Rational quality assessment procedure for less-investigated herbal medicines: Case of a Congolese antimalarial drug with an analytical report

    Rational quality assessment procedure for less-investigated herbal medicines: case of a Congolese antimalarial drug with an analytical report Dieudonn´e Tshitenge Tshitenge, Karine Ndjoko Ioset, Jos´e Nzunzu Lami, Josaphat Ndelo-di-Phanzu, Jean-Pierre Koy Sita Mufusama, Gerhard Bringmann PII: DOI: Reference:

S0367-326X(16)30055-7 doi: 10.1016/j.fitote.2016.03.012 FITOTE 3373

To appear in:

Fitoterapia

Received date: Revised date: Accepted date:

31 January 2016 9 March 2016 12 March 2016

Please cite this article as: Dieudonn´e Tshitenge Tshitenge, Karine Ndjoko Ioset, Jos´e Nzunzu Lami, Josaphat Ndelo-di-Phanzu, Jean-Pierre Koy Sita Mufusama, Gerhard Bringmann, Rational quality assessment procedure for less-investigated herbal medicines: case of a Congolese antimalarial drug with an analytical report, Fitoterapia (2016), doi: 10.1016/j.fitote.2016.03.012

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Rational quality assessment procedure for less-investigated herbal medicines: case of a Congolese antimalarial drug with an analytical report Dieudonné Tshitenge Tshitenge a,b, Karine Ndjoko Ioset a, José Nzunzu Lami b, Josaphat Ndelo-

RI P

T

di-Phanzu b, Jean-Pierre Koy Sita Mufusama a,b, Gerhard Bringmann a,* a

Institute of Organic Chemistry, University of Würzburg, Am Hubland, 97074 Würzburg, Germany

b

Faculty of Pharmaceutical Sciences, University of Kinshasa, B.P. 212 Kinshasa XI, DR Congo

SC

* Corresponding author. Tel.: +49 931 318 5323; fax: +49 931 318 4755.

NU

E-mail address: [email protected] (G. Bringmann).

ABSTRACT

MA

Herbal medicines are the most globally used type of medical drugs. Their high cultural acceptability is due to the experienced safety and efficiency over centuries of use. Many of them

ED

are still phytochemically less-investigated, and are used without standardization nor quality control. Choosing SIROP KILMA, an authorized Congolese antimalarial phytomedicine, as a

PT

model case, our study describes an interdisciplinary approach for a rational quality assessment of herbal drugs in general. It combines an authentication step of the herbal remedy prior to any

CE

fingerprinting, the isolation of the major constituents, the development and validation of an HPLC-DAD analytical method with internal markers, and the application of the method to

AC

several batches of the herbal medicine (here KILMA) thus permitting the establishment of a quantitative fingerprint. From the constitutive plants of KILMA, acteoside, isoacteoside, stachannin A, and pectolinarigenin-7-O-glucoside were isolated, and acteoside was used as the prime marker for the validation of an analytical method. This study contributes to the efforts of the WHO for the establishment of standards enabling the analytical evaluation of herbal materials. Moreover, the paper describes the first phytochemical and analytical report on a marketed Congolese phytomedicine.

Keywords: Herbal medicine Fingerprinting Authentication Validation

Page 1 of 22

ACCEPTED MANUSCRIPT Acteoside SIROP KILMA

T

Chemical compounds:

RI P

Chemical compounds studied in this article: Acteoside (PubChem CID: 5281800); Stachannin A (PubChem CID: 44258461); pectolinarigenin-7-O-glucoside (PubChem CID: 44258439);

SC

Isoacteoside (PubChem CID: 6476333).

NU

1. Introduction

About 80% of the people in Africa rely on herbal medicines for their primary healthcare [1-

MA

3]. Over a long time period they have proven their safety and curative properties, and various

scientific studies have validated their healing effects [4-8]. However, maintaining an acceptable level of quality is often a big challenge [2,9]. Despite the numerous regulations from the WHO,

ED

the standardization and the quality assessment of herbal materials are still concerns of our time

PT

[2,10-12].

The situation is dramatic in particular in many developing countries, where no standards are

CE

available despite the massive use of these products. The Democratic Republic of the Congo (DR Congo) is a typical case where regulatory statuses for herbal drugs do not exist, although no

AC

botanical identification of raw materials and no quality control procedures for herbal medicines are carried out, nonetheless many plant-made drugs are recommended in the primary healthcare system [11].

With respect to the chemical composition of herbal drugs and aiming at the development of analytical standards for quality control of phytomedicines, we elaborated a reliable approach from the sampling to the validation of analytical methods. By the involvement of the phytoproducer in the key steps, this procedure prevents all risk of biased fingerprinting. An authorized antimalarial drug, SIROP KILMA® from DR Congo was used as a proof of concept. The choice was guided by the curative profile of KILMA, proven by a preliminary clinical study [13]. The vacuum left by the lack of quality standards has led to choose it as a model case for the application of the procedure here described. Therefore, phytochemical and analytical investigations were performed on the syrup and on its constitutive plant materials. Using the

Page 2 of 22

ACCEPTED MANUSCRIPT procedure, markers were isolated and an analytical method was developed and optimized. The method was used for the authentication and fingerprinting of KILMA. After isolation, purification, and structural elucidation, one of the four characterized substances, acteoside (1)

T

was selected as a prime marker and used for the validation of the method. The validated method

RI P

was applied to quantify this marker and all major substances of the fingerprint. The analysis of three batches of KILMA gave a first impression on the standardization state of the manufacturing

SC

process, and the establishment of standards to be checked in routine. We expect that this work dealing with the particular case of the syrup KILMA will define a referential frame for other

NU

less-investigated herbal medicines.

MA

2. Material and methods 2.1. Chemicals and reagents

ED

Acetonitrile, methanol (MeOH), and trifluoroacetic acid (TFA) were of HPLC grade (Merck, Darmstadt, Germany). All other chemicals (like e.g., ethanol and ethyl acetate) were of analytical

PT

grade. Water was purified by a Millipore system (Bedford, MA, USA). Acteoside (compound 1, purity ≥ 99%) purchased from AppliChem (Darmstadt, Germany) was used as the internal prime

2.2. Methodology

CE

marker.

AC

To ensure the authenticity of the working materials, a close interactive collaboration with the phyto-producer was established, in particular for the sampling and the plant collection. The samples were bought directly from the producer to assure the initial composition. The three involved plants were harvested under the strict instructions of the producer, e.g., collection of mature plant materials at the same location, drying in the shade, with no sunlight. The plant species identification was professionally confirmed by a specialist. The authentication step consisted of the comparison of peaks related to the main metabolites upon the agreement of their retention times, their online UV spectra, and their ESI-MS profiles. Hydroalcoholic plant extracts were prepared following the producer‟s protocol to permit the comparison.

Page 3 of 22

ACCEPTED MANUSCRIPT Sample preparation was aiming at the reduction of the complexity of the drug and at a reliable monitoring of the chromatographic fingerprint, thus permitting a more-easy to handle analysis of herbal medicines in developing countries. The isolation and characterization of the

T

markers were performed by chromatography (HPLC) and spectroscopic (UV, MS, NMR)

RI P

techniques, respectively. The method validation was performed following Q2(R1)-ICH guidelines [14]; it was based on the quantification of an internal reference, acteoside, which was

SC

commercially available. The method was further applied to the quantification of three different batches of KILMA.

NU

2.3. Samples and plant materials

Three batches of KILMA (syrup) were purchased in February 2012 from the phytoproducer,

MA

based in Kinshasa (DR Congo).

Stem barks of Gardenia ternifolia (GTS) and Crossopteryx febrifuga (CFS), and leaves of Lantana camara (LCL) were collected during the same time period in Kimwenza (Kinshasa),

ED

according to the phyto-producer‟s directives. Botanical identification was done by Mr. Nlandu Lukebakio of the “Institut National pour l‟Etude et la Recherche Agronomiques” (INERA), at

PT

the University of Kinshasa, where a voucher specimen of each plant was deposited, under the numbers 843, 175, and 3653 for GTS, CFS, and LCL, respectively. Considering possible

CE

variations in the profiles of the metabolites, depending on several parameters, like e.g. the

AC

harvesting period, it was of great importance to collaborate with the phyto-producer. 2.4. Authentication and fingerprinting of KILMA, method development 2.4.1. Sample preparation From the syrup KILMA, 10 mL were sequentially evaporated to dryness under vacuum and lyophilized. The residue was dissolved in 96% ethanol (EtOH) to obtain a solution of 40 mg/mL. This alcoholic mixture was then filtrated on a polyamide resin phase preactivated three times with 200 µL of MeOH, eluted with 96% EtOH, evaporated in vacuo, and lyophilized. The residue was dissolved in MeOH, and filtered over a 0.22-µm PET syringe prior to injection into the HPLC system.

Page 4 of 22

ACCEPTED MANUSCRIPT Ethanolic extracts (80% EtOH) of the air-dried powders of the three plant materials were obtained by exhaustive maceration and evaporation in vacuo, and were then treated as described

T

above.

RI P

For authentication, the chromatogram of KILMA was compared with those from the respective plant extracts treated according to the protocol of the producer. In this analysis, each peak in the chromatogram of KILMA should be found in at least one of the plants. The similarity

SC

criteria were based on the agreement within the following triad: retention times as well as onlineUV and -MS data. A satisfactory authentication was the prerequisite for the establishment of the

NU

fingerprint for KILMA using the developed method.

MA

Resolutions of peaks (Rs) were determined following equation (1).

ED

“Rt1” and “Rt2” in (1) are the retention times of two peaks, and “W1” and “W2” are the widths of the tangent intersections of the respective peaks with the baseline (cf., green strip in

AC

CE

PT

Fig. 1).

Fig. 1. Peak shape parameters for acteoside (1). The purple strips represent the tangents and the green one the baseline, while the other strips indicate different levels of the width.

Page 5 of 22

ACCEPTED MANUSCRIPT 2.5. Isolation and characterization of markers From the fingerprint, eight characteristic peaks were selected, according to their originality,

T

their occurrence, and amount in various solvent extracts.

RI P

Prior to analytical and preparative HPLC analyses, 100 mL of KILMA were exhaustively defatted and extracted with n-hexane, and subsequently with ethyl acetate. The ethyl acetate phase was then evaporated to dryness in vacuo; 4.5 g of the residue thus obtained were then

SC

dissolved in MeOH. The solution was injected into a preparative HPLC system using a SymmetryPrep-C18 column (Waters, 19 x 300 mm, 7 μm) with a mobile phase system made of

NU

0.05% TFA in water (A) and in acetonitrile (B). With a flow rate of 10 mL/min, the following gradient was used; 0-16 min: 10-50% B; 16-20 min: 50-98% B; 20-24 min: 98% B. Spectra were

MA

acquired at 226, 300, and 328 nm. Eight peaks were isolated, of which the four most abundant ones were characterized.

ED

2.6. Validation

2.6.1. Matrix and standard preparation

PT

A matrix solution was obtained by diluting 2.5 mL of KILMA to a 25.0 mL methanolic solution in a volumetric flask, followed by a cleaning-up on a polyamide column and on a 0.22-

CE

µm PET syringe filter.

Stock solutions made of 3.0 mg of acteoside in 2 mL with MeOH (HPLC grade) were

AC

obtained in duplicate, sonicated for 10 min, and filtered over a 0.22-µm PET syringe filter. From the stock solutions, four concentration-level solutions for calibration curves were prepared: 66.67, 200.0, 533.3, and 666.7 µg/mL, with an injection volume of 15 µL in duplicate and during three consecutive days. For validation standards, five concentration levels were prepared in triplicate from the stock solutions, after vacuum-evaporation to dryness, dilution to 975 µL of the matrix solution, and sonication prior to injection: 66.67, 133.6, 233.3, 366.7, and 500.0 µg/mL. For every single run, 15 µL were injected in triplicate, during three consecutive days. According to the Q2(R1)-ICH guidelines, the following validation parameters for an analytical method, were determined using the one-way analysis of variance (ANOVA) approach: specificity, trueness, precision, accuracy, range, and linearity [14,15].

Page 6 of 22

ACCEPTED MANUSCRIPT 2.7. Quantitative determination, and first hints on the standardization state Starting from three batches, 4 mL of KILMA syrup were diluted to a 10-mL methanolic solution, followed by a two-step purging procedure on a polyamide phase and on a 0.22-µm PET

T

syringe filter. The injection volume was 15 µL. Using the validated method, five repetitions per

RI P

batch were carried out.

The standardization state of the manufacturing process was evaluated by comparison of the

SC

quantitative profiles from three batches of KILMA. This permitted to deduce the quantitative fingerprint of KILMA for the eight peaks in the chromatogram according to the validated

NU

method.

MA

3. Results

3.1. Method development and optimization

ED

An HPLC-DAD method was developed for the phytochemical screening of KILMA. After several trials of sample preparation and chromatographic parameters, suited conditions were

PT

obtained towards the finest baseline separation and resolution of main peaks in the

CE

chromatogram within a short run time. Optimal chromatographic conditions for the fingerprinting of KILMA were achieved on a

AC

Symmetry-C18 column (Waters, 4.6 x 250 mm, 5 μm) with the mobile phase prepared from Millipore water (A) and acetonitrile (B), both containing 0.05% (v/v) TFA, in gradient mode, 017 min: 20-50% B, 17-19 min: 50-98% B, 19-21 min: 98% B. Conditions were stabilized at 25 °C with a flow rate of 1 mL/min. The compounds were monitored at 226, 300, 328, and 360 nm. Unless otherwise indicated, all LC-UV and LC-MS analyses were carried out under the obtained optimized conditions described above. 3.2. Authentication and fingerprinting The results of the authentication of KILMA based on the three parameters described in chapter 2.5 are displayed in Table 1. They confirmed that every peak in KILMA could be detected in at least one plant extract. All of the main peaks in the KILMA profile were found in at least one of the plant extract chromatograms, which confirmed their contribution to the total

Page 7 of 22

ACCEPTED MANUSCRIPT mixture. The chromatographic fingerprint of KILMA is shown in Fig. 2, and its features are

NU

SC

RI P

T

summarized in Table 2.

MA

Fig. 2. Chromatographic fingerprint of KILMA (λ = 328 nm).

ED

This fingerprint describes the standard profile to be reproduced in routine during the quality control of KILMA. It is based on the agreement of three reliable qualitative parameters: the

PT

retention time, and the online UV and MS data of each peak.

KILMA

AC

CE

Table 1 Authentication of SIROP KILMA and plant extracts. Samples Peak retention Occurrence of peaks compared to the profile of KILMA (min) 7.6 10.4 11.4 12.2 16.6 17.0 24.3 +

+

+

+

+

+

+

+

-

-

-

-

-

-

C. febrifuga

+

+

-

-

+

+

+

L. camara

-

-

+

+

+

-

+

G. ternifolia

(+): Present; (-): absent.

Page 8 of 22

ACCEPTED MANUSCRIPT Table 2 Fingerprint of KILMA. Parameters 2

3

4

5

3.45

3.63

4.13

5.69

6.55

1.6

2.73

6.09

8

7.45

9.51

13.21

2.89

3.43

7.43

15.1

33.1

41.1

75.7

34.4

339.8

74.4

48.9

69.4

Area (mV.min)

3.5

4.7

15.3

7.4

53.8

13.4

14.8

17.7

-

-

835.2

624.2

624.2

462.0

476.0

-

MA

From the previous peak.

NU

Peak height (mV)

Mass (g/mol) a

7

SC

Resolutiona

6

T

1

RI P

Retention (min)

Peaks

ED

3.3. Isolation and characterization of markers From KILMA, eight compounds were isolated by preparative HPLC. Compounds 1, 2, 3,

PT

and 4 corresponded to peaks number 5, 7, 8, and 6 in the chromatographic fingerprint, respectively. Compounds 1, 2, and 3 were obtained in sufficient quantities to permit purification

CE

and full structural characterization. The isolated compound referring to peak number 6 was

AC

unambiguously identified as isoacteoside (4) [16]; the purity of 4 was determined to be 85%. HR-ESIMS of compound 1 gave an ion [M+Na]+ at m/z 647.19528, corresponding to a molecular formula of C29H36O15. Extensive 1D and 2D NMR experiments, especially HSQC, HMBC, and COSY, permitted the complete assignment of the proton and carbon signals. All data were in agreement with the known structure of acteoside, as described earlier [17]. The structure and the chemical shifts (1H and

13

C) of 1 are shown in Figs. 3 and 4. In the MS-MS

spectrum, a loss of 162 and 146, and the fragment peak at m/z 315 confirmed the fragmentation pathway as suggested in the literature [18]. The UV spectrum, with characteristic maxima at 240, 283, and 332 nm, also confirmed the identity of 1 as acteoside. Figs. 3 and 4 summarize the NMR features of the characterized markers, all likewise found in Lantana camara, and confirm the presence of the known compounds stachannin A (2), pectolinarigenin-7-O-glucoside (3), and isoacteoside (4) in the plant extracts. Only two other

Page 9 of 22

ACCEPTED MANUSCRIPT reports on the natural occurrence of 2 are known so far from the literature [19,20], this is now the third report on the isolation of 2 as a plant metabolite, and even the first discovery of 2 as a

T

constituent of L. camara.

RI P

The commercially available 1 was chosen as an internal prime marker for the validation of

AC

CE

PT

ED

MA

NU

SC

the developed analytical method.

Fig. 3. Constituents isolated from SIROP KILMA: acteoside (1), stachannin A (2), pectolinarigenin-7-O-glucoside (3), and (4). The compounds 1-4 refer to the peaks number 5-8 in the chromatographic fingerprint of KILMA presented in Fig. 2.

Page 10 of 22

CE

PT

ED

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

Fig. 4. Chemical shifts (δ) in ppm of acteoside (1), stachannin A (2), and pectolinarigenin-7-Oglucoside (3) (1H in red, and 13C in blue). Isoacteoside (4) is not displayed.

3.4. Validation Acteoside Chemical Reference Substance (CRS) was used for the validation of the analytical method according to the guidelines [14,15]. This task was successfully completed. The calibration standards and their chromatograms were obtained in duplicates and recorded within three days to provide three calibration curves characterized by their equations (Table S1 and Fig. S8, supplementary information: SI). Individual values of validation standards in an authentic matrix of KILMA were measured for five levels of concentrations, also from the matrix of KILMA itself considered as „background concentration‟ (as shown in SI: Table S2). The trueness of the method revealed a recovery between 98.27 and 102.8% for the five levels of

Page 11 of 22

ACCEPTED MANUSCRIPT concentrations (SI: Table S3). From the evaluation of the precision of the method using a oneway ANOVA approach [14,15], the standard deviations of the intermediate precision was between 1.32 and 4.32%, expressing the closeness of the obtained experimental data (SI: Table

T

S4). For all the five levels of concentrations used for calibration, the limits of acceptance and

RI P

limits of tolerance were calculated. These validation parameters enabled the establishment of the accuracy profile (SI: Table S5), which is illustrated in Fig. 5 showing the limits of the

PT

ED

MA

NU

SC

quantitative region of the method.

CE

Fig. 5. Accuracy profile displaying the reliable quantification area of 1 (µg/mL). The limit of quantification (LOQ) of acteoside in KILMA using the validated method is also shown.

AC

The specificity evaluation, hinting at the peak purity of acteoside (1) and the ability of the method to resolve markers, evaluated at several wavelengths was also performed (SI: Fig. S8). The linearity profile displaying the evolution of the upper and lower limits of acceptance and of tolerance was also determined (SI: Fig. S9). The lower and upper limits of quantification for 1 were found to be 116.9 and 528.7 µg/mL, respectively. The validated procedure proved to be specific, precise, and accurate, with a dosage range from 116.9 to 528.7 µg/mL in 1, and linear with a coefficient of correlation (r2) equal to 0.998. 3.5. Quantitative determination, and first hints at the standardization state Figs. 6a and 6b display chromatograms and content histograms of the three batches of KILMA according to the quantitative determination. These Figures illustrate the variation in content of each single peak in the fingerprint. The means of single-peak concentrations in the

Page 12 of 22

ACCEPTED MANUSCRIPT three batches of KILMA and the calibration at the five concentration levels was also performed (SI: Tables S6 and S7). In order to assess the analytical range of variation in the fingerprint, eight main peaks were measured. The content variation of 1 in the three batches appears to be the most

T

fluctuating parameter, especially for the first batch. Based on the range of variation of peak

PT

ED

MA

NU

SC

RI P

contents, a quantitative fingerprint was established.

AC

CE

Fig. 6a. Comparison of the chromatograms obtained for the quantitative determination of three batches of KILMA (λ = 328 nm).

Fig. 6b. Content of three batches of KILMA: quantitative fingerprint.

Page 13 of 22

ACCEPTED MANUSCRIPT 4. Discussion Experimental designs were not required in the optimization of the developed LC method, because an appropriate separation of most of the compounds was achieved in a short run time (23

T

min) and with a minimum resolution of 1.6. The method developed was found optimum for its

RI P

main purposes: permitting identification and quantification of markers.

The authentication step confirmed that KILMA is a true mixture of the three declared herbal

SC

extracts. The origin of each main peak from the individual plant was established, and the approximate extract ratio was found to be 4: 3: 4.4 for GTS, CFS, and LCL, respectively. This

NU

revealed that the major peak in KILMA is 1 (peak number 5), according to the described

MA

protocol, and it is from LCL. The compound was subsequently taken as the prime marker. In the fingerprint, each peak was characterized by its UV and MS data, by its retention time

ED

in LC, and by its approximate amount in the 100 mL of KILMA (SI: Table S7). Previous studies dealing with fingerprinting of phytopharmaceuticals had rarely taken profit from hyphenated liquid chromatography and mass spectrometry (LC-MS) techniques [7,10,21-

PT

26], neither in the monitoring of peaks in complex plant mixtures, nor in the similarity of

CE

fingerprints with the original plant extracts to access the so-called phytoequivalence [27]. The procedure presented in this report takes profit of LC hyphenated with UV and MS, in the

AC

assignment of similarity and stability of peaks in a complex plant mixture. As required, in this work a clear difference was made between a routine task of quality control and a process leading to the establishment of standards, which was the aim of the work described herein. No quality specification can be applied to a non-normalized product [28-30]. Establishing quality guidelines for KILMA, the samples were taken from the producer, freshly prepared, and kept under his storage specifications. This approach permitted to have compounds in their “original” state, and avoided biased authentication and fingerprinting. The procedure applied to the samples permitted having a reduced and controllable number of metabolites in the fingerprint, which facilitated the monitoring of the whole composition in routine.

Page 14 of 22

ACCEPTED MANUSCRIPT The results of the validation reveals that the developed procedure is specific, linear (r2 = 0.998), precise, and accurate, and the dosage range is reliable for 1 between 116.9 and 528.7 µg/mL. The analysis of 1 in a matrix of KILMA by the chosen standard addition eliminates

RI P

T

virtually all of the matrix effects in the quantification step.

The analysis of 1 by HPLC-UV also permitted the simultaneous quantitative estimation of other peaks in three batches of KILMA. The methodology combined qualitative assessment and

SC

content determination in three batches, and thus give rise to a quantitative fingerprint. Content

NU

variation in 100 mL of the syrup was also deduced from all the data. Moreover, quantitative fingerprinting on the investigated three batches of KILMA provided

MA

the first hints at the standardization state of the manufacturing process. The first batch had the lowest content of 1 compared to the second and the third ones, which were almost identical. The quantities of the seven other compounds were quite similar, which made the content of 1 in

ED

KILMA a decisive parameter in the quantitative fingerprint and for the standardization process. With respect to previous investigations concerning the quantification of acteoside in crude

PT

natural extracts [31-33], the developed LC method enables a rapid and accurate analysis, in

CE

particular in a complex mixture of three plant extracts, viz. KILMA. Summarizing, this study led to the first identification of 2 in L. camara. More importantly,

AC

this is the first study providing a validated LC method with an internal prime marker on a Congolese phytomedicine here developed for the SIROP KILMA as a model case. It is also the first description of the standardization state of a Congolese herbal medicine at all. The overall described procedure may also serve as a valuable example for the development and the reliable application of this rational quality assessment on uninvestigated phytomedicines in general.

Conflict of interest The authors declare that this article has no conflict of interest.

Acknowledgments The authors are grateful to the Excellence Scholarship Program BEBUC (www.foerderverein-uni-kinshasa.de), the Else-Kröner-Fresenius-Stiftung, the DFG-funded

Page 15 of 22

ACCEPTED MANUSCRIPT research network SFB 630, and the German Excellence Initiative to the Graduate School of Life Sciences at the University of Würzburg, for the financial support to D.T.T. Special thank is due to Dr. Kabala Dihuidi, producer of KILMA, for his collaboration in the sampling, and to Mr.

Appendix A. Supplementary information (SI)

RI P

T

Nlandu Lukebakio from INERA for the botanical identification of the plants.

SC

Supporting data for this article can be found online at http: …

NU

References

[1] World Health Organization, Quality Control Methods for Herbal Materials, Geneva,

MA

Switzerland, 2011.

[2] World Health Organization, WHO Traditional Medicine Strategy: 2014-2023, Geneva,

ED

Switzerland, 2013.

PT

[3] World Health Organization, the World Malaria Report, Geneva, Switzerland, 2014. [4] L. Tona, N.P. Ngimbi, M. Tsakala, K. Mesia, K. Cimanga, S. Apers, T. Bruyne, L. Pieters,

CE

J. Totté, A.J. Vlietinck, Antimalarial activity of 20 crude extracts from nine African medicinal plants used in Kinshasa, Congo, J. Ethnopharmacol. 68 (1999) 193–203.

AC

[5] S.C. Jain, E. Menghani, R. Jain, Biomarkers as a Tool for Validation of Herbs and Spices, Internet J. Food Safety 9 (2007) 1–6. [6] C.Y. Li, S.I. Tsai, A.G. Damu, T.S. Wu, A rapid and simple determination of protoberberine alkaloids in Rhizoma Coptidis by 1H NMR and its application for quality control of commercial prescriptions, J. Pharm. Biomed. Anal. 49 (2009) 1272–1276. [7] L. Franco, B.H.D. Oliveira, Determination of umckalin in commercial tincture and phytopreparations containing Pelargonium sidoides by HPLC: Comparison of sample preparation procedures, Talanta 81 (2010) 1368–1372. [8] K. Mesia, L. Tona, M.M. Mampunza, N. Ntamabyaliro, T. Muanda, T. Muyembe, T. Musuamba, T. Mets, K. Cimanga, J. Totté, L. Pieters, A.J. Vlietinck, Antimalarial efficacy

Page 16 of 22

ACCEPTED MANUSCRIPT of a quantified extract of Nauclea pobeguinii stem bark in human adult volunteers with diagnosed uncomplicated falciparum malaria. Part 1: a clinical phase IIA trial, Planta Med.

T

78 (2012) 853–860.

RI P

[9] R. Shobhen, N.M. Patel, P.M. Patel, A review on modification of analytical techniques in herbal research, Intern. J. Research Ajurveda Pharm. 2 (2011) 1483–1485.

SC

[10] E.P. Springfield, P.K.F. Eagles, G. Scott, Quality assessment of South African herbal medicines by mean of HPLC fingerprinting, J. Ethnopharmacol. 101 (2005) 75–83.

NU

[11] S. Skalli, R. Soulaymani Bencheikh, Pharmacovigilance of herbal medicines in Africa-

MA

Questionnaire study, J. Ethnopharm. 171 (2015) 99–108. [12] World Health Organization, WHO traditional medicine strategy: 2002-2005, Geneva, Switzerland, 2002.

ED

[13] C.D. Kabala, V.N. Nzingula, N. Maluantesa, N. Basilua, A.P. Mbenza, J.N. Lami, U. Mbangu, R.B. Suami, P.N. Mambanzulua, Evaluation de l‟activité antipaludique du SIROP

PT

KILMA®, Annales de Pharmacie 3 (2005) 103–106. (Also available upon request to the

CE

authors).

[14] E. Rozet, A. Ceccato, C. Hubert, E. Ziemons, R. Oprean, S. Rudaz, B. Boulanger, P. Hubert,

AC

Analysis of recent pharmaceutical regulatory documents on analytical method validation, J. Chromatogr. A 1158 (2007) 111–125. [15] European Medicines Agency, London, United Kingdom, 2006. Note for Guidance on Validation of Analytical Procedures: Text and Methodology. [16] K. Taoubi, M.T. Fauvel, J. Gleye, C. Moulis, I. Fouraste, Phenylpropanoid glycosides from Lantana camara and Lippia multiflora, Planta Med. 63 (1997) 192-193. [17] C. Andary, R. Wylde, C. Laffite, G. Privat, F. Winternitz, Structures of verbascoside and orobanchoside, caffeic acid sugar esters from Orobanche rapum-genistae, Phytochemistry 21 (1982) 1123–1127.

Page 17 of 22

ACCEPTED MANUSCRIPT [18] L. Li, R. Tsao, Z. Liu, S. Liu, R. Yang, J.C. Young, H. Zhu, Z. Deng, M. Xie, Z. Fu, Isolation and purification of acteoside and isoacteoside from Plantago psyllium L. by high-

T

speed counter-current chromatography, J. Chromatogr. A 1063 (2005) 161–169.

RI P

[19] I.P. Sheremet, N.F. Komissarenko, Khim, Flavonoid glycosides of Stachys annuae, Prirodn. Soedin. 5 (1971) 583–587.

SC

[20] D.C. Albach, Charlotte Held Gotfredsen, Søren Rosendal Jensen, Iridoid glucosides of Paederota lutea and the relationships between Paederota and Veronica, Phytochemistry 65

NU

(2004) 2129–2134.

[21] A.K. Shah, B.A. Avery, C.M. Wyandt, Content Analysis and Stability Evaluation of

MA

Selected Commercial Preparations of St. John‟s Work, Drug Dev. Ind. Pharm. 31 (2005) 907–916.

ED

[22] Y. Cao, L. Wang, X. Yu, J. Ye, Development of the chromatographic fingerprint of herbal preparations Shuang-Huang-Lian oral liquid, J. Pharm. Biomed. Anal. 41 (2006) 845–856.

PT

[23] P.S. Xie, S. Chen, Y. Liang, X. Wang, R. Tian, R. Upton, Chromatographic fingerprint analysis - a rational approach for quality assessment of traditional Chinese herbal medicine,

CE

J. Chromatogr. A 1112 (2006) 171–180.

AC

[24] D.Z. Yang, Y.Q. An, X.L. Jiang, D.Q. Tang, Y.Y. Gao, H.T. Zhao, X.W. Wu, Development of a novel method combining HPLC fingerprint and multi-ingredients quantitative analysis for quality evaluation of traditional Chinese medicine preparation, Talanta 85 (2011) 885– 890. [25] Y. Xu, K. Foubert, L. Dhooghe, F. Lemière, K. Cimanga, K. Mesia, S. Apers, L. Pieters, Chromatographic profiling and identification of two new iridoid-indole alkaloids by UPLCMS and HPLC-SPE-NMR analysis of an antimalarial extract from Nauclea pobeguinii, Phytochem. Lett. 5 (2012) 316–319. [26] O. Shehzad, I.J. Ha, Y. Park, Y.W. Ha, Y.S. Kim, Development of a rapid and convenient method to separate eight ginsenosides from Panax ginseng by high-speed countercurrent

Page 18 of 22

ACCEPTED MANUSCRIPT chromatography coupled with evaporative light scattering detection, J. Sep. Sci. 34 (2011) 1116–1122.

T

[27] Y. Liang, P. Xie, K. Chan, Quality control of herbal medicines, J. Chromatogr. B 812

RI P

(2004) 53–70.

[28] S. Li, Q. Han, C. Qiao, J. Song, C. Lung Cheng, H. Xu, Chemical markers for the quality

SC

control of herbal medicines: an overview, Chin. Med. 3 (2008) 1–16. [29] A. Sharma, C. Shanker, L.K. Tyagi, M. Singh, C.V. Rao, Herbal Medicine for Market

NU

Potential in India: An Overview, Acad. J. Plant Sci. 1 (2008) 26–36.

MA

[30] V.M. Shinde, K. Dhalwal, M. Potdar, K.R. Mahadik, Application of quality control principles to herbal drugs, Intern. J. Phytomed. 1 (2009) 4–8. [31] B. Blazics, Á. Alberti, S. Béni, L. Kursinszki, L. Tölgyesi, Á. Kéry, Identification and LC-

ED

MS/MS determination of acteoside, the main antioxidant compound of Euphrasia

(2011) 203–208.

PT

rostkoviana, using the isolated target analyte as external standard, J. Chromatogr. Sci. 49

CE

[32] P. Sarkhail, M. Nikan, P. Sarkheil, A.R. Gohari, Y. Ajani, R. Hosseini, A. Hadjiakhoondi, S. Saeidnia, Quantification of verbascoside in medicinal species of Phlomis and their genetic

AC

relationships, J. Pharm. Sci. 22 (2014) 1–9. [33] G. Cao, X. Wu, Q. Li, H. Cai, B. Cai, X. Zhu, Influence of processing procedure on the quality of Radix scrophulariae: A quantitative evaluation of the main compounds obtained by accelerated solvent extraction and high-performance liquid chromatography, J. Sep. Sci. 38 (2015) 390–394.

Page 19 of 22

ACCEPTED MANUSCRIPT Abbreviations

T

INERA: Institut National pour l‟Etude et la Recherche Agronomiques

RI P

HPLC: High Performance Liquid Chromatography DAD: Diode Array Detector

ICH: International Conference on Harmonization

NU

ANOVA: Analysis Of Variance

NMR: Nuclear Magnetic Resonance

ED

COSY: Correlation SpectroscopY

MA

SD: Standard Deviation RSD: Repeatability Standard Deviation

SC

CRS: Chemical Reference Substance

TOCSY: Total Correlation SpectroscopY

PT

HSQC: Heteronuclear Single Quantum Correlation HMBC: Heteronuclear Multiple Bond Correlation

CE

NOESY: Nuclear Overhauser Effect SpectroscopY

AC

HR-ESI-MS: High Resolution-Electrospray Ionization-Mass Spectrometry

Page 20 of 22

MA

NU

SC

RI P

T

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

Graphical abstract

Page 21 of 22

ACCEPTED MANUSCRIPT Tables

Table 2 Fingerprint of KILMA. Parameters

3 4.13 2.73 75.7 15.3 835.2

Peaks 4 5 5.69 6.55 6.09 2.89 34.4 339.8 7.4 53.8 624.2

6 7.45 3.43 74.4 13.4 624.2

7 9.51 7.43 48.9 14.8 462.0

8 13.21 15.1 69.4 17.7 476.0

AC

CE

PT

Retention (min) Resolutiona Peak height (mV) 33.1 Area (mV.min) 3.5 Mass (g/mol) a From the previous peak.

2 3.63 1.6 41.1 4.7 -

ED

1 3.45

MA

NU

SC

RI P

T

Table 1 Authentication of SIROP KILMA and plant extracts. Samples Peak retention Occurrence of peaks compared to the profile of KILMA (min) 7.6 10.4 11.4 12.2 16.6 17.0 24.3 KILMA + + + + + + + G. ternifolia + C. febrifuga + + + + + L. camara + + + + (+): Present; (-): absent.

Page 22 of 22