In Vitro Biological Activity of Natural Products from the Endophytic Fungus Paraboeremia selaginellae against Toxoplasma gondii

Toxoplasma gondii is an apicomplexan pathogen able to infect a wide range of warm-blooded animals, including humans, leading to toxoplasmosis. Current treatments for toxoplasmosis are associated with severe side-effects and a lack efficacy to eradicate chronic infection. Thus, there is an urgent need for developing novel, highly efficient agents against toxoplasmosis with low toxicity. For decades, natural products have been a useful source of novel bioactive compounds for the treatment of infectious pathogens. In the present study, we isolated eight natural products from the crude extract of the endophytic fungus Paraboeremia selaginellae obtained from the leaves of the plant Philodendron monstera. The natural products were tested for inhibiting Toxoplasma gondii proliferation, and their cytotoxicity was evaluated in different human cell lines. Six natural products showed antitoxoplasma activity with low or no cytotoxicity in human cell lines. Together, these findings indicate that biphenyl ethers, bioxanthracenes, and 5S,6S-phomalactone from P. selaginellae are potential candidates for novel anti-toxoplasma drugs.


Introduction
Toxoplasma gondii is an obligate intracellular protozoan parasite member of the phylum Apicomplexa, which includes known human pathogens such as Plasmodium sp., Eimeria sp., Neospora, Babesia, Theileria, and Cryptosporidium spp., with which it shares significant biological similarities [1]. Beyond these organisms, the study of T. gondii has experimental advantages since its basic biology and the methodology for the genetic manipulation and quantification of its different stages are well established. Thus, T. gondii is considered a major model for the study of apicomplexan biology and for anti-apicomplexan drug target validation [2]. T. gondii infections are among the most common human zoonoses, leading to toxoplasmosis disease [3]. T. gondii is considered one of the world's most successful parasites due its ability to infect a wide range of warm-blooded vertebrate intermediate hosts [4]. T. gondii is estimated to chronically infect one-third of the world's human population and is acquired mainly through two ways: by ingesting oocysts shed from feline hosts (the definitive hosts) in contaminated food or water and by the consumption of raw or undercooked meat containing viable tissue cysts [5]. Waterborne and food-borne outbreaks of toxoplasmosis have been reported from countries with diverse cultural, social,

Isolation of Compounds from Paraboeremia selaginella
We isolated an endophytic fungus from fresh surface-sterilized leaves of the ornamental plant Philodendron monstera. The isolated strain was identified as Paraboeremia selaginella by the internal transcribed spacer (ITS) sequence with 99.56% identity in comparison with the ITS database of the National Center for Biotechnology Information. From the crude extract of a culture of Paraboeremia selaginella grown on solid rice medium, eight compounds were isolated by chromatographic methods and structurally elucidated by complementary spectroscopic analyses ( Figure 1). All eight compounds have previously been reported from other sources but are reported here for the first time as natural products occurring in the genus Paraboeremia.

Anti-T. gondii Activity
The eight natural products isolated from P. selaginellae were tested for anti-T. gondii activity. Interestingly, A-F showed activity against T. gondii growth, with IC 50 values of 5.75, 22.16, 27.22, 7.38, 17.99, and 5.13 µM, respectively (Table 1 and Figure 3). Therefore, we further explored the in vitro cytotoxicity of the natural compounds in different human cell lines.

Cytotoxicity Assays
First, we evaluated the cytotoxicity of compounds A to F in an MTT assay against Hs27 human fibroblasts (same cell type used for the T. gondii proliferation assay). The results of the MTT assay are shown in Figure 4 and Table 2. A-E showed no cytotoxicity at 100 µM against Hs27 cells. Only F showed moderate cytotoxicity with a cytotoxic concentration CC 50 = 81 ± 2.16 µM.
These compounds also were tested against the THP-1, Huh-7, and Hek 293 cell lines in a Resazurin assay. The mean IC 50 values of the Resazurin assay are shown in Table 3. While compounds A-C showed no cytotoxic effect in concentrations < 100 µM against any of the tested cell lines, D had only a weak cytotoxic activity against the Hek 293 cell line with an IC 50 of 93.8 µM. E showed moderate cytotoxic activity against all of the three tested cell lines and thus was the most cytotoxic of the tested compounds. F showed no or weak cytotoxic effects against the Huh-7 and Hek-293 cell lines. The cytotoxic effect against the THP-1 cell line was higher, with an IC 50 of 24.3 µM. The graphs for the Resazurin assay are shown in Figure 5.

Determination of Anti-Bacterial Activity
In our ongoing research for antibacterial and particularly antitubercular compounds, A to H were also tested in a minimal inhibitory concentration assay against S. aureus ATCC 700699, P. aeruginosa ATCC 87110, and M. tuberculosis H37Rv. Compounds A to H had no inhibitory effect on S. aureus ATCC 700699 and P. aeruginosa ATCC 87110, except for compound F, which showed a weak inhibitory effect on P. aeruginosa ATCC 87110 with an MIC 90 of 100 µM. Compounds A, B, C, F, G, and H had no inhibitory effect on the growth of M. tuberculosis, and compounds D and E showed a weak inhibitory effect with an MIC 90 of 50 and 100 µM, respectively. The results are shown in Table 4, highlighting that compounds A to F had a specific anti-toxoplasma effect and were devoid of broad, unspecific antimicrobial activity.

Discussion
Natural products have played an important role in the history of drug discovery for infectious disease. In the quest for new anti-T. gondii drugs, natural products have been proven to exhibit high potential for the discovery and development of new lead compounds with strong anti-T. gondii activity [25,26]. In this study, we isolated eight natural products from the crude extract of the endophytic fungus P. selaginellae. A previous report on the inhibitory activity of one of these compounds (phomalactone, F) against the apicomplexan parasite Plasmodium falciparum with an IC 50 of 84.32 µM [27] prompted us to test the natural products for anti-T. gondii activity. Interestingly, six compounds showed activity against T. gondii proliferation with no or low cytotoxicity in different human cell lines and no or low antibacterial activity against a gram-positive, a gram-negative, and a mycobacterial representative, revealing reasonable anti-T. gondii specificity and promising therapeutic windows. These results establish diphenyl ethers, bioxanthracenes, and lactones from P. selaginellae as potential candidates for further preclinical development of novel antitoxoplasma therapeutics.
Some of the isolated compounds share similar structural elements, which give insights into a structure-activity relationship of the natural products against the tested T. gondii strain ME49. Compounds A, B, and C are biphenyl ether derivatives that differ either in the position of the methoxy group or in the number of substituted hydroxyl groups. The most potent of these compounds is A (IC 50 = 5.75 µM), followed by B (IC 50 = 19.35 µM) and C (IC 50 = 27.22 µM). While A only differs from B by a switch in the position of the methoxylated hydroxyl group from position 2 to 4, it differs from C only by an additional hydroxyl group in position 2 , which it shares with B. Because of the higher potency of A in the toxoplasma proliferation assay compared to B and C, the position of the methoxy group in 4 and the amount and position of hydroxyl groups in 2 and 3 both are likely to have an influence on the antitoxoplasma activity of these derivatives. This suggestion, nevertheless, needs further experimental evidence. Furthermore, diphenyl ethers A, B, and C are structurally related to triclosan, a well-known broad spectrum antifungal and antibacterial agent targeting lipid synthesis [28]. It has been shown that triclosan also inhibits the growth of apicomplexans by inhibition of the enoyl reductase ENR (FabI) enzyme, the second reductive step in the type II fatty acid biosynthesis pathway. Nevertheless, due the poor solubility of triclosan, there is considerable interest in finding novel potent triclosan analogs with improved properties such as solubility, activity, and toxicity [29,30]. The mechanism of action of A, B, and C may be similar to that of triclosan, but further studies are necessary to explore and confirm their mode of action and cellular target. Furthermore, in vitro and in vivo pharmacokinetic characterization is needed to reveal whether any of the compounds reported here has superior properties compared to triclosan.
Compounds D and E represent bioxanthracenes belonging to the ES-242 class and share the same structure, differing only in position 4 by the hydroxyl group that is present only in E. The IC 50 values in the toxoplasma proliferation assay were 7.38 µM and 17.99 µM for D and E, respectively, suggesting a reduction in the antitoxoplasma activity if position 4 is substituted by a hydroxyl group. The bioxanthracenes D and E were previously isolated from Verticillium spp. and are well-known to act as N-methyl-D-aspartate receptor antagonists [31]. Both compounds were also found to be active against the apicomplexan parasite P. falciparum with IC 50 values of 8.44 and 13.22 µM, respectively [32]. Interestingly, the activities of D and E against T. gondii in this study were comparable to their reported activity in P. falciparum with IC 50 values of 7.38 µM and 17.99 µM (see Table 1). Nevertheless, the mechanism of action of D and E on apicomplexans is still unknown and is probably independent from their activity as NMDA receptor antagonists [32].
Compounds F and G are small δ-lactonic molecules; 5S,6S-phomalactone (F) differs from methyltriaceticlactone (G) in the length of the sidechain in position 6, the hydroxyl group in position 5, and in the absence of the methyl group that is present in methyltriaceticlactone in position 2. Interestingly, antitoxoplasmal activity was observed for F, but not for G, suggesting that one or more of these structural differences and not only the presence of the δ-lactonic base structure plays a crucial role in the bioactivity against T. gondii. Phomalactone (F) is a frequent fungal metabolite and was first isolated from the plant-pathogenic fungus Nigrospora sp. [16]. It has a wide range of activities such as antifungal, immunomodulating, insecticide, nematocidal, and phytotoxic activity [15,19,[33][34][35]. In addition, it has been found to be active against the apicomplexan parasite P. falciparum, with an IC 50 of 84.32 µM [27]. In the present study, we tested F for inhibition of T. gondii proliferation and, interestingly, it showed a more potent activity with an IC 50 of 5.13 µM (see Table 1). No target or mode of action has been suggested for phomalactone in P. falciparum, and the target of this compound in T. gondii also remains elusive and has to be determined in the future. Importantly, the newly identified natural products with inhibitory activity against T. gondii showed very little in vitro toxicity and should be evaluated in in vivo infection model systems in the future. In general, this study highlights the potential of endophytic fungi as a promising source for novel antitoxoplasma compounds.

General Experimental Procedures
Optical rotations were measured on a Jasco P-2000 polarimeter (Jasco, Pfungstadt, Germany). UV-spectra were obtained by the use of a Dionex P580 system in combination with a diode array detector (UVD340S) and an Eurosphere 10 C18 column (125 mm × 4 mm). ECD spectra were measured on a JASCO J-810 spectropolarimeter. VCD spectra were recorded on a BioTools Chiral-IR-2X at a resolution of 4 cm −1 under ambient temperature for 18 × 3000 scans. Samples were dissolved in CDCl 3 , and the solution was placed in a 100 µm BaF 2 cell. 1D and 2D NMR spectra were recorded on a Bruker Avance III ( 1 H, 600 MHz; 13 C 150 MHz) spectrometer. Mass spectra were measured on a Finnigan LCQ Deca (Thermo Quest, Egelsbach, Germany) mass spectrometer and for HRESIMS, on a UHR-QTOF maXis 4G (Bruker Daltonics, Bremen, Germany) mass spectrometer. Semipreperative HPLC was performed on a Lachrom-Merck Hitachi system (pump L7100, UV-detector L7400, Eurospher 100 C18 column 300 mm × 8 mm, Knauer, Everswinkel, Germany). VLC and non-vacuum-column chromatography were accomplished using Macherey Nagel silica gel 60M (0.04-0.063 mm). Precoated TLC silica gel 60 F254 plates (Merck, Darmstadt, Germany) were used for tracking separation using detection under UV light at 254 and 365 nm wavelengths or spraying anisaldehyde-sulfuric acid reagent. Sephadex LH20 (GE Healthcare Bio.Sciences AB, Uppsala, Sweden) was used as a stationary phase for column chromatography. The measurement of optical rotations was accomplished by using spectral grade solvents.

Fungal Material
The fungus was obtained from the leaves of the plant Philodendron monstera as an endophyte. A single leaf was surface sterilized by soaking it with 70% ethanol for 30 s and letting it dry under sterile conditions. With a heat-sterilized scalpel, the leaf was cut into pieces and put onto a YPD agar plate, which was enriched with 100 mg/L chloramphenicol to suppress bacterial growth. After seven days of incubation at room temperature, distinct fungal growth was observed on the plate. A 1 cm 2 piece of the fungus was cut out of the agar medium using a heat-sterilized scalpel under sterile conditions and was transferred onto a new sterile YPD agar plate to isolate a pure organism. The isolated strain was identified as Paraboeremia selaginella by the internal transcribed spacer (ITS) sequence with 99.56% identity in comparison with the ITS database of the National Center for Biotechnology Information (GenBank Accession ON231784).

Fermentation and Extraction
The fungus was fermented on solid rice medium. Ten Erlenmeyer flasks were used; 100 g of rice and 100 mL of demineralized water were added to each flask and autoclaved at 121 • C for 15 min. Under sterile conditions, 1 cm 2 of fungal material was cut out of an agar plate using a sterile scalpel and transferred onto the autoclaved rice medium. The fungus was grown for 4 weeks under static conditions at room temperature. Each flask was soaked with 250 mL of ethylacetate for at least 12 h. The rice medium was then cut into small pieces and shaken for 8 h at 150 rpm. The liquid crude extract was filtrated into round flasks and evaporated using a rotary evaporator to yield 14.66 g of dry crude extract.

Preparation of Compounds for T. gondii Proliferation Assay
The purified natural products A-F and pyrimethamine [36] were dissolved in DMSO as 10 mM stocks and stored at −20 • C. The compounds were diluted in Iscove's Modified Dulbecco's medium (Gibco-Thermo Fisher Scientific, Braunschweig, Germany) immediately prior to use.

Parasites and Cell Culture for T. gondii Proliferation Assay
T. gondii ME49 tachyzoites (ATCC/LGC Standards GmbH, Wesel, Germany) were cultured in human foreskin fibroblast Hs27 cells (ATCC/LGC Standards GmbH, Wesel, Germany) as host cells as described previously [37]. The cells were maintained in Iscove's modified Dulbecco's medium (Gibco-Thermo Fisher Scientific, Braunschweig, Germany) supplemented with 10% heat-inactivated fetal bovine serum (Invitrogen, Karlsruhe, Germany) and 50 mM 2-mercaptoethanol (Gibco-Thermo Fisher Scientific, Braunschweig, Germany) and were grown in a humidified incubator at 37 • C with 5% CO 2 in air atmosphere. For toxoplasma propagation, 25 cm 2 cell culture flasks, containing a confluent monolayer of Hs27 cells, were infected with 5 × 10 6 T. gondii tachyzoites after medium change. After three days, the supernatant of the cell culture containing parasites was harvested and transferred to a 15 mL centrifuge tube and centrifuged at 700 rpm for five minutes and resuspended in cell culture medium. The number of parasites was counted using a hemocytometer.

T. gondii Proliferation Assay
Microtiter plates (96-well) with a final volume of 200 µL per well were used for the assay. Hs27 fibroblast monolayers were infected with 3 × 10 4 freshly harvested tachyzoites per well (MOI = 1) and incubated for 48 h at 37 • C, after which various concentrations of the tested compounds (0.04, 0.09, 0.19, 0.39, 0.78, 1.5, 3.12, 6.25, 12.5, 25, 50 µM) in culture medium were added to the cells. Pyrimethamine (0.007, 0.01, 0.03, 0.06, 0.125, 0.25, 0.5, 1 µM) was added under identical conditions as a positive drug control [37]. Hs27 cells were pre-stimulated for 24 h with IFNγ (300 U/mL) and infected with T. gondii cells without further treatment as the growth inhibition control. After 48 h, proliferating toxoplasma parasites were radioactively labelled with tritiated uracil (5 mCi, Hartmann Analytic, Braunschweig, Germany) and diluted 1:30 (10 µL per 200 µL total culture volume per well) in order to determine parasite proliferation [38]. After 28-30 h, the microtiter plates were frozen at −20 • C. To evaluate the assay, the microtiter plates were thawed at room temperature. Cells were transferred to glass-fiber filters (Printed Filtermat A 102 mm × 258 mm, PerkinElmer, Waltham, MA, USA) using a cell harvester (Basic96 Harvester, Zinsser Analytic, Skatron Instruments, Northridge, CA, USA). The filters were dried for 20 min at 130 • C in a drying cabinet and were then soaked in 10 mL of scintillation fluid (Betaplate Scint, PerkinElmer, Waltham, MA, USA) and shrink-wrapped in plastic covers (Sample Bag for Betaplate, PerkinElmer, Waltham, MA, USA). The filters were then clamped in cassettes and evaluated using a beta-counter device (Betaplate Liquid Scintillation Counter 1205, LKB-WALLAK, Melbourne, Australia) to measure the Cherenkov radiation, which refers to the amount of incorporation of tritiated uracil into the RNA of T. gondii. IC 50 values, the concentration of inhibitors necessary to inhibit the growth of tachyzoites by 50%, were determined for each experiment with the use of Prism GraphPad version 9.2.0 software.

Cell Viability Assay against Hs27 Cells
The 3-[4,5-dimethylthiazole-2-yl]-2,5-diphenyltetrazolium bromide (MTT) test was used to assess cell viability of the isolated active compounds against Hs27 cells. The MTT assay is a colorimetric reaction based on the enzymatic reduction of MTT to MTT-formazan, which is catalyzed by mitochondrial succinate dehydrogenase [39].

Determination of the Minimal Inhibitory Concentration against Different Pathogenic Bacteria
Testing for antibacterial activity was done as described previously [37]. Briefly, a single colony of Methicillin-resistant Staphylococcus aureus (MRSA strain Mu50, ATCC 700699) or Pseudomonas aeruginosa (strain PAO1, ATCC 87110) were grown in Mueller-Hinton broth (MHB) at 37 • C shaking at 120 rpm to reach an optical density of approx. 0.4. The cell suspension was adjusted to 10 6 CFU/mL, of which 50 µL was seeded into a prepared 96-well polystyrene round-bottom plate containing test compounds diluted in MHB in a 1:1 serial dilution ranging from 100 µM to 0.78 µM. The plates were incubated at 37 • C statically for 24 h, and readout was performed using the BacTiter Glo assay (Promega) following the manufacturer's instructions. Briefly, BacTiter Glo reagent was added to a white flat-bottom 96-well plate, and an equal volume of bacteria suspension was added to each well and mixed carefully. After 5 min, the luminescence was measured using a TECAN plate reader. The growth was calculated in regard to the vehicle (DMSO) and sterile control. Moxifloxacin and cefuroxime were used as a positive and negative control, respectively. All compounds were tested in triplicate.
For the testing against M. tuberculosis H37Rv, the Minimal Inhibitory Concentration (MIC) was determined in 96-well microtiter plates containing a total volume of 100 µL employing a resazurin reduction assay [41]. Briefly, a 96-well plate was prepared containing 7H9 medium supplemented with 10% ADS (0.81% NaCl, 5% BSA, 2% dextrose), 0.5% glycerol, and 0.05% tyloxapol. Compounds were two-fold serially diluted with the highest tested concentration of 100 µM. A M. tuberculosis culture was pre-grown to an OD 600 nm of approx. 0.4-0.6 by shaking at 37 • C in PETG square bottles (ThermoFisher Scientific, Braunschweig, Germany) containing 10 mL supplemented 7H9 medium. The cell density was adjusted to an OD 600 nm of 0.08 (10 6 CFU/mL, and 5 × 10 4 CFU were added to each well). Rifampicin and DMSO were used as a positive and solvent control, respectively. The 96-well plates were incubated for 5 days at 37 • C and 5% CO 2 in humidified atmosphere. Afterwards, 10 µL of a 100 mg/mL resazurin solution was added to each well and resuspended carefully. After another 24 h at room temperature, the cells were fixed by adding 100 µL of a 10% formalin solution to each well. The readout was performed using a TECAN plate reader at 535 nm excitation and 590 nm emission. The growth was calculated in relation to the solvent control being 100% growth. The experiment was performed in triplicate.

Cytotoxicity Assay against Different Human Cell Lines
The cytotoxicity study was carried out using the THP-1 (human monocytic leukemia cell line), Huh-7 (Human liver carcinoma cell line), and HEK293 (human embryonic kidney cell line) cell lines as described before [37]. The THP-1 cells were cultured using RPMI 1640 medium containing 2 mM L-glutamine and supplemented with 10% fetal calf serum (FCS) and 1% sodium pyruvate. Huh-7 cells were cultured using a 1:1 mixture of RPMI 1640 medium containing 2 mM L-glutamine and 10% FCS medium and DMEM containing 10% FCS and 1% sodium pyruvate. The HEK-293 cells were cultured with DMEM including 2 mM L-glutamine and supplemented with 1% NE amino acids, 1% 1.0 mM sodium pyruvate and 10% FCS. All three cell lines were then incubated at 37 • C in an atmosphere of 5% CO 2 under humid conditions for 2 weeks while renewing the medium twice weekly. Subsequently, the cells were suspended and adjusted to a density of 2 × 10 5 cells/mL. In a 96-well flat-bottom microtiter plate, the cells were adjusted to a total volume of 100 µL containing 2-fold serial dilutions of the tested compounds A-F ranging from 100 to 1.56 µM. Cycloheximide (4, 2, 1, 0.5, 0.25, 0.13, 0.06, 0.03 µg/mL) was used as a positive control. After an incubation time of 48 h at 37 • C in an atmosphere of 5% CO 2 under humid conditions, 10 µL resazurin solution (100 µg/mL) was added to each well and incubated for another 4 h. The fluorescence was then quantified using a Tecan Infinite 200pro microplate reader (excitation 540 nm, emission 590 nm). The residual growth was calculated relative to non-inoculated conditions (0% growth) and controls treated with DMSO (100% growth).