Search for Antiprotozoal Activity in Herbal Medicinal Preparations; New Natural Leads against Neglected Tropical Diseases

Sleeping sickness, Chagas disease, Leishmaniasis, and Malaria are infectious diseases caused by unicellular eukaryotic parasites (“protozoans”). The three first mentioned are classified as Neglected Tropical Diseases (NTDs) by the World Health Organization and together threaten more than one billion lives worldwide. Due to the lack of research interest and the high increase of resistance against the existing treatments, the search for effective and safe new therapies is urgently required. In view of the large tradition of natural products as sources against infectious diseases [1,2], the aim of the present study is to investigate the potential of legally approved and marketed herbal medicinal products (HMPs) as antiprotozoal agents. Fifty-eight extracts from 53 HMPs on the German market were tested by a Multiple-Target-Screening (MTS) against parasites of the genera Leishmania, Trypanosoma, and Plasmodium. Sixteen HMPs showed in vitro activity against at least one of the pathogens (IC50 < 10 µg/mL). Six extracts from preparations of Salvia, Valeriana, Hypericum, Silybum, Arnica, and Curcuma exhibited high activity (IC50 < 2.5 µg/mL). They were analytically characterized by UHPLC/ESI-QqTOF-MSMS and the activity-guided fractionation of the extracts with the aim to isolate and identify the active compounds is in progress.


Activity Screening of HMPs against Protozoan Parasites
A total number of 58 extracts were prepared from 53 HMPs and tested in vitro at two test concentrations, 2 and 10 µg/mL, for growth inhibitory (GI) activity against the following parasites: Trypanosoma brucei rhodesiense (Tbr), T. cruzi (Tc), Leishmania donovani (Ld), and Plasmodium falciparum (Pf), according to the procedures described in Section 3.3. The identity of the HMPs is depicted in Table A1, Appendix. The results of these biological assays are reported in Figure 1 (numerical data are reported in Table A2, Appendix). In case of samples displaying activity in this assay, IC50 values against the susceptible parasites as well as for cytotoxicity against mammalian cells (L6 rat skeletal myoblasts) were determined. The results of the IC50 determinations and the respective selectivity indices (SI = IC50 (L6)/IC50 (parasite)) are reported in Table 1. UHPLC-UV chromatograms and LC-MS data of the active extracts, i.e., their suggested dereplicated constituents, are depicted in the section Dereplication Data, Appendix.     Sixteen HMPs showed growth inhibition (GI) activity in vitro against at least one of the pathogens (>50% GI at 10 µg/mL), six extracts of which displayed high activity (IC50 < 2.5 µg/mL).
Generally, Plasmodium falciparum was found to be the most sensitive parasite to the tested HMPs and T. cruzi the least sensitive, against which activities hardly ever exceeded 45% of inhibition. Pf (10µg/mL) 91 In particular, promising results were obtained with Arnica montana and Salvia officinalis as the most active preparations against the etiologic agent of East African Human Trypanosomiasis (sleeping sickness). Moreover, the highest antimalarial activities were determined for the extracts of Curcuma longa, Silybum marianum, and Hypericum perforatum. It is noteworthy that the preparation of Valeriana officinalis showed antileishmanial activity with an IC50 value of 2.1 µg/mL and was the only preparation that also displayed moderate activity against T. cruzi. Beside these "first line samples", some preparations presenting moderate activity (IC50 values 2.5-6 µg/mL) and showing high SI values, i.e., ID_38, can still be considered interesting samples which will be subject to following studies.

Antitrypanosomal Activity
The extract of Arnica montana L. (Asteraceae) (IC50 = 1.12 µg/mL and SI = 10.86) was included in the study as a herbal positive-control, since the anti-trypansosomal activity of its main sesquiterpene lactone constituents was previously described by our group [8,9]. The tincture under study was analyzed by UHPLC/ESI-QqTOF-MSMS and found to contain the main sesquiterpene lactones of the helenalin type known from this plant (see Figure A6 and Table A8, Appendix). Thus, the positive result found with this Arnica tincture in the present research is in agreement with the strong antitrypanosomal activity of its major constituents.
With an IC50 value of 1.86 µg/mL against Tbr, the extract ID_39 of Salvia officinalis L. (Lamiaceae) appears to be a promising hit for further evaluation. The antitrypanosomal and antimalarial activity of other species of the genus Salvia have been previously reported, however, the majority of hitherto isolated and active compounds also showed nonselective toxicity [10][11][12][13][14]. Our tested extract presented a favorable value of SI = 17.3. Therefore, the fractionation of this extract and the evaluation of the resulting biologically active fractions appears interesting. In-depth studies with the aim to identify its antitrypanosomal constituent(s) based on multivariate data analyses in a similar manner, as it was recently described by Ellendorf et al. [15], and to target isolation of the relevant natural products are in progress.
The only preparation that showed activity against T. cruzi (IC50 = 5.86 µg/mL and SI = 7.9) was the ethanolic extract ID_26 of Valeriana officinalis L. (Caprifoliaceae-including all former Valerianaceae- [16]). To the best of our knowledge, no previous reports exist on any antitrypanosomal activity of this plant. Even though the biological values of this sample somewhat exceeded our general criteria range for promising activity (IC50 < 4 µg/mL and SI > 10), which is probably due to the generally lower sensitivity of this intracellular parasite to drugs, it appears interesting to study extract ID_26 further for the antitrypanosomal activity of its single constituents. Valerian is very widespread in Europe, Asia, and North America and is easy to cultivate, so it might become a promising source of new therapeutics against T. cruzi infections.

Antileishmanial Activity
The preparation ID_26 of Valeriana officinalis L. also displayed interesting antileishmanial activity. A much higher IC50 value (≈225 µg/mL) was previously reported for the chloroform extract of V. officinalis [17]. Several isolated constituents of V. wallichii were tested against Leishmania major and showed activity in a range (from lowest and highest value) but also showed unfavorable toxicity results [18].
In contrast to the published results, our extract displayed an IC50 value of 2.1 µg/mL against Ld and SI = 18, which constitutes grounds to select this preparation for further evaluation against this parasite.

Antiplasmodial Activity
Extract ID_14, obtained from a commercial preparation of turmeric, Curcuma longa L. (Zingiberaceae), showed promising antiplasmodial activity with an IC50 value of only 0.59 µg/mL, which is significantly lower than previously published data on this plant [19]. In the literature, the rhizomes of different species of Curcuma are mentioned to be used in traditional medicine, attributing the therapeutic properties largely to their polyphenolic curcuminoid constituents [19]. Specifically, the most abundant curcuminoid, curcumin, exhibited in vitro and in vivo antimalarial activity with IC50 values ranging from 1.84-3.5 µg/mL [20,21] and an absence of significant toxicity. Furthermore, Arteminisin (ART)-based combination therapies with curcumin have been investigated as a new hope for malaria therapy. Curcumin was found to synergize with ART but also to prime the immune system to protect against recrudescence in mice infected with Plasmodium berghei [22]. Based on the significantly lower IC50 value of the present extract compared to the published data, experiments are in progress to investigate whether curcumin/curcuminoids are the only active compounds of our extract or whether further synergistic contributions can come from other constituents of the total extract.
The sample ID_25, representing a preparation of Silymarin, a standardized mixture of closely related flavolignans obtained from milk thistle (Silybum marianum (L.) Gaertn. (Asteraceae)), showed promising antiplasmodial activity with IC50 = 2 µg/mL and SI = 27.3. Milk thistle is one of the oldest medicinal plants and is nowadays used for the treatment of liver damage, hepatitis, and cirrhosis. Moreover, Silybin dihemisuccinate, a derivative of silybin, is used as a clinical antidote for acute Amanita mushroom poisoning [23]. In spite of a large number of studies on the pharmacological activity of milk thistle existing in the literature, there is no information, to date, on the antiplasmodial activity of this plant or its constituents, apart from some tests carried out with the synthetized flavolignans 2,3-dehydrosilibinin and 8-(1;1)-DMA-kaempferide [24]. Since the IC50 value and SI found in our study were quite favorable and the therapeutic window of the extract is known to be wide, we consider S. marianum extract a promising and a useful hit to combat Pf. The isolation of the different flavolignans present in the extract and their biological activity tests are in progress.
The ethanolic extract ID_40 of Hypericum perforatum L. (Hypericaceae) showed high activity against Pf with an IC50 value of 2.27 µg/mL and with favorable selectivity values. Five phloroglucinol derivatives of H. erectum were reported to possess antiplasmodial activity [25], however, they were not found in our active extract (see Figure A5 and Table A7, Appendix). Hyperforin, which was present in our sample, and some derivatives were also reported active against Pf. For instance, the lithium salt of Hyperforin showed IC50 = 2.1 µM [26]. Thus, bioguided fractionation and biological evaluation of isolated compounds compared to the total extract are in progress in order to evaluate whether synergism or the additive effects of single constituents may account for the conspicuous activity of the total extract. Thus, we would justify the use of the complete extract as antiplasmodial instead of the single compounds.

Source Material and Reagents
The source material used for this study consists of a selected array of 53 approved or registered HMPs of liquid and solid dosage marketed in Germany and purchased from commercial sources (Table A1, Appendix). Solvents used for extraction procedures and LC-MS determinations were of reagent grade and HPLC grade, respectively.

Liquid Dosage Form Preparations
Liquid preparations containing low-polarity extracts were obtained by liquid/liquid extraction with low polarity solvents (CH2Cl2 or Ethyl acetate) and subsequently filtered over Na2SO4. Extractions were finally concentrated under vacuum using rotary evaporators at 40 °C for large volumes of solvent (>5 mL) or blow-dried with Nitrogen in case of small volumes. Purely aqueous preparations were directly freeze-dried using a lyophilizer.
Some preparations were submitted to two different procedures (methods A and B in Table 2) in order to get a higher range of possible active constituents and facilitate their identification in case of positive biological activity.

Solid Dosage Form Preparations
In the case of hard-shelled capsules and tablets, a number of five pieces or the contents of them were grinded to fine powder and then extracted with 40 mL of the selected solvent (see Table 3) on a magnetic stirrer for 10 min at room temperature. Extractions were vacuum-filtered and concentrated using a rotary evaporator. In case of soft gel capsules containing liquid or oily extracts, their contents was tested without previous extraction.
Samples displaying significantly more than 50% growth inhibition at 10 µg/mL and significant inhibition at 2 µg/mL were considered interesting for further study and, in such cases, the IC50 values against the susceptible parasites and cytotoxicity against L6 cells were determined.

IC50 Determination and Cytotoxicity Assays
The IC50 values of the selected samples yield an exact characterization of their effect against the corresponding parasites. The IC50 values were calculated from the sigmoidal growth inhibition curves resulting of a double determination at seven different concentrations [9]. Melarsoprol, benznidazole, miltefosine, and chloroquine were used as a reference drugs for Tbr, Tc, Ld, and Pf, respectively.
Cytotoxicity was assessed with a similar IC50 protocol using non-infected rat skeletal myoblasts (L6 cells) and Podophyllotoxin as a reference drug [9]. The selectivity index (SI) was calculated as the ratio between the IC50 of the L6 cells and IC50 of the tested parasites. Those samples that showed at least 10-fold selectivity for the parasites were considered interesting hits for following steps of the study.

Analytical Profiling of the Samples by UHPLC/ESI-QqTOF-MSMS
Analytical characterization of the metabolite profile of the samples was performed by UHPLC coupled with mass spectrometry to obtain detailed information of their constituent profile.
The extracts were injected at a concentration of 2 mg/mL. Chromatographic separations were performed on a Dionex Ultimate 3000 RS Liquid Chromatography System on a Dionex Acclaim RSLC 120, C18 column (2.1 × 100 mm, 2.2 µm) with a binary gradient (A: water with 0.1% formic acid; B: acetonitrile with 0.1% formic acid) at 0.8 mL/min: 0 to 0.2 min: isocratic at 5% B; 0.2 to 9.7 min: linear from 5% B to 100% B; 9.5 to 12.5 min: isocratic at 100% B; 12.5 to 12.6 min: linear from 100% B to 5% B; 12.6 to 15 min: isocratic at 5% B. The injection volume was 10 µL. Eluted compounds were detected using a Dionex Ultimate DAD-3000 RS over a wavelength range of 200-400 nm and a Bruker Daltonics micrOTOF-QII time-of-flight mass spectrometer equipped with an Apollo electrospray ionization source in positive mode at 4 Hz over a mass range of m/z 100-1000 using the following instrument settings: nebulizer gas nitrogen, 4 bar; dry gas nitrogen, 9 L/min, 220 °C; capillary voltage 4500 V; end plate offset −500 V; transfer time 100 µs; collision gas nitrogen; collision energy and collision RF (Radio Frequency) settings were combined to each single spectrum of 1250 summations as follows: 624 summations with 80 eV collision energy and 130 Vpp + 313 summations with 16 eV collision energy and 130 Vpp + 313 summations with 16 eV collision energy and 130 Vpp. Internal dataset calibration (HPC (High Precision Calibration) mode) was performed for each analysis using the mass spectrum of a 10 mM solution of sodium formiate in 50% isopropanol that was infused during LC re-equilibration using a divert valve equipped with a 20 µL sample loop.

Dereplication of Active Extracts
A partial dereplication of the active extracts was performed by comparison of the obtained UV chromatograms and mass spectra data with the literature and online databases. Moreover, peaks with known m/z from the literature were also searched using Extract Ion Chromatogram (EIC) monitoring. Nevertheless, some peaks could not be assigned unambiguously and it should be noted that for final assignments, the use of a complementary technique after isolation, such as NMR, would be necessary.

Conclusions
This screening of a comparatively small number of HMPs led to the discovery of six extracts, namely, from Curcuma longa L., Silybum marianum (L.) Gaertn, Valeriana officinalis L., Salvia officinalis L., Arnica montana L., and Hypericum perforatum L., which showed antiprotozoal activity against at least one of the tested parasites with promising IC50 values and a favorable SI, representing a rate of 10% of tested extracts. Seven extracts also showed moderate antiprotozoal activity with IC50 values ranging between 2.7 and 5.3 µg/mL. This considerable rate of active hits confirms the interesting potential of HMPs in the search for new antiprotozoal agents. Due to the urgent need of research and development of new therapies against protozoan NTDs and the high efforts necessary to develop new drugs de novo, further investigation of the antiprotozoal activity of these and other HMPs may lead to potentially new antiprotozoal compounds of natural origin and facilitate the development of new effective drugs from sustainable and inexpensive natural sources that have already been proven safe.
It should also be considered that natural products in complex mixtures such as plant extracts are often found to show additive or even synergistic effects [27], which could justify the use, in some cases, of the complete extract instead of isolated compounds. Deeper investigations in this direction with the preparations from Hypericum perforatum L. and Curcuma longa L. are in progress.

Acknowledgments
Financial support of this study by Apothekerstiftung Westfalen-Lippe is gratefully acknowledged. NLM receives financial support from Heinrich Böll Fundation in the form of a promotion-fellowship (Promotionsstipendium), which is most gratefully acknowledged. We thank J. Sendker, IPBP Münster, for performing the MS analyses and M. Cal and S. Keller (Swiss TPH) for performing the antiparasitic assays.
This project is an activity within the Research Network Natural Products against Neglected Diseases (ResNetNPND): http:// ResNetNPND.org/.

Author Contributions
N.L.M. performed the phytochemical part of the study and prepared the manuscript. M.K. and R.B. performed the biological assays. This study was initiated and coordinated by T.J.S., who supervised the chemical work and finalized the manuscript.

Conflicts of Interest
The authors declare no conflict of interest.    Tables 1  and 2).