Chromatographic fingerprinting technique is a useful and
facile tool for the quality evaluation of herbal medicine and
their related products. This technique has been introduced
and accepted by WHO as a strategy for the quality assess-
ment of herbal medicines in early 1990s.1)Among those
commonly used methods, HPLC-DAD-MS technique has
been used widely for the simultaneous qualitative and quanti-
tative analysis of herbal medicine and their related products,
with on-line UV and mass spectrometric information avail-
able for each individual peak in the chromatogram. In this re-
gard, HPLC-DAD-MS fingerprint analysis has recently been
considered as a more comprehensive and powerful technique
for establishment of chromatographic fingerprinting.2—4)
Fresh Houttuynia cordata THUNB. has been a well known
traditional Chinese medicinal material widely used in China,
Japan and also listed in the Chinese Pharmacopoeia (2005
edition).5)It possesses a variety of pharmacological activities
including anti-platelet aggregation, antibacterial, antitumor,
antimicrobial, anti-inflammatory, antileukemic, immunomod-
ulatory and recently, demonstrated its efficacy in anti-
SARS.6—10)In general, the chemical components of H. cor-
data comprised of three major types, namely: essential oil,
phenols and alkaloid components.11)However, most of the
previous related studies were mainly focused on the chem-
istry of essential oil, which has been considered responsible
for the claimed clinical efficacy.12)Recent pharmacological
studies also revealed that the flavonoid components in H.
cordata possess antineoplastic, antioxidant, antimutagenic
and free radical scavenging capacity.13—15)Similarly, the al-
kaloid components demonstrated significant potent anti-
platelet and cytotoxic activities,7,16)whereas chlorogenic acid
had significant antipyretic and antibiotic properties.17)Al-
though a variety of pharmacological activities associated
with these groups of chemical in H. cordata was demon-
strated, the common used quality standard of H. cordata is
still relied merely on the content of undecanone which served
as a marker compound for the quality evaluation standard as
stipulated in the Chinese Pharmacopoeia (2005 edition).5)
This phenomenon stimulated our thought to develop a chro-
matographic fingerprinting technique to associate various
pharmacological activities with the chemical profile and thus
the quality of H. cordata and its related products. Moreover,
as the crude herbs are generally cultivated for use over
China, a vast variety of external factors can therefore affect
the chemical composition, and thus the pharmacological ac-
tivity of that particular herbal medicine. These factors in-
clude geographical origin for cultivation, drying methods
used, different medicinal portions, etc.18—19)In this regard, a
facile and comprehensive means for evaluating the overall
quality of H. cordata is urgently needed.
In literature, related analyses on H. cordata species are
particularly focused on TLC, GC, GC-MS and a few of them
on HPLC.20—24)Recently, a method using GC-MS finger-
printing for quality evaluation was established by our re-
search group.25)In that part of work, GC-MS was employed
to analyze for the volatile components in H. cordata but not
yet covered with other phenolic and alkaloid-typed compo-
nents. In fact, there still exists an analytical void for simulta-
neous detection of alkaloids, phenols and other major chemi-
cal compounds in H. cordata for comprehensive quality eval-
uation purpose.26,27)Among the various chromatographic
methods available, HPLC-DAD-MS is proven particularly
useful for the purpose.
The present study aims at developing a HPLC-DAD-MS
fingerprint of H. cordata and characterized the major active
chemical constituents of H. cordata. The developed finger-
printing comprises of all the major chemical components of
H. cordata and useful in reflecting the quality and associated
with the clinical efficacy of H. cordata. The developed
HPLC-DAD-MS fingerprint was then applied to differentiate
H. cordata among various cultivation areas, different drying
methods and also demonstrated applicable to distinguish
among different medicinal portions.
with a photodiode array detector (DAD). An Alltima C18analytical column
(150mm?4.6mm, 5mm, Alltech Associates, Inc. U.S.A.) coupled with a
C18guard column (7.5mm?4.6mm, 5mm, Alltech Associates, Inc., U.S.A.)
An Agilent 1100 series HPLC system was equipped
∗ To whom correspondence should be addressed.e-mail: email@example.com © 2005 Pharmaceutical Society of Japan
Establishment of HPLC-DAD-MS Fingerprint of Fresh Houttuynia cordata
Jiang MENG,a,bKelvin Sze-Yin LEUNG,aZhihong JIANG,aXiaoping DONG,bZhongzhen ZHAO,*,aand
aSchool of Chinese Medicine, Hong Kong Baptist University; Kowloon Tong, Hong Kong Special Administrative Region,
P . R. China: bChengdu University of TCM; Chengdu, 611730, Sichuan, P . R. China: and cInstitute of Medicinal Plant
Development, Peking Union Medical College & Chinese Academy of Medical Sciences; Beijing, 100094 P . R. China.
Received August 8, 2005; accepted September 14, 2005
A HPLC-DAD-MS fingerprint method of fresh Houttuynia cordata THUNB. was developed basing on the con-
sistent chromatographic features among 11 batches of authentic samples. Major chemical components including
phenolic compounds, flavones and alkaloids were simultaneously analyzed. Eleven common peaks in the finger-
print were chosen and identified by comparing their UV and ESI-MS data with the authentic compounds. The
unique properties of this HPLC-DAD-MS fingerprint were successfully applied to analyze and differentiate sam-
ples from different geographical origins, processing methods and various medicinal parts of H. cordata. The re-
sults showed that these variations will give rise to differences in identities and/or abundance of chemical com-
pounds, indicating that a comprehensive quality evaluation of those major ingredients in H. cordata is critical to
assess and represent its overall quality.
fingerprint; Houttuynia cordata; HPLC-DAD-MS; quality control
Chem. Pharm. Bull. 53(12) 1604—1609 (2005) Vol. 53, No. 12
was used at room temperature. The mobile phase consisted of 0.2% acetic
acid in water (A) and acetonitrile (B) using a gradient program of 20% (B)
in 0—30min, 20—21% (B) in 30—50min, 21—70% (B) in 50—80min.
The flow rate was 0.6ml/min. DAD detector was set at 280nm for acquiring
chromatograms. The injection volume was 10ml. The data were obtained
and processed with the software of Chemstation for LC3D (Hewlett Packard,
The mass spectrometer was equipped with an ESI source. The interface
and MS parameters were as follows: dry gas, N2(7l/min); dry gas tempera-
ture, 400°C; for positive ion mode: sprayer voltage, 5000V; orifice voltage:
101V; focusing ring voltage: 380V; for negative ion mode: S.V .: ?4000V;
O.V .: ?101, F.R.V: ?380V; scan range, m/z 130—800. All data acquired
were processed by Macc Chrom 1.1 software (Applied Biosystem, CA,
Solvents and Chemicals
Methanol (Analytical grade), glacial acetic
acid (analytical grade) and acetonitrile (HPLC grade) were purchased from
E. Merck (Darmstadt, Germany). Deionized water was obtained from a
Milli-Q water system (Millipore, Bedford, MA, U.S.A.).
All the standard reference chemical compounds were extracted, isolated
and purified from fresh H. cordata in our laboratory. Their purities were
shown to be higher than 96% by HPLC analysis, and their structures were
elucidated by comparison their spectroscopic data (ESI-MS, 1H- and 13C-
NMR, HMBC, HMQC and H–H COSY) with references.17,28—36)The de-
tailed procedures for isolation and spectrometric identification of these com-
pounds will be reported in another paper.
Samples of fresh H. cordata were collected from the
cultivation areas in Sichuan, Guangdong provinces and retailers in Hong
Kong of China. All the samples were authenticated by one of the authors,
Prof. Zhong-Zhen ZHAO. Freeze-dried processes were as follows: soon after
collection, samples were placed in a ?20°C freezer. After pre-cooling for at
least 12h, samples were placed in the freeze-drying system (Model 77530,
LABCONCO Co., Switzerland) for further 48h, and finally sealed in air-
Samples were prepared by two methods: lyophilization (cryochem) for the
fresh and air-dried for the dry.5)Herbs collected from Emei were divided and
respectively analyzed according to different medicinal portions (leaf, stem,
root and herb). All samples were crushed (CERT-04, FARGO, China) and
had passed through a 40 mesh sieve. The ground powders were stored at
about 4°C before use.
0.5g fine powder was accurately weighed and ex-
tracted with 50ml of 50% methanol in ultrasonic processor (1875HTAG,
CREST, U.S.A.) for 30min and filtered. The original solvent weight was re-
stored. The extract was filtered through a 0.2mm membrane filter. An aliquot
of 10ml solution was injected for HPLC analysis.
Results and Discussion
Optimization of HPLC Systems
tector (DAD) was used in the current study. The detection
wavelength at 280nm gave the best abundance for target
compounds within the chromatographic windows.
Method precision was based on replicated analyzing of
samples, with reported relative standard deviations (RSD) of
0.27% and 0.11% for relative retention time (RRT) and rela-
tive peak area (RPA) of all peaks, respectively. The method
reproducibility was studied through six replicated sample so-
lutions extracted from a single batch of H. cordata. The cor-
responding RSD of RRT and RPA were reported less than
3% over 2d of investigation. The stability test was performed
with a sample solution over 24h of standing period. The
RSD of the RRT and RPA were found less than 0.25% and
2.95%, respectively. The result indicated that the developed
method was validated and applicable for sample analysis.
HPLC Fingerprint of Fresh H. cordata
representative chromatographic fingerprint for fresh H. cor-
data, 11 authentic batches of fresh H. cordata acquired from
EMei (major production area in Sichuan province, P. R.
China) (EM-1 to EM-11) were analyzed using the established
HPLC method. Among the acquired chromatograms, repre-
Photodiode array de-
To establish a
sentative peaks existing in all batches of samples were as-
signed as common peaks for H. cordata. Altogether there
were eleven common peaks and compounds identified in the
fingerprint (Figs. 1, 2). Peak 7 (Quercitrin) indicated the
highest content among others, with good resolution and
demonstrated pharmacological activities in other studies.26,37)
Therefore, it has been chosen as the reference peak. The RRT
and RPA of all common peaks with respect to this reference
compound were obtained. The entire HPLC profiles with the
information on RRT and RPA of the common peaks will be
used to identify and assess the quality of H. cordata.
HPLC-MS Analysis for Components Identification
order to identify structures of the major chemical compo-
nents in H. cordata, the sample was analyzed by HPLC-
DAD-MS techniques. ESI in both negative and positive
modes were performed. The results showed that ESI in nega-
tive mode was particularly sensitive to the flavones while ESI
in positive mode was useful to detect the alkaloids. By study-
ing on the characteristic mass spectra of these peaks and
comparing with the UV and ESI-MS spectra with the respec-
tive authentic compounds, 11 common peaks in H. cordata
were designated and identified (Table 1). Among the 11
chemical compounds identified, Compounds 2 and 3 are the
first time observed in H. cordata and successfully isolated.
Compound 2 has a considerably high content of about 4% in
our preliminary quantitative analysis. In literature, this com-
pound has already reported antioxidant and DPPH radical-
scavenging activities.38)The results of isolation will be re-
Analyzing Fingerprinting of H. cordata from Various
The concept of “phytoequivalence”
is a very important task in herbal fingerprints analysis.39)The
pharmacological activities of a herbal medicine always vary
with its cultivation origin. These differences can be associ-
ated with and retrieved in the chemical profiles of a finger-
print pattern. Therefore, it is essential to perform analysis of
December 2005 1605
Fingerprints: (1) Chlorogenic Acid; (2) Quercetin-3-O-b-D-galactopyra-
nosyl-7-O-b-D-glucopyranoside; (3) Quercetin 3-O-a-L-rhamnopyranosyl-7-
O-b-D-glucopyranoside; (4) Rutin; (5) Hyperin; (6) Isoquercitrin; (7)
Quercitrin; (8) Afzelin; (9) Quercetin; (10) Piperolactam A; (11) Aristolac-
Chemical Structures of the Identified Compounds in the HPLC
similarity and difference of the overall fingerprint pattern. In
this part, we have compared the samples from four different
locations: Emei (EM-1 to EM-11), Yaan (YA-1 to YA-11),
GuangDong (GD-1 to GD-5) and Hong Kong (HK-1 to HK-
5) of P. R. China. Their total ion chromatograms are shown
in Fig. 3. In general, the HPLC profiles among samples from
the four locations are similar to each other. Comparing the
chemical components among various cultivation areas (Table
2), compound 1 was found highest in content in YA and GD,
whilst compound 7 was the highest content in HK and EM.
The RPA of compound 1 in YA and GD were 4 times higher
than that of HK, about 2 times higher than that of EM. On
the other hand, the RPA of compound 8 in HK was generally
doubled of the others. Thus, fresh H. cordata of different
production areas can be distinguished by assessing the RPA
of compound 1, which differentiates YA and GD from HK
and EM. The highest content of compound 8 in HK among
others helped to distinguish itself from EM. Likewise, YA
and GD can differentiate from each other by considering the
level of compound 2.
Comparative Fingerprint among Various Medicinal
Parts of Fresh H. cordata
It has been well known that dif-
ferent medicinal portions always possess variety in pharma-
cological activities.39)In some circumstances, Chinese medi-
cine practitioner may use different medicinal portions of the
herb for the same curative purpose. It is therefore important
to learn about any difference in composition among various
medicinal portions of H. cordata.
The total ion chromatograms for different medicinal por-
tions of H. cordata are given in Fig. 4. It is obvious that the
overall chemical profile among these fingerprints vary con-
siderably for different medicinal portions. Considering the
chemical profile for each medicinal part (Table 3), compound
3 was not available in leaf portion whilst compounds 10 and
11 were absent in root. These unique features help to identify
themselves from the others. Although the chemical profiles
1606 Vol. 53, No. 12
Fig. 2. HPLC Fingerprint of 11 Batches of Fresh H. cordata Collected from Emei
Table 1.The On-Line Detected Chromatographic and MS Data of the Eleven Identified Compounds in the HPLC Fingerprints
tR(min) [M?H]?(m/z) [M?H]?(m/z) Identification
of entire herb and stem are similar to one another, the dis-
tinctively high contents of compounds 1 and 2 in stem region
differentiated itself from the entire herb. The RPA for each
medicinal portion is given in Table 3. The corresponding
data for entire herb is shown in Table 2 (EM).
Distinguish between Fresh and Dried H. cordata
general, dried H. cordata was preferentially used as a medici-
nal material and the practice was recorded in previous edi-
tions of Chinese pharmacopoeia. However, the fresh counter-
part has been added in the latest 2005 edition. Our recent
pharmacological studies showed that a methanolic extract of
fresh herb gave better activities than its dried counterpart in
term of anti-inflammatory activity, relieving cough and an-
tibiosis (The result will be reported in another paper). In this
regard, the corresponding fingerprint between these two
types of processed herbs with different processing methods
was compared. 11 fresh (EM-1 to EM-11) and 5 dried (EM-
27 to EM-31) samples were compared in terms of their fin-
gerprinting profiles. Comparing the chemical components
December 2005 1607
A: stem; B: root; C: leaf; D: entire herb.
HPLC Fingerprints of Different Parts of H. cordata Collected from
A: HongKong; B: GuangDong; C: Yaan; D: Emei.
HPLC Fingerprints of H. cordata Collected from Various Sources
Table 2. The Relative Retention Time (RRT) and Relative Peak Area (RPA) of Characteristic Peaks in YA, GD, HK and EM
Peak/ YA (n?11) GD (n?5)HK (n?5) EM (n?11)
no.RRTRPA RRT RPARRT RPARRTRPA
RRT and RPA are the ratio of retention time and peak area of each peak with respect to peak 7, respectively. The value is mean?S.D.
between the fresh and dried herbs (Tables 2, 3), most of the
contents in the fresh herb are generally higher than that of
the dried counterpart (Fig. 5). Probably the differences were
attributed to the lost or decomposition during the drying
process.40,41)This observation also helps to interpret their dif-
ferences in pharmacological activities.
Comparison between GC/MS and HPLC-DAD-MS
Fingerprints of H. cordata
generated from the present developed direct HPLC-DAD-MS
fingerprint analysis are generally comparable and consistent
with our previously studies using direct GC-MS fingerprint
method on H. cordata.25)However, the analysis in previously
established GC-MS fingerprint only focused on those volatile
components. These volatile compounds are always unstable
and easily decomposed during the course of sample prepara-
tion and analysis. With the aims to streamline the analysis
and more comprehensively representing the quality of H.
cordata, a new HPLC-DAD-MS fingerprinting method has
been developed. The method gives an overall picture about
the various major types of chemical components present in
H. cordata and more comprehensively reflecting its quality.
The validated method has been proven simple, sensitive and
selective with good accuracy and reproducibility. In other
words, the HPLC-DAD-MS method can be readily utilized as
a comprehensive quality evaluation tool for simultaneous de-
termination of phenols, flavones and alkaloid components in
H. cordata. Furthermore, the method can further be extended
The results and conclusion
its application to authenticate and assess the quality of re-
lated medicinal products for H. cordata.
A HPLC-DAD-MS fingerprint method has been developed
for the analysis of H. cordata. The simulative mean chro-
matogram of 11 batches of samples from Emei of Sichuan
was obtained using a standardized procedure. The eleven
common peaks were identified by the HPLC-DAD-MS with
respect to the authentic chemical reference compounds. The
entire HPLC profiles with 11 common peaks were success-
fully used to authenticate H. cordata and differentiate among
different cultivations areas and different portions of herb. The
results showing that the developed method is a straightfor-
ward, sensitive and selective tool with good accuracy and re-
producibility. This HPLC-DAD-MS fingerprint method can
be readily utilized as a suitable quality control method for
quality evaluation of H. cordata and its related products.
Grant, Hong Kong Baptist University (FRG/03-04/II-39). The authors would
like to thank Chi-Leung Chan, School of Chinese medicine, Hong Kong
Baptist University, for the technical support in the operation of LC-MS.
This work was supported by the Faculty Research
References and Notes
1) World Health Organization, Guidelines for the Assessment of Herbal
Medicines, Munich, 28.6.1991, WHO, Geneva, 1991.
2) Lazarowych N. J., Pekos P., Drug Inf. J., 32, 497—512 (1998).
3)Hasler A., Sticher O., Meier B., J. Chromatogr. A., 605, 41—48
1608Vol. 53, No. 12
Table 3. The RRT and RPA of Characteristic Peaks for Samples of Emei among Different Medicinal Parts of H. cordata and Processing Methods
Fresh herbDry herb
Leaf (n?5) Root (n?5) Stem (n?5) Entire (n?5)
RRTRPA RRT RPARRT RPA RRTRPA
RRT and RPA are the ratio of retention time and peak area of each peak with respect to peak 7, respectively. The value is mean?S.D.
A: dry entire herb; B: fresh entire herb.
HPLC Fingerprints of Dried and Fresh Herbs of H. cordata Collected from Emei
Chuang W. C., Wu H. K., Sheu S. J., Chiou S. H., Chang H. C., Chen
Y. P., Planta Med., 61, 459—465 (1995).
The Pharmacopoeia Committee of China. “The Chinese Pharma-
copoeia,” Vol. I, The Chemical Industry Publishing House, Beijing,
China, 2005, p. 155.
Nishiya H., Ishiwata K., Komatsu K., Nakata O., Kitamura K., Fujii
S., Chem. Pharm. Bull., 36, 1902—1904 (1988).
Jong T. T., Jean M. Y., J. Chin. Chem. Soc., 40, 301—303 (1993).
Proebstle A., Neszmelyi A., Jerkovich G., Wagner H., Bauer R., Nat.
Prod. Lett., 4, 235—240 (1994).
State Administrative Bureau of TCM: ?http://www.cntcm.com.cn/?, 11
Chang J. S., Chiang C. C., Liu L. T., Wang K. C., Lin C. C., Am. J.
Chin. Med., 29, 303—312 (2001).
Bauer R., Proebstle A., Lotter H., Phytomedicine, 2, 305—308 (1996).
Kwon H. D., Cha I. H., Lee W. K., Song J. H., Park I. H., Nihon Eiyo
Shokuryo Gakkai Shi, 1, 208—213 (1996).
Hoang T. H., Ha V . B., Tran Q. H., Ha V . H., Tap Chi Duoc Hoc, 10,
Choi C. W., Kim S. C., Hwang S. S., Choi B. K., Ahn H. J., Lee M. Y.,
Park S. H., Kim S. K., Plant Sci., 163, 1161—1168 (2002).
Chen Y. Y., Liu J., Chen C. M., Chao P. Y., Chang T. J., J. Nutr. Sci. Vi-
taminol., 49, 327—333 (2003).
Kim S. K., Ryu S. Y., No J., Choi S. U., Kim Y. S., Arch. Pharm. Res.,
24, 518—521 (2001).
Jung H. A., Park J. C., Chung H. Y., Kim J., Choi J. S., Arch. Pharm.
Res., 22, 213—218 (1999).
Qin J., Chen T., Chen S. L., Lu Q., J. Instrum. Anal., 22, 85—87
Li L., Hu G. L., Lee S. CF., Wang X. R., Acta Agriculturae Universi-
ties JiangXiensis, 23, 487—491 (2001).
Choe K. H., Kwon S. J., Jung D. S., Analytical Science & Technology,
4, 285—288 (1991).
Liang M. M., Qi M. L., Zhang C. B., Zhou S., Fu R. N., Huang J. X.,
Anal. Chim. Acta, 531, 97—104 (2005).
Qi M. I., Ge X. X., Liang M. M., Fu R. N., Anal. Chim. Acta, 527,
23) Li X. N., Hu Y., Zhou Z. F., Liang Y. Z., Computers and Applied
Chemistry, 19, 234—236 (2002).
Choe K. H., Kwon S. J., Jung D. S., Eum K. D., Punsok Kwahak, 1,
Meng J., Leung K. S. Y., Jiang Z. H., Dong X. P., Zhao Z. Z., Chem.
Pharm. Bull., 53, 1484—1489 (2005).
Kawamura T., Hisata Y., Okuda K., Noro Y., Tanaka T., Yoshida M.,
Sakai E., Natural Medicines, 48, 208—212 (1994).
Fuse J. I., Kanamori H., Sakamoto Y. S., Natural Medicines, 48, 307—
Ye M., Yan Y. N., Tiao L., Ni X. M., Chinese J. Chinese Materia Med-
ica, 27, 115—117 (2002).
Lee J. Y., Moon S. O., Kwon Y. J., Rhee S. J., Park H. R., Choi S. W.,
Food Science and Biotechnology, 13, 176—184 (2004).
Yu D. Q., Yang J. S., “Analytical Chemistry Handbook,” 2nd ed., Vol.
VII, Chemical Industry Press, Beijing, 1999, p. 838.
Yi K., Shi J. G., Zhou G. X., Xie M. Y., Chinese J. Chinese Materia
Medica, 27, 43—45 (2002).
Gross G. A., Merier B., Sticher O., Phytochemistry, 31, 1391—1394
Li J. B., Ding Y., Chinese J. Chinese Materia Medica, 26, 479—481
Hua Y., He L., Wang H. Q., Chinese J. Chinese Materia Medica, 28,
Yu D. L., Guo J., Liao Y. H., Xu L. Z., Yang S. L., Acta Bot. Sin., 41,
Achenbach H., Frey D., Waibel R., J. Nat. Prod., 54, 1331—1336
Kang L. P., Qi L. H., Zhang J. P., Zhou B., J. Medical Colleges of PLA,
18, 32—35 (2003).
Cakir A., Mavi A., Yildirim A. D., Mehmet E., Harmandar M., Kazaz
C., J. Ethnopharmacol., 87, 73—83 (2003).
Zeng Z. D., Liang Y. Z., Xu C. J., Anal. Bioanal. Chem., 381, 913—
Sakai E. J., Shibata T., Kawamura T., Hisata Y., Noro Y., Yoshida M.,
Tanaka T., Natural Medicines, 50, 45—48 (1996).
Yamaguchi S., Kawamura T., Noro Y., Kawamura K., Tanaka T., Nat-
ural Medicines, 55, 139—142 (2001).
December 2005 1609