Phytochemical Study and Antiglioblastoma Activity Assessment of Plectranthus hadiensis (Forssk.) Schweinf. ex Sprenger var. hadiensis Stems

Glioblastoma (GB) is the most malignant form of primary astrocytoma, accounting for more than 60% of all brain tumors in adults. Nowadays, due to the development of multidrug resistance causing relapses to the current treatments and the development of severe side effects resulting in reduced survival rates, new therapeutic approaches are needed. The genus Plectranthus belongs to the Lamiaceae family and is known to be rich in abietane-type diterpenes, which possess antitumor activity. Specifically, P. hadiensis (Forssk.) Schweinf. ex Sprenger has been documented for the use against brain tumors. Therefore, the aim of this work was to perform the bioguided isolation of compounds from the acetonic extract of P. hadiensis stems and to investigate the in vitro antiglioblastoma activity of the extract and its isolated constituents. After extraction, six fractions were obtained from the acetonic extract of P. hadiensis stems. In a preliminary biological screening, the fractions V and III showed the highest antioxidant and antimicrobial activities. None of the fractions were toxic in the Artemia salina assay. We obtained different abietane-type diterpenes such as 7α-acetoxy-6β-hydroxyroyleanone (Roy) and 6β,7β-dihydroxyroyleanone (DiRoy), which was also in agreement with the HPLC-DAD profile of the extract. Furthermore, the antiproliferative activity was assessed in a glioma tumor cell line panel by the Alamar blue assay. After 48 h treatment, Roy exerted strong antiproliferative/cytotoxic effects against tumor cells with low IC50 values among the different cell lines. Finally, we synthesized a new fluorescence derivative in this study to evaluate the biodistribution of Roy. The uptake of BODIPY-7α-acetoxy-6β-hydroxyroyleanone by GB cells was associated with increased intracellular fluorescence, supporting the antiproliferative effects of Roy. In conclusion, Roy is a promising natural compound that may serve as a lead compound for further derivatization to develop future therapeutic strategies against GB.


Introduction
Cancer is one of the main causes of morbidity and mortality worldwide. The World Health Organization (WHO) reported approximately 14.1 million new cases in 2012 and 8.2 million deaths. These figures are expected to increase due to population growth and aging, with an estimate of more than 20 million new cancer cases diagnosed per year and 13 million deaths by 2030 [1,2].
Among the different cancers, brain and central nervous system (CNS) tumors, including glioblastoma (GB), are the 12th most common tumor type according to the WHO report in 2020. It is the most frequent and malignant tumor of the CNS affecting mainly adults at the age of 40-60 [3,4].
Current treatments available for GB comprise surgical resection followed by concomitant chemoradiotherapy plus chemotherapy (temozolomide, TMZ) and the use of biological drugs, such as the therapeutic antibody bevacizumab. However, these treatments are associated with side effects and multidrug resistance (MDR), eventually yielding relapses. In fact, GB has a median survival rate of 15-18 months after diagnosis and a 5% chance of survival in the 5 years postdiagnosis. Therefore, new therapeutic approaches are needed to overcome these problems [3,5,6].
Medicinal plants are a rich source of bioactive natural products with a high structural diversity and pleiotropic effects. Therefore, they represent a potential source of anticancer lead compounds. The screening, isolation, and identification of secondary metabolites from medicinal plants have served as scaffolds for the development of derivatives with optimized potency [7]. Compared to synthetic compound libraries, naturally occurring molecules typically have higher molecular masses, a large number of sp 3 carbons and oxygen atoms, but fewer nitrogen and halogen atoms, higher numbers of H-bond acceptors and donors, lower partition coefficients, and a greater molecular rigidity. All these characteristics make natural products valuable in drug discovery for their ability to tackle protein-protein interactions; moreover, they represent the major source of oral drugs beyond Lipinski's rule of five [8].
Some of the most promising plant-derived anticancer compounds include diterpenes, such as paclitaxel and docetaxel [7]. Due to the importance of terpenes as anticancer drugs, the study of botanical families rich in these compounds is a promising approach to find new bioactive compounds against GB.
The genus Plectranthus belongs to the Lamiaceae family and consists of 300 species distributed in Africa, Asia, and Australia. The most frequently cited traditional use of Plectranthus species is related to their medicinal properties [9].
The first Plectranthus species to be described was P. hadiensis (Forssk.) Schweinf. ex Sprenger in 1775. However, it is botanically still poorly defined, resulting in identification problems until today [10][11][12]. In order to resolve these issues, some investigations have attempted to differentiate P. hadiensis varieties and its counterfeits by using molecular biology techniques [13,14].
In fact, the World Flora Online database suggests more than 30 possible names referring to this species, e.g., P. ramosior (Benth.) van Jaarsv., P. cyaneus Gürke, Ocimum hadiense Forssk. or Coleus zeylanicus (Benth.) L.H.Cramer which is considered as a variety of P. madagascariensis (P. madagascariensis var. ramosior Benth.). P. hadiensis itself also has two varieties, i.e., var. hadiensis and var. tomentosus [12,15]. A recent text-mining study that applied the novel degrees of publication (DoP) method identified P. hadiensis as an understudied species [16]. It was, therefore, claimed as being interesting for future laboratory studies, especially for pharmacological bioassays, isolation procedures, and drug discovery strategies.
Among all phytochemical classes, diterpenes are the most prominent biologically active group of secondary metabolites present in this genus, whose presence is considered as a chemotaxonomic marker [38]. In particular, abietane diterpenes have attracted attention regarding their biological activities, namely because of their antiproliferative, antitumoral, and/or cytotoxic potential [39,40].
Concerning its antitumoral properties, this plant has been described to be useful for the treatment of abdominal, brain, breast, intestinal, prostate, skin, throat, and uterine cancers as well as of leukemia [37]. Methanolic and ethanolic P. hadiensis shoot extracts revealed cytotoxicity towards brine shrimp [20,22], and stem extracts were tested against HeLa cells [26]. The terpenoid fraction (monoterpenes and sesquiterpenes) isolated from the shoot had antiproliferative activity against human colon cancer cells (HCT-15) but was only minimally cytotoxic towards noncancerous normal cells. The upregulation of caspase-3 activity and proapoptotic BAX and the downregulation of antiapoptotic BCL-2 and COX-2 indicated apoptosis induction by the mitochondria-dependent pathway [17].
Regarding the antitumoral effects of different abietane diterpenes present in Plectranthus species, recent results showed that abietane diterpenoids DiRoy (3) and Roy (1) induced apoptosis in glioma cell lines. These compounds exhibited antitumoral effects in primary H7PX glioma cells by inducing DNA double-strand breaks, G2/M cell cycle arrest, and apoptosis as indicated by increased levels of phosphorylated γ-H2AX, decreased mitochondrial membrane potential, and altered expression of pro-and antiapoptotic genes (BAX, BCL2, TP53, and CAS3) [43]. Furthermore, Roy (1) and DiRoy (3) were cytotoxic against human CCRF-CEM leukemia cells and A549 lung adenocarcinoma cells by altering pro-and antiapoptotic gene expression and induction of apoptosis [44]. A recent study showed that both the acetonic extract of P. hadiensis var. hadiensis leaves and its major compound, Roy (1), were active against the aggressive triple-negative breast cancer cell line, MDA-MB-231S, with IC 50 values of 25.6 µg/mL and 2.15 µg/mL, respectively [45].
Having the remarkable anticancer activity of this plant in mind, we performed bioactivity-guided isolation of the acetonic extract of P. hadiensis var. hadiensis. stems, quantified the isolated compounds, and assessed their antiproliferative/cytotoxic effects in GB cell lines. genes (BAX, BCL2, TP53, and CAS3) [43]. Furthermore, Roy (1) and DiRoy (3) were cytotoxic against human CCRF-CEM leukemia cells and A549 lung adenocarcinoma cells by altering pro-and antiapoptotic gene expression and induction of apoptosis [44]. A recent study showed that both the acetonic extract of P. hadiensis var. hadiensis leaves and its major compound, Roy (1), were active against the aggressive triple-negative breast cancer cell line, MDA-MB-231S, with IC50 values of 25.6 μg/mL and 2.15 μg/mL, respectively [45].

Results and Discussion
In the present investigation, we focused on the acetonic stem extract of P. hadiensis since results from leaf extracts have been previously published [45].

Results and Discussion
In the present investigation, we focused on the acetonic stem extract of P. hadiensis since results from leaf extracts have been previously published [45].

Phytochemical Study of Plectranthus hadiensis Stems
P. hadiensis stems (1.507 kg) were extracted with acetone five times until exhaustion using an ultrasound-assisted extraction method. At the end, a residue of 46.07 g was obtained resulting in a final extraction yield of 2.96% (weight of extract (g)/weight of dry plant (g)). After extraction, several fractionations were carried out by dry flash column chromatography yielding six fractions (I-VI). Subsequently, a qualitative analysis of the fraction composition of the acetonic leaf extract [45], acetonic stem extract, and fractions I-VI from the stems was carried out by using high-performance liquid chromatography coupled to a diode-array detector (HPLC-DAD). The samples were either injected alone or authentic standards (Roy and DiRoy) known to be present in the plant were coinjected [41].
Compounds from fractions III and IV were isolated using liquid column chromatography and preparative thin-layer chromatography (PTLC). In total, six compounds were isolated but due to the scarce quantities only two of them, i.e., Roy (1) and DiRoy (3), could have been characterized and quantified by NMR, FTIR, melting point and UV spectroscopy (Supplementary Materials).
The HPLC-DAD quantification of the major compounds in P. hadiensis leaves, stems, and stem fractions was also performed based on the calibration curves for Roy (1) and DiRoy (3) ( Table 1). Table 1. Quantitative analysis of compounds (1) and (3) in Plectranthus hadiensis extracts and fractions using HPLC-DAD.

Compounds
Linear The quantitative analysis by HPLC-DAD (Table 2) showed a huge difference in the amounts of Roy (1) between the extract of the leaves and the extract of the stems (5.37 mg/g versus 0.40 mg/g) which was 13.4 times higher for the latter. The extract from the stems contained double the amount of DiRoy (3) compared to the leaves (2.15 mg/g versus 1.12 mg/g). Roy (1) and DiRoy (3) were both present in fractions II-V. The fraction richest in Roy (1) was fraction V (75.68 mg/g), while the richest in DiRoy (3) was fraction III (50.33 mg/g). Fractions III and V showed the highest antioxidant and antimicrobial activity (see Section 3.1). Fraction I only contained small quantities of Roy (1) (1.03 mg/g), while none of the compounds were detected in fraction VI. All fractions were subjected to antioxidant, general toxicity, and antimicrobial assays (Section 2.3.1).

Studies on the Antioxidant and Antimicrobial Activity and Toxicity of the Fractions
Fractions III and V showed the highest antioxidant and antimicrobial activities, which could be explained by the presence of Roy (1) and DiRoy (3) (Section 2.1). The antimicrobial activities of fractions III and V were even higher than those of the positive control. None of the samples showed general toxicity in the Artemia salina assay (Table 3; Figure 4).  The synthesis of the BODIPY derivative (13) started with the base-catalyzed reaction of succinic anhydride (15) with ortho-hydroxymethylphenyl-BODIPY (14) [48]. Having BODIPY (13) successfully prepared, the conjugated BODIPY-abietane (12) was obtained using a regioselective coupling mediated by EDC (1-ethyl-3-(3-dimethylaminopropyl) carbodiimide) and 4-DMAP (4-dimethylaminopyridine).

Studies on the Antioxidant and Antimicrobial Activity and Toxicity of the Fractions
Fractions III and V showed the highest antioxidant and antimicrobial activities, which could be explained by the presence of Roy (1) and DiRoy (3) (Section 2.1). The antimicrobial activities of fractions III and V were even higher than those of the positive control. None of the samples showed general toxicity in the Artemia salina assay (Table 3; Figure 4).

Assessment of the Cytotoxic Effect of Plectranthus hadiensis var. hadiensis Extracts and Phytochemicals towards GB Cell Lines
The cytotoxic/antiproliferative potential of P. hadiensis extracts and phytochemicals, i.e., the extracts from stems and leaves, fractions of the extract (EI, EII, EIII, EIV, EV and EVI), and isolated compounds (Roy (1) and DiRoy (3)) was assessed using the Alamar blue assay. A panel of five glioblastoma cell lines was treated for 24, 48, and 72 h with a serial dilution range of concentrations (100, 50, 25, 6.25, 1.5625, and 0.390625 μg/mL) of these P. hadiensis products. Even at the lowest concentrations, Roy (1) induced the most pronounced antiproliferative/cytotoxic effect across all cell lines ( Figure S15, Supplementary Materials). In fact, DiRoy (3) did not significantly affect cell viability once the proliferative capability of cells dropped below 50% relative to the control, except for the results in the U118 cell line (most sensitive), in which treatment with the highest concentrations led to a decrease of 60% in cell viability. The significant difference in the bioactivity of Roy (1) and DiRoy (3) may be related to the chemical structure of these compounds and their physicochemical properties, such as their molecule polarity. Based on the calculated Log P of the two structures (1.08 for Roy (1) and 0.85 for DiRoy (3)), it is reasonable to hypothesize that Roy (1) could pass the membrane, and reach its target, more easily than DiRoy (3), even if the exact mechanism of action of these molecules in glioblastoma cell lines and their organelles is currently under study. Nonetheless, the treatment with fraction EII and EV exhibited a greater cytotoxic effect against GB cells as these fractions have a higher quantity of DiRoy (3) in their chemical composition. Furthermore, treatment with both P. hadiensis extracts obtained either from the stems or leaves did not affect the proliferative capacity of brain tumor cells with the exception of the highest concentration (100 μg/mL), which was probably due to the mixture of compounds that masks the bioactivity of certain compounds alone such as Roy (1).
Therefore, all results indicate that Roy (1) is the most promising phytochemical to be further explored and as a lead compound to further develop new therapeutic strategies against GB.

Assessment of the Cytotoxic Effect of Plectranthus hadiensis var. hadiensis Extracts and Phytochemicals towards GB Cell Lines
The cytotoxic/antiproliferative potential of P. hadiensis extracts and phytochemicals, i.e., the extracts from stems and leaves, fractions of the extract (I, II, III, IV, V and VI), and isolated compounds (Roy (1) and DiRoy (3)) was assessed using the Alamar blue assay. A panel of five glioblastoma cell lines was treated for 24, 48, and 72 h with a serial dilution range of concentrations (100, 50, 25, 6.25, 1.5625, and 0.390625 µg/mL) of these P. hadiensis products. Even at the lowest concentrations, Roy (1) induced the most pronounced antiproliferative/cytotoxic effect across all cell lines ( Figure S15, Supplementary Materials). In fact, DiRoy (3) did not significantly affect cell viability once the proliferative capability of cells dropped below 50% relative to the control, except for the results in the U118 cell line (most sensitive), in which treatment with the highest concentrations led to a decrease of 60% in cell viability. The significant difference in the bioactivity of Roy (1) and DiRoy (3) may be related to the chemical structure of these compounds and their physicochemical properties, such as their molecule polarity. Based on the calculated Log P of the two structures (1.08 for Roy (1) and 0.85 for DiRoy (3)), it is reasonable to hypothesize that Roy (1) could pass the membrane, and reach its target, more easily than DiRoy (3), even if the exact mechanism of action of these molecules in glioblastoma cell lines and their organelles is currently under study. Nonetheless, the treatment with fraction II and V exhibited a greater cytotoxic effect against GB cells as these fractions have a higher quantity of DiRoy (3) in their chemical composition. Furthermore, treatment with both P. hadiensis extracts obtained either from the stems or leaves did not affect the proliferative capacity of brain tumor cells with the exception of the highest concentration (100 µg/mL), which was probably due to the mixture of compounds that masks the bioactivity of certain compounds alone such as Roy (1).
Therefore, all results indicate that Roy (1) is the most promising phytochemical to be further explored and as a lead compound to further develop new therapeutic strategies against GB.  (1), which had the highest antiproliferative/cytotoxic effect against GB cells, was conjugated with a fluorescent probe to assess its biodistribution in the cells. Using fluorescence and confocal microscopy and acquired data and image analysis, we observed a pronounced fluorescence in the cells treated with both concentrations of Roy-BODIPY (12), 6.25 µg/mL ( Figure 5B,D) and 25 µg/mL ( Figure 5A,C). Moreover, Roy-BODIPY (12) accumulated in the cytoplasm of cells instead of the nucleus (pointed out with N in each image) ( Figure 5). Using confocal microscopy, a distinctive accumulation in cytoplasmic vesicles was visible (marked with an arrow) ( Figure 5C,D). Therefore, the increased uptake of Roy (1) in cytoplasmic vesicles supported the finding of the higher antiproliferative/cytotoxic activity of this phytochemical ( Figure 5). Furthermore, a possible explanation for these results could be the fact that this compound may target the mitochondria and induce the mitochondriadependent intrinsic pathway of apoptosis. Further studies should be performed to unveil the exact mechanism of action of Roy (1).

Biodistribution in U87 Cells of Roy Labelled with BODIPY
To complement the results obtained for the cell viability assay, Roy (1), which had the highest antiproliferative/cytotoxic effect against GB cells, was conjugated with a fluorescent probe to assess its biodistribution in the cells. Using fluorescence and confocal microscopy and acquired data and image analysis, we observed a pronounced fluorescence in the cells treated with both concentrations of Roy-BODIPY (12), 6.25 μg/mL (Figure 5B,D) and 25 μg/mL ( Figure 5A,C). Moreover, Roy-BODIPY (12) accumulated in the cytoplasm of cells instead of the nucleus (pointed out with N in each image) ( Figure 5). Using confocal microscopy, a distinctive accumulation in cytoplasmic vesicles was visible (marked with an arrow) ( Figure 5C,D). Therefore, the increased uptake of Roy (1) in cytoplasmic vesicles supported the finding of the higher antiproliferative/cytotoxic activity of this phytochemical ( Figure 5). Furthermore, a possible explanation for these results could be the fact that this compound may target the mitochondria and induce the mitochondriadependent intrinsic pathway of apoptosis. Further studies should be performed to unveil the exact mechanism of action of Roy (1).   (Lisbon, Portugal). The plant name was checked at http://www.worldfloraonline.org/ (World Flora Online, accessed 12 January 2022). After collecting, the plant was air dried (25 • C) under the shade and stored in cardboard boxes protected from light and humidity to maintain its stability. Leaves and stems were separated at the beginning of this study. Experimental results from the leaves were previously published [45].

Chemicals
All chemicals and reagents used, including the solvents, were of analytical grade. The solvents were distilled before use. Acetone

Extract Preparation
Dried and powdered P. hadiensis var. hadiensis stems (1.507 kg) were extracted until exhaustion with acetone (31.95 L × 5, 30 min each) at room temperature for 2 weeks using an ultrasound-assisted extraction method [49]. The ultrasonic bath was operated at 35 Hz with a maximum input power of 320 W. The extract was filtered using Whatman No. 5 paper and the filtrate was concentrated using a rotary evaporator under reduced pressure at low temperature (40 • C). The extraction yield (% w/w) was determined by Equation (1):

Bioassay-Guided Fractionation
The extract residue was subjected to chromatographic fractionation. Dry flash column chromatography was carried out with silica gel 60 (0.063-0.200 nm) for column chromatography (Ref.: 1.007734.1000, Merck) using eluents Hex:EtOAc 80:20 and 85:15 following the procedure and silica quantities described in [50,51] and collecting the fractions in test tubes of 70 mL. It was performed with a funnel of 10 cm diameter × 8.

Isolation and Chemical Characterization Procedure
Flash liquid column chromatography was performed to obtain isolated compounds from stem fractions following previously described procedures [53][54][55]. A glass column filled with silica gel 60 (0.040-0.063 nm) was used for column chromatography (Ref.: 1.09385.1000, Merck; Ref.: 27623.323, VWR Chemicals), and 20 mL test tubes were used to collect the fractions. The vacuum system consisted of a 90 • glass adaptor linked to an aquarium air pump as a pressure source. This assembly can be inserted into the column and clamped in place. Flash liquid column chromatography was performed with two columns; the columns (one of 4 cm diameter × 46 cm height × 500 mL capacity and another one of 2.5 cm diameter × 66 cm height × 200 mL capacity) were purchased from Normax (Marinha Grande, Portugal).
TLC was used to monitor this process and to detect the compounds which were compared to standards (Roy(1) and DiRoy(3)) as explained in Section 2.1. To purify and isolate compounds from the obtained fractions, homemade PTLC plaques containing silica gel 60 for PTLC (Ref.: 1.07747.1000, Merck) were used following the procedure indicated in [56]. Briefly, the sample was dissolved in the minimum volume of solvent possible. Then, the sample was applied to the bottom of a homemade preparative silica UV indicator plate (1.5 cm from the bottom) as a thin straight line (2-4 mm) using a Pasteur pipette. The plate was run with 100 mL of freshly prepared mobile phase previously chosen. When the solvent front was a few milliliters from the top, the plate was removed from the tank and air dried in the fume cupboard. Next, UV-absorbing bands were marked out with a pencil, scraped off onto tin foil, and put into a sintered glass funnel attached to a glass Buchner flask to which vacuum was applied. The filter removed the silica and the solution with the compound was evaporated on a rotary evaporator. Finally, isolated compounds were confirmed by TLC coapplication with authentic standards (Roy, DiRoy) before characterizing their structure.

BODIPY-7α-acetoxy-6β-hydroxyroyleanone derivative (12) Synthesis
Succinic anhydride (42 mg, 0.42 mmol) was added to a mixture of BODIPY (14) (100 mg, 0.28 mmol) and triethylamine (79 µL, 0.56 mmol) in anhydrous dichloromethane (4 mL). The reaction mixture was stirred at room temperature until complete consumption of the starting material was observed by TLC. The solution was diluted with DCM and aqueous HCl 1M was added until pH 3. The resulting mixture was washed with H 2 O, the layers were separated, and the organic layer was washed with brine, dried, and concentrated. The crude material of (13) was used in the next step without further purification (112 mg, 88%). 1

Compound Characterization Nuclear Magnetic Resonance (NMR)
1 H-NMR and 13 C-NMR analyses were carried out in CDCl 3 and CD 3 OD at room temperature on a Bruker ® Biospin Fourier spectrometer ( 1 H-NMR and 13 C-NMR spectra were recorded at 300 MHz and 75 MHz, respectively) operating at 315K at the Faculty of Pharmacy (University of Lisbon, Lisbon, Portugal) and on a Bruker ® Advance 400 spectrophotometer ( 1 H-NMR and 13 C-NMR spectra were recorded at 400.13 MHz and 100.61 MHz, respectively) operating at 293K at the Faculty of Sciences (University of Lisbon).

Fourier Transform Infrared Spectroscopy (FTIR)
Infrared spectra were obtained using the PerkinElmer ® Spectrum TWO 400 (instrument number: L1600400; PerkinElmer Inc, Waltham, MA, USA) equipped with an attenuated total reflectance (ATR) device.
The samples were evaluated by FTIR in a PerkinElmer ® Spectrum TWO 400 (PerkinElmer Inc, Waltham, MA, USA) equipped with an attenuated total reflectance (ATR) device. The ATR system was cleaned before each analysis using dry paper and scrubbing it with methanol and water (50:50). The air in the analysis room of the FTIR-ATR spectrum was used as background to verify the cleanliness and to evaluate the instrumental conditions and room interferences due to H 2 O and CO 2 . The spectra were obtained collecting four scans of each sample, between 4000 and 600 cm −1 with a resolution of 4 cm −1 .
The component yield for Roy and DiRoy in the extracts was determined as follows: samples of the extracts (or fractions) were injected at 10 mg/mL. The resulting chromatograms, showing the presence of several products with different Rt and areas, were analyzed. The presence of the compounds of interest was confirmed by doing a coinjection of the extract and the pure compound. UV spectra and Rt, together with the coinjection outcome, allowed us to identify which peak of the sample corresponded to Roy and DiRoy. The respective areas were replaced to the X value on their calibration curves. The obtained value indicated the amount of Roy (or DiRoy) in 10 mg/mL of the sample. Next, the quantity of Roy (or DiRoy) in 1 mg/mL of the sample was calculated, dividing by 10. Finally, the quantity of Roy (or DiRoy) in 1 g/mL of the sample was obtained by multiplying it by 1000.

Bioactivity Assays Determination of Antioxidant Activity (DPPH)
The antioxidant activity of the fractions was measured by the DPPH method as described in [58]. Briefly, 10 µL of each sample was added to a 990 µL solution of DPPH (0.002% in methanol). The mixture was incubated for 30 min at room temperature. The absorbance was measured at 517 nm against a corresponding blank and the antioxidant activity was calculated using Equation (2): AA was the percentage of antioxidant activity, A DPPH was the absorption of DPPH against the blank, and A Sample was the absorption of the sample or control against the blank. Tests were carried out in triplicate at a sample concentration at 1 mg of sample/mL. The reference standard used for this procedure was quercetin, used in the same conditions as the samples.

Antibacterial Activity: Well Diffusion Method
The antibacterial activity of the fractions was evaluated against Gram-positive S. aureus [32]. The samples were reconstituted in DMSO at a concentration of 1 mg/mL. Stock solutions of vancomycin as the reference antibiotic were also prepared at 1 mg/mL DMSO.
In aseptic conditions, Petri dishes containing 20 mL of solid Mueller-Hinton for bacteria were inoculated with 0.1 mL of bacterial suspension matching a 0.5 McFarland standard solution and uniformly spread on the medium surface using a sterile swab. Wells of approximately 5 mm in diameter were made in the medium, using a sterile glass Pasteur pipette and 50 µL of each extract was added into the wells. Vancomycin as positive control against Gram-positive bacteria and DMSO as negative control were used in the assay. Plates were incubated at 37 • C for 24 h. The antimicrobial activity was evaluated by measuring the diameter (in mm) of the inhibition zone formed around the wells and compared to the controls.

Brine Shrimp Lethality Bioassay
In order to evaluate the general toxicity of the different samples, a lethality test using Artemia salina brine shrimp was performed as described [59]. Concentrations of 10 ppm of each fraction were tested. The number of dead larvae was recorded after 24 h and used to calculate the lethal concentration (% dead animals), according to Equation (3): In Vitro Assays

2.
Cell Viability U87, A172, U118, U373, and H4 cells (1 × 10 4 cells/well) were seeded in 96-well plates, 24 h before treatment. Cells were treated with the same concentrations of the extract from P. hadiensis leaves, the extract from P. hadiensis stems, the fractions of the extracts (I, II, III, IV, V, VI), and the isolated compounds (Roy (1) and DiRoy (3)), or with the vehicle control (DMSO), and further incubated for 24, 48, and 72 h. Cell viability was assessed using a modified Alamar blue assay [60]. Briefly, a solution of DMEM-HG medium with 10% (v/v) of a stock solution of resazurin salt dye at a concentration of 0.1 mg/mL was prepared, which was further added to each well after 24, 48, and 72 h. After a 4 h incubation at 37 • C and 5% CO 2 , the absorbance of the plate was read at 570 and 600 nm in a BioTek reader (BioTek Instruments, Inc., Winooski, VT, USA). The absorbance results were obtained by the Gen5 program. Cell viability was then calculated in accordance with Equation (4): Half-maximal inhibitory concentration (IC 50 ) values were further calculated using GraphPad Prism Software v.7.04 (GraphPad Software Inc., San Diego, CA, USA) [61].

Conclusions
We reported on the bioguided isolation of compounds from the acetonic P. hadiensis var. hadiensis stems extract and the in vitro antiglioblastoma activity of the extract, fractions, and isolated constituents. Leaves and stems samples from P. hadiensis var. hadiensis were extracted five times using acetone in an ultrasound-assisted extraction method. Roy (1) and DiRoy (3) compounds were isolated from the P. hadiensis var. hadiensis stem extract.
By using a HPLC-DAD analysis, we found significative differences between extracts from leaves compared with those from stems, mainly in their content of Roy (1). This was verified by TLC (authentic sample) and HPLC-DAD. Roy (1) was mainly present in the leaves while the stems were richer in DiRoy (3). Moreover, this is the first paper that provides a HPLC-DAD quantification of P. hadiensis compounds in acetonic extracts of the leaves and stems.
The activity of the pure compound Roy (1) was higher than any of the activities found in extracts, fractions, and the compound DiRoy (3), even at the lowest concentrations across all the tested cell lines (U87, A172, H4, U118, and U373).
The significant difference in the bioactivity of Roy (1) and DiRoy (3) may be related to the chemical structure of these compounds and their physicochemical properties, such as their molecule polarity. The fractions V and II showed the highest activity in cell lines among the fractions tested. Moreover, fractions III and V showed the highest antioxidant and antimicrobial activities. The higher values shown in the biological assays in fraction V can be explained by the fact that this fraction was the richest in Roy (1).
Taken together, the results indicate that Roy (1) is a promising natural compound that may serve as a lead compound to generate semisynthetic derivatives with improved activity against GB.