Key molecular scaffolds in the development of clinically viable α-amylase inhibitors

ABSTRACT

The escalating cases of type II diabetes combined with adverse side effects of current antidiabetic drugs spurred the advancement of innovative approaches for the management of postprandial glucose levels. α-Amylase is an endoamylase responsible for the breakdown of internal α-1,4-glycosidic linkages in dietary starch, producing oligosaccharides. Subsequently, α-glucosidase degraded these oligosaccharides to monosaccharides, which are absorbed into the bloodstream and become available to the body. The inhibitors of α-amylase reduced the digestibility of carbohydrates accompanied by delayed glucose absorption, leading to decreased blood glucose levels after meals and thus, inhibition of the enzyme seems to be a crucial strategy for diabetes management and improving overall glycemic control in diabetic patients. The present review article emphasizes the therapeutic promise of recently discovered potential α-amylase inhibitors, highlighting their in vitro, in silico and in vivo profiles. Ultimately, we addressed the contemporary challenges and potential routes ahead in the search for safe and reliable α-amylase inhibitors for clinical use, summarizing the most recent research in the field.

Graphical abstract

1. Introduction

Diabetes mellitus, a chronic progressive metabolic condition featured by postmeal or postprandial hyperglycemia, is ranked the most challenging global health concern. This abnormality in postmeal blood glucose levels is induced by the complex interplay of abnormalities in insulin secretion by β-cells in the pancreatic islets, insulin insensitivity, and a combination of genetic, environmental, and behavioral factors [1,2]. The risks of several complications like nephropathy, retinopathy, neuropathy, neurodegenerative, and cardiovascular disorders intensify with time in untreated diabetic patients [3–5]. Reducing diet-dependent blood glucose elevations is a renowned strategy for its control and management [6]. This can be achieved by limiting dietary carbohydrate digestion in the intestinal tract by inhibiting carbohydrate hydrolases (α-amylase and α-glucosidase) [7]. In the present review article, we focus on the clinically viable molecular scaffolds available in the literature possessing potential α-amylase inhibitory activity. So, we restrict our further discussions to α-amylases.

α-Amylases (α-1,4-glucan-4-glucanohydrolases, EC 3.2.1.1) are ubiquitous, and the familiar sources include humans and animals, plants, bacteria, yeast, and fungi [8]. α-Amylase family belongs to the class 13 of glycoside hydrolases (GH 13) comprising about 30 enzymes, accelerating the nonselective endo-hydrolysis of internal α-1,4-glycosidic bonds in carbohydrates retaining the α-anomeric configuration [9,10]. They exhibited optimum activity in the pH range of 6.7–7.0 and were better tolerant of the basic conditions than acidic [11]. The different hydrolytic patterns of marketed amylases resulted from structural variations and enzyme purity [12].

2. Human pancreatic α-amylase: the target

Human pancreatic α-amylase (HPA) is a calcium (Ca²⁺)-containing metalloenzyme of 57.6 kDa molecular weight and consists of 496 amino acid units arranged in a single peptide chain [13,14]. The enzyme contains three domains; the largest domain, A, contains residues 1–99 and 169–404. The smallest domain, B, having residues 100–168, serves as a Ca²⁺ binding site in which Asn100, Arg158, Asp167, and His201 interact with calcium to stabilize the structure. The remaining amino acid residues 405–496 constituted domain C which is loosely attached with A and B domains. The active site of the enzyme contains the catalytic triad Asp197, Glu233 and Asp300, which are involved in the hydrolysis of starch [15]. The hydrolytic reactions are believed to proceed through a double displacement mode. The anomeric center of the sugar undergoes a nucleophilic attack by the Asp197 residue, facilitated by the acidic catalytic roles of Glu233 and Asp300, resulting in the formation of a β-glycosyl enzyme intermediate. Later, this intermediate undergoes basic hydrolysis using Glu233 and Asp300 residues.

3. α-amylase inhibition

Complex carbohydrates such as starch are polymeric materials that cannot enter the bloodstream. α-Amylases hydrolyze the dietary starch molecules into simple oligosaccharides such as maltose, maltotriose, and α-limit dextrins [16]. Another enzyme, α-glucosidase, degraded the oligosaccharides into the monosaccharides like α-D-glucose, which entered the bloodstream and available to the body [17]. Thus, the blood glucose levels are dependent on the degree of carbohydrate hydrolysis. Considering the hydrolytic role of α-amylases in humans, their inhibition emerged as a crucial strategy in the realm of diabetes management and improving overall glycemic control in diabetic patients. The inhibitors diminish the hydrolysis of (α-1,4) glycosidic bonds and act as carbohydrate blockers [18], thus reducing the digestibility of carbohydrates accompanied by delayed glucose absorption, leading to decreased blood glucose levels after meals (Figure 1). To combat this and with the aim of improving life quality, several natural or synthetic drugs have been developed targeting these enzymes. Synthetic drugs, viz. miglitol, voglibose, and acarbose, have been exploited to inhibit these enzymes. However, these drugs displayed severe side effects viz. constipation, drug resistance, diarrhea, weight gain, and other gastrointestinal disturbances [19,20]. The hunt to augment the pool of available inhibitors stimulated researchers’ interest in developing safer α-amylase inhibitor molecules for regulating diabetes.

Figure 1.

Inhibition of starch digestion.

4. Classification of α-amylase inhibitors

α-Amylase is a key target for the regulation and management of diabetes. Generally, the inhibitors are grouped into two categories, namely, proteinaceous or peptide-based and non-proteinaceous inhibitors. Seven kinds of proteinaceous inhibitors were documented, six of them are of plant origin (the Kunitz type, knottin-like type, the cereal type, the γ- thionin-like type, the thaumatin-like type, and the lectin-like type) [21] and 7^th is of microbial type [22]. The non-proteinaceous inhibitors include, carbohydrate mimics [23], polyphenols [24], terpenoids [25], sulfonylureas [26], imidazoles [27], thiosemicarbazones [28], triazoles [29,30], thiazolidines [31,32], and metal complexes [33,34].

5. Scope of the review

Marketed α-amylase inhibitors have various side effects associated with them. Still, there is a need for more effective, long-lasting, and safer antidiabetic medications. In this review, we strived to compile the current state of α-amylase inhibitors, focusing on the recently discovered inhibitors in the last five years (2019–2024) with excellent α-amylase inhibition potential. The main focus is laid on the investigations documenting the in vitro and in vivo examinations. The literature studies were conducted using various search engines, including SciFinder, Reaxys, Google Scholar, and Web of Science. The contemporary challenges and future scopes for the development of effective clinically viable α-amylase inhibitors are also discussed. With the recent advancements in the inhibitory potential along with gaps in existing understanding of the α-amylase inhibitors, this article could pave the way for innovative solutions offering a distinct and advanced viewpoint merging structural and mechanistic viewpoints emphasizing clinical applicability. This up-to-date analysis is useful for the researchers engaged in the development of new α-amylase inhibitors.

6. Recent reports on α-amylase inhibitors

Antibiotic drugs tetracycline, kanamycin, lincomycin, erythromycin, and azithromycin have displayed efficacy against α-amylase in the in vitro assay (Figure 2). The drugs have inhibited the enzyme in a dose-dependent fashion with IC₅₀ values of 47.8 ± 3.2, 65.1 ± 1.6, 51.3 ± 4.1, 65.7 ± 3.8, and 103.6 ± 6.2 μM, respectively, compared with the acarbose (19.9 ± 0.6 μM) [35]. The most potent drug, tetracycline (1), has exhibited the mixed mode of inhibition with a K_i value of 8.4 ± 0.8 μM. The administration of tetracycline using in vivo experiments has resulted in significant blood glucose reductions compared with the control. Additionally, the drugs have exhibited binding energies of −8.8, −7.1, −6.7, −6.4, and −7.5 kcal/mol, respectively, with pig α-amylase (PDB ID: 1PIF).

Figure 2.

α-amylase inhibitors with IC₅₀ values.

Taxifolin (2), a flavanonol, has inhibited α-amylase competitively and reversibly with an IC₅₀ value of 1.555 mg/mL compared to acarbose (0.135 mg/mL) (Figure 2) [36]. Fluorescence quenching titrations have established the interaction between α-amylase and taxifolin, and the mode of quenching has been identified as static. The numerical values of the binding constant K_a and number of sites (n) are 6.65 and 1.39, respectively, showing a robust binding between the enzyme and taxifolin. The administration of 50 mg/kg of body weight taxifolin in rats has exhibited significant postprandial glucose reductions compared to control rats after one hour.

Anthocyanins, pelargonidin 3-O-rutinoside (Pg3R) (3), and malvidin 3-O-arabinoside (M3A) (4) have inhibited human pancreatic amylase with IC₅₀ values of 45.8 and 78.2 μM, respectively, compared to acarbose (IC₅₀ = 0.73 μM, K_i = 0.68 μM) [37]. Kinetic studies have revealed the competitive mode of inhibition for M3A, whereas mixed inhibition for Pg3R and acarbose (Figure 2), with a binding constant K_i of 24.8 and 13.2 μM. The ligand exhibited excellent size-wise binding of M3A (0.210) compared to acarbose (0.199) and Pg3R (0.159). Both the compounds M3A and Pg3R have been selected for the in vivo starch tolerance test in mice. Reduced postprandial glucose levels have been observed at 25 and 50 mg/kg doses. M3A has exhibited a strong effect compared to Pg3R at a dosage of 25 mg kg⁻¹. Since many side effects are associated with acarbose, authors have ruled out the side effects of blueberry anthocyanin extracts in clinical trials.

Tetrahydrobenzo[b]thiophen-2-yl)urea hybrids (5 and 6) have inhibited α-amylase with IC₅₀ values of 30.37 ± 0.58 and 56.14 ± 0.71 μM compared to acarbose (70.82 ± 0.89 μM) [38]. The hypoglycemic effects were further confirmed by in vivo assay, showing a lowering of blood glucose levels in the rats administered with these compounds at 5 mg/kg doses after 2 h (Figure 2). The liver cell lines, human normal hepatocyte (LO2), and human liver cancer (HepG2) have been screened using MTT assay, which revealed the non-cytotoxic nature of the compounds. The cell lines LO2 and HepG2 treated with 5, 6, and acarbose have exhibited cell viability of (94.7 ± 2.4, 95.5 ± 1.8, 100.4 ± 4.0%) and (96.3 ± 2.5, 96.0 ± 1.6, 95.2 ± 2.3%), respectively.

Aminothiazol-4(5H)-one derivative (7) has inhibited the pancreatic α-amylase with IC₅₀ value 5.98 ± 1.71 μM compared with acarbose (1.56 ± 0.71 μM) [39]. The antihyperglycemic effect of 7 has been examined by conducting in vivo experiments with alloxan-induced diabetic rats. There has been a notable reduction in intestinal α-amylase activity (21.3%) and blood glucose levels (39%) compared to the untreated rats on administration of 7 at 30 mg/kg daily for 21 days (Figure 2).

Rhodanine-clubbed spiro[pyrrolidine-2,3′-oxindole] derivatives (8a-d) have exhibited satisfactory in vitro inhibition with IC₅₀ values of 1.49 ± 0.10, 1.50 ± 0.07, 1.57 ± 0.10 and 1.59 ± 0.08 µM compared with the standard drug acarbose (1.56 ± 0.07 µM) (Figure 2). Administration in alloxan-induced diabetic rats on a daily basis has resulted in a significant reduction in the serum glucose levels by 6.8 (8a), 9.9 (8b), 12.3 (8c) and 5.5% (8d), respectively, compared with the untreated rats. After 1 h of administration, the glucose levels have been reduced to 251, 269, 281, and 267 mg/dL compared to untreated (315 mg/dL) [40].

1,3,4-Oxadiazole-2-thiol derivatives 9, 10, and 11 have demonstrated comparable α-amylase inhibitory activity with IC₅₀ values of 44, 52.15 and 52.11 µg/mL compared with acarbose (34.72 µg/mL) [41]. The in vivo studies with Drosophila melanogaster flies have shown a reduction in blood glucose levels of 9 (male: 1.2 mg/dL; female: 2.47 mg/dL), 10 (male: 1.22 mg/dL; female: 2.59 mg/dL) and 11 (male: 1.13 mg/dL; female: 2.43 mg/dL) compared to acarbose (male: 0.98 mg/dL; female: 1.8 mg/dL) at a given concentration of 18 μg/g feed (Figure 2).

Chondroitin sulfate (CS), a natural polysaccharide containing alternate D-glucuronic acid and N-acetyl-D-galactosamine units, has been extracted from different species and exhibited varied effects on the α-amylase activity [42]. The shark and porcine CSs have inhibited α-amylase in a dose-dependent manner with IC₅₀ values of 11.97 and 14.42 mg/ml, respectively, whereas cattle CS has not displayed any inhibition. This difference in inhibition potential has been attributed to the structural differences in CSs of different origins. Kinetic assessments have displayed a noncompetitive mode of inhibition. A bathochromic shift in the fluorescence emission spectra has disclosed the interaction between CSs and enzymes, leading to a conformational change. Thermodynamic parameters have gestured toward the hydrophobic interaction with the enzyme. The dissociation constant has confirmed the stronger binding of pig CS with the enzyme. The administration of shark and pig CSs in normal mice showed a reduction in blood glucose levels from 10.76 ± 1.92 to 5.78 ± 0.33 and 6.52 ± 0.71 mmol⁻¹ for shark and pig CSs after 120 min. These effects have also been observed in STZ-induced diabetic mice exhibiting a reduction in blood glucose levels from 20.9 ± 3.83 to 10.8 ± 1.73 and 13 ± 2.00 mmol/L after 120 min of administration.

Thiazolidinedione decorated quinoline hybrids 12 and 13 have exhibited 87.47 ± 0.94 and 81.08 ± 0.12% inhibition toward α-amylase w.r.t. standard drugs rosiglitazone (79.08 ± 0.24%), pioglitazone (90.60 ± 0.10%) and acarbose (92.55 ± 0.28%), respectively [43]. The administration of 12 and 13 in the streptozotocin-induced diabetic rats have demonstrated a significant reduction in the blood glucose levels by 43.7 ± 0.91 and 45.6 ± 0.28%, respectively, at the dosages of 50 mg/kg body weight, comparable with the standard pioglitazone (47.6 ± 0.31%) and rosiglitazone (44.9 ± 0.24%). A step further, blood lipid profiles after administering 12 and 13 have shown a prominent decrease in the cholesterol, LDL, VLDL, and triglyceride concentrations (Figure 2).

Thiazolidinedione-morpholine hybrid 14 has inhibited the enzyme IC₅₀ value of 19.51 ± 1.34 µM compared with acarbose (10.32 ± 1.02 µM) [44]. The in vivo examination has shown prominent dose-dependent blood glucose reductions in experimental mice to 19.2, 16.5, and 13.4 mmol/L compared with the standard glibenclamide (18.9, 15.7, and 12.5 mmol/L) at 500 µM/kg dosage after 1^st, 2^nd and 3^rd (Figure 2).

Benzothiazine derivatives 15 and 16 have competitively inhibited the α-amylase with IC₅₀ values of 75.17 and 29.10 μM, respectively, compared to acarbose (28.8 μM) [45]. Inhibition constants (K_i) and dissociation constant (K_i^’) for 15 and 16 have been 0.216, −1.619 and 1.128, 0.4721, respectively, w.r.t. acarbose (0.896, 1.67), highlighting the stronger enzyme inhibition. The diabetic mice administered with 15 showed reduced levels of fasting blood glucose (112 mg/dL), cholesterol (85 mg/dL), triglycerides (129 mg/dL), creatinine (1.9 mg/dL), Hb1Ac (4%), and improved insulin secretion (25 μM/L). Furthermore, 15 has exhibited a lower necrosis rate in different organs (liver, kidney, and pancreas) than 16 (Figure 2).

Swertisin 17, a flavonoid glycoside, has exhibited a weaker α-amylase inhibition effect than acarbose, with IC₅₀ values of 1.894 and 0.137 mg/ml, respectively [46]. Computational and kinetic investigations have confirmed the competitive mode of inhibition. The interaction between α-amylase and swertisin has been confirmed by a decrease in the fluorescence intensity measured at λ_ex of 280 and λ_em 350 nm. The absorption maxima at 205 nm have exhibited a reduction in the absorbance intensity with increasing concentration of inhibitor, again establishing the interaction. The amide band of the peptide (enzyme) at 1667 cm⁻¹ has been shifted to 1671 cm^−1, confirming the role of the carbonyl group in the binding. α-Amylase and swertisin complex appeared aggregated and swelled compared with the free enzyme. Administration of swertisin (13.5 mg/kg body weight) in rats has revealed a meaningful reduction in the blood glucose levels after 30, 60 and 90 min (Figure 2).

Thiazolidinedione derivatives 18, 19 and 20 have inhibited α-amylase with IC₅₀ values of 18.02, 10.26, and 20.01 µg/mL compared to standard acarbose (24.1 µg/mL) [47]. All the compounds have significantly reduced blood glucose levels at a dosage of 250 µg/kg compared to the normal group. Two derivatives, 18 and 20, have exhibited decreases in blood glucose levels at the 15^th and 30^th treatments (Figure 3). In contrast, 19 and 20 have demonstrated a noteworthy diminution in glucose levels after the 30^th treatment. Compound 20 has significantly decreased blood glucose without mortality at 50 µg/kg. Additionally, the antihyperlipidemic effect has been improved at a dose of 50 µg/kg.

Figure 3.

Promising α-amylase inhibitors.

The methanolic extract of Quercus coccifera (Oaktree) leaves extracts has exhibited better α-amylase inhibition (IC₅₀ value of 0.17 mg/mL) compared with acarbose (0.59 mg/mL) [48]. The in vivo studies have shown reduced blood glucose levels to 146.8 mg/dL after 28 days (comparable to glibenclamide) on administering 200 mg/kg/day in the Swiss albino mice with normal body weight.

Thiazolidine-2,4- dione hybrids 21a-e and 22a-e have strongly inhibited α-amylase compared to acarbose (10.32 ± 1.03 μM) [49]. In vitro glucose uptake results revealed improved cellular ability to take glucose with compounds 22a-b and 21b-c. STZ-induced diabetic mice administered with 22a-b and 21b-c displayed dose-dependent blood glucose level reduction, and the effects persisted up to 4 h (Figure 3). Compound 21b has exhibited maximum glucose reductions (46.20 and 46.49%) at dosages of 2 and 4 mg/kg body weight, and over the 14-day protocol, 54.87 and 56.19% glucose reductions observed compared with standard (55.23%). Furthermore, diabetic-induced rats treated with compounds 22a-b and 21b-c plus glibenclamide have exhibited prominent reductions in high triglycerides, total cholesterol, very-low-density lipoprotein, and low-density lipoprotein levels and a rise in high-density lipoprotein levels, consequently regulating lipid levels, as compared to the diabetic group. Glycosylated hemoglobin levels have been significantly reduced for 22a and 21b-c at a dosage of 4 mg/kg and 2 mg/kg body weight compared to the diabetic group, except for compound 22b. The compound 21b has exhibited 80-fold higher activity against α-amylase.

Curcumin-fused aldohexoses have inhibited α-amylase with IC₅₀ values of 17.9 ± 0.5 (23), 15.7 ± 0.7 (24), 19.8 ± 1.6 μM (25) compared with acarbose (8.5 ± 0.5 μM) [50]. The non-linear Michaelis-Menten regression has displayed a mixed type of inhibition. The higher inhibition constant K_I (enzyme-substrate complex) than K_i (free enzyme) has shown a stronger affinity of curcumin derivatives for α-amylase. The reduction in the fluorescence intensity has obtained evidence for the interaction of derivatives with enzyme observed at 349 nm in the free enzyme. The enhancement in the Stern-Volmer constant (K_SV) with temperature has demonstrated the existence of dynamic quenching. Compound 24 showed a high binding constant (10⁷ M⁻¹), and a single binding site confirmed stronger affinity. Negative values of thermodynamic parameters ΔH°, ΔS°, and ΔS° have demonstrated exothermic and spontaneous binding with Van der Waals and hydrogen interactions. The alterations in the FT-IR region 1600–1700 cm⁻¹ (amide region-1) with increased concentrations have shifted to a higher wavelength, highlighting the rearrangement of H-bonding to the carbonyl group and confirming the inhibition. In vivo experiments have exhibited comparable hypoglycemic activity of curcumin and 24 at 100 and 50 mg/kg dosages, respectively, reducing glucose entry into the bloodstream (Figure 3).

Metformin (26) is a well-established and commercially used α-amylase inhibitor. The Metformin derivatives (27–31) have inhibited α-amylase with IC₅₀ values of 127.63 ± 21.88, 67.97 ± 7.98, 253.44 ± 15.43, 120.46 ± 16.52 and 243.72 ± 11.11 μg/mL compared to Metformin (482.91 ± 21.86 μg/mL) [51]. Compound 31 exhibited the highest reduction in blood glucose levels, 40 and 46%, at 50 and 100 mg dosages in streptozotocin-induced diabetic mice (Figure 3). Additionally, 31 has surpassed the Metformin side effect of lipid profile alteration since it reduced triglyceride levels by 15% (50 mg) and 17% (100 mg) and 10% decrease in cholesterol levels at 50 mg. Diabetes is often linked with lipoprotein abnormalities such as decreased HDL (high-density lipoprotein) and enhanced LDL (low-density lipoprotein). The derivatives 30 and 31 have improved HDL levels by (61 & 65%), and (40.5 & 52%) at 50 and 100 mg doses, respectively. Furthermore, LDL levels decreased to 8.4% compared to the control group at a 100 mg dosage. Compounds 30 and 31 have displayed decreased γ-glutamyl transferase activity, suggesting reduced liver damage and, thus, beneficial over-marketed Metformin.

Benzodioxol carboxamide derivatives 32 and 33 have inhibited α-amylase with IC₅₀ values of 0.85 ± 0.28 and 0.68 ± 0.25 µM, respectively, compared to acarbose (1.55 ± 0.85 µM) [52]. Compound 33 has been explored using in vivo hypoglycemic effects in STZ-induced diabetic mice (Figure 3). The in vivo examinations have shown substantial blood glucose reductions from 252.2 mg/dL to 173.8 mg/dL after administration of 5 doses of 33 (10 mg/kg).

Thiazolidinedione-1,3,4-oxadiazole derivatives 34-38 have strongly inhibited α-amylase with IC₅₀ values of 18.61 ± 0.22, 22.60 ± 0.34, 23.19 ± 0.31, 21.12 ± 0.23, 18.42 ± 0.21 μM compared with acarbose (24.35 ± 1.44 μM) [53]. The male and female flies treated with 34, 35 and 38 have exhibited blood glucose reductions of (27.7, 28.4, and 29.05%) and (12.1, 7.03, and 37%), respectively. The acarbose has an inhibition of 62%. By increasing the dosage of the compounds, the antihyperglycemic activity can be improved. Additionally, compound 38 has exhibited greater cell toxicity (83%) on NIH-3T3 cells in the MTT assay (Figure 4).

Figure 4.

α-amylase inhibitors with heterocyclic skeleton.

Rosiglitazone-based heterodimers 39 and 40 have exhibited greater in vitro α-amylase inhibition than acarbose [54]. The compounds 39, 40 and acarbose displayed an IC50 value of 24.02, 29.88 and 33.08 μM, respectively (Figure 4). The derivatives have also exhibited an increase in inhibition when concentration was increased. The compounds have shown antioxidant activity comparable to ascorbic acid. Under in vivo diabetic models, the rats treated with 39 and sitagliptin have shown a significant decrease in plasma glucose levels, suggesting its antihyperglycemic activity.

Devi et al. [55] have screened naphtho[2,3-d]imidazole-4,9-dione appended 1,2,3-triazoles against α-amylase which displayed IC₅₀ values ranging from 17.83 ± 0.14 to 26.00 ± 0.17 μg/mL compared to standard acarbose (18.81 ± 0.05 μg/mL). The derivative 40 has demonstrated the highest in vitro activity (17.83 ± 0.14 μg/mL) and docked in the active site of A. oryzae α-amylase (PDB ID: 7TAA) to get insight into the binding interactions (Figure 4). Molecular dynamics simulations have supported the stability and favorable interactions of compound 40 with the enzyme’s active site. The compounds have shown DPPH free radical scavenging activity similar to that of standard BHT. The study has shown favorable drug-likeness and potential for oral absorption without CNS toxicity in ADME evaluations.

Singh and colleagues [56] have reported the α-amylase inhibition activity of various thiazolidine-2,4-dione derivatives and found that 3-ethoxy-4-methoxy substituted derivative (41) has exhibited the strongest inhibition (IC₅₀ = 10.19 ± 0.25 μM) compared to acarbose (IC₅₀ = 22.57 ± 2.30 μM). Two other compounds, 42 (4-Br, 23.91 ± 0.89 μM) and 43 (2,4-Cl, 24.07 ± 1.56 μM) have also significantly inhibited the enzyme activity (Figure 4). Under in vivo trials, compound 41 has significantly reduced the blood glucose and lipid levels in STZ-induced diabetes in the Wistar rats. It has significantly reduced the levels of many stress indicators and inflammatory cytokines and the levels of hemoglobin, monocytes, basophils, and other immune cells. Additionally, under a high glucose-induced diabetic milieu, compounds 41 and 43 significantly reduced ROS formation in the PINC-1 cell line and were found to exhibit cytotoxicity toward the pancreatic cell line (PANC-1).

Thiazolidinedione hybrids 44, 45, 46 have inhibited α-amylase with IC₅₀ values of 113.91 ± 1.4, 128.84 ± 0.23, and 86.06 ± 1.1 μM, respectively, compared to acarbose (26.89 ± 3.12 μM) [57]. In vitro glucose uptake assays have demonstrated significant glucose uptake capacity using yeast cells (Saccharomyces cerevisiae) with compounds 44, 45, 46 showing uptake of 55.23 ± 0.14, 52.25 ± 0.14 and 58.23 ± 0.13% compared to standard insulin (54.62 ± 0.13%) and pioglitazone (62.92 ± 0.12%). In vivo oral glucose tolerance tests in Wistar rats have demonstrated a rise in plasma glucose levels within the first 30 mins, followed by a decline up to 90 mins and a further improvement after 120 mins. Further, compound 46 has reduced blood glucose levels (114 ± 1.17 mg/dL), comparable to pioglitazone (102.2 ± 0.79 mg/dL) (Figure 4).

Fluorinated 1,3,4-oxadiazole derivatives 47, 48 and 49 have inhibited α-amylase with IC₅₀ values of 54.83, 64.95, and 64.78 µg/mL, respectively, compared to acarbose (35.17 µg/mL) [58]. The medication exposure of Drosophila melanogaster flies for 10 days has not yielded apparent changes in the mortality compared to acarbose and control, neglecting the harmful effects (Figure 4). The compounds have also exhibited antihyperglycemic effects in the hemolymph homogenate. The effects of 48 were comparable to the acarbose.

3,5-Disubstituted-thiazolidine-2,4-dione hybrids have displayed IC₅₀ values of 17.10 ± 0.015 and 9.2 ± 0.092 μM for 50 and 51 against α-amylase w.r.t. acarbose (22.57 ± 2.30 μM) [59]. The compounds have been found to be nontoxic toward the PANC-1 and INS-1 cell lines. The authors have investigated their effects on high glucose-induced α-amylase in PANC-1 cells and measured minimum fractional index (MFI) values of 4361 ± 257.83 and 2887 ± 95.87 for 50 and 51, respectively, w.r.t. control (70.7 ± 2.88 and 78.5 ± 1.93) highlighting increased expression of digestive enzymes (Figure 5). In the case of animal pancreas, the MFI values of 676 ± 4.85 and 2248 ± 101.44 for STZ + 50 and STZ + 51, respectively, w.r.t. control (83.5 ± 3.55). Additionally, streptozotocin-induced diabetic models in Wistar rats have been employed to investigate compounds’ in vivo antidiabetic effects. The blood glucose levels of 50 and 51 administered compounds have exhibited no variation compared with pioglitazone. After 60 mins of treatment, glucose levels have been reduced at the proposed dosage of 10 mg/kg of body weight and continued for 120 mins. The change in the blood glucose levels and body weights have been measured on 1, 4, 14, and 35 days. There has been a reduction in blood glucose levels on the 35^th day of the treatment, more dominant with 51 treated animals. The animals have also demonstrated improved PPAR-gamma levels. The cholesterol levels and levels of liver functioning in three enzymes, serum-glutamic-oxaloacetic transaminase, serum glutamic-pyruvic transaminase, and alkaline phosphatase, have also been improved in the treated animals. Additionally, the stress markers malondialdehyde and superoxide dismutase levels have been significantly reduced. The effects have been further confirmed by investigating blood parameters histology of the pancreas.

Figure 5.

Heterocyclic α-amylase inhibitors.

Thabet and colleagues [60] have investigated the α-amylase inhibitor activity of paracetamol-sulfonamide derivatives and found that compound 54 (IC₅₀ = 0.98 ± 0.015 μM) has the highest potency compared to standard drug acarbose (0.43 ± 0.009 μM), followed by compound 53 (1.55 ± 0.022 μM) and compound 52 (1.59 ± 0.023 μM). The docking studies of derivatives 52, 53, and 54 revealed binding energies of −10.19, −11.32, and −11.63 kcal/mol, respectively, as compared to the standard (−17.19 kcal/mol) with receptor α-amylase (PDB ID: 2QV4). Based on in silico toxicity projections, these compounds are comparatively less hazardous than acarbose, and furthermore, it has been discovered that the most potent diazo-sulfonyl paracetamol derivatives, 52, 53, and 54 have been ineffective against BBB-barrier toxicity, neurotoxic, cardiotoxic, immunotoxic, mutagenic, cytotoxic, ecotoxic, clinical, and nutritional toxicity with excellent probability values (Figure 5).

Luo and colleagues [61] have reported α-amylase (from hog pancreas) activity of carbazole- oxadiazole hybrids and found varied degrees of inhibitory activity with IC₅₀ values ranging from 45.531.50 to 126.14 ± 6.33 μM compared to acarbose (24.68 ± 1.10 μM). Compound 55 has been recognized as the most potent inhibitor (45.53 ± 1.50 μM), functioning as a mixed-type inhibitor with binding constant (K_α-amylase and K_{α-amylase-6e}) values of 45.05 and 63.28 μM, respectively. Additional mechanistic investigations, such as molecular docking and atomic force microscopy (AFM), have indicated that compound 55‘s inhibitory effect may be credited to hydrophobic contacts with α-amylase (S. cerevisiae, PDB ID: 3BAJ). Additionally, the sucrose loading test and ADME calculations of 55 have anticipated excellent oral bioavailability, postprandial blood glucose reduction, and negligible toxicity to HEK-293 cells (Figure 5).

Singh and coauthors [62] have evaluated the α-amylase activity of thiazolidine-2,4-dione tethered 1,2,3-triazoles, and it has been observed that five derivatives 56–60 exhibited comparable inhibition with the standard drug acarbose (IC₅₀ = 0.028 μmol/mL). The compound 56 with a 7-chloroquinolinyl substituent has demonstrated noteworthy inhibition (IC₅₀ = 0.040 μmol/mL). The in vitro inhibition data have been used for QSAR modeling, and multiple QSAR models have been built using the divided and undivided datasets. The best predictive model has shown robust prediction potential for unknown compounds (R²_ext = 0.8509), and the descriptors IDDE, GGI6, and L3s were positively correlated. The molecular docking has confirmed the excellent binding of 56 at the active site of α-amylase (PDB: 7TAA) with the help of hydrophobic interactions and molecular dynamics, revealing stability over a period of 100 ns. Further, ADMET studies have revealed favorable toxicity profiles and good oral bioavailability (Figure 5).

Mortada and colleagues [63] have examined the α-amylase inhibition of two pyrazole derivatives, 61 and 62, and observed that they have exhibited comparable inhibition with acarbose (IC₅₀ = 115.6 ± 0.574 µM) with IC₅₀ values of 95.85 ± 0.92 and 120.2 ± 0.68 µM (Figure 5). Molecular docking has displayed the binding energies of −6.7 and −5.6 kcal/mol for 61 and 62 at the receptor (PDB = 2GJP) active site compared with acarbose (−6.2 kcal/mol). Van der Waals interactions have been responsible for binding the two derivatives at the binding site. The ADMET property values fell within the recommended range.

Joshi and colleagues [64] have identified a competitive inhibitor of α-amylase 63 with a K_i and IC₅₀ value of 1.76 μM and 5.37 ± 0.25 μM demonstrating better efficacy compared to acarbose (6.40 ± 0.14 μM). Molecular docking of 63 with α-amylase (PDB ID: 1RPK) has displayed the role of H-bonds, hydrophobic and π-interactions with the active site amino residues. The 63-enzyme complex has been stabilized after 20 ns of the molecular dynamic simulation and persisted throughout the 100 ns. The in silico pharmacokinetic profiling and toxicity profiles of compound 63 have established its safety features (Figure 5).

Alzahrani and coauthors [65] discovered the outstanding α-amylase inhibitory activity of bis-indolylmethane analogs with IC₅₀ values ranging from 0.8 ± 0.05 to 15.3 ± 0.1 μM in contrast to acarbose (8.9 ± 0.1 μM). Analogues 64 (0.80 ± 0.05) and 65 (0.90 ± 0.10 μM) have displayed greater potency over acarbose, whereas the compounds 66, 67 and 68 displayed good to excellent potency (Figure 6). Docking simulations of 64 and 65 against α-amylase (PDB ID: 6GXV) have unveiled impressive binding affinity −25.40 and −24.62 kcal/mol, respectively, and have binding modes similar to acarbose. The imine group (-C=N) present in the structure has established robust binding affinity for both nucleophilic and electrophilic substituents within the active site. Additionally, the derivatives have also displayed promising ADMET predictions.

Figure 6.

Promising α-amylase inhibitors.

Semwal and Meera [66] have modified the kafirin, a prolamin protein in sorghum, employing papain and infrared thermal treatments (PIR) and have observed that the modified protein displayed superior radical scavenging properties and improved inhibition of the α-amylase. Kafirin has exhibited 56.98, 60.86, and 61% α-amylase inhibition at 5, 10 and 15 mg/mL, whereas the PIR sample has displayed 67.06, 69.38 and 71% inhibition at 5, 10 and 15 mg/mL. This enhancement in the inhibitory potential has been attributed to the additional π-π hydrophobic contacts resulting from thermal and enzymatic treatments. Adding modified kafirin to corn starch has raised the resistant starch levels from 5.09 to 21.04%, indicating that this protein could be utilized as a natural carbohydrate blocker for developing diabetic-friendly food formulation.

Khan and colleagues [67] have reported the α-amylase inhibitory potential of a series of Schiff bases with IC₅₀ values ranging from 0.80 ± 0.05 to 20.10 ± 0.40 μM, compared with acarbose having (10.30 ± 0.20 μM). The outcomes of this research have been noteworthy since nine scaffolds (69 to 77) demonstrated better potency than acarbose. Compound 77 has exhibited the highest inhibition (0.80 ± 0.05 μM), almost 13-fold more potent than the reference (Figure 6). Further insights about the binding mechanism gained via molecular docking studies with α-amylase (PDB ID = 4BFH) and H-bonding interactions have been found to be responsible for the efficient binding at the active site. The authors have also examined the toxicity profiles of the examined derivatives.

In another study, Ertunga and colleagues [68] evaluated the α-amylase inhibition potential of water-soluble Zn(II) phthalocyanine (78) and observed it as a moderate inhibitor with IC₅₀ value of 3.97 ± 0.32 µM compared to acarbose (1.64 ± 0.36 µM). The kinetic experiments suggested the competitive nature of inhibition with an inhibition constant (K_i) of 6.92 µM. The literature comparison reflected it as one of the first studies on the α-amylase inhibition of phthalocyanines (Figure 7).

Figure 7.

Promising α-amylase inhibitors.

Feunaing and colleagues [69] have identified 11 compounds (flavonoids, triterpenoids, ellagitannins, and other phenolic compounds) in the extract of Terminalia macroptera (Combretaceae) and studied their α-amylase inhibition. The phenolic compound myricetin-3-O-rhamnoside 79 has been found to be the most active against α-amylase with IC₅₀ value of 65.17±0.43 µg/mL compared to acarbose (32.25 ± 0.36 µg/mL) whereas 3,3′,4′-tri-O-methyl ellagic acid-4-O-β-D-glucopyranoside has the least potency (110.13 ± 0.80 µg/mL). Pentacyclic triterpenoids have been the least potent of the 11 phytochemicals, with IC₅₀ values greater than 145.91±0.56 µg/mL (Figure 7). Generally, the pure compounds have been more potent than the extract (164.71±1.12 µg/mL), gesturing toward the absence of a synergistic effect. The most potent constituent has a binding energy of −7.83 kcal/mol compared with acarbose (−14.46 kcal/mol).

Sciacca and coauthors [70] have screened the naturally occurring neolignan obovatol and its derivatives against α-amylase and have found that compound 80 (IC₅₀ = 23.6 ± 2.0 µM) and four neolignan derivatives 81 (12.7 ± 0.9 µM), 83 (27.0 ± 2.3 µM), 82 (35.4 ± 1.2 µM) and 84 (6.1 ± 1.0 µM) have exhibited greater efficacy with respect to acarbose (34.6 µM). Kinetic investigations have revealed a competitive mode of inhibition for the potent derivatives (Figure 7). Docking studies have shown the binding affinities of −4.48, −4.83, −5.60, −4.99, and −5.50 kcal/mol with α-amylase (PDB ID: 4W93) compared with acarbose (−8.33 kcal/mol), where H-bonding and π-π hydrophobic interactions have been found to be main interactions responsible for the binding. All the compounds have displayed moderate solubility and good intestinal absorption, and all the Lipinski rules have been obeyed except for those for 80 and 84 (lipophilicity >4.15). 84 has exhibited the highest binding (K_a = 3.10 ± 0.22 × 10⁴, 4.78 ± 0.54 × 10⁴ and 7.60 ± 0.07 × 10⁴ L/mol at 298.15, 303.15 and 308.15 K, respectively). Surface plasmon resonance imaging (SPRI) has also confirmed the excellent binding of 84 with the lowest value of dissociation constant (K_D = 8.34 ± 0.03 ×10⁻⁶ M), and the binding affinity has been decreased in the order of 84 > 81 > 80 > acarbose. This comprehensive study has promised exciting structural features for antidiabetic drug development.

Jan and colleagues [71] have screened a library of thiosemicarbazones against α-amylase and have observed that compounds 85, 86, 87, and 88 have been more potent than the acarbose (IC₅₀ = 21.55 ± 1.31 µM). The ligands have been docked into the binding pocket of the receptor (PDB ID: 3BAJ), and the binding energy has been found to be −7.2, −7.0, −6.9, and −7.1 kcal/mol for 85, 86, 87, and 88, respectively (Figure 7). TD-DFT studies have been conducted to understand the chemical nature of all the compounds, and it has been found that the frontier molecular orbitals of these three compounds exhibited similarity in their energy gap i.e., 6.07, 6.08, 5.88, and 6.07 eV for 85, 86, 87 and 88, respectively.

Xu and colleagues [72] have extracted the betanin (89) and phyllocactin (90) monomers from Hylocereus polyrhizus peel in excellent purity and evaluated them against α-amylase. The acarbose equivalent (ratio of IC₅₀ values between acarbose and the sample) of both the constituents betanin and phyllocactin have been found to be 0.540 ± 0.027 and 0.581 ± 0.042, respectively (Figure 7). Kinetic studies have evidenced a mixed type of inhibition by both the inhibitors. Thermodynamic parameters have been examined to investigate the binding of monomer-enzyme complexes. The values of ΔH and ΔS have been found to be negative and gestured toward the dominancy of H-bonds and Van der Waal forces for binding with enzymes in an enthalpy-driven process. The feasibility of the binding has been confirmed by the negative values of Gibb’s free energy (ΔG). Finally, both the compounds have been docked with the enzyme, and binding affinities of −8.5 and −8.7 were obtained for betanin and phyllocactin.

Monisha and coauthors [73] have reported the α-amylase inhibition of 1 h-isochromene and pyran carbonitrile and observed that the compounds 91 and 92 were more potent than the acarbose (IC₅₀ = 144.44 ± 1.60 µg/mL). The docking of ligands with porcine pancreatic α-amylase (PDB ID: 1OSE) displayed high binding energies for 91 (−7.738 kcal/mol) and 92 (−7.535 kcal/mol) compared to acarbose (−7.49 kcal/mol). Finally, the stability of the docked complexes has been established with the help of molecular dynamic simulations. H-bonds have been observed as the main binding interaction for the ligand at the active site. All the derivatives have obeyed the Lipinski rule of five. Compounds 91 and 92 have shown higher intestinal absorption and skin permeability (Figure 8).

Figure 8.

Hybrid α-amylase inhibitors.

Fluorinated thiazole – thiosemicarbazones hybrids (93 & 94) have significantly reduced blood glucose levels, more efficient than the standard drug, pioglitazone [74]. The compounds 93 and 94 have poor α-amylase inhibitory activity compared to standard acarbose (IC₅₀ = 0.108 ± 0.005 µg/mL). Afterwards, compounds 93 and 94 exhibited superior efficacy against PPAR-γ with IC₅₀ values of 1.594 ± 0.028 and 1.631 ± 0.023 ng/mL compared to pioglitazone (1.912 ± 0.11 ng/mL). ADMET predictions have shown negative toxicity and improved oral bioavailability and membrane permeability (Figure 8).

Isobenzofuranone derivative (97) has exhibited 11-fold higher α-amylase inhibition (IC₅₀ = 85.39 ± 0.04 μM) compared to acarbose (950.9 ± 0.34 μM) [75]. Compounds 96, 98, and 95 have also inhibited the enzyme significantly with IC₅₀ values of 805.6 ± 0.47, 919.2 ± 0.21, and 680 ± 0.01 μM, respectively. The most potent derivative, 97 has demonstrated noncompetitive inhibition with an inhibition constant value of 138.1 ± 9.80 μM. The STZ-induced diabetic rats administered with 97 for two weeks have reduced glucose levels from 329.24 ± 5.64 mg/dL to 109.17 ± 7.97 mg/dL compared to acarbose 327.6 ± 4.21 to 123.34 ± 5.32 after 14^th day (Figure 8). The histopathological investigation has also shown diminished pathological abnormalities linked with diabetes in the liver and kidney.

Naz and coauthors [76] have evaluated the benzimidazole-based indole/thiazole hybrids against α-amylase and observed good to moderate α-amylase inhibition with IC₅₀ values ranging from 4.90 ± 0.10 to 15.30 ± 0.60 µM compared to acarbose (10.30 ± 0.20 µM). The analog 104, having -OMe and -NO₂ substitutions, exhibited the highest enzymatic inhibition (4.90 ± 0.10 µM). Remarkably, eight derivatives have outperformed the standard drug (Figure 8). Molecular docking and ADMET studies have been used to examine the binding mechanism and the pharmacokinetic properties of the examined derivatives. The exciting outcomes of this research could provide a direction to the medicinal chemists working on diabetes regulation.

7. Conclusion

In conclusion, α-amylase, a ubiquitous metalloenzyme, emerged as a promising target for diabetes management, especially by reducing post-meal hyperglycemia. The recent research for the development of α-amylase inhibitors extends from natural products to synthetic compounds, with a notable focus on heterocyclic frameworks. Literature studies evidence the therapeutic promise of heterocyclic scaffolds in managing glucose metabolism-related disease. The most potent inhibitors discovered recently have been identified, and their potential therapeutic benefits have been discussed. The majority of the research on α-amylase inhibitors is still in the preclinical phase. Thus, extensive clinical trials are required to overcome the existing challenges so that α-amylase inhibitors can be an effective addition to the arsenal of antidiabetic remedies. The present article will be beneficial for the researchers working in the field of diabetes management.

8. Future perspective

Despite the encouraging advancements in the therapeutic applications of α-amylase inhibitors, many of them suffer serious challenges. Several are nonspecific, have poor bioavailability, cause liver problems, and have specific gastrointestinal effects like abdominal discomfort, flatulence and bloating. So, there is an urgent need to develop inhibitors with good oral bioavailability, negative toxicity, and site specificity. Currently, most of the research on α-amylase inhibitors is still at the preclinical stage. To ensure safety and efficacy, future research should focus on improving pharmacokinetic properties. A step further for long-term diabetes management, more clinical trials are required.

The increasing global prevalence of diabetes is expected to stimulate the development of new therapeutic agents. The technological and scientific innovations led to significant advancements in the field of α-amylase inhibitors. Integrating advanced computational tools like machine learning and artificial intelligence is expected to revolutionize the drug discovery process, facilitating the prediction of inhibitor efficacy and safety profiles.

Funding Statement

This work was not funded.

Article highlights

Introduction

• Diabetes mellitus is a chronic progressive metabolic condition featured by postmeal or postprandial hyperglycemia and is the most challenging global health concern.

• Reducing diet-dependent blood glucose elevations is a renowned strategy for its control and management.

α-Amylase

• α-Amylases are ubiquitous, and the familiar sources include humans and animals, plants, bacteria, yeast, and fungi.

• Dietary carbohydrate digestion in the intestinal tract can be limited by inhibiting carbohydrate hydrolases (α-amylase and α-glucosidase).

• Human pancreatic α-amylase (HPA) is a calcium-containing metalloenzyme of 57.6 kDa molecular weight and consists of 496 amino acid units arranged in a single peptide chain.

• The active site of the enzyme contains the catalytic triad Asp197, Glu233 and Asp300, which are involved in the hydrolysis of the starch.

α-Amylase inhibition

• α-Amylase inhibitors diminish the hydrolysis of (α-1,4) glycosidic bonds and act as carbohydrate blockers, thus reducing the digestibility of carbohydrates accompanied by delayed glucose absorption, leading to decreased blood glucose levels after meals.

• The inhibitors are grouped into two categories, namely, proteinaceous or peptide-based and non-proteinaceous inhibitors.

• Natural products, including flavonoids, alkaloids, and polyphenols, significantly inhibited α-amylase.

Synthetic α-Amylase Inhibitors

• Heterocyclic scaffolds evidenced promise in the management of glucose metabolism-related disease and have been extensively explored for their α-amylase inhibitory potential.

• The heterocyclic skeleton exhibited multiple interactions within the enzyme’s active site.

Clinical Studies

• Clinical trials for both natural and synthetic inhibitors are limited yet essential for assessing safety, efficacy, and pharmacokinetics in human subjects.

Conclusion

• α-Amylase, a ubiquitous metalloenzyme, emerged as a promising target for diabetes management, especially by reducing post-meal hyperglycemia.

• The present review article emphasizes the therapeutic promise of recently discovered potential α-amylase inhibitors, highlighting their in vitro, in silico and in vivo profiles.

• The most potent inhibitors discovered recently have been identified, and their potential therapeutic benefits have been discussed.

Future Perspectives

• There is an urgent need to develop inhibitors with good oral bioavailability, negative toxicity, and site specificity. Future research should focus on improving pharmacokinetic properties.

• Advances in computational drug design and personalized medicine will accelerate the discovery of novel α-amylase inhibitors.

• Advancements in computational techniques will expedite the discovery of novel, effective inhibitors.

Disclosure statement

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.