai Jiao Tong University, Shanghai, China
2 Deparment of Pharmacy, Qilu Meical University, Zibo, China
3 Department of Anorectal, Longhua Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai 200032, China
4 Solebury School, New Hope, PA, USA

Correspondence Address:
Mingfeng Qiu
School of Pharmacy, Shanghai Jiao Tong University, 800, Dongchuan Road, Shanghai 200240


Aim: This study aims to establish the quality assessment methods of Pollen Typhae. Materials and Methods: High performance liquid chromatography (HPLC) fingerprint analysis, hierarchical cluster analysis (HCA), and principal component analysis (PCA) were used for quality evaluation of Pollen Typhae from different origins together with microscopic identification. Then, the quantity of 43 crude Pollen Typhae samples in the market was collected and analyzed. Results: In true and false test, four False Pollen Typhae samples, 13 Net Pollen Typhae (NPT) samples, and 26 Grass Pollen Typhae (GPT) samples were identified by microscopic identification. In quality test, the amounts and percentages of Qualified Pollen Typhae, Unqualified Pollen Typhae were 24 (55.81%) and 19 (44.19%), respectively with typhaneoside and isorhamnetin-3-O-neohesperidoside determined by HPLC according to China Pharmacopeia. We analyzed 43 samples from 20 regions and established their fingerprints, then selected 31 peaks as characteristic peaks and calculated their relative peak areas. To express the HPLC fingerprints quantitatively, peak 16, 18, 22, 23, and 26 were verified as typhaneoside, isorhamnetin-3-O-neoheptanoside, rutin, quercetin, and isorhamnetin. The similarity of correlation coefficients in chromatogram was 0.954 ± 0.007 and 0.922 ± 0.004 for NPT and GPT, respectively, while 0.67 ± 0.008 for 43 samples. The analysis of HCA and PCA can distinguish true or false, qualified or unqualified of Pollen Typhae. Conclusion: HPLC fingerprint combined with HCA and PCA provides a very efficient and comprehensive method for quality evaluation of Pollen Typhae. Abbreviations used: HCA: Hierarchical cluster analysis; PCA: Principal component analysis; FPT: False Pollen Typhae; NPT: Net Pollen Typhae; GPT: Grass Pollen Typhae; QPT: Qualified Pollen Typhae; UPT: Unqualified Pollen Typhae; RSDs: The relative standard deviations; CASE: Computer Aided Similarity Evaluation; TCM: Traditional Chinese medicine.

How to cite this article:
Ma X, Zou H, Pan Y, Su J, Qiu Y, Qiu M. Quality assessment of pollen typhae by high-performance liquid chromatography fingerprint, hierarchical cluster analysis, and principal component analysis.Phcog Mag 2019;15:177-182

How to cite this URL:
Ma X, Zou H, Pan Y, Su J, Qiu Y, Qiu M. Quality assessment of pollen typhae by high-performance liquid chromatography fingerprint, hierarchical cluster analysis, and principal component analysis. Phcog Mag [serial online] 2019 [cited 2019 May 26 ];15:177-182
Available from:

Full Text



The study revealed that the proposed multivariate analysis by high-performance liquid chromatography combined with a fingerprint method and analysis of principal component analysis is an efficient and comprehensive method for crude Pollen Typhae.


Pollen Typhae is the dry Pollen of the genus Typha (Typhaceae) including Typha angustifolia L. and Typha orientalis Presl., which is widely distributed near the pond and shallow water in the Northern hemisphere.[1] Modern pharmacological studies have found that it has well effects in improving microcirculation, antiatherosclerosis, and antitumor.[2],[3],[4] In clinical practice, it has been widely used to prevent and treat the coronary heart disease and hyperlipidemia. Its mechanism is mainly related to reducing prothrombin time, activating coagulation factors, and increasing cyclic adenosine monophosphate levels.[5],[6],[7] Pollen Typhae contains various compounds, such as flavonoids, amino acids, steroids, fatty acids, sugars, and other types of compounds.[8] Flavonoids are main active compounds and have antioxidant, anti-inflammatory, and antigen toxic effects.[9],[10],[11],[12] Typhaneoside and isorhamnetin-3-O-neohesperidoside are the significant components of Pollen Typhae extract. The pharmacological effects related mainly to repairing endothelial cell damage, inhibiting uterine contraction, and protecting cells from oxidative stress.[13],[14],[15],[16] Rutin exhibits strong 1,1-diphenyl-2-picrylhydrazyl radical scavenging activity.[17] Quercetin can clear highly reactive species such as peroxynitrite and the hydroxyl radical.[18] Isorhamnetin is known to exert beneficial effects on the prevention of obesity.[19],[20] Isorhamnetin-3-O-α-L-rhamnosyl (1-2)-β-glucoside, quercetin, hentriacontanol-6B-sitosterol, and its palmitate play important roles in reducing lipid levels and preventing atherosclerosis.[21] Linoleic acid can strongly inhibit the binding of the Myc-Max heterodimer to E-box DNA sites.[22] Narcissin can increase the mice myocardial cell extraction of 86Rb.[23]

In previous studies, thin layer chromatography and gas chromatography-mass spectrometry are insufficient to assess the quality of Pollen Typhae. High-performance liquid chromatography (HPLC) is one of the most powerful technologies to control the quality of traditional Chinese medicine (TCM) with its simplicity, repeatability, and accuracy. HPLC fingerprint is a comprehensive and quantifiable method for identification. It is based on the research of chemical composition system of TCM, and one of the most effective analytical methods for multicomponent complex TCM.[24],[25],[26],[27],[28],[29],[30] Merely, a few researches have reported to evaluate the Pollen Typhae using HPLC. However, it is inadequate to determine only one or two markers to evaluate the overall quality of Pollen Typhae completely. Therefore, it is necessary to establish a comprehensive and systematic standard to assess the quality of Pollen Typhae. HPLC fingerprint analysis or together with hierarchical cluster analysis (HCA) and principal component analysis (PCA) for the quality assessment of Pollen Typhae have not been reported before. It may be a comprehensive evaluation method for Pollen Typhae.

In our study, we developed a simple, holistic, and reliable chromatographic HPLC fingerprint combined with PCA and HCA methods to quantitatively analyze the Pollen Typhae. This combinative method has the ability to identify false from true and unqualified from qualified and assess the quality of Pollen Typhae more comprehensively.

 Materials and Methods

Materials and reagents

We collected 43 batches of crude Pollen Typhae samples from 20 regions in China [Table 1]. Typhaneoside and isorhamnetin-3-O-neoheptanoside were provided by the Chinese Food and Drug Inspection Institute (97.0%, 93.2%, Beijing, China). Isorhamnetin, rutin, and quercetin were purchased from Chengdu Herbpurity Co., Ltd., (98.0%, 98.41%, 98.41%, Chengdu, China). HPLC grade acetonitrile was purchased from the ANPEL Scientific Instrument Co., Ltd. (Shanghai, China). Methyl alcohol was from Shanghai Lingfeng Chemical Regent Co., Ltd. (Shanghai, China), and formic acid was from Sinopharm Chemical Reagent Co., Ltd. Chloral hydrate and glycerinum were from Chemical Reagent Co., Ltd. Other chemicals were all analytical grade. Water was purified by Milli-Q System (Millipore, Merck, USA).{Table 1}


HPLC analysis was conducted with an Agilent 1200 HPLC System (Agilent Technologies, USA). The thermostatic water bath was from Shanghai Purdue HH-ZK4 Series with 220 V, 50 Hz. The analytical balance from Metller Toledo JB/T Series was used for sample weighing. OLYMPUS inverted microscope of IX51 was from Japan. Micro slides, micro coverslips of Sail Brand was made in China. The alcohol burner dissecting needle, tweezers, and lighter were dedicated to the experiment.

Microscopic identification analysis

A volume of 10 mg Pollen Typhae was dipped with dissecting needle and placed in the middle of a clean micro slide. An appropriate amount of chloral hydrate solution was added to the Pollen and mixed homogeneously. The micro slide with the sample was baked to nearly dry on the outer flame of alcohol burner. A volume of 2 mL of dilute glycerin was added and covered with a micro coverslip. Excess glycerin was absorbed by blotting paper from a side. Samples were performed in triplicate. Number the slides and record the samples information. The status of the samples was observed and recorded by the optical microscope.

Preparation of sample solution and standard solution

Each sample of crude Pollen Typhae was soaked for 1 h and extracted with eight times of methyl alcohol for 45 min at 80°C and constant pressure. All extracts were performed in triplicate and stored at −4°C for further use. The stock solution including typhaneoside and isorhamnetin-3-O-neoheptanoside, isorhamnetin, rutin, and quercetin were prepared and diluted properly. All of them were passed through 0.22 μm filters.

High-performance liquid chromatography conditions

The column was a TC-C18 column (4.6 mm × 250 mm, 5 μm), and the mobile phase was water-formic acid (1000:0.75) (A) and acetonitrile (B). The gradient procedure was set as follows: 5%–15% B in 0–15 min; 15%–17% B in 15–20 min; 17%–20% B in 20–30 min; 20%–77% B in 30–55 min; 77% B in 55–60 min. The flow rate was 1 mL/min, the column temperature was 30°C, the detection wavelength was 254 nm, and the injection volume was 20 μL.

Data analysis

HCA and PCA were performed with SPSS version 19.0 software (SPSS for Windows 19.0, SPSS Inc., USA) and SIMCA-P 11.5 software (Umetrics, Ume a, Sweden). The “average linkage between groups” method was applied and the cosine was selected as a measurement in HCA. The resulting three-dimensional matrix containing any specified peak index (retention time), peak intensity information (variables), and sample names (observations) was then output to the SIMCA-P software for PCA. Using the Computer-Aided Similarity Evaluation (CASE) software calculated the correlation coefficients for the entire chromatographic model and generated simulated average chromatograms and characteristic peaks.[31]

 Results and Discussion

Microscopic identification analysis

We processed 43 samples into microscopy specimens and analyses were performed in triplicate. Pictures of microscopy specimens were taken by the OLYMPUS inverted microscope [Figure 1]. Under the microscope, round or oval Pollen grains could be seen in Net Pollen Typhae (NPT) [Figure 1]a and [Figure 1]d. Mesh carved lines could be observed on the surface of the Pollen grain. The peripheral contour line of Pollen grain was smooth with wavy or convex gear [Figure 1]g. A small amount of plant tissue such as pollen, calyx, and duct could be seen in some samples called Grass Pollen Typhae (GPT) [Figure 1]b and [Figure 1]e. Moreover, a mass of plant tissue and unknown crystal could be observed in the False Pollen Typhae (FPT) [Figure 1]c and [Figure 1]f. There were a total of 26 GPT samples, 13 NPT samples, and four FPT samples in 43 samples [Table 1]. The above results showed that the GPT, NPT, and FPT were existed in the market. The FPT accounted for a proportion of 9.3%. Therefore, the evaluation of the quality of Pollen Typhae is necessary. The microscopic identification analysis is a good way to tell false from true of Pollen Typhae and tell NTP from GTP in true samples.{Figure 1}

Method validation

As shown in [Table 2], the relative standard deviations (RSDs) of precision test ranged from 0.15%–0.80% (n = 6). The RSD of repeatability test varied from 0.75% to 1.15% (n = 6). The stability of the solutions was determined at 0, 2, 4, 8, 12, 24, and 48 h, and the RSD of the stability teat ranged from 0.58% to 1.15%. The average recovery rate of the typhaneoside and isorhamnetin-3-O-neoheptanoside were 96.54% ± 0.03% and 105.16% ± 0.02%, respectively. Moreover, the RSD varied from 0.93% to 2.18% (n = 6). The above results revealed the repeatability and accuracy of the analysis conditions.{Table 2}

Determination of typhaneoside and isorhamnetin-3-O-neoheptanoside

The determination of typhaneoside and isorhamnetin-3-O-neoheptanoside in samples from different regions were analyzed by HPLC according to China Pharmacopoeia (2015 edition)[31] [Table 3]. By comparing the retention time with standard substances, peaks 16 and 18 were identified as typhaneoside and isorhamnetin-3-O-neoheptanoside, respectively [Figure 2]. The contents of typhaneoside and isorhamnetin-3-O-neoheptanoside varied from 0% to 0.58% (g/g) and from 0% to 0.55% (g/g), respectively. The samples, whose total percentage of typhaneoside and isorhamnetin-3-O-neoheptanoside did not reach 0.5%, were called Unqualified Pollen Typhae (UPT). Moreover, the rest of the samples were called Qualified Pollen Typhae (QPT). In this study, the quantity and the proportions of UPT and QPT samples were 19 (44.19%) and 24 (55.81%), respectively. UPT accounted for nearly half of the proportion.{Table 3}{Figure 2}

High performance liquid chromatography fingerprint analysis

To evaluate the quality of Pollen Typhae comprehensively, we analyzed 43 samples from different regions and constructed a standard HPLC fingerprint [Figure 2]a and [Figure 2]b. We selected and marked 31 common peaks as characteristic peaks. Compared retention time and ultraviolet spectrum with standard references, peak 16, 18, 22, 23, and 26 were in conformity with typhaneoside, isorhamnetin-3-O-neoheptanoside, rutin, quercetin, and isorhamnetin. Peak 18 (isorhamnetin-3-O-neoheptanoside) was selected as the reference peak because its proportion of areas accounted of 10%. In addition, the similarity of the chromatogram was evaluated by CASE software.

The correlation coefficients of similarity in chromatogram were 0.954 ± 0.007, 0.922 ± 0.004 for NPT and GPT, respectively [Figure 2]c and [Figure 2]d, while 0.67 ± 0.008 of 43 samples. The results above showed that the quality of the samples in the market had a certain difference due to the wide distributions and different varieties.

Hierarchical cluster analysis of the samples

In the light of the chromatograms of all samples, the natural cluster was found by HCA to assess the differences of Pollen Typhae from 20 regions. When the clustering coefficient was 5, 43 samples were split into eight groups [Figure 3]a. Compared above, the results of these eight groups were consistent with that of microscopic identification and measurement. In summary, HCA can differentiate the superior or inferior of the samples.{Figure 3}

Principal component analysis of the samples

PCA was considered as a data reduction technique to produce a visual scatterplot for the qualitative assessment of resemblances and differences among the samples. To further classify the true samples of Pollen Typhae without FPT, PCA was performed [Figure 3]b and [Figure 3]c. In the figure, the black dots stood for UPT samples, the green dots (GPT), and the blue dots (NPT) were on behalf of QPT samples. They can be found to be clearly separated from each other. The GPT and NPT could be basically separated with R2Ycum = 0.559 represented interpretation and Q2Y = 0.366 stood for forecast [Figure 3]b. In the three-dimensional-PCA figure, the three groups could be completely separated [Figure 3]c, which might suggest of some differences between UPT and QPT. And QPT can be divided into GPT and NPT.


In this article, 43 samples of Pollen Typhae from 20 regions were analyzed by HPLC fingerprint analysis, HCA and PCA together with optical microscope. In previous studies, traditional microscopic identification can tell false from true, tell GPT from NPT, but it cannot confirm whether the crude Pollen Typhae is qualified or not. Two compounds determination only can evaluation its quality partly. Either of them cannot evaluate the quality of Pollen Typhae comprehensively. Because of the high similarity of chromatograms among samples, it is necessary to introduce novel methods to solve this problem. The HPLC fingerprint is a good choice to evaluate the quality of the samples comprehensively. The HCA and PCA can analyze the complex data column of HPLC fingerprint. We managed to divide 43 samples into different groups including UPT group and QPT group. Moreover, QPT group was divided into two groups of GPT and NPT.

In conclusion, HPLC coupled with multivariate analysis was developed to evaluate the quality of Pollen Typhae comprehensively. And, HPLC-HCA or HPLC-PCA can be used to classify the samples of Pollen Typhae. The methods have been successfully applied in the evaluation of Pollen Typhae samples in the market. The study revealed that the proposed multivariate analysis by HPLC combined with a fingerprint method, and analysis of PCA is an efficient and comprehensive tool for crude Pollen Typhae.


It was supported by the Shanghai Science and Technology Commission Project 14401900100, 1540190170, and 17401900900.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Qin F, Sun HX. Immunosuppressive activity of pollen typhae ethanol extract on the immune responses in mice. J Ethnopharmacol 2005;102:424-9.
2Chung S, Park S, Yang CH. Unsaturated fatty acids bind Myc-Max transcription factor and inhibit myc-max-DNA complex formation. Cancer Lett 2002;188:153-62.
3Zhao J, Zhang CY, Xu DM, Huang GQ, Xu YL, Wang ZY, et al. The antiatherogenic effects of components isolated from pollen typhae. Thromb Res 1990;57:957-66.
4Zhao J, Zhang CY, Xu DM, Huang GQ, Xu YL, Wang ZY, et al. Further study of pollen typhae's effects on the production of tPA and PGI, by cultured endothelial cells. Thromb Res 1989;56:677-85.
5Akkol EK, Süntar I, Keles H, Yesilada E. The potential role of female flowers inflorescence of Typha domingensis pers. In wound management. J Ethnopharmacol 2011;133:1027-32.
6Chen Y, Yu H, Wu H, Pan Y, Wang K, Jin Y, et al. Characterization and quantification by LC-MS/MS of the chemical components of the heating products of the flavonoids extract in pollen typhae for transformation rule exploration. Molecules 2015;20:18352-66.
7Yan X, Zhao Y, Luo J, Xiong W, Liu X, Cheng J, et al. Hemostatic bioactivity of novel pollen typhae carbonisata-derived carbon quantum dots. J Nanobiotechnology 2017;15:60.
8Ding M, Jiang Y, Yu X, Zhang D, Li J, Wang H, et al. Screening of combinatorial quality markers for natural products by metabolomics coupled with chemometrics. A case study on pollen typhae. Front Pharmacol 2018;9:691.
9Bouhlel I, Skandrani I, Nefatti A, Valenti K, Ghedira K, Mariotte AM, et al. Antigenotoxic and antioxidant activities of isorhamnetin 3-O neohesperidoside from Acacia salicina. Drug Chem Toxicol 2009;32:258-67.
10Cao S, Ni B, Feng L, Yin X, Dou H, Fu J, et al. Simultaneous determination of typhaneoside and isorhamnetin-3-O-neohesperidoside in rats after oral administration of pollen typhae extract by UPLC-MS/MS. J Chromatogr Sci 2015;53:866-71.
11Yu XA, Azietaku JT, Li J, Cao J, An M, He J, et al. Simultaneous determination of eight flavonoids in plasma using LC – MS/MS and application to a pharmacokinetic study after oral administration of pollen typhae extract to rats. J Chromatogr B 2017;1044-1045:158-65.
12Lei X, Zhou Y, Ren C, Chen X, Shang R, He J, et al. Typhae pollen polysaccharides ameliorate diabetic retinal injury in a streptozotocin-induced diabetic rat model. J Ethnopharmacol 2018;224:169-76.
13Tao W, Yang N, Duan JA, Wu D, Guo J, Tang Y, et al. Simultaneous determination of eleven major flavonoids in the pollen of Typha angustifolia by HPLC-PDA-MS. Phytochem Anal 2011;22:455-61.
14Chen P, Cao Y, Bao B, Zhang L, Ding A. Antioxidant capacity of Typha angustifolia extracts and two active flavonoids. Pharm Biol 2017;55:1283-8.
15Liu SJ, Chen PD, Dai GL, Ju WZ, Xie LY, Xu J, et al. Analysis of isorhamnetin-3-O-neohesperidoside in rat plasma by liquid chromatography/electrospray ionization tandem mass spectrometry and its application to pharmacokinetic studies. Chin J Nat Med 2013;11:572-6.
16Du LY, Zhao M, Tao JH, Qian DW, Jiang S, Shang EX, et al. The metabolic profiling of isorhamnetin-3-O-neohesperidoside produced by human intestinal flora employing UPLC-Q-TOF/MS. J Chromatogr Sci 2017;55:243-50.
17Yang J, Guo J, Yuan J.In vitro antioxidant properties of rutin. LWT Food Sci Technol 2008;41:1060-6.
18Boots AW, Haenen GR, Bast A. Health effects of quercetin: From antioxidant to nutraceutical. Eur J Pharmacol 2008;585:325-37.
19Feng XT, Duan HM, Li SL. Protective role of pollen typhae total flavone against the palmitic acid-induced impairment of glucose-stimulated insulin secretion involving GPR40 signaling in INS-1 cells. Int J Mol Med 2017;40:922-30.
20Lee MS, Kim Y. Effects of isorhamnetin on adipocyte mitochondrial biogenesis and AMPK activation. Molecules 2018;23. pii: E1853.
21Feng XT, Chen Q, Xie Z, Liang X, Jiang ZH, Zhao W, et al. Pollen typhae total flavone improves insulin resistance in high-fat diet and low-dose streptozotocin-induced type 2 diabetic rats. Biosci Biotechnol Biochem 2014;78:1738-42.
22Wang W, Guo Z, Xu Z, Meng Q, Chen C, Zhang Y, et al. Effect of pollen typhae on inhibiting autophagy in spinal cord injury of rats and its mechanisms. Int J Clin Exp Pathol 2015;8:2375-83.
23Wang EJ, Jin Y, Wang L, Shao-Chun LI. Inhibitory effect of pollen typhae on thrombosis in rats. Acad J PLA Postgrad Med Sch 2008;29:227-8.
24Li Y, Wu T, Zhu J, Wan L, Yu Q, Li X, et al. Combinative method using HPLC fingerprint and quantitative analyses for quality consistency evaluation of an herbal medicinal preparation produced by different manufacturers. J Pharm Biomed Anal 2010;52:597-602.
25Xia HL, Liu L, Jia-Na NI. Simultaneous determination of amygdaloside, typhaneoside and isorhamnetin-3O-neohesperidoside in hongteng granules by HPLC. Chin Tradit Pat Med 2013;35:525-8.
26Du Y, Li Q, Liu J, Yin Y, Bi K. Combinative method using multi-components quantitation by single reference standard and HPLC fingerprint for comprehensive evaluation of Rhodiola crenulata Ohba H. Anal Methods 2014;6:5891-8.
27Yudthavorasit S, Wongravee K, Leepipatpiboon N. Characteristic fingerprint based on gingerol derivative analysis for discrimination of ginger (Zingiber officinale) according to geographical origin using HPLC-DAD combined with chemometrics. Food Chem 2014;158:101-11.
28Wu XD, Chen HG, Zhou X, Huang Y, Hu EM, Jiang ZM, et al. Studies on chromatographic fingerprint and fingerprinting profile-efficacy relationship of Saxifraga stolonifera meerb. Molecules 2015;20:22781-98.
29Liu Z, Wang D, Li D, Zhang S. Quality evaluation of Juniperus rigida sieb. et Zucc. Based on phenolic profiles, bioactivity, and HPLC fingerprint combined with chemometrics. Front Pharmacol 2017;8:198.
30Jiang Z, Zhao C, Gong X, Sun X, Li H, Zhao Y, et al. Quantification and efficient discovery of quality control markers for Emilia prenanthoidea DC. by fingerprint-efficacy relationship modelling. J Pharm Biomed Anal 2018;156:36-44.
31China Pharmacopoeia Committee. Pharmacopoeia of the People's Republic of China. China Beijing: Chemical Industry Press; 2015. p. 353.