Antiproliferative and Carbonic Anhydrase II Inhibitory Potential of Chemical Constituents from Lycium shawii and Aloe vera: Evidence from In Silico Target Fishing and In Vitro Testing

Lycium shawii Roem. & Schult and resin of Aloe vera (L.) BURM. F. are commonly used in Omani traditional medication against various ailments. Herein, their antiproliferative and antioxidant potential was explored. Bioassay-guided fractionation of the methanol extract of both plants led to the isolation of 14 known compounds, viz., 1–9 from L. shawii and 10–20 from A. vera. Their structures were confirmed by combined spectroscopic techniques including 1D (1H and 13C) and 2D (HMBC, HSQC, COSY) nuclear magnetic resonance (NMR), and electrospray ionization-mass spectrometry (ESI-MS). The cytotoxic potential of isolates was tested against the triple-negative breast cancer cell line (MDA-MB-231). Compound 5 exhibited excellent antiproliferative activity in a range of 31 μM, followed by compounds 1–3, 7, and 12, which depicted IC50 values in the range of 35–60 μM, while 8, 6, and 9 also demonstrated IC50 values >72 μM. Subsequently, in silico target fishing was applied to predict the most potential cellular drug targets of the active compounds, using pharmacophore modeling and inverse molecular docking approach. The extensive in silico analysis suggests that our compounds may target carbonic anhydrase II (CA-II) to exert their anticancer activities. When tested on CA-II, compounds 5 (IC50 = 14.4 µM), 12 (IC50 = 23.3), and 2 (IC50 = 24.4 µM) showed excellent biological activities in vitro. Additionally, the ethyl acetate fraction of both plants showed promising antioxidant activity. Among the isolated compounds, 4 possesses the highest antioxidant (55 μM) activity followed by 14 (241 μM). The results indicated that compound 4 can be a promising candidate for antioxidant drugs, while compound 5 is a potential candidate for anticancer drugs.


Introduction
Cancer is one of the most dreadful diseases in the whole world. Due to this disease, nearly 8.2 million people died in 2012 and approximately 14.1 million new cases were reported [1]. Despite on the antioxidant activity of L. shawii stem extracts. The biological activities of the above plants encouraged our group to further examine the phytochemical composition and biological activities of L. shawii and A. vera resin which led to the isolation of 20 known compounds. The isolated compounds were scrutinized for their cytotoxic and antioxidant behavior, and the anticancer mechanism of active compounds was predicted by in silico target fishing.
Pharmaceuticals 2020, 13, x FOR PEER REVIEW 3 of 18 Pharmaceuticals 2020, 13, x; doi: FOR PEER REVIEW www.mdpi.com/journal/pharmaceuticals asthma, ulcers, sores, cold, acne, and burns [36]. Traditionally, Aloe vera resin is used in the management of diabetes, obesity, and other infectious diseases [37]. Previous findings investigated beneficial effects of different parts (leaves and fruits) of L. shawii but, to the best of our knowledge, there is no report on the antioxidant activity of L. shawii stem extracts. The biological activities of the above plants encouraged our group to further examine the phytochemical composition and biological activities of L. shawii and A. vera resin which led to the isolation of 20 known compounds. The isolated compounds were scrutinized for their cytotoxic and antioxidant behavior, and the anticancer mechanism of active compounds was predicted by in silico target fishing.

Cytotoxic Activity
The in vitro cytotoxic activity of each compound isolated from L. shawii was examined against breast cancer cells. Results of an MTT assay showed that almost all the compounds exhibited a concentration-dependent growth inhibition of MDA-MB-231 breast cancer cells. At 25 μM (the

Cytotoxic Activity
The in vitro cytotoxic activity of each compound isolated from L. shawii was examined against breast cancer cells. Results of an MTT assay showed that almost all the compounds exhibited a concentration-dependent growth inhibition of MDA-MB-231 breast cancer cells. At 25 µM (the lowest concentration tested), compound 5 showed maximum (50.53%) loss of cell viability as compared to the rest of the compounds. While 1 exhibited 76% cell viability, 2, 7, 9, 8, and 3 demonstrated 81-89% Pharmaceuticals 2020, 13, 94 4 of 18 cell viability, respectively. Thus, at 25 µM, compound 5 showed a significant cytotoxic effect, while the rest of the compounds did not produce substantial antiproliferative activities. However, at 100 µM (the maximum concentration used to treat cancer cells), compound 1 exhibited maximal inhibition (92.1%) of cell proliferation, followed by compounds 7, 2, 3, 5, 6, 4, 9, and 8. Therefore, it is evident that at low concentrations (25 µM and 50 µM), compound 5 is the most active cytotoxic agent [47,48], while at concentrations of 75 µM and above, compound 1 is the most effective cytotoxic compound. Gaweesh et al. (2015) studied the cytotoxic activity of different fractions of L. shawii against HepG2, MCF7, and HCT116 cancer cells and reported that the CH 2 Cl 2 fraction exhibits potent inhibition of the cell growth of MCF7 (breast cancer line) with IC 50 value of 11 ± 0.195 µg/mL [49]. These active compounds might be potential sources of the anticancer activity of the extracts. Compound 9 was previously isolated by our group [35] which inhibited the cell proliferation in MDA-MB-231 breast cancer cells in a dose-dependent manner. At 25 µM, compound 9 did not inhibit cell growth significantly; however, at concentrations above 50 µM, 9 exhibited over 50% cell growth inhibition in MDA-MB-231 cells. At 100 µM, compound 9 showed approximately 70% loss of cell viability, which is considerable. The results are demonstrated in Figures 2 and 3. lowest concentration tested), compound 5 showed maximum (50.53%) loss of cell viability as compared to the rest of the compounds. While 1 exhibited 76% cell viability, 2, 7, 9, 8, and 3 demonstrated 81-89% cell viability, respectively. Thus, at 25 μM, compound 5 showed a significant cytotoxic effect, while the rest of the compounds did not produce substantial antiproliferative activities. However, at 100 μM (the maximum concentration used to treat cancer cells), compound 1 exhibited maximal inhibition (92.1%) of cell proliferation, followed by compounds 7, 2, 3, 5, 6, 4, 9, and 8. Therefore, it is evident that at low concentrations (25 μM and 50 μM), compound 5 is the most active cytotoxic agent [47,48], while at concentrations of 75 μM and above, compound 1 is the most effective cytotoxic compound. Gaweesh et al. (2015) studied the cytotoxic activity of different fractions of L. shawii against HepG2, MCF7, and HCT116 cancer cells and reported that the CH2Cl2 fraction exhibits potent inhibition of the cell growth of MCF7 (breast cancer line) with IC50 value of 11 ± 0.195 μg/mL [49]. These active compounds might be potential sources of the anticancer activity of the extracts. Compound 9 was previously isolated by our group [35]     lowest concentration tested), compound 5 showed maximum (50.53%) loss of cell viability as compared to the rest of the compounds. While 1 exhibited 76% cell viability, 2, 7, 9, 8, and 3 demonstrated 81-89% cell viability, respectively. Thus, at 25 μM, compound 5 showed a significant cytotoxic effect, while the rest of the compounds did not produce substantial antiproliferative activities. However, at 100 μM (the maximum concentration used to treat cancer cells), compound 1 exhibited maximal inhibition (92.1%) of cell proliferation, followed by compounds 7, 2, 3, 5, 6, 4, 9, and 8. Therefore, it is evident that at low concentrations (25 μM and 50 μM), compound 5 is the most active cytotoxic agent [47,48], while at concentrations of 75 μM and above, compound 1 is the most effective cytotoxic compound. Gaweesh et al. (2015) studied the cytotoxic activity of different fractions of L. shawii against HepG2, MCF7, and HCT116 cancer cells and reported that the CH2Cl2 fraction exhibits potent inhibition of the cell growth of MCF7 (breast cancer line) with IC50 value of 11 ± 0.195 μg/mL [49]. These active compounds might be potential sources of the anticancer activity of the extracts. Compound 9 was previously isolated by our group [35] which inhibited the cell proliferation in MDA-MB-231 breast cancer cells in a dose-dependent manner. At 25 μM, compound 9 did not inhibit cell growth significantly; however, at concentrations above 50 μM, 9 exhibited over 50% cell growth inhibition in MDA-MB-231 cells. At 100 μM, compound 9 showed approximately 70% loss of cell viability, which is considerable. The results are demonstrated in Figures 2 and 3.  Compounds 10-12 were isolated from Aloe vera resin and tested for their growth inhibitory potential against MDA-MB-231 cancer cells; 12 demonstrated the highest cytotoxic activities at three different concentrations (Figure 3). At 25 µM, 12 showed~14.0% of cytotoxic effect, 11 exhibited no cell growth inhibition, while compound 10 depicted a growth-promoting effect on MDA-MB-231 cells. At 50 µM, compounds 10-12 exhibited cytotoxic activity in the range of 67-99%, while at 75 µM, 12 showed 44%, while 11 and 10 depicted 64% and 77% cell viability, respectively, indicating that compound 12 is potent at this concentration. At 100 µM, 12 exhibited the highest activity, followed by 10 and 11. Based on the results, we can say that 1, 7, 2, 3, 12, and 5 were the most potential hits. The calculated IC 50 values revealed that 5, 1-3, 7, and 12 are potent cytotoxic compounds with IC 50 values ranging from 31 to 60 µM, while the IC 50 values of 8, 6, and 9 were 72, 73, and 76 µM, respectively. The IC 50 values of 1-12 are shown in Table 1.

Antioxidant Activity
The antioxidant activity of crude extracts/fractions and isolated compounds was tested using DPPH radical scavengers. The tests were performed at different concentrations to calculate the IC 50 value. The ethyl acetate, BuOH, and CH 2 Cl 2 fractions of L. shawii showed promising inhibitory potential of 76%, 72%, and 60%, respectively. The IC 50 values for ethyl acetate, BuOH, and CH 2 Cl 2 fractions of L. shawii were 378 ± 1.50, 650 ± 1.50, and 735 ± 2.00 µg/mL, respectively, indicating that the ethyl acetate fraction is most potent as compared to standard ascorbic acid (53 ± 1.32 µg/mL). The crude MeOH extract of L. shawii demonstrated mild activity (50 ± 1.50%), while aqueous and n-hexane fractions were inactive. According to Gaweesh et al. (2015) [49], the EtOAc fraction of aerial parts of L. shawii exhibited antioxidant activity with an IC 50 value of 55.4 ± 3.48 µg/mL. The difference in activity may be due to the composition of the EtOAc fraction using only the stem instead of a mixture (leaves and stem). In the case of A. vera, only the ethyl acetate fraction exhibited moderate activity of 51%, while other fractions of this plant did not show promising results. The results are summarized in Table 2. All the isolated compounds were screened in a DPPH radical scavenging assay in order to test their antioxidant potential. The results demonstrated that compound 4 exhibited potent antioxidant activity with an IC 50 value of 55 ± 2.00 µM, followed by 14 (IC 50 = 241 ± 1.50 µM), 6 (IC 50 = 645 ± 1.50), and 13 (IC 50 = 762 ± 2.00 µM) as compared to standard ascorbic acid. Compound 4 showed higher activity than 9 (IC 50 = 30 µg/mL) reported by our group [35]. The results are tabulated in Table 3. It is a well-accepted notion that the presence and position of -OH groups in a molecule can increase or decrease the antioxidant activity [50]. Among diterpenes, compound 1 showed moderate activity which may be due to the presence of three exocyclic double bonds. Comparing anthraquinones, 6 possesses two -OH groups at the meta position of the right hand benzene ring, and one -OH at the C-8 position meta to methyl group at C-6. Compounds 5, 7, and 8 have similar basic skeletons with the absence of the meta -OH group which showed that the meta -OH groups play an important role in the antioxidant activity of anthraquinones. Comparing 4 and 14, 4 bears two -OH groups at the meta position of ring A, two -OH at ortho in ring B, one -OH at C-3 in ring C. Compound 14 has two -OH groups at the meta position of the benzene ring and one -OH at the C-3 position at cyclohexane. Due to the presence of two additional ortho (ring B)-OH groups, compounds showed higher DPPH radical scavenging activity. It is thus suggested that the chelating ability of the ortho and meta -OH groups in 4 played a greater role in the antioxidation property. All the isolated compounds were screened in a DPPH radical scavenging assay in order to test their antioxidant potential. The results demonstrated that compound 4 exhibited potent antioxidant activity with an IC 50 value of 55 ± 2.00 µM, followed by 14 (IC 50 = 241 ± 1.50 µM), 6 (IC 50 = 645 ± 1.50), and 13 (IC 50 = 762 ± 2.00 µM) as compared to standard ascorbic acid. Compound 4 showed higher activity than 9 (IC 50 = 30 µg/mL) reported by our group [35]. The results are tabulated in Table 3. It is a well-accepted notion that the presence and position of -OH groups in a molecule can increase or decrease the antioxidant activity [50]. Among diterpenes, compound 1 showed moderate activity which may be due to the presence of three exocyclic double bonds. Comparing anthraquinones, 6 possesses two -OH groups at the meta position of the right hand benzene ring, and one -OH at the C-8 position meta to methyl group at C-6. Compounds 5, 7, and 8 have similar basic skeletons with the absence of the meta -OH group which showed that the meta -OH groups play an important role in the antioxidant activity of anthraquinones. Comparing 4 and 14, 4 bears two -OH groups at the meta Pharmaceuticals 2020, 13, 94 7 of 18 position of ring A, two -OH at ortho in ring B, one -OH at C-3 in ring C. Compound 14 has two -OH groups at the meta position of the benzene ring and one -OH at the C-3 position at cyclohexane. Due to the presence of two additional ortho (ring B) -OH groups, compounds showed higher DPPH radical scavenging activity. It is thus suggested that the chelating ability of the ortho and meta -OH groups in 4 played a greater role in the antioxidation property.
The results of the Swiss Target Prediction server are summarized in Table S1 (supporting information) which revealed that compound 1 may target Cyp19A1 with 0.11 probability, which is not an anticancer drug target, while compound 2 depicted 0.09 to <0.05 probabilities for its predicted drug targets. Thus, the results for compound 2 were considered insignificant. Compounds 3 and 5 showed ≥0.10 and~0.10 probabilities, respectively, for their predicted targets. Among the suggested targets, poly [ADP-ribose] polymerase (PARP)-1 is a potential drug target in TNBC, and carbonic anhydrase II (CA-II), CA-I, CA-XII, CA-IX, and estrogen receptor (ER) α and β are potential anticancer drug targets. However, ER is not expressed in TNBC, thus it was excluded from our study. Compound 7 showed 0.11 probability for its probable targets. Among the suggested targets, CA-II is a target of interest in cancer treatment. Compound 12 showed probabilities in the range of 0.09-0.04 for its suggested targets, in which CA-II was also included, however its probability was lower than 0.05. Based on these results and an extensive literature survey, PARP and CA-II were also included in our docking experiments.

Pharmacophore Modeling
The Swiss Target Prediction (STP) server applies 2D-similarity searching to predict the target of the query compound. A number of compounds are present in this server with their actual biological activities and binding mechanism. In STP, 2D structures of the query compounds are matched with the 2D structures of the compounds present in its database and their probabilities are calculated and based on the calculated probabilities, and the target for the query compound is predicted. We also applied pharmacophore modeling to select the drug targets of our compounds with cytotoxic potential. It was hypothesized that the biological targets of those drugs that matched with the pharmacophore model can act as a target of our compounds. Two pharmacophore models, namely, M1 and M2, were generated by using compounds 1-3, 5, 7, and 12. M1 was generated by aligning the 3D structures of compounds 1-3, while M2 was created by aligning the 3D structures of compounds 5, 7, and 12. M1 possessed three hydrophobic (Hyd) and one H-Bond acceptor (AccP) feature while M2 contained three Hyd and two AccP features. The pharmacophore models are displayed in Figure 4.

Molecular Docking Studies
Compounds 1-3, 5, 7, and 12 were potential hits in our cell-based anticancer assay. Structure-based inverse docking analysis was carried out to predict their binding mechanism. Compounds were docked at the active site or ligand binding sites of the selected drug targets. The docking scores of the compounds suggests that CA-II is the most potential target for compounds 1 (−10.75), 2 (−9.64), 5 (−21.98), and 12 (−13.88), while c-KIT is the best target for compound 3 (−12.62). After CA-II, PARP-1 is an excellent candidate for 1, 2, 7, and 12, while PI3K and c-KIT are probable drug targets for compounds 3 and 5, respectively. Moreover, c-KIT is a good drug target for compounds 2, 7, and 12, while PARP1 is the fourth good target for compounds 3 and 5. PI3K was identified as the second most probable drug target for compound 3, while moderate for 1, 7, 2, and 5, and least for 12. The docking score suggests that the receptors EGFR, Notch 1, and IGFR have variable potency with these compounds, however, binding interactions suggest that compound 7 possesses good binding interactions with Notch 1, while compound 5 has higher binding interactions with EGFR. As per the docking scores, PTEN was the least favorable target for compounds 1-3, 5, and 7, while CDK4 was ranked as the worst target for all the compounds ( Figure 5). The docking results are presented in Table S2.
CA-II was exposed as the best target for compounds 1, 2, 5, and 7. Compound 1 binds with the docking score −10.75 and interacts with three water molecules and the side chain of Thr200 by H-bonding. Compound 2 interacts with the side chain of Asn67 and one water molecule via H-bonding. The -OH groups of compound 3 accept and donate H-bonds to the side chains of Asn67 and Glu69, respectively. Moreover, two water molecules provide H-bonding to the compound. The docked view of compound 5 depicts that its -OH groups donate H-bonds to the side chain of Thr200 and accepts H-bonds from the amino group of Thr199, while its ring mediates hydrophobic interactions with the side chain of Leu198. The docking score (−21.98) of 5 suggests that this compound possesses the highest binding potential for CA-II in silico. The side chain of Phe131 provides π-π interactions to the ring of compound 7. The docking score of 7 (−3.26) and its binding interactions suggest that this compound is the least active as compared to the rest of the compounds. The carbonyl moiety of compound 12 mediates bidentate interactions with the side/main chain amino groups of Thr199. Moreover, hydrophobic interaction was observed between the ring of the compound and the side chain of Thr199. The docking score of 12 (−13.88) is less than the docking score of compound 5, while higher than the docking scores of compounds 1-3 and 7. Some of the docked conformations of 12 showed that the compound may bind with the Zn atom present in the active site. The reference known drug, acetazolamide, interacts with ZN via metal-ligand interaction, and its carbonyl group also accepts H-bonds from the side chain of Thr199. The sulphate oxygen and ring nitrogen also mediate several water-mediated interactions. The active site of human CA-II is presented in Figure 6. The ligand (1GO), an acetazolamide derivative, complexed in the X-ray structure of CA-II, is also displayed, which interacts with ZN, His94, His96, His119, Gln92, Phe131, Thr199, Thr200, and two water molecules. Our compounds, except 12, do not bind with the zinc (ZN) present in the active site; however, they interact within the vicinity. The binding interactions of compounds are shown in Figure 7. The docking results are tabulated in supporting information Table S2.
Pharmaceuticals 2020, 13, x; doi: FOR PEER REVIEW www.mdpi.com/journal/pharmaceuticals based inverse docking analysis was carried out to predict their binding mechanism. Compounds were docked at the active site or ligand binding sites of the selected drug targets. The docking scores of the compounds suggests that CA-II is the most potential target for compounds 1 (−10.75), 2 (−9.64), 5 (−21.98), and 12 (−13.88), while c-KIT is the best target for compound 3 (−12.62). After CA-II, PARP-1 is an excellent candidate for 1, 2, 7, and 12, while PI3K and c-KIT are probable drug targets for compounds 3 and 5, respectively. Moreover, c-KIT is a good drug target for compounds 2, 7, and 12, while PARP1 is the fourth good target for compounds 3 and 5. PI3K was identified as the second most probable drug target for compound 3, while moderate for 1, 7, 2, and 5, and least for 12. The docking score suggests that the receptors EGFR, Notch 1, and IGFR have variable potency with these compounds, however, binding interactions suggest that compound 7 possesses good binding interactions with Notch 1, while compound 5 has higher binding interactions with EGFR. As per the docking scores, PTEN was the least favorable target for compounds 1-3, 5, and 7, while CDK4 was ranked as the worst target for all the compounds ( Figure 5). The docking results are presented in Table S2. CA-II was exposed as the best target for compounds 1, 2, 5, and 7. Compound 1 binds with the docking score −10.75 and interacts with three water molecules and the side chain of Thr200 by Hbonding. Compound 2 interacts with the side chain of Asn67 and one water molecule via Hbonding. The -OH groups of compound 3 accept and donate H-bonds to the side chains of Asn67 and Glu69, respectively. Moreover, two water molecules provide H-bonding to the compound. The docked view of compound 5 depicts that its -OH groups donate H-bonds to the side chain of Thr200 and accepts H-bonds from the amino group of Thr199, while its ring mediates hydrophobic interactions with the side chain of Leu198. The docking score (−21.98) of 5 suggests that this compound possesses the highest binding potential for CA-II in silico. The side chain of Phe131 provides π-π interactions to the ring of compound 7. The docking score of 7 (−3.26) and its binding interactions suggest that this compound is the least active as compared to the rest of the compounds. The carbonyl moiety of compound 12 mediates bidentate interactions with the side/main chain amino groups of Thr199. Moreover, hydrophobic interaction was observed between the ring of the compound and the side chain of Thr199. The docking score of 12 (−13.88) is less than the docking score of compound 5, while higher than the docking scores of compounds 1-3 and 7. Some of the docked conformations of 12 showed that the compound may bind with the Zn atom present in the active site. The reference known drug, acetazolamide, interacts with ZN via metal-ligand interaction, and its carbonyl group also accepts H-bonds from the side chain of Thr199. The sulphate oxygen and ring nitrogen also mediate several water-mediated interactions. The active site of human CA-II is presented in Figure 6. The ligand (1GO), an acetazolamide derivative, complexed in the X-ray structure of CA-II, is also displayed, which interacts with ZN, His94, His96, His119, Gln92, Phe131, Thr199, Thr200, and two water molecules. Our compounds, except 12, do not bind with the zinc (ZN) present in the active site; however, they interact within the vicinity. The binding interactions of compounds are shown in Figure 7. The docking results are tabulated in supporting information Table S2.   The binding modes also demonstrated the reason for higher binding scores. Compounds 5 and 7 had higher numbers of H-bonding and hydrophobic interactions as compared to 1-3 and 12. Moreover, compounds 5, 7, 12, and 3 depicted higher docking scores than the reference compound sunitinib, which is a drug complexed in the X-ray crystal structure of c-KIT (used in docking studies). Thr670, CYS673, LEU595, VAL603, and a water molecule (HOH289) played important roles in the stabilization of compounds 5, 7, and 12 in the active site. These residues provided hydrogen bonding and π-H interactions to the compounds.
Compounds 7 and 5 displayed higher binding scores than the positive control of EGFR (afatinib) and IGFR (ibutamoren mesylate). While compounds 5, 7, 12, and 3 scored better than crenigacestat (inhibitor of Notch 1), compounds 12, 5, and 7 were better than inhibitors of CDK4 (omacetaxine mepesuccinate), CA-II (acetazolamide) and PARP1 (niraparib). The docking results are tabulated in Table S1. 1, 2, 5, 7, and 12 Based on docking results, compounds 1, 2, 5, 7, and 12 were tested for their potential against CA-II in vitro. The quantity of compound 3 was not sufficient, thus it was not tested. The in vitro testing results showed that compound 5 is the most active compound with IC 50 values of 14.4 µM, followed by compounds 12 and 2. Both the compounds (12 and 2) also showed very good activity in the range of 23 and 24 µM, respectively. The docking score also indicated that compounds 5 and 12 are the most active compounds. In docking, compounds 1 and 3 showed good binding potential after 5 and 12, however, in vitro results demonstrated that compound 1 is inactive. Compound 7 showed the least binding potential in silico and was found to be inactive against CA-II in vitro. The results are summarized in Table 4. The structure-activity relationship of the compounds is discussed in docking results. The experimental findings correlate well with the docking results.

Materials and Methods
This study consisted of extraction and isolation of bioactive compounds from L. shawii and A. vera and determination of their cytotoxic and antioxidant potential. The cytotoxic assay was conducted on TNBC cell lines, followed by in silico molecular targeting of active compounds. Pharmacophore modeling and molecular docking approaches were used in computational analysis. The computational results were validated by in vitro testing of active compounds against the most suitable predicted targets.

Plant Material and Identification
The whole plant materials of L. shawii and A. vera resin were purchased from market (Souq, Nizwa) and identified by the plant taxonomists Saif Al-Hatmi (Oman Botanical Garden, Muscat, Oman (OBGM)) and Syed Abdullah Gillani (Department of Biological Sciences and Chemistry (DBSC), University of Nizwa, Oman), respectively. Voucher specimens of A. vera (No. AFS-08/2016) and L. shawii (No: BSHR-05/2015) were deposited in the herbarium of OBGM and DBSC, respectively.
Similarly, the shade-dried powdered resin of A. vera was exhaustively extracted with MeOH (2 L) at room temperature (3 × 15 days). Evaporation of the MeOH in vacuo at 45 • C yielded a crude methanol extract, which after suspension in water was successively fractionated into n-hexane, CH 2 Cl 2 , EtOAc, and n-BuOH [36]. After taking TLC, CH 2 Cl 2 and EtOAc fractions were combined.
The combined material was subjected to CC and eluted with n-hexane, n-hexane-EtOAc, EtOAc, MeOH/EtOAc, and finally washed with 20% MeOH/EtOAc with 10% increments in polarity to afford several fractions which were subsequently subjected to further repeated CC with different concentrations of n-hexane-EtOAc-MeOH as eluent to obtain 11 compounds, 10-20 [36].
Cells were seeded at a density of 1 × 10 4 cells per well in 96-well microtiter culture plates. After overnight incubation, normal growth medium was removed and replaced with either fresh medium (untreated control) or different concentrations of respective compounds in growth medium diluted from a 2 mM stock. After 24 h of incubation, MTT solution was added to each well (0.1 mg/mL in DMEM) and incubated further for 4 h at 37 • C. Upon termination, the supernatant was aspirated and the MTT formazan, formed by metabolically viable cells, was dissolved in a solubilization solution containing DMSO (100 µL) by mixing for 5 min on a gyratory shaker. The absorbance was measured at 540 nm (reference wavelength 690 nm) on an Ultra Multifunctional Microplate Reader (Bio-Rad, Hercules, California, USA). Absorbance of control (without treatment) was considered as 100% cell survival. Each treatment had three replicate wells [56]. The IC 50 values of the active compounds were calculated using nonlinear regression through GraphPad Prism 4 software (La Jolla, CA, USA).

Assay Protocol for DPPH Radical Scavenging Activity
Free radical scavenging activity of the test fractions/compounds was determined by measuring the change in absorbance of DPPH (l, l-Diphenyl-2-picrylhydrazyl radical, Sigma-Aldrich, St Louis, MO, USA) at 515 nm by the microplate reader (SpectraMax M2, Molecular Devices, CA, USA) as previously described [57]. Ascorbic acid (Sigma-Aldrich, St Louis, MO, USA) was used as a standard with 90% inhibition and IC 50 value of 53 µg/mL. The IC 50 values of the tested compounds were calculated using nonlinear regression through GraphPad Prism 4 software (La Jolla, CA, USA).

Selection of Known Drugs
By extensive literature survey, 62 known anticancer drugs were selected and their 3D structures were taken from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/), which was used as a library in pharmacophore searching. Moreover, during docking, those drugs were considered as positive control or reference compounds. The docking results of compounds 1-3, 5, 7, and 12 were compared with the docking scores of these drugs. The selected drugs for each target are tabulated in Table 5.

Molecular Docking
The X-ray crystal structures of the target proteins including EGFR, c-KIT, IGF1R, Notch 1, PTEN, PI3K, CDK4, PARP1, and CA-II were retrieved from Research Collaboratory for Structural Bioinformatics Protein Data Bank (RCSB-PDB) (https://www.rcsb.org/). The data are summarized in Table 6. For docking, protein files were prepared by addition of protons, partial charges, and the removal of cocrystallized ligands and heteroatoms. The role of water molecules was deduced by visualizing the protein-ligand interactions. Only water molecules within the vicinity of 3.0 Å of active ligands were retained in the file during docking, otherwise they were removed from the protein structure. The structures of the compounds and selected drugs were energy minimized using MMFF94x forcefield and gradient: 0.05. During energy minimization, hydrogens were added, and partial charges were applied.
Molecular docking was performed by MOE using Triangle Matcher placement method, Rescoring1: London dG, Refinement: Forcefield, and Rescoring2: Affinity dG. As a default parameter, 30 docked conformations were selected to be saved for each compound after docking. After docking, we applied conformation sampling method to select the best docked orientation of compound. For this purpose, each conformation of all the docked compounds were visualized and based on the protein-ligand interactions, docking score, and rank, best conformation was selected for analysis. The images in 2D were captured through MOE ligand binding interaction while 3D images were taken by Chimera [60].

Carbonic Anhydrase II Inhibition
The total reaction volume of 200 µL included 20 µL of test compounds prepared in DMSO, followed by the addition of 140 µL of the HEPES-tris buffer, 20 µL of purified bovine erythrocyte CA-II (0.1 mg/mL) prepared in buffer, and 20 µL of a solution of 4-nitrophenyl acetate [70]. A 20 µL amount of test compound was incubated with the enzyme (EC 4.2.1.1, Sigma-Aldrich, St. Louis, MO, USA) for 15 min in a 96-well flat-bottom plate. The rate of product formation was monitored with the addition of 20 µL of 4-NPA as substrate, prepared in methanol at the final concentration of 0.7 mM, at 25 • C for 30 min with regular intervals of 1 min, by using microplate readers (Bio-Rad, Molecular Devices, CA, USA). HEPES-tris was used as a buffer for the reaction at the final concentration of 20 mM at pH 7.4. The percent inhibition was calculated by using the following formula: % Inhibition = 100 − (OD test well/OD control) × 100

Statistical Analysis for Cytotoxic Activities
Results are expressed as mean ± SEM of at least three independent observations. Student's t-test was used to statistically examine significant differences. Analysis of variance was performed using ANOVA. p-Values < 0.05 were considered statistically significant.

Conclusions
Twenty compounds (isolated from L. shawii and A. vera) were scrutinized for their anticancer potential in triple-negative breast cancer cell lines (MDA-MB-231). Among the tested compounds, compounds 5, 1-3, 7, and 12 were retrieved as most potential hits. We used extensive in silico target fishing techniques to predict their biological targets in the human genome. For this purpose, 2D-cheminformatic tools and 3D-pharmacophore modeling were used that suggested that carbonic anhydrase II (CA-II), poly [ADP-ribose] polymerase (PARP)-1, and proto-oncogene tyrosine-protein kinase Kit (c-KIT) can be possible targets for the active hits. The in silico findings were validated by in vitro testing of compounds on CA-II, which showed that 5, 12, and 2 are excellent inhibitors of CA-II. Moreover, antioxidant activities of compounds were examined, which demonstrated that compound 4 possesses the highest antioxidant potential. These results indicated that constituents of L. shawii and A. vera are promising drug candidates for triple-negative breast cancer and should be investigated in detail.