Chemical and Biological Review of Endophytic Fungi Associated with Morus sp. (Moraceae) and In Silico Study of Their Antidiabetic Potential

The chronic nature of diabetes mellitus motivates the quest for novel agents to improve its management. The scarcity and prior uncontrolled utilization of medicinal plants have encouraged researchers to seek new sources of promising compounds. Recently, endophytes have presented as eco-friendly leading sources for bioactive metabolites. This article reviewed the endophytic fungi associated with Morus species and their isolated compounds, in addition to the biological activities tested on their extracts and chemical constituents. The relevant literature was collected from the years 2008–2022 from PubMed and Web of Science databases. Notably, no antidiabetic activity was reported for any of the Morus-associated endophytic fungal extracts or their twenty-one previously isolated compounds. This encouraged us to perform an in silico study on the previously isolated compounds to explore their possible antidiabetic potential. Furthermore, pharmacokinetic and dynamic stability studies were performed on these compounds. Upon molecular docking, Colletotrichalactone A (14) showed a promising antidiabetic activity due to the inhibition of the α-amylase local target and the human sodium-glucose cotransporter 2 (hSGT2) systemic target with safe pharmacokinetic features. These results provide an in silico interpretation of the possible anti-diabetic potential of Morus endophytic metabolites, yet further study is required.


Introduction
Type 2 diabetes mellitus (Type 2 DM) is considered one of the most prevalent metabolic disorders, affecting approximately 90% of diabetic patients [1,2]. Many medicinal plants are used in managing diabetes [3,4]. The advantage of the use of medicinal plants is due to their availability, cost-effectiveness, and higher safety [5][6][7][8]. The extensive and uncontrolled utilization of medicinal plants may add to the ecological burden in terms of overutilization of endangered and rare species and in disturbing the ecological balance [9]. In this context, endophytes associated with medicinal plants present an eco-friendly alternative source of bioactive metabolites, given that endophytes may cross talk with the host medicinal plants in terms of their biosynthetic routes or that they may be the original producers of some active ingredients, or at least may provide the host organisms with extra chemical defense to cope with the surrounding stress conditions [10][11][12][13]. The abundance of endophytic fungi within the host medicinal plants may be associated with the pharmacological actions linked to the plant part used [14]. For example, endophytic fungal metabolites associated with Syzygium cumini L. showed significant amylase inhibitory activity, which could be

Endophytic Fungi Associated with Morus Species
Reviewing the literature as shown in (Figures 1 and 2), 115 endophytic fungal isolates were reported from Morus alba leaf, stem, and root tissues; 95 (82.6%) isolates were identified, and 20 (17.4%) isolates were reported as unidentified. The most abundant

Reported Biological Studies on Endophytic Fungal Extracts
A few studies have reported the biological evaluation of the endophytic fungal metabolites associated with Morus genera. The endophytic fungi crude EtOAC extracts of Aspergillus sp. A204, Colletotrichum sp. C103, and Penicillium sp. P306 associated with M. alba showed a broader antifungal spectrum [45]. M. alba endophytic fungi EtOAc extracts of Phoma sp. MJ76 and Chaetomium sp. showed inhibition of human immunodeficiency virus-1 (HIV-1) replication using β-galactosidase and p24 antigen in vitro assays on cell lines developed from human cervical epithelial carcinoma (TZM-bl cells) and peripheral blood mononuclear cells (PBMC) [49]. The EtOAC extract of M. nigra endophytic fungus Botryosphaeria fabicerciana (MGN23-3) showed antioxidant activity using a DPPH assay and selective antibacterial activity on gram-positive bacteria using an in vitro plate dilution method revealed by determination of minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) [51].

Pharmacokinetic Profiling
As shown in ( Figure 4 and Table 2), the predicted pharmacokinetic properties of the evaluated compounds revealed their high potential for gastrointestinal (GI) absorption due to their reasonable solubility. Additionally, nearly all compounds lacked permeability to the blood-brain barrier and cytochrome P2D6 (CYP2D6) inhibition, thus expanding their safety profiles, except compound (9). In accordance with our analysis, the high absorbability increases the potential for targeting hSGT2. Additionally, this lack of central presence adds to the benefits of these compounds through the elimination of possible side effects owing to central permeability.

Molecular Modelling
Based on the pharmacokinetic results, the antidiabetic potential of the compounds was investigated locally and systematically through screening of action against α amylase, α/β glucosidase enzymes, and human sodium-glucose cotransporter 2 (hSGT2). Autodock Vina successfully performed the docking process of the select compounds in three targets, α amylase (PDBID: 2QV4), β glucosidase enzymes (PDBID: 2XWE), and hSGT2 (PDBID: 7VSI) for screening of their potential in antidiabetic therapy while MOE08 was used for α glucosidase (PDBID: 3A4A) after unsuccessful attempts using Vina. The docking protocol was validated through docking of the co-crystallized ligand in each receptor, followed by comparing the co-crystallized pose and docked pose and calculation of RMSD between them. As shown in ( Figure 5), α amylase with co-crystallized acarbose showed an RMSD of 1.22 Å, while α and β glucosidase RMSD was 0.67 and 1.95 Å, respectively. Similarly, empagliflozin, which is co-crystallized in hSGT2, had an RMSD of 1.01 Å.
All compounds demonstrated favorable binding to the three selected targets as evidenced by the obtained negative values of docking scores in kcal/mol in (Table 3). For comparative analysis of the antidiabetic abilities of the evaluated compounds, acarbose was used as a reference for inhibitory activity on α amylase and β of glucosidase, while empagliflozin was used as an hSGT2 reference inhibitor. 7VSI) for screening of their potential in antidiabetic therapy while MOE08 was used for α glucosidase (PDBID: 3A4A) after unsuccessful attempts using Vina. The docking protocol was validated through docking of the co-crystallized ligand in each receptor, followed by comparing the co-crystallized pose and docked pose and calculation of RMSD between them. As shown in (Figure 5), α amylase with co-crystallized acarbose showed an RMSD of 1.22 Å, while α and β glucosidase RMSD was 0.67 and 1.95 Å, respectively. Similarly, empagliflozin, which is co-crystallized in hSGT2, had an RMSD of 1.01 Å. All compounds demonstrated favorable binding to the three selected targets as evidenced by the obtained negative values of docking scores in kcal/mol in (Table 3). For comparative analysis of the antidiabetic abilities of the evaluated compounds, acarbose was used as a reference for inhibitory activity on α amylase and β of glucosidase, while empagliflozin was used as an hSGT2 reference inhibitor.

α Amylase Interaction
Both acarbose and tested compounds demonstrated negative bind scores hinting at the favorable interactions with the enzyme. Acarbose showed the highest affinity with a score of −9.70 kcal/mole, while (14), (3), (4), and (2) demonstrated the highest affinities of −8.80, −8.50, −8.50, and −8.40 kcal/mol, respectively. Upon inspection of 2D interactions, it becomes clear that the hydrophilic nature of acarbose enables it to form multiple hydrogen bonds with several α amylase residues such as Trp59, Tyr62, Gln63, His101, Asn105, Ala106, Val107 Thr163, Arg195, Glu233, and Asp300 ( Figure 6). Although the compounds were able to interact with common amino acids such as Trp59, Tyr62, Thr163, Glu233, and Asp300, they were less able due to the more lipophilic characteristics of the compounds. However, the top-scoring compounds compensated for this deficiency through the formation of hydrogen bonds with other amino acids in the binding site, namely, Leu162, Leu165, Asp197, Asp198, and Asp305 (Figures 7 and 8).

α and β Glucosidase Infarction
α/β glucosidases contributed to the treatment of type 2 DM by breaking down the glycosidic by the α isoform and the aryl and alkyl glycosides, disaccharides, and small oligosaccharides by the β isoform [32,59]. The binding of acarbose to the α and β isoforms was −8.97 and −8.70 kcal/mol, respectively. Although the compounds showed favorable binding in both cases, the binding was stronger in the β isoform in nearly all instances suggesting a partial preference for β rather than α. The hydrophilic nature of acarbose enables it to form multiple hydrogen bonds with several α glucosidase residues such as Tyr72, Tyr158, Phe178, Arg213, Asp215, Asp242, Gln279, Pro312, His351, Asp352, and Arg442 ( Figure 8). Additionally, ionic interactions were observed as well with Tyr158 and Asp242 in addition to one hydrophobic bond with Arg315 ( Figure 9). Among the tested compounds, only (20) and (19) were the ones with the closest scores of −6.96 and −6.58 kcal/mole, respectively. Although (20) maintained two similar ionic interactions with Asp215 and Arg462, its less hydrophilic nature only accommodated the formation of a lower number of hydrogen bonds than acarbose, which explains its lower score. This observation becomes more evident upon inspection of the interactions of (19), which has fewer hydrophilic groups capable of the formation of hydrogen bonding (Figures 10 and 11).
On the other hand, the docking results against β glucosidase were more intriguing. (15) marginally outperformed acarbose with scores of −9.10 and −8.70 kcal/mol, respectively. Additionally, compounds (14) and (16) attained nearly similar scores, achieving −8.60 and −8.50 kcal/mol, respectively. A closer inspection of the interactions explains why (15) achieved this score. Upon closer inspection, it binds more tightly to β glucosidase, the distance of the hydrogen bonds formed is optimal, ranging from 2.19 to 3.59 Å, and the hydrophobic bond was 3.76 Å with Tyr313. In contrast, acarbose formed hydrogen bonds ranging from 2.71 Å to 4.14 Å. Additionally, the binding of acarbose creates unfavorable binding and steric tension with Trp179 and Glu340 ( Figure 12). The combined effects of these two factors rationalize the marginal superiority of (15) over acarbose.    The impact of this unfavorable binding and hydrogen bond distance becomes more evident when viewing interactions of compounds (14) and (16) (Figures 13 and 14). In the case of (14), despite the short distance hydrogen bonds, there is unfavorable interaction with Glu235. On the other hand, there are no unfavorable interactions with (16), but the hydrogen bond distances are longer.

α and β Glucosidase Infarction
α/β glucosidases contributed to the treatment of type 2 DM by breaking down the glycosidic by the α isoform and the aryl and alkyl glycosides, disaccharides, and small oligosaccharides by the β isoform [32,59]. The binding of acarbose to the α and β isoforms was −8.97 and −8.70 kcal/mol, respectively. Although the compounds showed favorable binding in both cases, the binding was stronger in the β isoform in nearly all instances suggesting a partial preference for β rather than α. The hydrophilic nature of acarbose enables it to form multiple hydrogen bonds with several α glucosidase residues such as Tyr72, Tyr158, Phe178, Arg213, Asp215, Asp242, Gln279, Pro312, His351, Asp352, and Arg442 ( Figure 8). Additionally, ionic interactions were observed as well with Tyr158 and Asp242 in addition to one hydrophobic bond with Arg315 ( Figure 9). Among the tested compounds, only (20) and (19) were the ones with the closest scores of −6.96 and −6.58 kcal/mole, respectively. Although (20) maintained two similar ionic interactions with Asp215 and Arg462, its less hydrophilic nature only accommodated the formation of a lower number of hydrogen bonds than acarbose, which explains its lower score. This observation becomes more evident upon inspection of the interactions of (19), which has fewer hydrophilic groups capable of the formation of hydrogen bonding (Figures 10 and  11).  . Two-dimensional binding interaction of acarbose with α glucosidase. Figure 9. Two-dimensional binding interaction of acarbose with α glucosidase.

hSGT2 Interaction
Glucose reabsorption via the kidney is one of the contributing factors in type 2 DM, and as such, targeting this process is an intriguing prospect in antidiabetic therapy [60]. Human sodium-glucose co-transporter proteins are responsible for this machination and as such hSGT2 (PDBID: 7VSI) was selected, which also contained co-crystallized empagliflozin and was used for validation and comparison [61]. As shown in (Figure 15), the sugar moiety of empagliflozin is involved in many hydrogen bond interactions with Asn75, Phe98, Glu99, Ser287, and Lys321. Additionally, several hydrophobic interactions were also observed with His80, Leu84, Val95, Phe98, Tyr290, and Phe453. This plethora of interactions resulted in empagliflozin scoring −11.60 kcal/mol.  On the other hand, the docking results against β glucosidase were more intriguing. (15) marginally outperformed acarbose with scores of −9.10 and −8.70 kcal/mol, respectively. Additionally, compounds (14) and (16) attained nearly similar scores, achieving −8.60 and −8.50 kcal/mol, respectively. A closer inspection of the interactions explains why (15) achieved this score. Upon closer inspection, it binds more tightly to β glucosidase, the distance of the hydrogen bonds formed is optimal, ranging from 2.19 to 3.59 Å, and the hydrophobic bond was 3.76 Å with Tyr313. In contrast, acarbose formed hydrogen bonds ranging from 2.71 Å to 4.14 Å. Additionally, the binding of acarbose creates unfavorable binding and steric tension with Trp179 and Glu340 (Figure 12). The combined effects of these two factors rationalize the marginal superiority of (15) over acarbose. Although no compound was able to outperform empagliflozin, the closest binding was observed with (6) and (19), both scoring −8.80 kcal/mol. Several members scored −8.70 kcal/mole (7,8) and (20), and only (14) scored −8.50 kcal/mol. The 2D interactions of both (6) and (19) reveal their interactions with His80 and Tyr290 (Figures 16 and 17). Individually, (6) interacted with certain five amino acids as empagliflozin (Asn75, His80, Phe98, Tyr290, and Gln457) in addition to Leu283 while (19) interacted with only four similar amino acids (His80, Glu99, Ser287, and Tyr290) and Val157. explains why (15) achieved this score. Upon closer inspection, it binds more tightly to β glucosidase, the distance of the hydrogen bonds formed is optimal, ranging from 2.19 to 3.59 Å, and the hydrophobic bond was 3.76 Å with Tyr313. In contrast, acarbose formed hydrogen bonds ranging from 2.71 Å to 4.14 Å. Additionally, the binding of acarbose creates unfavorable binding and steric tension with Trp179 and Glu340 ( Figure 12). The combined effects of these two factors rationalize the marginal superiority of (15) over acarbose. The impact of this unfavorable binding and hydrogen bond distance becomes more evident when viewing interactions of compounds (14) and (16) (Figures 13 and 14). In the case of (14), despite the short distance hydrogen bonds, there is unfavorable interaction with Glu235. On the other hand, there are no unfavorable interactions with (16), but the hydrogen bond distances are longer.

hSGT2 Interaction
Glucose reabsorption via the kidney is one of the contributing factors in type 2 DM, and as such, targeting this process is an intriguing prospect in antidiabetic therapy. [60] Human sodium-glucose co-transporter proteins are responsible for this machination and as such hSGT2 (PDBID: 7VSI) was selected, which also contained co-crystallized empagliflozin and was used for validation and comparison [61]. As shown in (Figure 15), the sugar moiety of empagliflozin is involved in many hydrogen bond interactions with Asn75, Phe98, Glu99, Ser287, and Lys321. Additionally, several hydrophobic interactions

Molecular Dynamic Simulations and Generalized MMGBSA Calculations
Extensive investigation of the binding modalities and stability under realistic physiological settings was performed using molecular dynamic simulations. The proteins were simulated for 50 ns with and without the compounds using the Schrodinger Maestro package. The root mean square deviation (RMSD) of the protein-ligand complexes was calculated to ascertain the stability of the binding interactions, while the root mean square deviation (RMSD) of the ligands was used to assess the conformational changes the ligands undergo over the estimated simulation time period. Additionally, the root mean fluctuation (RMSF) of the amino acid residues and their contact with ligands was computed.
Analysis of the free amylase's trajectory reveals a relatively uniform behavior, as demonstrated by the nearly plateaued RMSD value of 1.40 Å (Figure 18). On the other hand, the effect of binding of compound (14) is observed as a consistent decrease in RMSD, indicating restriction of enzymatic movement and binding stability. Similarly, the same conclusion can be drawn when comparing the RMSF values of amino acid residues in the presence and absence of (14) and finding that fluctuations are restricted. In addition, (14) demonstrated conformational uniformity throughout the entire procedure with an RMSD of 0.80 Å.

Molecular Dynamic Simulations and Generalized MMGBSA Calculations
Extensive investigation of the binding modalities and stability under realistic physiological settings was performed using molecular dynamic simulations. The proteins were simulated for 50 ns with and without the compounds using the Schrodinger Maestro package. The root mean square deviation (RMSD) of the protein-ligand complexes was calculated to ascertain the stability of the binding interactions, while the root mean square deviation (RMSD) of the ligands was used to assess the conformational changes the ligands undergo over the estimated simulation time period. Additionally, the root mean fluctuation (RMSF) of the amino acid residues and their contact with ligands was computed.
Analysis of the free amylase's trajectory reveals a relatively uniform behavior, as demonstrated by the nearly plateaued RMSD value of 1.40 Å (Figure 18). On the other hand, the effect of binding of compound (14) is observed as a consistent decrease in RMSD, indicating restriction of enzymatic movement and binding stability. Similarly, the same conclusion can be drawn when comparing the RMSF values of amino acid residues in the presence and absence of (14) and finding that fluctuations are restricted. In addition, (14) demonstrated conformational uniformity throughout the entire procedure with an RMSD of 0.80 Å. As shown in Figure 19, α-amylase trajectory analysis revealed the interaction of (14) with Trp59 and Glu233 continuously in addition to the appearance of several other interactions with Asp197 and Ala198. As shown in Figure 19, α-amylase trajectory analysis revealed the interaction of (14) with Trp59 and Glu233 continuously in addition to the appearance of several other interactions with Asp197 and Ala198.
Trajectory analysis of the free α and β glucosidase shows relative homogeneity in behavior as demonstrated by its near plateau of RMSD at around 1.40 and 1.45 Å, respectively (Figures 20 and 21). On the other hand, the effect of binding of compounds (20) and (15) is observed as a consistent decrease in RMSD, implying the restriction of enzymatic movement and the stability of binding. Similarly, the same observation can be drawn when examining the RMSF values of the amino acid residues in the presence and absence of compounds, in which residues show limitation in fluctuations when (20) and (15) were present. Finally, both (20) and (15) exhibited conformational uniformity throughout the process as well with RMSD values of 0.75 and 0.70 Å, respectively.
Analysis of the various interactions of (20) across the whole simulation duration (Figure 22) illustrated the consistency with previous docking, in that the two vicinal hydroxyl groups were involved with Asp69 in addition to His112 and Arg442 throughout the simulation. On the other hand, several hydrophobic interactions of (15) were revealed with Tyr313 and Phe347. Trajectory analysis of the free α and β glucosidase shows relative homogeneity in behavior as demonstrated by its near plateau of RMSD at around 1.40 and 1.45 Å, respectively (Figures 20 and 21). On the other hand, the effect of binding of compounds (20) and (15) is observed as a consistent decrease in RMSD, implying the restriction of enzymatic movement and the stability of binding. Similarly, the same observation can be drawn when examining the RMSF values of the amino acid residues in the presence and absence of compounds, in which residues show limitation in fluctuations when (20) and (15) were present. Finally, both (20) and (15)     Trajectory analysis of the free α and β glucosidase shows relative homogeneity in behavior as demonstrated by its near plateau of RMSD at around 1.40 and 1.45 Å, respectively (Figures 20 and 21). On the other hand, the effect of binding of compounds (20) and (15) is observed as a consistent decrease in RMSD, implying the restriction of enzymatic movement and the stability of binding. Similarly, the same observation can be drawn when examining the RMSF values of the amino acid residues in the presence and absence of compounds, in which residues show limitation in fluctuations when (20) and (15) were present. Finally, both (20) and (15)    Similarly, analysis of molecular dynamic simulation of the hSGT2 without any ligand demonstrated a plateau RMSD around 3 Å, while both (6) and (19) reduced RMSD to 2.40 and 2.10 Å, respectively. Their binding was also reflected in RMSF values as shown in (Figure 23). Finally, both compounds (6) and (19)  Analysis of the various interactions of (20) across the whole simulation duration ( Figure 22) illustrated the consistency with previous docking, in that the two vicinal hydroxyl groups were involved with Asp69 in addition to His112 and Arg442 throughout the simulation. On the other hand, several hydrophobic interactions of (15) were revealed with Tyr313 and Phe347.
(A)  Similarly, analysis of molecular dynamic simulation of the hSGT2 without any ligand demonstrated a plateau RMSD around 3 A 0 , while both (6) and (19) reduced RMSD to 2.40 and 2.10 A 0 , respectively. Their binding was also reflected in RMSF values as shown in (Figure 23). Finally, both compounds (6) and (19) exhibited conformational uniformity throughout the process as well with RMSD values of 0.50 and 0.90 A 0 , respectively. Interactions of (6) and (19) were also analyzed throughout the simulation interval ( Figure 24); interactions with Phe98 and Tyr290 were the most frequent in both cases. Similarly, analysis of molecular dynamic simulation of the hSGT2 without any ligand demonstrated a plateau RMSD around 3 A 0 , while both (6) and (19) reduced RMSD to 2.40 and 2.10 A 0 , respectively. Their binding was also reflected in RMSF values as shown in (Figure 23). Finally, both compounds (6) and (19) exhibited conformational uniformity throughout the process as well with RMSD values of 0.50 and 0.90 A 0 , respectively. Interactions of (6) and (19) were also analyzed throughout the simulation interval ( Figure 24); interactions with Phe98 and Tyr290 were the most frequent in both cases. Interactions of (6) and (19) were also analyzed throughout the simulation interval ( Figure 24); interactions with Phe98 and Tyr290 were the most frequent in both cases. However, individually, the hydrophilic nature of (6) enabled the formation of hydrogen bonds Thr153 and Asp158. However, individually, the hydrophilic nature of (6) enabled the formation of hydrogen bonds Thr153 and Asp158.  Another tool for assessing the stability under solvated conditions as in physiological systems is the calculation of binding free energy. Among these tools, Molecular Mechanics Generalized-Born Surface Area (MM-GBSA) is one of the most frequently used methods deployed. The difference in solvent (water) interaction energy with the free receptor, free ligand, and complex is used to calculate the GB and SA energy terms. The molecular mechanics energy obtained from the interaction between the receptor and the ligand under the considered force field is used to compute MM [62]. The lower the predicted binding free energy of a ligand-protein complex, the more stable the complex will be and the greater the ligand's activity and potency (Table 4). For all simulations, the complexes maintained close energy scores at the beginning and end. This consistency of the binding energies of the targets to different compounds hints at stable binding throughout the simulation.

Eligibility Criteria for the Review
Studies were selected according to the isolated bioactive compounds from endophytic fungi associated with Morus species from 2008 to July 2022 and the biological activities conducted on these compounds during this period. The search spanned several databases such as PubMed and Web of Science.

Pharmacokinetic Profiling
The ADME profile provided by the SwissADME website (www.swissadme.ch; accessed on 6 September 2022) is an excellent web-based tool for the prediction of pharmacokinetic parameters [63,64]. Compounds were imported and predicted as demonstrated by the previous literature [65].

Molecular Dynamic Simulations and Generalized MMGBSA Calculations
The Schrodinger Desmond package was utilized for simulations of molecular dynamics utilizing the "OPLS4" forcefield for 50 ns, as detailed in previous studies [79]. The solvation was performed using "TIP3P" water molecules using an "Octadecahedron" solvation box. The binding free energy of the examined protein-ligand complexes was computed using the MM-GBSA method, which integrated molecular mechanics (MM) force fields with a Generalized Born and Surface Area continuum solvation solvent model using the Schrödinger Prime software [80,81].

Conclusions
The chronic nature of diabetes mellitus and its crippling effects on the quality of life drives the research for the identification of new agents to improve antidiabetic management. Traditional medicine provides an enormous source of medicinal plants and phytochemicals with established use. However, the environmental burden of using these plants increases the importance of finding alternative sources of bioactive molecules from eco-friendly endophytic fungi. Taking advantage of the antidiabetic effects of Morus plants, this study sought to explore the Morus endophytic fungal metabolites responsible for this property. The previous literature revealed a total of twenty-one compounds under this criterion. The pharmacokinetic properties of the compounds were calculated to narrow down the potential targets and ascertain their safety. The compounds showed safe properties with high intestinal absorption, low blood-brain barrier permeability, and no interactions with cytochrome P2D6. Expanding on these data, we evaluated the compounds' antidiabetic properties through their capability to affect local and systemic targets in the form of α/β glucosidase and human sodium-glucose cotransporter 2 (hSGT2), respectively. The compounds showed promising potential against all targets with varying degrees in terms of binding scores as well as the stability of such interactions. One of the most promising agents is Colletotrichalactone A (14); it inhibited α amylase and both isoforms of glucosidase with a greater preference for β than α. Moreover, it was among the top-scoring agents that inhibited hSGT2. This highlights its potential in antidiabetic management locally and systematically. Another candidate is Colletotrichalactone B (15) which outperformed acarbose inhibition on β glucosidase. The result of our study provides an in silico interpretation of the antidiabetic potential of Morus endophytic metabolites as well as providing sufficient evidence for future research on these agents and linking their pharmacological actions to the host, assuming that endophytic fungi are a more eco-friendly leading source of promising bioactive compounds than plant sources.