Introduction

Fungal Endophytes obtained from unique environmental habitats offer a pool of potentially useful medicinal entities and have proven to be rich sources of bioactive natural products (Xia et al. 2018; Hardoim et al. 2015; Kaul et al. 2012; Pimentel et al. 2011; Sarasan et al. 2017). In the recent investigations of bioactive metabolites, the marine-derived fungi from invertebrates (e.g. sponges and soft corals) represent an outstanding source of novel bioactive metabolites having functional potentiality as drugs or drug leads (Haefner 2003; Saleem et al. 2007; Yin and Keller 2011). Such marine hosts are expected, therefore, to harbor marine-derived fungi having the capability to deliver various secondary metabolites (Suryanarayanan 2012). Large numbers of these metabolites have been included in the medical applications (Nicoletti and Trincone 2016; Pejin et al. 2013). Penicillium species represent the major source of antibiotics and mycotoxins (Frisvad et al. 2004), synthesized by different biosynthetic pathways, including terpenes, polyketides and alkaloids (Leitao 2009).

The fungus Penicillium sp. MMA, among others isolated from Sarcphyton glaucom got from Red Sea, exhibited high antibacterial, anti-yeast and cytotoxic activities. The strain was accordingly applied to upscale fermentation to isolate and identify its produced metabolites using different chromatographic techniques. Nine diverse metabolites were obtained: veridicatol (1), aurantiomide C (2), aspterric acid (3), 3,4-dihydroxy benzoic acid (4), linoleic acid (5), ergosterol (6), β-sitosterol (7), β-sitosterol glucoside (8) and cerebroside A (9) (Fig. 1). Their chemical structures were verified by NMR spectroscopic and mass spectrometric means and literature comparison. Biologically, the antimicrobial, antioxidant, antitumor and antibiofilm activities of the produced metabolites were investigated in comparison with the original extract. The computational calculations with respect to the physicochemical properties, ADME, acute oral toxicity were performed for compounds 1–4. Taxonomically, the studied fungus was characterized based on morphological and molecular biology (18srRNA) strategies (Fig. 2).

Fig. 1
figure 1

Chemical structures of the fungal metabolites 1–9

Full size image
Fig. 2
figure 2

Penicillium sp. MMA phylogenetic tree

Full size image

Materials and methods

General experimental procedure

The NMR spectra were measured on a Bruker AMX 300 (300.135 MHz), a Varian Unity 300 (300.145 MHz) and a Varian Inova 500 (125.820 MHz) spectrometer. ESI MS was recorded on a Finnigan LCQ with quaternary pump Rheos 4000 (Flux Instrument). EI mass spectra were recorded on a Finnigan MAT 95 spectrometer (70 eV) with perfluorkerosine as reference substance for EI HRMS. Flash chromatography was carried out on silica gel (230–400 mesh). Rf-values were measured on Polygram SIL G/UV254 (Macherey–Nagel & Co.). Size exclusion chromatography was done on Sephadex LH-20 (Pharmacia).

Sampling and Isolation of the producing fungus

Collection of Sarcophyton sp. and isolation of the desired endophytic fungus strain were performed as same as done in our recently reported work (El-awady et al. 2019), then kept at 4 °C (Debbab et al. 2009). Four pure isolates were obtained and deposited in the Microbial Biotechnology Department, NRC, Egypt, until investigation.

Pre-screening

The four fungal isolates (F1–F4) obtained were fermented in a small scale on rice-solid media at 30 °C for 7 days. After incubation, their culture media were individually soaked in ethyl acetate, followed by decantation, filtration and in vacuo concentration till dryness. Then extract of the isolates were (a) biologically tested as antimicrobial, antioxidant, antitumor agents (using Ehrlich’s antitumor activities) and (b) chemically screened (during TLC, visualized by UV and spraying reagents). The fungal isolated F4 was the most interesting among the four isolates through the biological and chemical screening criteria and selected, therefore, for full taxonomical characterization and large-scale fermentation to isolate its desired bioactive metabolites.

Genetic identification

The molecular identification of the selected fungal strain F4 was carried out by genomic DNA extraction using Qiagen DNeasy Mini Kit following the manufacturer’s instructions. The PCR amplification was performed using two primers ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′), the reaction mixture was as follows: 1 µg fungal genomic DNA, 1 µL (20 µM of each primer), 10 mM dNTPs mixture, 2 units of Taq DNA polymerase enzyme and 10 µL 5 × reaction buffer) with the following PCR thermal profile: 94 °C for 5 min, followed by 35 cycles of 94 °C for 30 s, 55 °C for 30 s, 72 °C for 90 s and a final extension step at 72 °C for 5 min. The PCR product was purified using JeneJET purification kit (ThermoFisher Scientific) and shipped for sequencing by Macrogen, South Korea (Blunt et al. 2007). The 18S rRNA gene sequence was aligned using BLAST available at NCBI database (GenBank C, https://www.ncbi.nlm.nih.gov/Genbank/National Institute of Biotechnology Information, Bethesda, Maryland, USA). The phylogenetic tree was constructed using neighbor-joining tree method using the software MEGA7, identifying it as Penicillium sp. MMA.

Large-scale cultivation

Well-grown colonies of Penicillium sp. MMA were inoculated into 100 mL of International System Project (ISP2) medium composition (g L−1): Malt extract (10); Yeast extract (4), glucose (4), 50% natural sea water, pH 6. and applied to cultivation on shaker at 30 °C for 3 days. The grown seed culture was served to inoculate 5 × 1 L Erlenmeyer flasks, each containing 100 g commercial rice and 150 mL 50% natural sea water. The seeded culture medium was then applied to static incubation at 28 °C for 14 days (Bara et al. 2013). After harvesting, the obtained yellowish-brown culture was soaked in ethyl acetate, followed by decantation and filtration. The solid culture residue was further re-soaked in methanol followed by filtration and concentration in vacuo; then the water residue obtained was extracted by ethyl acetate. The organic extracts were combined and concentrated to dryness giving 8.1 g.

Isolation and purification

The fungal extract was fractionated using column chromatography on silica gel (3 × 100 cm), eluted by cyclohexane-CH2Cl2–MeOH gradient. According to TLC visualization using UV (254/366 nm) and consequent spraying with anisaldehyde/sulphuric acid, four fractions were obtained: I (3.73 g), II (1.51 g), III (0.92 g) and FIV (1.11 g). Fraction I was re-purified on silica gel column (2 × 60 cm) and then Sephadex LH-20 (DCM/50% MeOH) to afford three colourless solids: ergosterol (6, 25.0 mg), β-sitosterol (7, 8.2 mg) and linoleic acid (5, 55.2 mg). Like fraction I, Fraction II was fractionated by silica gel column into three sub-fractions F-IIa (0.32 g), F-IIb (0.41 g) and F-IIc (0.21 g). Purification of FIIa on Sephadex LH-20 (DCM/40% MeOH) afforded veridicatol (1, 18.1 mg), while FIIb gave aurantiomide C (2, 18 mg) as further colourless solid. An application of sub-fraction F-IIc to PTLC (DCM/5% MeOH) followed by Sephadex LH-20 (MeOH) delivered a colourless solid of 3,4-dihydroxy-benzoic acid (4, 4.2 mg). Purification of fraction III through silica gel column (2 × 60 cm) and elution with DCM-MeOH gradient afforded aspterric acid (3) as major crude compound (225 mg), which after re-purification on Sephadex LH-20 (MeOH) was afforded (200.1 mg) as colourless solid. The polar fraction IV was fractioned on silica gel column using DCM-MeOH gradient, followed by purification on Sephadex LH-20 (MeOH) to afford two colourless solids of β-sitosterol glucoside (8, 22.1 mg) and cerebroside A (9, 25.0 mg). Spectroscopic data of the isolated compounds (1–9) are found in “Supplementary Data” file.

Biological activity

Antimicrobial activity

40 μL for each of compounds 1–9 (dissolved in CH2Cl2/10% MeOH, 1 mg/mL) were soaked on paper discs (6 mm ∅) and dried under sterilized conditions. Then, they were placed on inoculated agar plats and incubated for 24 h at 37 °C for bacterial and 48–72 h (30 °C) for the fungal isolates. The disc diffusion test has been done according to Bauer et al. (1966). Inhibition zones were measured in mm and recorded. The microbes Bacillus subtilis, Staphylococcus aureus, Pseudomonas areuginosa, Escherichia coli, Candida albicans and Aspergillus niger were served. The isolates were obtained from the Microbial Biotechnology Department, NRC, Egypt. The nutrient agar medium (g/l): Beef extract 3; peptone,10; and agar, 20 (pH 7.2) were served for growing of bacteria and yeasts, while the test fungal strains were grown on Czapek-Dox medium.

Antioxidant assaying

The free radical scavenging activity (RSA) was measured by the decolouration of an ethanolic solution of DPPH radical and evaluated spectrophotometrically at 517 nm according to Brand-Williams et al. (1995).

Antitumor activity assaying against Ehrlich cells

The in vitro antitumor activity testing, based on Ehrlich’s, was performed according to Bennett et al. (1976).

Cytotoxic activity

Five human cancer cell lines were tested using the MTT assay (Mosmann 1983): Hepatocellular (HePG-2), Epitheliod (Hela), Epdermoid (HEP2), Mammary gland (MCF-7) and colorectal (HCT-116) carcinoma. The cell lines were obtained from American Type Culture Collection (ATCC) via Holding company for biological products and vaccines (VACSERA), Cairo, Egypt.

Biofilm inhibitory assaying

To evaluate anti-biofilm efficiency, microtitre plate assay (MTP) was carried out against four clinical microbes (P. aeruginosa, S. aureus, E. coli and B. subtilis) using 96-well flat-bottom polystyrene titre plates according to (Christensen et al. 1985; Hamed et al. 2020).

Study of the physicochemical properties, ADME-parameters, acute oral toxicity and toxicity targets

The physico-chemical properties and ADME parameters of compounds 1–4 were performed according to Daina et al. (2017). Prediction of rodent oral toxicity and indication of possible toxicity targets were estimated by ProTox web server according to Drwal et al. (2014).

Results and discussion

Pre-screening

Four fungal strains (F1–F4) isolated from Sarcophyton sp. were biologically screened. Anti-microbially, using paper-disk diffusion assay, isolates F1–F4 showed high similarity in their potential activity against Staphylococcus aureus, Bacillus subtilis, Pseudomonas aeruginosa and Candida albicans. In contrast to the high activity of the fungal isolates F1 and F2 against Escherichia coli, isolates F2 and F3 were exceptionally inactive against the latter. Nevertheless, the four isolates displayed no antifungal activity against A. niger (see supplementary data, Table S1). The organic extracts of the four isolates were further investigated for antioxidant and antitumor by Ehrlich’s activities, revealing that F4 > F2 > F1 > F3 in their antioxidant potentiality,;meanwhile, they were ordered as F4 > F1 > F2 > F3 according to their antitum or potency (see supplementary data, Table S2). Based on this study, the fungal isolate F4 was remarked to be the most active from the antioxidant (79.1%) and antitumor (67.2%) activities point of view in addition to its potential antimicrobial activity as well. Therefore, the strain F4 was selected for full taxonomy, large-scale fermentation and studied for its produced bioactive metabolites in both of chemical assignments and biological activities.

Genetic identification

The genomic DNA of the selected fungal isolate F4 was extracted, amplified and the sequence of 18S rRNA gene was obtained and aligned to identify the similarity score with other known sequences available in the GenBank database using BLAST tool (https://www.blast.ncbi.nlm.nih.gov/Blast). The obtained result confirmed a very close similarity of the 18S rRNA gene sequence obtained with 100% homology of the isolate coded MMA with Penicillium sp. The phylogenetic analysis and tree were constructed using the neighbor-joining method by MEGA 7 program and based on the genetic analysis and similarity score the selected fungus MMA was identified as Penicillium sp. MMA and deposited in GenBank with the accession no. MK026953.

Isolation and Structure identification

The fungus Penicillium sp. MMA was up-scale cultivated on solid rice medium, worked up and its bioactive metabolites produced were purified using different chromatographic techniques (see the experimental section) delivering nine compounds. Structures of the obtained metabolites were assigned by the study of their NMR (1D, 2D) spectroscopy and mass spectrometric means (see supplementary data), identifying them as viridicatol (1) (Shaaban et al. 2016; Hamed et al. 2019), aurantiomide C (2) (Hamed et al. 2019), aspterric acid (3) (Shimada et al 2002), 3,4-dihydroxy benzoic acid (4) (Syafni et al. 2012), linoleic acid (5) (Shaaban 2004; Hamed et al. 2019), ergosterol (6) (Nagia et al. 2012), β-sitosterol (7), β-sitosterol glucoside (8) (Hamed et al. 2019) and cerebroside A (9) (Hamed et al. 2019). According to our searches in literature and different databases, the obtained compounds in this investigation were reported previously either from Penicillium sp. or Aspergillus sp. However, they are reported herein for the first time from Penicillium sp. MMA which has the highest similarity to Penicillium crustosum FRR 1669 (ex-type) (Nicoletti and Trincone 2016; Yu et al 2019) based on our genomic characterization mentioned above. In accordance, originality of the compounds investigated herein has been assured.

Viridicatol (1) belongs to 4-arylquinolin-2(1H)-ones, showing cytotoxicity toward human cervix (KB, KBv200), lung (A549), liver (HEPG2, SMMC7721), breast (MCF7), leukemia (K562) and gastric (SGC7901) tumor cell lines (Hamed et al. 2019; Luckner & Mothes 1962; Austin & Myers 1964). Aurantiomide C (2) was alternatively reported to show a potent anticancer activity against hepatocellular carcinoma (BEL-7402) and leukaemia (P388) cell lines (Xin et al. 2007). Aspterric acid (3) is a carotane-type sesquiterpene, a potent inhibitor of pollen development in Arabidopsis thaliana (Shimada et al. 2002). Aspterric acid (3) reported no considerable cytotoxicity in human cell lines (Yan et al. 2018). Benzoic acid derivatives (e.g. 4) are medicinally served as antiseptic, expectorant, antifungal, antipyretic and keratolytic agents (Shaaban 2004; Chapman & Hall Chemical Database 2018). Linoleic acid and its derivatives represent, on the other hand, the main essential unsaturated fatty acids (EFA), which belong to Omega 6 fatty acids (Abdel-Razek et al. 2017; Gaullier et al. 2005), are necessary to human body physiological processes. EFA have several medicinal applications and in treatment of cardiovascular diseases, skin permeability, insulin resistance, cancer, depression and plasmodial activity (Undurti 2008; Melariri et al. 2012). EFA were reported as well to reduce the nerve and breast pains in addition to their reductive efficiency of blood pressure and rheumatoid arthritis (Abdel-Razek et al. 2017).

Sterols represent the first choice of potential natural preventive dietary products. Ergosterol (6) is a common fungal sterol found displaying broad and significant biological activities so that up to date, 8247 articles have been cited in Scifinder referring to its biological importance (scifinder.cas.org). Particularly, ergosterol was reported to increase vitamin D concentrations in serum and liver of mice (Baur et al. 2019). Ergosterol has anti-inflammatory efficiency through the reduction of nitric oxide formation in LPS-stimulated RAW264.7 cells (Patjana et al. 2019). So, it might be served as novel functional food in prevention of allergic diseases (Kawai et al 2019).

β-Sitosterol is a phytosterol, structurally like cholesterol and is well spread in plants, fungi and animals (Peshin and Kar 2017). As a secondary metabolite, it is used as health-promoting constituent of natural foods. According to European foods and safety authority (EFSA) and US Food and Drug Administration (USFDA), it has been recommended to consume 1.5–4 g/day of phytosterol to reduce blood pressure and the risk of heart attack. β-sitosterol-β-d-glucoside has been proposed as a valuable compound for the improvement of new drugs to cure several inflammations accompanied by nitric oxide overproduction (Peshin and Kar 2017). It powerfully inhibits the activity of interleukin 6 of motivated macrophages (Kontogiorgis et al. 2010; Gautam and Navneet 2012). Cerebrosides (e.g. 9) (sphingolipids) (Barreto-Bergter et al. 2011) are playing an important role in major cellular processes (Youssef et al. 2016) and ecological activities of eukaryotic cells (Koga et al. 1998).

Biological activity studies

Antimicrobial activity

The antimicrobial potentiality of obtained metabolites 1–9 (using paper-disk method), revealed that aspterric acid (3) is the most potent against all studied microorganisms: S. aureus (10 mm), B. Subtilis (18 mm), P. aeruginosa (18 mm), E. coli (10 mm), C. albicans (12 mm) and A. niger (20 mm). Veridicatol (1) is weakly to moderately active against B. Subtilis, P. aeruginosa, E. coli and A. niger; meanwhile. ergosterol (6) was inactive against yeast, fungi and the G positive S. aureus. As expected, linoleic acid (5) was weakly to moderately active against Gram-positive and Gram-negative bacteria and fungi (8–10 mm), while being potentially active against C. albicans (16 mm) (Table 1).

Table 1 Antimicrobial activity of compounds (1–9) (Φ mm)
Full size table

Antioxidant activity

The antioxidant activity of compounds 1–9 based on DPPH assay (200 µg/mL) revealed that aurantiomide C (2) is the most potent antioxidant agent, showing maximum DPPH scavenging activity (75.26%), followed by aspterric acid (9) (50.16%), then veridicatol (1) (42.29%). In contrast, the remaining compounds are of very less antioxidant activity (Table 2).

Table 2 DPPH scavenging activity (%) of compounds 1–9
Full size table

In vitro cytotoxicity

As there is a matching correlation between the natural compounds’ antioxidant activity and their cytotoxicity (Li et al 2007), veridicatol (1), aurantiomide C (2) and aspterric acid (3) were accordingly selected to investigate their efficiency as anticancer agents. So, they were studied against liver (HePG-2), breast (MCF-7), colon (HCT-116), human epithelial type 2 (HeP2) and cervical (Hela) cancer cell lines, in comparison with doxorubicin as reference control. Accordingly, veridicatol (1) showed the maximum potency against the five cell lines (IC50: 6.27–13.87 µg/mL), followed by aurantiomide C (2) (IC50: 9.51–37.63 µg/mL); meanwhile, aspterric acid (3) showed the lowest cytotoxicity (IC50: 26.71–51.10 µg/mL (Table 3).

Table 3 In vitro Cytotoxicity (IC50 µg/ml) of compounds (1–3) against different cell lines
Full size table

Anti-biofilm activity

Recently, bacterial biofilm is considered as one of the major public health problems. This process gives bacteria the ability to maintain their attachment to the living and non-living surface by production of a complex exopolymers formed from proteins, polysaccharides and nucleic acids (Chapman et al. 2002; Branda et al. 2005; Steinberg and Kolodkin-Gal 2015). Biofilm participates in a large task in the persistence of pathogenic bacteria (Rabin et al. 2015) and is considered as a source of pathogenic bacteria that are involved in numerous infectious diseases (Donlan and Costerton 2002; Wingender and Flemming 2011).

Based on MTT assay, the biofilm inhibition action of the fungus extract and corresponding pure compounds was investigated against P. aeruginosa, S. aureus, E. coli and B. subtilis. Preliminary antibiofilm results of the crude extract displayed potent biofilm inhibition against P. aeruginosa up (43%) and low biofilm inhibition against S. aureus (19%). However, it showed no activity against E. coli and B. subtilis. β-Sitosterol (7: C1) and Veridicatol (1: C3) reduced the biofilm formation of B. subtilis up to 28% and 35%, respectively (Fig. 3a).S. aureus biofilm formation was inhibited over 64% by β-Sitosterol (7: C1). In contrast, aurantiomide C (2: C6), ergosterol (6: C8) and aspterric acid (3: C9) have low inhibitory activity (Fig. 3). In case of E coli, the highest percentage of biofilm inhibition was shown by Veridicatol (1: C3), Aurantiomide C (2: C6) and ergosterol (6: C8) followed by (-Sitosterol (7: C1) and (-Sitosterol glucoside (8: C2) (Fig. 3). On the other hand, P. aeruginosa biofilm displayed a very low response to β-Sitosterol (7: C1), aurantiomide C (2: C6) and ergosterol (6: C8), while it did not display any response to other compounds (Fig. 3).

Fig. 3
figure 3

Biofilm inhibition (%) of the fungus extract and pure compounds (β-Sitosterol (7: C1), β-Sitosterol glucoside (8: C2), Veridicatol (1: C3), linoleic acid (5: C5), aurantiomide C (2:C6), ergosterol (6: C8), aspterric acid (3: C9), Crude extract: Cr). Test organisms: P. aeruginosa, S. aureus, E. coli and B. subtilis biofilm were assessed by crystal violet staining. Data presented represents mean ± SD of three independent experiments

Full size image

Based on our updated search in literature, aspterric acid (3) and linoleic acid were reported previously as antifungal agents. Particularly, they showed significant antifungal activities against crop and plant pathogens (Liang et al. 2019; Walters et al. 2004). Fortunately, aspterric acid (3) is reported herein first as potent antibacterial agent especially against B. subtilis. Moreover, linoleic acid showed a weak activity against Gram-positive and Gram-negative bacteria.

On the other hand, the quinolinone and quinazoline moieties, viridicatol (1) and aurantiomide C (2), respectively, were reported previously to exhibit potent anticancer activities (Xin et al. 2007; Hamed et al. 2019). The potential cytotoxic activity of veridicatol against colon (HCT-116), human epithelial type 2 (HeP2) and cervical (Hela) cancer cell lines is reported herein for the first time; meanwhile, the cytotoxicity of aurantiomide C (2) against the reported five cell lines is reported herein for the first time. It is worthy to refer herein that, viridicatol (1), aurnatiomide C (2) and aspterric acid (3) are reported to first time as potential antioxidant agents, and this is strongly matching with their reported subsequent anticancer activities. Finally, the investigated antibiofilm activity of the afforded compounds against P. aeruginosa, S. aureus, E. coli and B. subtilis has been discussed herein for first time so far.

Physicochemical properties and ADME parameters

Swiss ADME web-based tools (Daina et al 2017) were used to estimate the physicochemical properties and ADME behaviors of compounds 1–4. According to drug-likeness rules, the compounds passed Lipinski, Veber and Ghose rules with zero violation, except compound 4 did not pass the filter of Ghose rule attributable to three violations including MW < 160, MR < 40 and number of atoms < 20. The compounds have oral bioavailability (0.56/0.55) and could be possible oral drugs (Table 4). In the same manner, the Bioavailability Radar plot is adopted for a rapid estimation of drug-likeness, taking into account six physicochemical properties (Fig. 4) (Daina et al. 2017; Vuppala et al. 2013). Evidently, compound 2 from the first glance displayed all the properties in the optimal range (pink area).

Table 4 Physico-chemical Properties, drug-likeness and lead-likeness parameters of compounds 1–4
Full size table
Fig. 4
figure 4

Bioavailability Radar plot of the predicted compounds: 1–4

Full size image

Lipophilicity is an important physicochemical characteristic quantified by the partition coefficient Log Po/w between water and n-octanol that gives a good indicator of permeability across the cell wall (Rutkowska et al. 2013; Potts and Guy 1992). Compounds 1–4 demonstrated Log Po/w values below 5, ranging from 0.80 to 2.61, suggesting good permeability and absorption across the cell membrane of infected cells. Based on ESOL topological model, compounds 1–4 are soluble (Daina et al. 2017). For defining lead-likeness, they passed the rule of three (RO3), except compound 4 that has one violation against this rule. So, compounds 1, 2 and 3 could be possible lead compounds. For synthetic accessibility score (SAscore) that estimated on similarity of fragments and complexity penalties, compound 4 is the easiest one (1.07), while compound 3 is the most difficult (5.09).

The pharmacokinetic parameters of compounds 1–4 were visualized by vector machine algorithm (SVM) model (Daina et al. 2017). As shown in Table 5, compounds 2 and 3 are non-inhibitors against all the isoenzymes, while compound 1 display selective inhibitory toward CYP1A2 and CYP3A4. The latter (CYP3A4) is also inhibited by compound 4. Based on BOILED-Egg model (Brain or Intestinal Estimate D permeation method, WLOGP vs TPSA) that adapted according to (Daina et al. 2017 and Daina and Zoete 2016) and illustrated in Fig. 5, compounds 1–4 demonstrated high human gastrointestinal absorption (GI). Compounds 2 and 3 are P-gp substrates (PGP + , blue dots), while the others are none. Compounds 2 and 4 are non-permeant for blood–brain barrier (BBB) and hence they have no adverse effects on central nervous system (CNS), while compounds 1 and 3 are BBB permeant (TPSA < 75 Å2).

Table 5 Pharmacokinetics parameters of compounds 1–4
Full size table
Fig. 5
figure 5

The BOILED-Egg plot of the predicted compounds 1–4

Full size image

Predicting the skin permeability coefficient (Kp) of compounds 1–4 was done as described by Potts and Guy (Potts and Guy 1992; Atta et al 2019), where the more negative the log Kp, the less skin permeant is the compound (2).

Prediction of acute oral toxicity and indication of toxicity targets

Based on the ProTox web server, has been indicated the possible toxicity targets of compounds 1–4 (Drwal et al. 2014). Fortunately, the predicted toxicity classes were ranged from four to five. Compound 3 displayed the highest LD50 value (3520 mg/kg), so it has the lowest toxic effect. Also, all the predicted compounds have not any toxic fragments as manifested in Table 6. Regarding with toxicity targets, compound 3 indicated binding to androgen receptor with zero average pharmacophore fit (Table 7).

Table 6 In silico acute oral toxicity prediction of compounds 1–4
Full size table
Table 7 In silico indication of toxicity targets of compounds 1–4
Full size table

It is worthy to refer herein and based on our updated searching in literature that, the presented computational studies for compounds 1–4 have been discussed to first time so far, which can be helpful for future drug administration and exploration.