See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/24210395

Composition of Indigo naturalis

Article  in  Planta Medica · April 2009

DOI: 10.1055/s-0029-1185447 · Source: PubMed

CITATIONS READS

15 1,747

4 authors, including:

Matthias Hamburger
University of Basel
514 PUBLICATIONS   10,164 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Terpenoids and phenolics of Micromeria persica View project

Ethnoveterinary medicine & local knowledge on home made remedies for maintaining and improving animal health in organic farming View project

All content following this page was uploaded by Matthias Hamburger on 07 September 2015.

The user has requested enhancement of the downloaded file.

860 Letter
Composition of Indigo naturalis
Inken Plitzko, Tobias Mohn, Natalie Sedlacek,
Matthias Hamburger
Institute of Pharmaceutical Biology, University of Basel, Basel, Switzer-
land
Abstract
!
A proposal for a European Pharmacopoeia monograph concern-
ing Indigo naturalis has recently been published, whereby the in-
digo (1) and indirubin (2) content should be determined by
HPLC‑UV. This method was tested, but problems were seen with
the dosage of indigo due to poor solubility. A quantitative assay
for indigo based on 1H‑NMR was developed as an alternative.
The HPLC and qNMR assays were compared with eight Indigo
naturalis samples. The HPLC assay consistently gave much lower
indigo concentrations because solubility was the limiting factor
in sample preparation. In one sample, sucrose was identified by
1
H‑NMR as an organic additive. Simple wet chemistry assays for
undeclared additives such as sugars and starch were tested with
artificially spiked Indigo naturalis samples to establish their lim-
its of detection, and sulfate ash determinations were carried out
in view of a better assessment of Indigo naturalis in a future Euro-
pean monograph.
Key words
Indigo naturalis · indigo · indirubin · qNMR · HPLC · Pharmaco-
poeia monograph
Abbreviations
! naturalis samples purchased in China, Germany, and Switzerland
FID: free induction decay
PLE: pressurized liquid extraction
qNMR: quantitative 1H‑NMR
TCM: traditional Chinese medicine
Supporting information available online at
http://www.thieme-connect.de/ejournals/toc/plantamedica
Indigo naturalis (Qingdai) is used in traditional Chinese medicine
(TCM) to treat chronic diseases such as psoriasis. Interest in the
drug was fostered when activity against myelotic leukemia was
discovered [1, 2]. Indigo naturalis contains indigo (1) and indiru-
bin (2). The latter compound inhibits cyclin-dependent kinases
such as CDK1/cyclin B and CDK5/p25 and is considered the main
active principle of Indigo naturalis [3, 4].
Indigo naturalis is prepared from five indigoferous plants, Baphi-
cacanthus cusia Brem. (Acanthaceae), Polygonum tinctorium Ait.
(Polygonaceae), Indigofera tinctoria L. (Fabaceae), Isatis indigotica
Fort., and Isatis tinctoria L. (Brassicaceae), via a fermentative ex-
traction process [5]. Recently, the European Pharmacopoeia Com-
mission initiated the preparation of monographs for several TCM
drugs, and a draft monograph for Indigo naturalis was proposed
[6]. It is largely based on the monograph of the Chinese Pharma-
copoeia, but a major difference is the quantitative assay of indigo
(1) and indirubin (2), where RP-HPLC‑UV is used instead of a
photometric determination. The specifications remain the same,
with a content of not less than 2.0 % for 1 and 0.13 % for 2.
Plitzko I et al. Composition of Indigo naturalis … Planta Med 2009; 75: 860–863
Downloaded by: IP-Proxy CONSORTIUM:SchweizerHochschulen (UniBasel), Universität Basel. Copyrighted material.
Fig. 1 Structures of 1–4.
Given our ongoing studies on indigoferous Isatis species [7–10],
the draft monograph, and in particular the HPLC assay, drew our
attention because of the extremely poor solubility of indigo. An-
other point that needed some investigation was the fact that
> 97% of this drug remained undefined. This could indeed be a
problem, given various reports on adulteration of TCM herbal
products with synthetic drug substances [11]. We here report an
alternative method for determination of indigo (1) content in In-
digo naturalis and analysis of commercial samples for organic and
inorganic constituents (see l " Fig. 1 for structural formulas).
We first analyzed indigo and indirubin in eight commercial Indigo
with the HPLC‑UV assay of the monograph proposal. The results
are given in l " Fig. 2 and Table 1S (Supporting Information). The
indigo content varied between 1.1 % and 1.4 %, and the indirubin
content between 0.16 % and 0.37 %. However, we realized that
even after careful sonification insoluble residues of indigo re-
mained during extraction of the samples with CHCl3/MeOH
(1 : 1). To investigate the extraction step further, we carried out
extraction experiments with this solvent mixture using pressur-
ized liquid extraction (PLE). PLE is a technique that previously
provided exhaustive extraction (> 97 %) for various compound
classes in medicinal plants [12, 13]. PLE experiments showed that
indigo was not exhaustively extracted from Indigo naturalis when
using up to four consecutive extraction cycles with CHCl3/MeOH
(1 : 1) at 70 °C (Fig. 1S, Supporting Information).
As an alternative, we determined the indigo content by quantita-
tive 1H‑NMR (qNMR) [14]. Samples were suspended in DMSO-d6
and 1,3,5-trimethoxybenzene used as internal standard. Com-
plete solubility of indigo was achieved by thermostatting of the
samples at 80 °C during measurements. The indigo content as de-
termined by qNMR varied between 1.4 % and 2.7 % and was equal
to or substantially higher than the values obtained by HPLC
(l" Fig. 2 and Table 1S, Supporting Information). Analysis of an in-
digo reference sample showed that the standard deviation of the
qNMR assay was 2%. Hence, the low contents determined with
the HPLC assay of the draft monograph appear to be due to in-
complete solubility of indigo. The indirubin (2) content could
not be determined reliably by qNMR due to its low concentration

1
Fig. 3 H‑NMR spectra of Indigo naturalis sample GreenLife and sucrose.
in Indigo naturalis. Because the solubility of 2 is much higher than
that of 1 [15], the proposed HPLC‑UV method provides correct re-
sults.
The qNMR experiments allowed a simultaneous search for other
organic compounds that could be contained in the samples as
adulterants [11] or formulating agents. In seven of eight samples,
only indigo (1) and traces of indirubin (2) were detected (Figs. 2S
and 3S, Supporting Information). One sample (GreenLife), how-
ever, showed a complex pattern of resonances in the spectral
range δ = 3.2–5.2 reminiscent of a carbohydrate. The compound
was finally identified as sucrose by comparison with the spec-
trum of an authentic reference (l " Fig. 3). Because the draft
monograph does not test for carbohydrates, we explored the pos-
sibility of using simple wet chemical assays for this purpose. Indi-
go naturalis (sample LIAN) was spiked with potato starch and glu-
cose [20 %, 10 %, and 5 % (w/w)], respectively, and the limits of de-
tection of these additives by I2 and Fehlingʼs reagent were deter-
Letter 861
Fig. 2 Quantification of indigo (1) by HPLC and
qNMR and of indirubin (2) by HPLC in Indigo natu-
ralis samples.
Downloaded by: IP-Proxy CONSORTIUM:SchweizerHochschulen (UniBasel), Universität Basel. Copyrighted material.
mined (Supporting Information). The detection limits for both
were 5 %, and sample GreenLife tested positive for reducing sug-
ars after treatment with 1 N HCl.
The remaining, insoluble part of the Indigo naturalis samples had
to be inorganic. Sulfate ash determination according to Pharma-
copoeia Europea gave values up to 69 %, and calcium as well as
carbonate tested positive [16] (Supporting Information). The typ-
ical production procedure is to add lime to the indigo vat at the
end of the fermentation process [17]. It has been shown by elec-
tron microscopy that indigo is adsorbed to the surface of CaCO3
particles during this process [18].
Indigoferous plants such as Isatis tinctoria and I. indigotica con-
tain various alkaloids [7, 8] that may possibly also be found in In-
digo naturalis. We therefore analyzed the MeOH/CHCl3 (1 : 1) ex-
tracts from the HPLC assay by LC‑MS, modified from a published
Plitzko I et al. Composition of Indigo naturalis … Planta Med 2009; 75: 860–863
862 Letter

method [15] (l " Fig. 4). Isatin (3) was detected in all samples, and

varying amounts of tryptanthrin (4) were found in the samples

China1, China2, China3, and GreenLife.
In summary, our analyses showed that the draft monograph
needs some improvements. The quantitative assay for indigo is
not adequate. With this assay, none of the eight samples met the
minimal requirements of 2% indigo. However, even with the
qNMR assay, four out of eight samples were below the 2 % limit
of the draft monograph. Hence, a reassessment of the minimal
requirements is warranted, but certainly the analysis of a wider
range of Qingdai samples is needed. A determination of sulfate
ash is needed in the monograph because of the high content
(> 97%) of inorganic matter. Simple wet chemistry assays allow

Downloaded by: IP-Proxy CONSORTIUM:SchweizerHochschulen (UniBasel), Universität Basel. Copyrighted material.

detection of starch and reducing sugars in Indigo naturalis at con-
centrations as low as 5%. LC‑MS analysis of Indigo naturalis sam-
ples showed some characteristic differences in minor alkaloids,
which may be indicative of the plant species from which they
were produced. Tryptanthrin, for instance, was found only in four
of eight samples. Among the indigoferous plants used to prepare
Qingdai, this alkaloid has so far been reported only in Isatis tinc-
toria and I. indigotica.

Materials and Methods

!
The powdered drugs were from the following sources (sample
name in parentheses): Fujian Xianyou Dye Factory (China1),
Chendu Jianjiang Pharmaceutical Factory (China2), Chengdu
Zongseng Herbal Pieces Co. Ltd. (China3), Complemedis (Comple-
medis), Milestone International AG (GreenLife), Kaiser-Apotheke
(Kaiser), LIAN Chinaherb (LIAN), and Sinomed Hirslanden (Si-
nomed).
Indigo (1) (> 95 %) and 1,3,5-trimethoxybenzene (5) (> 99%) were
purchased from Sigma-Aldrich. Indirubin (2) and tryptanthrin (4)
had been previously synthesized at the Institute of Pharmaceuti-
cal Biology, University of Jena, according to published procedures
[4, 19]. Isatin (3) was from VEB Berlin-Chemie. Purity was ≥ 99 %
as determined by HPLC and NMR. Deuterated solvents (> 99.5 %)
were purchased from Armar Chemicals, and HPLC-grade solvents
were from Scharlau.
HPLC analyses according to the monograph proposal and PLE ex-
periments with the Indigo naturalis sample Sinomed are de-
scribed in the Supporting Information.
NMR spectra were obtained with a 500-MHz Avance III (Bruker)
equipped with a 5-mm BBO probe. For processing and evaluation,
Topspin 2.0 was used. Measuring at 80 °C in DMSO-d6 ensured
sufficient solubility of 1. In order to define the recycle delay re- Fig. 4 LC‑MS of minor alkaloids.
quired for quantitative conditions, the relaxation time T1 was de-
termined. For this purpose synthetic 1 and 1,3,5-trimethoxyben-
zene (5) were measured using the pulse program t1ir1d (Bruker)
whereby the inversion recovery delay was changed in steps of of 0.3 Hz prior to Fourier transformation. A digital filter (baseopt;
100 ms (0–5 s). The T1 value of the aromatic protons of 5 was Bruker) improved the automatic baseline correction, which was
5.9 s, and the maximum T1 of 1 was 3.6 s. performed before signal integration.
Quantitative 1H‑NMR was carried out using the pulse program In a glass vial, 0.25 mg of reference 1 or 2.5 mg of the eight Indigo
zgpr (Bruker) with a recycle delay of 30 s (≥ 5 × T1) and a pre-scan naturalis samples were taken up in 980 µL DMSO‑d6 under heat-
delay of 15 µs. The carrier frequency was set to 3.038 ppm, and ing to 80 °C and sonification. 1,3,5-Trimethoxybenzene stock so-
the receiver gain was automatically adjusted to accomplish effec- lution (20 µL, 47.7 mmol/L) was added as reference, and the sam-
tive water suppression. Two dummy scans, 128 scans, and a ple was transferred into an NMR tube. Reference signals for inte-
sweep width of 20 ppm, covering 64 K data points, provided the gration were δ = 7.52 (1) and δ = 6.11 (5). All samples were pre-
FID. Processing of the spectra included zero filling (x1) and multi- pared and measured in triplicate.
plication of the FID by an exponential decay apodization function

Plitzko I et al. Composition of Indigo naturalis … Planta Med 2009; 75: 860–863
 Letter 863

Supporting information 10 Mohn T, Hamburger M. Glucosinolate pattern in Isatis tinctoria and I.

Tables of the quantitative analysis of indigo and indirubin and of
sulfate ash determination; figures of ASE experiments, 1H‑NMR
spectra of indigo reference, and Indigo naturalis samples; and
protocols for Pharmacopoeia tests are available as Supporting In-
formation.
Acknowledgements
! qHNMR illustrated with taxol. J Nat Prod 2007; 70: 589–595
A major part of this study was carried out as the Master thesis of
N. Sedlacek. Thanks are due to Dr. X. Yang for procurement of In-
digo naturalis samples from China and for translation of Chinese
publications.
indigotica seeds. Planta Med 2008; 74: 885–888
11 Yee SK, Chu SS, Xu YM, Choo PL. Regulatory control of Chinese proprie-
tary medicines in Singapore. Health Policy 2005; 71: 133–149
12 Benthin B, Danz H, Hamburger M. Pressurized liquid extraction of me-
dicinal plants. J Chromatogr A 1999; 837: 211–219
13 Basalo C, Mohn T, Hamburger M. Are extraction methods in quantita-
tive assays of pharmacopoeia monographs exhaustive? A comparison
with pressurized liquid extraction. Planta Med 2006; 72: 1157–1162
14 Pauli GF, Jaki BU, Lankin DC. A routine experimental protocol for
15 Mohn T, Potterat O, Hamburger M. Quantification of active principles
and pigments in leaf extracts of Isatis tinctoria by HPLC/UV/MS. Planta
Med 2007; 73: 151–156
16 [no authors]. European Pharmacopeia, 6th edition. Strasbourg: Direc-

Downloaded by: IP-Proxy CONSORTIUM:SchweizerHochschulen (UniBasel), Universität Basel. Copyrighted material.

torate for the quality of medicines of the Council of Europe; 2008
17 Yang T, Wang Y, Xu F, Sun Q, Hu Y. Fermentation with microorganism for
References indigo. Chin Tradit Patent Med 2005; 27: 914–915
1 Stöger E, Friedel F. Aktualisierung, Monographie „Indigo naturalis“. Arz- 18 Wei Q, Yang M, Xu R, Sun Q, Hu Y. New conception about the processing
neibuch der chinesischen Medizin. Stuttgart: Deutscher Apotheker of Indigo naturalis. Chin Tradit Patent Med 2004; 26: 116–118
Verlag; 2006: 11 19 Friedländer P, Roschdestwenski N. Über ein Oxidationsprodukt des Indi-
2 Chen DH, Xie JX. Chemical constituents of traditional Chinese medicine goblaus. Ber Dtsch Chem Ges 1915; 48: 1841–1847
Qing Dai. Chinese Tradit Herbal Drugs 1984; 1: 6–8
3 Leclerc S, Garnier M, Hoessel R, Marko D, Bibb JA, Snyder GL, Greengard P,
Biernat J, Wu YZ, Mandelkow EM, Eisenbrand G, Meijer L. Indirubins in- received September 19, 2008
hibit glycogen synthase kinase-3β and CDK5/P25, two protein kinases revised January 19, 2009
involved in abnormal tau phosphorylation in Alzheimerʼs disease. accepted January 22, 2009
J Biol Chem 2001; 276: 251–260
4 Hoessel R, Leclerc S, Endicott JA, Nobel ME, Lawrie A, Tunnah P, Leost M,
Bibliography
Damiens E, Marie D, Marko D, Niederberger E, Tang W, Eisenbrand G,
DOI 10.1055/s-0029-1185447
Meijer L. Indirubin, the active constituent of a Chinese antileukaemia
Published online March 18, 2009
medicine, inhibits cyclin-dependent kinases. Nat Cell Biol 1999; 1:
60–67 Planta Med 2009; 75: 860–863
5 [no authors]. Monograph of the Chinese Pharmacopoeia, Engl. edition. © Georg Thieme Verlag KG Stuttgart · New York ·
Beijing: The Pharmacopoeia Commission of PRC; 2005 ISSN 0032‑0943
6 [no authors]. Monograph proposal „natural indigo“. Pharmeuropa
2008; 20 (1): 118 Correspondence
7 Hamburger M. Isatis tinctoria – from the rediscovery of an ancient me- Prof. Dr. Matthias Hamburger
dicinal plant towards a novel anti-inflammatory phytopharmaceutical. Department of Pharmaceutical Sciences
Phytochem Rev 2002; 1: 333–344 Institute of Pharmaceutical Biology
University of Basel
8 Oberthür C, Schneider B, Graf H, Hamburger M. The elusive indigo pre-
Klingelbergstrasse 50
cursors in woad (Isatis tinctoria L.) – identification of the major indigo 4056 Basel
precursor, isatan A, and a structure revision of isatan B. Chem Biodivers Switzerland
2004; 1: 174–182 Phone: + 41 6 12 67 14 25
9 Oberthür C, Graf H, Hamburger M. The content of indigo precursors in Fax: + 41 6 12 67 14 74
Isatis tinctoria leaves – a comparative study of selected accessions and Matthias.Hamburger@unibas.ch
post-harvest treatments. Phytochemistry 2004; 65: 3261–3268

Plitzko I et al. Composition of Indigo naturalis … Planta Med 2009; 75: 860–863
View publication stats