Fruitful decade of fungal metabolites as anti-diabetic agents from 2010 to 2019: emphasis on α-glucosidase inhibitors

In recent years the prevalence of diabetes has increased globally and by 2040 the number of diabetic people has been estimated to increase to 642 million. Various classes of drugs are available to treat Type ll diabetes. However, these drugs are associated with certain side effects. α-Glucosidase is an intriquing target enzyme to treat Type II diabetes, and α-glucosidase inhibitors are considered as first-line drugs for Type ll patients. Fungi, in general, produce natural products with some amazing chemical diversity and many fungal metabolites have illustrated a wide range of biological and pharmacological effects. In this review the focus is on describing the α-glucosidase effects and their potential as anti-diabetic agents of various metabolites isolated from fungi.


Introduction
Diabetes mellitus (DM) is a metabolic disorder associated with insulin resistance and the inability of the pancreatic b-cells to produce insulin, which leads to hyperglycemia. Moreover, hyperglycemia is associated with polyuria, weight loss, ketoacidosis, polydipsia, and other life-threatening health conditions (Usman et al. 2019). AGIs are an intriguing class of pharmaceutical drugs most often considered as firstline antidiabetic drugs for Type ll patients (Hossain and Pervin 2018;Usman et al. 2019). However, in some studies, it has been reported that AGIs can be employed as second-line antidiabetic drugs if these inhibitors are used as combination therapy with metformin (Chan et al. 2018).
The a-glucosidase enzyme (EC 3.2.1.20) has been considered as an important therapeutic target to treat carbohydrate mediated illnesses. It is well known that the secretion of a-glucosidase occurs in the small intestine and this enzyme catalyzes the cleavage of disaccharides and oligosaccharides into monosaccharides in the final step of carbohydrate digestion. Moreover, the conversion of complex carbohydrates into monosaccharides enhances the glucose body level (Abbas et al. 2019). Numerous studies have demonstrated that a-glucosidase inhibitors slow down the absorption and digestion of carbohydrates and therefore reduce the postprandial blood glucose concentrations which thus require less demand for insulin. a-Glucosidase inhibitors are considered mild compared to other oral antihyperglycemic agents because of their confined operation in the intestine rather than regulating various complex biochemical operations within the body (Abbas et al. 2019).
Fungi are considered one of the richest sources of natural products among living organisms because fungi have a unique metabolic system and can synthesise various types of natural products with quite intriquing chemical diversity (Srivastava 2019). After the discovery of penicillin (penicillin F) in 1929 by Alexander Fleming, substantial research on fungi lead to the isolation of thousands of new fungal metabolites with a diverse range of biological and pharmacological effects (Srivastava 2019). Besides, penicillins (antibacterial), echinocandin B (antifungal), cyclosporin A (immunosuppressive, and lovastatin (cholesterol-lowering) are all fungal originated and marketed pharmaceutical drugs. Furthermore, these factors evidently illustrate the significance of fungal metabolites to be a sustainable resource for new pharmaceutical agents.
Currently, acarbose and miglitol are two commercially available drugs with activity described as aglucosidase inhibitors (AGIs). These pharmaceutical drugs furthermore inhibit the absorption of carbohydrates from the gut and thus these anti-diabetic drugs are either administered alone or in combination with insulin (Bhatia et al. 2019;Hung et al. 2012). However, serious gastrointestinal damage and liver injuries have been associated with the use of these AGIs, and this has restricted their clinical usage (Yin et al. 2014;Kao et al. 2016;Usman et al. 2019). Therefore, there is a crucial need to discover and develop new and safer anti-diabetic drugs with low toxicity. Natural products for this purpose would be ideal if they could combat such diseases without creating other secondary health issues. Plant secondary metabolites have especially been widely studied for their potential anti-diabetic properties. It is thus conceivable that diverse compounds derived from fungal sources could be developed or transformed into new therapeutics against diabetes. This review article describes some small molecules isolated in the last decade (2010 to 2019) from various fungi and developed as inhibitors of a-glucosidase, and thus represent potential anti-diabetic drug leads.

Alkaloids
Alkaloids are nitrogen containing natural products and have been recognized substances in the treatment of human diseases for many years (Pervaiz et al. 2016;Rehman and Khan 2016). Literature indicated that a significant number of alkaloids from plant sources have been identified as a-glucosidase inhibitors (Yin et al. 2014). On the other hand, a number of alkaloids have been reported from various fungi possessing a diverse range of biological effects (Mahmood et al. 2010). Some alkaloids have also been reported from various fungi, which illustrated a-glucosidase inhibition. Valle et al. reported that benzomalvin A (1), quinolactacins A1 (2), A2 (3) in a mixture between B (4) and asperphenamate (5) (Fig. 1) were produced from the fungus Penicillium spathulatum (Valle et al. 2016).
Isoindolinone type alkaloids, viz., the sterenins A-C (6-8) (Ito- Kobayashi et al. 2008;Wang et al. 2014) and K-M (9-11) Wang et al. 2014) (Fig. 2) were reported from the fungus Stereum sp. and tested for their potential as a-glucosidase inhibitors. The chemical diversity generated among these alkaloids is mainly due to the different substituent groups on nitrogen (Wang et al. 2014). These compounds possess a-glucosidase effects with IC 50 : 3.31 to 36.6 lM (Table 1). Notably, alkaloid 8 illustrated potent inhibition with IC 50 : 3.31 lM. SAR studies demonstrated that the different substituents on nitrogen play a most important role and when there is no substituent (R = H) the activity was enhanced (alkaloid 8). On the other hand, various alky substituents attached to the nitrogen dramatically decreased activities (Wang et al. 2014). Penicidone C (12) was reported from Penicillium sp. (He et al. 2019;Ge et al. 2008) and demonstrated a-glucosidase inhibition with a low IC 50 : 51.9 lM (He et al. 2019).
Chermesinone A (15), isolated from the fungus Penicillium chermesinum, illustrated a-glucosidase effects with IC 50 : 24.5 lM (Huang et al. 2011). In this regard, compound 15 was also reported from the   (He et al. 2019) and was shown to possess a-glucosidase effects with IC 50 : 33.8 lM.
p-Terphenyls p-Terphenyls bearing a C-18 tricyclic or polycyclic aromatic core demonstrate a huge chemical diversity generated among these compounds by suitable changes in the middle aromatic ring or the linkages between the rings . Over 230 p-terphenyl analogs have been reported as well as a number of isolated derivatives . The majority of these compounds were reported from fungi Quang et al. 2003;Liu et al. 2004;Lee et al. 1996;Nagasawa et al. 2014). The three p-terphenyls 20-22 (Fig. 4) were isolated from the fungus Penicillium chermesinum and illustrated a-glucosidase effects with IC 50 values of 2.5, 4.9, and 0.9 lM, respectively (Table 1). Notably, the activity of these compounds was higher than the reference genistein (IC 50 : 9.8 lM, Huang et al. 2011). Previously, compound 20 was reported from the fungus Penicillium raistrickii (Belofsky et al. 1998). p-Terphenyls 23-32 were produced by the fungus Sarcodon leucopus and their structures were determined via extensive NMR techniques . All compounds displayed a-glucosidase effects with IC 50 values ranging from 0.58 to 35.0 lM. Among these, sarcoviolin b (23) showed good and potent effects with an IC 50 : 0.58 lM followed by an isomer of 24. Compounds 24 and 27-32 illustrated moderate inhibition with IC 50 values ranging from 1 to 10 lM (Table 1). On the other hand p-terphenyls 25 and 26 possess relatively weak effects with IC 50 values of 35.0 and 19.0 lM, respectively. An SAR analysis demonstrated that the configuration at N-1b and C-2b greatly effects the a-glucosidase activity. For instance, metabolites 23 and 32 having the cis N-1b and C-2b displayed better activity than metabolites 24 and 30 bearing a trans configuration .

Depsidones
The depsidone talaromyone B (54) was reported from the fungus Talaromyces stipitatus (Cai et al. 2017) while purpactin A (55) was obtained from the fungi T. stipitatus (Cai et al. 2017) and Penicillium sp. Nishida et al. 1991). In addition, tenellic acid A (56) (Fig. 7) was isolated from the fungi T. stipitatus (Cai et al. 2017) and Camposporium quercicola (Wang et al. 2008) and its structure was established via NMR spectroscopic methods as well as employing the Mosher's protocol. Compounds 54-56 illustrated moderate a-glucosidase activity (Table 2)  Six further depsidones viz., botryorhodines E-G (57-59), botryorhodine A (60), B (61), D (62) were reported from the fungus Meyerozyma guilliermondii and all illustrated significant a-glucosidase potentials with IC 50 : ranging from 2.1 to 15.4 lM ( Table 2). Compound 62 was the most active with an IC 50 value of 2.1 lM followed by compounds 58, 61 and 59, which is significantly lower than that of acarbose (IC 50 = 553.7 lM). An SAR analysis suggests that the hydroxymethyl group at C-3 increases the a-glucosidase effects, whereas, the presence of a methyl group at C-3 0 exerts no additional effect on the a-glucosidase inhibitory power of compounds 57-62 (Chen et al. 2015a). Another fungus Trichoderma sp. produces botryorhodines G (59) and H (63) (Fig. 8) and compound 59 possess moderate a-glucosidase effects with an IC 50 : 54.1 lM. On the other hand, compound 63 illustrated potent a-glucosidase effects with IC 50 : 8.1 lM and its activity was higher than the standard acarbose IC 50 : 703.8 lM . Moreover botryorhodines C (64) and D (62) possess significant a-glucosidase effects with IC 50 : 11.2 and 10.3 lM  respectively and these metabolites were reported from the fungi Trichoderma sp.  and Botryosphaeria rhodian (Abdou et al. 2010). All these compounds illustrated aglucosidase activity with IC 50 values ranging from 8.1 to 54.1 lM (Table 2). Compared to the reference drug acarbose (IC 50 = 703.8 lM), it can be concluded that compounds 62-64 are potent anti-diabetic depsidones. An SAR study showed that C-3 groups could affect a-glucosidase effects (compound 63 vs. 62 vs. 59) while on the other hand the C-3 0 methyl group exerts no effect on the activity (compound 64 vs. 62) .

Isocoumarins
Isocoumarins are a class secondary metabolites bearing a lactone core and these compounds feature a wide range of chemical diversity with a most diverse range of biological effects. Notably, some isocoumarins have been entered into clinical trials for cancer and other diseases (Hampl et al. 2011;Yin et al. 2001;Salloum et al. 2000;Pochet et al. 2004). The fungus Talaromyces amestolkiae produced the library of isocoumarins (65-79) ( Fig. 9) which were evaluated for their a-glucosidase inhibition. Isocoumarins 66, 70, 72 and 74 illustrated the most encouraging effects among all the tested compounds with IC 50 values ranging from 17.2 to 89.4 lM ( Table 2). Furthermore metabolites 65, 75, 78 and 79 are interestingly, five- fold more potent (IC 50 range: 140.8 to 193.1 lM) than acarbose (IC 50 : 958.3 lM) ). Since metabolites 69, 71, 73, 76 and 77 illustrated moderate inhibition with IC 50 ranging from 266.3 to 431.4 lM, respectively, it may be concluded that the 4-CH(OH)CH 3 moiety in metabolites 68, 69 and 73 decreases the activity level when compared with the activties with metabolites 70 and 71 bearing a 4-hydroxymethyl group. The compounds comprising an alkyl chain with the carbonyl group (as in compound 74) appear to possess higher activity in competition with compounds 75-79 which only have an hydroxyl moiety at the corresponding position   (Saeed 2016;Baba et al. 1991). Asperisocoumarin E (81) featured an isopentenyl group having two adjacent carbonyl moieties and asperisocoumarin F (82) represents an isocoumarin dimer via an ether linkage (Xiao et al. 2016). Furthermore, the same fungus Aspergillus sp. produced other isocoumarin analogues viz., asperisocoumarins C (83), E (84), F (85), G (86), I (87), J (89) (Fig. 10) along with compound 88 and their structures were all established via extensive NMR spectroscopic techniques. It was found that metabolites 83, 86, 87, and 89 illustrated significant a-glucosidase activities with IC 50 ranging from 38.1 to 78.1 lM. Since metabolites 84 and 85 displayed only moderate inhibition (Table 3), it may be concluded that either an epoxide and/or an hydroxyl group at C-10/C-11 could enhance the a-glucosidase effects (Cai et al. 2018).
b-Resorcylic acid derivatives b-Resorcylic acid analogs have been reported from a number of natural sources and these compounds were reported to display a diverse range of biological activities viz., antimicrobial (Yang et al. 2006), antimalarial (Xu et al. 2010), cytotoxic (Buayairaksa et al. 2011) and kinases and ATPases inhibitions (Shen et al. 2015). In this regard, the fungus Lasiodiplodia sp. produced b-resorcylic acid analogs 97-99 (Fig. 12) and their structures were established by 1D, 2D NMR, and X-ray crystallography in addition to the Mosher protocol (Chen et al. 2015b). All the compounds were evaluated for their a-glucosidase effects and illustrated inhibition with IC 50 values ranging from 15.2 to 24.6 lM (Table 3) and interestingly, their activities were better than the standard acarbose (IC 50 = 368 lM) (Chen et al. 2015b). Moreover compound 100, which was reported from the fungus Lasiodiplodia sp. (Chen et al. 2015b) and the yeast Saccharomyces cerevisiae , proved to be a most potent a-glucosidase inhibitor with an IC 50 : 10.1 lM (Chen et al. 2015b).
Compound 101, initially reported as a plant metabolite viz., from Euphorbia splendenfs (Lee et al. 1982) and later from the fungus Lasiodiplodia sp. (Chen et al. 2015b), illustrated a-glucosidase effects with IC 50 : 32.5 lM (Chen et al. 2015b). Compound 102 was obtained from the fungus Lasiodiplodia sp. (Chen et al. 2015b;Aldridge et al. 1971), and compound 103 from the fungi Lasiodiplodia sp. (Chen et al. 2015b) as well as ZZF36 (Yang et al. 2006). Both compounds inhibited a-glucosidase activity with IC 50 : 13.6 and 35.9 lM respectively (Chen et al. 2015b). SAR studies for compounds 97-103 demonstrated that the C-3 hydroxyl moiety in these compounds enhances the inhibitory effects (compound 99 vs. 100 and compound 101 vs. 102)  (Table 3). Furthermore, when comparing the activity of compound 97 and 102, it may be assumed that the aliphatic macrocyclic acid system is also necessary for the activity (Chen et al. 2015b). Another 12-membered ring system comprising a b-resorcylic acid analog, lasiodiplactone A (104), reported from the fungus Lasiodiplodia theobromae comprises an interesting and unique 12/6/6/5 tetracyclic system. Compound 104 inhibited a-glucosidase with an IC 50 : 29.4 lM which is better than the clinical acarbose (IC 50 = 367 lM) ).
The xanthone derivatives (?)-and (-)-ascomlactone A (117a and 117b) (Fig. 16) are produced by the fungus Ascomycota sp. and their structures were established based on spectroscopic methods. The activity studies revealed that as a racemic mixture, these metabolites inhibited the activity of a-glucosidase with an IC 50 value of 36.1 lM. On the other hand, and interestingly, the pure enantiomer, (-)-ascomlactone A (117b) illustrated better inhibition with an IC 50 value of 27.9 lM, whereas, the enantiomer (?)ascomlactone A (117a) demonstrated slightly weaker inhibition (IC 50 = 63.7 lM). When compared to the activity of acarbose (IC 50 = 477.0 lM), both pure enatiomers as well as the racemate is more potent . Chrysoxanthone (118) was obtained from the fungus Penicillium chrysogenum and illustrated aglucosidase effects with an IC 50 of 0.04 mM (Wang et al. 2018). The fungus Daldinia eschscholtzii produced a small library of tetralones viz., daldiniones B-D (119-121) and tetralones 122-124 (Fig. 17) and their structures were established by NMR, GIAO based NMR and ECD spectroscopic methods. All these compounds illustrated a-glucosidase effects with IC 50 values ranging from 21 to 155 lM (Table 5). Among these compounds, metabolite 124 was the most active (IC 50 : 21 lM) followed by compounds 121 (IC 50 : 35 lM) and 119 (IC 50 : 38 lM) . Moreover, Daldinia eschscholtzii also produced the chromone analogs, 125-128, which were also tested for their aglucosidase effects. All these compounds inhibited aglucosidase with IC 50 : ranging from 13 to 100 lM (Table 5). Notably, chromone analogs 125 (IC 50 : 13 lM) and 128 (IC 50 : 15 lM) proved to be the most active metabolites among the tested chromones .
Further butenolides, compound 142 and rubrolide S (143) (Fig. 20) were obtained from the fungus A. terreus and demonstrated a-glucosidase effects with IC 50 values of 24.8 and 1.2 lM, respectively. Additional enzyme kinetic studies were performed with compound 143, which revealed that it is an anticompetitive inhibitor with a Ki value of 1.42 lM . Another fungus varietal of Aspergillus sp., viz., Aspergillus flavipes produced antidiabetic avipesolides A-C (144-146) and butanolide 147 and these compounds inhibit a-glucosidase with IC 50 values ranging from 34 to 95 lM (Table 6).
Appropriate enzyme kinetic studies have also been carried out which revealed that compound 147 (K i : 0.43 lM) is a noncompetitive inhibitor while compounds 144-146 were demonstrated to be more competitive (K i : 2.5, 3.4, and 9.2 lM respectively) (Wang et al. 2016).
The fungus Penicillium chrysogenum produced diphenl ethers, chrysines B (159), C (160), compound 161, and methyl chloroasterrate (162), together with benzophenone, mono-chlorosulochrin (163) (Fig. 23). The group demonstrated that all these compounds illustrated a-glucosidase effects with IC 50 : ranging from 0.15 mM to 0.35 mM ( Table 6). Most of the compounds were more potent than the standard acarbose with an IC 50 of 0.28 mM (Wang et al. 2018). The fungus Aspergillus flavus produced compounds 164-167 whose structures were established via spectroscopic techniques. Compounds 166 and 167 exhibited significant inhibition potential with IC 50 values of 1.5, and 2.3 lM, respectively. Furthermore, metabolites 164 and 165 were only moderately active with IC 50 values of 165 and 129 lM, respectively. However, their effects were higher than acarbose which had an IC 50 : 840.2 lM) .

Conclusion
Fungi are known as prolific producers of diverse secondary metabolites. Fungal metabolites range from simple small molecules to more complex systems viz., proteins and polypeptides with a wide range of biological activities. Since the scurge of diabetes has increased worldwide, it is estimated that by 2040, the number of diabetic people will increase to 642 million around the globe. There is thus a critical need to find new antidiabetic drugs with less side effects. In spite of the fact that drug treatment for diabetes has improved over the last decade, drug resistance too has become an important issue in diabetic drug discovery and consequently new and improved strategies will have to be sought. One such strategy would be to either inhibit or decrease the manufacture of glucose in the small intestine. a-Glucosidase inhibitors can reduce the digestion of carbohydrates and thus could be considered as one of the most effective strategies to reduce post-prandial hyperglycemia. a-Glucosidase inhibitors are thus an important group of therapeutic agents to treat diabetes which currently comprises of three drugs viz., acarbose, miglitol, and voglibose (Derosa and Maffioli 2012). Numberous studies have been conducted to analyze the clinical efficacy and safety of acarbose, miglitol, and voglibose a-glucosidase with respect to glycemic control, atherosclerosis, and inflammation. It has been reported that these three a-glucosidase inhibitors are considered to be safe and effective both in monotherapy as well as in combination with additional anti-diabetic drugs (Dash et al. 2018;Derosa and Maffioli 2012).
In this context, the last decade has been noted to be a most fruitful period in isolating low molecular weight antidiabetic compounds from fungi. During this decade (2010 to 2019), more than 200 natural products have been isolated from various fungal sources and screened for their a-glucosidase inhibitory activity. Among the alkaloids, compound 8 exhibited an IC 50 value of 3.3 lM, which showed that the pyrolidine-2-one system coupled with a phenolic function are important functional groups for the development of antidiabetic drugs. Moreover, among p-terphenyls as a-glucosidase inhibitors, sarcoviolin b (23, IC 50 = 0.58 lM), 3,3 00 -dihydroxy-6 0 -Odesmethylterphenyllin (22, IC 50 = 0.9 lM) and concrescenin A (33, IC 50 : 0.9 lM) were found to be the most potent a-glucosidase inibitors. Depsides 43-45 showed a-glucosidase inibiton with IC 50 : 3-7.6 lM, and thus offer their strong candidature to be studied further for development as antidiabetic drugs. Among the depsidones, botryorhodine D (62, IC 50 = 2.1 lM) has been identified as a potential a-glucosidase inhibitor. Since two more depsides (63 and 64) also showed significant inhibtion, this fact revealed that the dioxipanone system is important and is also supported by other functional groups attached at C-3 to improve the a-glucosidase activity. A reasonable number (25) of isocoumarins have been idetified from various fungi as a-glucosidase inhibitors. Notably compound 93 (IC 50 : 0.027 mM) was found to be 35 fold more potent than the standard drug acarbose which indicated that a substituted tetrahydrofuran system might be playing a key role in the inhibitory activity. Among the benzofurans, 6-demethylpenisimplicissin (108, IC 50 : 9.5 lM) and 1 00 -epihydroxydihydrovermistatin (109, IC 50 : 8.0 lM) have also shown their potential as antidiabetic drug candidates. However, among the few quinone exmples, only compound 115 (IC 50-= 7.2 lM) exerted potential inhibition of a-glucosidase being higher than the positive control genistein (IC 50 = 13.6 lM), whereas, chromone analogs 125 (IC 50 : 13 lM) and 128 (IC 50 : 15 lM) proved to be the most active metabolites. Butanolide polyketides are another major class identified as antidiabetic fungal metabolites. A total of 29 compounds forming this group were isolated from various fungi during the last decade and almost all showed remarkable a-glucosidase inhibitory activity. Rubrolide S (143, IC 50-= 1.2 lM) with a Ki value of 1.42 lM has been identified as the most potent inhibitor. Further studies on this compound may lead to the development of a new and novel antidiabetic agent. Among the diphenyl ether dervatives, peniciaculin A (166, IC 50 : 1.5 lM) and expansol D (167, IC 50 : 2.3 lM) showed their potential as future drug candidates to treat diabetes mellitus. Other than these metabolites, 6 0 -Methyl-[1,1 0 -biphenyl]-3,3 0 ,4 0 ,5-tetraol (172, IC 50 : 2.2 lM), meroterpenoids 179 (IC 50 : 4.5 lM) and 180 (IC 50 : 3.1 lM), sesquiterpenoid; 7-deoxy-7,14-didehydrosydonol (205, IC 50 : 7.5 lM), tripalmitin (208, IC 50 : 3.75 lM) and dothiorelone I (211, IC 50 : 5.4 lg/mL) are also considered as powerful candidates as aglucosidase inhibitors. The above information clearly demonstrates that fungi are one of the most vital sources of novel substances with diverse structural features, which can be further explored as new and noval antidiabetic agents.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.