Comprehensive Metabolite Profiling and Enzyme Inhibitory Potential of Psidium guajava L. Hydro Alcoholic Extract: An In vitro and In Silico Approach

ABSTRACT

Psidium guajava L. is traditionally used for gastrointestinal issues, along with its use in lowering blood glucose. The present study evaluates the phytochemical composition, in vitro antioxidant potential, and antidiabetic properties of P. guajava L. hydro alcoholic extract (PGHE). TPC and TFC were quantified from PGHE. LC–ESI–QTOF–MS/MS and HPTLC were used for metabolite profiling. In vitro antioxidant, enzyme inhibition assays, and in silico molecular interactions were investigated for PGHE. The yield of extract was 6.16%. It exhibited high phenolic (147.2 ± 0.98 mg GAE/g) and flavonoid content (70 ± 1.67 mg QUE/g). HRLC–ESI–QTOF–MS/MS profiling tentatively identified 555 metabolites, with some bioactive compounds, that is, myricetin, quercetin, ellagic acid, and gallic acid. In vitro antioxidant assays showed potent radical scavenging activity, although lower than the standard antioxidant ascorbic acid. Enzyme inhibition assays revealed PGHE's α‐glucosidase and α‐amylase inhibition with IC₅₀ of 53.32 ± 0.25 and 79.28 ± 1.1 µg/mL, respectively. HPTLC confirmed the presence of gallic acid (0.028%), quercetin (0.093%), and ellagic acid (0.047%) with R _f values 0.592, 0.745, and 0.503. Molecular docking studies against α‐amylase showed that ellagic acid had the highest binding affinity (−7.6 kcal/mol), better than the standard drug acarbose. The study links chemical diversity with functional efficacy and molecular interaction, offering a detailed in vitro and in silico validation of guava's glucose‐lowering potential.

1. Introduction

Diabetes is a chronic, debilitating condition that arises when the body experiences hyperglycemia, a condition in which blood glucose levels are elevated. Root causes include insulin resistance, insufficiency, or both. If untreated, hyperglycemia can have devastating effects on the brain system and blood vessels. It can cause kidney and nerve damage, cardiovascular illness, foot ulcers, amputation of limbs, and eye disease that can cause blindness or visual loss, among other short‐ and long‐term problems. According to data, it is one of the health problems with the highest global rise in the 21st century, making it a major global malady [1, 2]. In addition to the individual immune system and genetic makeup, it is also known that unhealthy lifestyle choices, including stress, bad eating habits, and inactivity, have a substantial impact on the recent rapid rise in diabetes cases. According to the World Health Organization (WHO), over 25 million Indians over the age of 18 are anticipated to be prediabetic, and 77 million are anticipated to have Type 2 diabetes mellitus (T2DM). By 2045, it is projected to increase to more than 134 million [3]. To manage hyperglycemia, many medications are frequently used, including sulfonylureas, amylin analogs, α‐glucosidase inhibitors, d‐phenylalanine derivatives, meglitinides, GLP agonists, DPP‐4 inhibitors, and thiazolidinediones. Metformin (MET) is typically the first line of treatment for people with T2DM, accounting for 90%–95% of all instances of diabetes mellitus. MET lowers blood glucose levels by preventing gluconeogenesis, glycogenolysis, and glucose absorption from the small intestine. Acarbose (ACR), another medication, slows the conversion of carbs into glucose, hence lowering intestinal glucose absorption. However, MET is reasonably effective in improving insulin sensitivity in adults and is generally well tolerated; around 25% under treatment experience gastrointestinal side effects, of which 5% completely unable to tolerate MET [4]. Long‐term use of the drug has also been associated with vitamin B₁₂ deficiency and may lead to anemia or neuropathy [5]. Many people have trouble managing their blood sugar; therefore, doctors often advise using multiple medications. Therefore, it has become urgently necessary to produce antidiabetic medications that are both safe and effective and have fewer adverse effects to combat the global epidemic of diabetes.

Of late, the use of alternative methods of treatment, like the use of herbal products to combat various diseases, has gained momentum. Traditional plant‐based medicines offer a promising alternative due to their affordability, accessibility along with their potential for multitarget therapeutic effects, such as enhancing insulin secretion and improving glucose uptake [6]. Studies on active exploration of medicinal plants and various plant parts to identify functional bioactive compounds with antibacterial, anticancer, and antioxidant properties using advanced techniques like LC–MS/MS, especially when coupled with QTOF and HPTLC, play a crucial role in detecting and characterizing phytometabolites for potential therapeutic applications [7]. In India, the use of plant‐based remedies to cure a variety of illnesses is widespread and has been practiced for ages. Due to its safety and lack of hazardous side effects, self‐medication with natural products and medicinal plants has considerably grown, especially in low‐income and developing nations. According to the WHO, 75%–80% of the population completely relies on conventional medical systems for their initial course of treatment. As sources of polypharmacological therapy for numerous disorders, natural products have gained tremendous popularity lately. It is possible to delay the formation of diabetes by reducing oxidative stress in the body and blocking the enzymes involved in the digestion of carbohydrates. Thus, combining antioxidants with antidiabetic drugs is a viable strategy for treating oxidative stress illnesses like diabetes. Among the plethora of medicinal plants, Psidium guajava L. (guava) is a common plant that demonstrated antihyperglycemic effects, along with anti‐inflammatory, antidiarrheal, and antiviral activities and has been reported by traditional health practitioners in Manipur in regulating blood glucose levels [8, 9, 10]. Kossi et al. [11] have reported that the hydroalcoholic (hydroethanolic) extract of P. guajava L. leaves markedly enhanced glucose uptake by skeletal muscle cells more than MET and glibenclamide, and exhibited a strong dose‐dependent antioxidant activity. A combined in vitro and in silico study of fractions of P. guajava L. leaf extract, especially those containing ellagic acid and quinalizarin, demonstrated moderate multitarget inhibition against α‑amylase, tyrosinase, and hyaluronidase, while docking analyses confirmed strong binding affinities to these enzyme targets [12]. Methanolic and aqueous extracts of guava leaves demonstrated potent in vitro inhibition of α‑amylase and α‑glucosidase enzymes, highlighting their potential to modulate postprandial hyperglycemia [13]. Lorena et al. [14] reported on the use of high‐resolution LC–MS/MS (QTOF) to identify key phenolic and flavonoid compounds such as catechin, quercetin, epigallocatechin, and gallic acid in an aqueous and ethanolic guava leaf decoction, demonstrating strong antioxidant activity and enzyme inhibition (acetylcholinesterase and HMG‑CoA reductase), non‑cytotoxicity toward Caco‑2 cells, and measurable intestinal permeation of several compounds. However, P. guajava L. has been reported for various pharmacological effects along with phytochemical profiling, a comprehensive evaluation integrating high‐resolution metabolite profiling, quantification, in vitro antioxidant, and enzyme inhibition assays, and in silico docking of confirmed constituents linking chemical diversity with functional efficacy and molecular interaction still remains.

The present study reports on the simultaneous detection and identification of bioactive metabolites, hypothesizing that the bioactive metabolites present in P. guajava L. exhibit significant glucose‐lowering effects or hypoglycemic potential, thereby offering a safer and effective alternative for diabetes management.

2. Materials and Methods

2.1. Chemicals

Quercetin, α‐glucosidase, α‐amylase, ACR, 2,2‐diphenyl‐1‐picrylhydrazyl (DPPH), 2,2′‐azino‐bis (3‐ethylbenzothiazoline‐6‐sulfonic acid) (ABTS), and gallic acid were obtained from Sigma‐Aldrich (MO, USA). HPLC‐quality ethanol and methanol were procured from Merck (Darmstadt, Germany). In addition, sodium acetate, sodium carbonate, sodium hydroxide, p‐nitrophenyl‐α‐d‐glucopyranoside (PNPG), starch, 2,4‐dinitrosalicylic acid (DNSA), and sodium potassium tartrate were procured from Sigma‐Aldrich. The Milli‐Q system (Millipore, Bedford, MA, USA) provided the ultrapure water utilized in the tests.

2.2. Plant Material and Extraction

Leaves of P. guajava L. (PG; family: Myrtaceae) were procured from Leimaram Village in Nambol, Manipur, India (latitude: 24°43′37.5″ N, longitude: 093°46′14.5″ E; elevation: 855m). The collected plant was identified and authenticated by taxonomists in the Institute of Bioresources and Sustainable Development (IBSD), Imphal, and the voucher specimen (IBSD/HBM/KHKL/2022‐23) has been stored for future reference. The plant parts were cleaned, shade, and weighed. One hundred grams of coarsely ground PG leaves were macerated in 70% hydro alcoholic solvent (methanol) for 72 h. Then, the mixture was filtered through Whatman filter paper (Grade 1), and the filtrate was concentrated using a rotary vacuum evaporator (IKA, Japan). The extract was dried in a water bath and stored till further use. The P. guajava L. hydro alcoholic extract's (PGHE) percentage yield was calculated with the help of the following formula.

$$\mathrm{Percentage}\mathrm{yield}\mathrm{of}\mathrm{extract}=\frac{\mathrm{Weight}\mathrm{of}\mathrm{sample}}{\mathrm{Weight}\mathrm{of}\mathrm{extract}}\times 100$$

2.3. Evaluation of Antioxidant Activity by DPPH and ABTS Radical Scavenging Assay

The antioxidant activities of PGHE of leaves were measured on the basis of the scavenging activities of the stable DPPH radical [15]. Briefly, 100 mM of DPPH in methanol and the plant extract (mg/mL) solution in methanol were prepared. Equal volumes (0.1 mL) of plant extract solution (50–300 µg/mL) and freshly prepared DPPH radical solution in methanol were added to the 96‐well titer plate. The reaction mixture was incubated at room temperature for 30 min in the dark. Absorbance was calculated at 517 nm using a SpectraMax M5^e reader (CA, USA). The percentage inhibition of DPPH was calculated using the absorbance as follows:

$$\begin{array}{ccc}\hfill \%\mathrm{Inhibition}& =& \frac{\left(\mathrm{Absorbance}\mathrm{control}-\mathrm{Absorbance}\mathrm{sample}\right)}{\mathrm{Absorbance}\mathrm{control}}\hfill \\ & & \times 100\hfill \end{array}$$

The free radical scavenging activity of PGHE leaves was determined by ABTS radical cation decolorization assay [15]. ABTS (7 mM) and potassium persulfate solution (PPS) (2.45 mM) were prepared separately. ABTS⁺ cation radical was produced by the reaction of ABTS and PPS in an amber bottle, stored in dark conditions at room temperature for 24 h before use. The working ABTS was prepared by diluting the ABTS⁺ cation radical with methanol (v/v) to obtain an absorbance of 0.700 at λ _734 nm (use methanol or potassium buffer saline pH 7 for dilution). The absorbance was read at λ _734 nm after 30 min. Ascorbic acid was used as a standard, and the percentage inhibition was calculated using the equation above. The experiments were carried out in triplicate.

2.4. Determination of Total Phenolic and Flavonoid Contents

The total phenolic content (TPC) and total flavonoid content (TFC) values were determined spectrophotometrically as described by Pieczykolan et al. [16] and were expressed as gallic acid equivalent (mg of GAE/g of extract).

The TFC was determined following the method of Mahboubi et al. [17] with slight modifications and was expressed as quercetin equivalents (mg QUE/g). The experiments were carried out in triplicate.

2.5. Evaluation of Antidiabetic Activity by In Vitro Method

2.5.1. α‐Glucosidase Inhibitory Assay

The α‐glucosidase inhibitory activity was evaluated as described by Bhatia et al. [18] with slight modifications. 20–400 µg/mL of PGHE (50 µL) was reacted with 0.1 M sodium phosphate buffer (pH 6.8) (125 µL) and 5 mM pNPG solution (50 µL). The reaction was incubated at 37°C for 15 min before adding 10 µL α‐glucosidase (1.0 U/mL) solution in sodium phosphate buffer (pH 6.8). 0.1 N sodium carbonate (10 µL) was added to terminate the reaction and absorbance was measured at λ _405 nm. ACR was used as a positive control/standard, while all the components in the assay, minus the plant extract, served as the negative control for calculations. The percentage inhibition of three replicates was calculated using the above equation.

2.5.2. α‐Amylase Inhibitory Assay

The α‐amylase inhibitory activity was performed as described by Oyedemi et al. [19]. 20–400 µg/mL of PGHE (50 µL) was reacted with 0.02 M sodium phosphate buffer (pH 6.9) (125 µL) and 1% (w/v) potato starch solution (50 µL). The reaction was incubated at 37°C for 15 min before adding 10 µL salivary amylase (1.0 U/mL) solution in sodium phosphate buffer (pH 6.8). 3,5‐Dinitrosalicylic acid (DNS) reagent (10 µL) was added to terminate the reaction, and absorbance was measured at λ _405 nm. ACR was used as a positive control/standard, while all the components in the assay, minus the plant extract, served as the negative control for calculations. The percentage inhibition of three replicates was calculated using the above equation.

2.6. Chemical Profiling by LC–ESI–QTOF–MS/MS

Mass spectrometry was used to examine the metabolites in the PGHE using a 1290 Infinity HPLC and a 6545 QTOF (Agilent Technologies, Santa Clara, CA, USA) with a dual ESI source. A 2.1 × 100 mm C18 RRHD column (Zorbax Eclipse Plus) with 1.8 µm particle size was used to separate the metabolites using a gradient method following the set parameters‐

Temperature	35°C
Injection volume	10 µL
Flow rate	0.2 mL/min
Autosampler temperature	15°C

0.1% formic acid in water and in acetonitrile were used for the mobile phase. For the first 15 min, 90% of formic acid is in water and in acetonitrile, followed by 50% (15–25 min), 25% (25–26 min), and 90% (26–30 min). Optimization for MS scans was carried out as follows:

Gas source temperature	320°C
Drying gas rate	8 L/min
Nebulizer pressure	35 psig
Spray voltage	±3500 V
Cone voltage	±1000 V
Sheath gas temperature	350°C

The range of positive and negative ionization modes with m/z used was 50–3000. In both positive and negative modes, scan rates of 1.0 scans per second were used for the data capture of MS and MS/MS. For data acquisition, the Mass Hunter workstation program was used.

2.7. Targeted Metabolite Detection and Quantitation Using HPTLC

The content of standard marker compounds in the extracts was determined using a Camag HPTLC instrument (Muttenz, Switzerland). The stock solution of the standards (1 mg/mL) and extract (10 mg/mL) were prepared in methanol. The solutions were vortexed and filtered through before analysis using a 0.45 µL syringe filter. Then the sample and standards were applied on the aluminum‐backed HPTLC plates of silica gel 60 F254 (Merck) using CAMAG Automatic TLC Sampler (ATS). The plates were then developed in a pre‐saturated twin trough chamber using a suitable mobile phase comprising toluene:ethyl acetate:methanol (3:4:1; v/v/v). The compounds were detected by visualizing and photographing the plates using CAMAG TLC Visualizer 2 for illumination and detection at UV254 nm and 366 nm. Densitometric evaluation was carried out using the CAMAG TLC Scanner 4 through vision CATS integration software for analyzing the absorbance and integrating the peak area.

2.8. Molecular Docking Studies

The 3D crystal structure of human salivary amylase (PDB ID: 1SMD) was obtained from the Protein Data Bank (https://www.rcsb.org) with a resolution of 1.6 Å. The inhibitors, ligand complexes like chloride and calcium ions, and nonessential water molecules were eliminated using Discovery Studio Visualizer (v21.1.020298). The macromolecule was, however, prepared on AutoDockTools 1.5.6. The 3D structure of the 07 phytometabolites identified were retrieved in the sdf format from the LC–ESI–QTOF–MS/MS data (https://pubchem.ncbi.nlm.nih.gov). Optimization of the force field in the ligand was conformed accordingly. The downloaded compounds and protein structures were all uploaded to the PyRx tool, a virtual screening software (Kumar et al. 2020). The active site grid box in the enzyme was generated by using the PyRx tool. The coordinates and size were adjusted simultaneously by adjusting the boundary line of the grid or alternatively by entering the values, respectively [20]. The H‐bond interactions with amino acids, van der Waals and π‐interactions, and bond length were estimated after the binding interactions [21].

2.9. Statistical Analysis

Dose–response data from antioxidant assays (ABTS and DPPH) and enzyme inhibition assays (α‐glucosidase and α‐amylase) were analyzed using nonlinear regression in GraphPad Prism version 10.1.0 (GraphPad Prism Software, Boston, USA). Percentage inhibition at each concentration was fitted to a two‐parameter logistic (2PL) model, with the minimum (bottom) constrained to 0% and the maximum (top) constrained to 100%, while IC₅₀ and Hill slope were estimated as free parameters. A global curve‐fitting strategy using an extra sum of squares F‐test was employed to compare reference compounds with the plant extract (PGHE) and to assess whether the extract exhibited comparable potency across assays. The null hypothesis specified that groups shared the same IC₅₀ and Hill slope, while the alternative hypothesis allowed these parameters to vary. Model comparison was conducted using the extra sum‐of‐squares F‐test, with statistical significance set at α = 0.05 and p value (p < 0.0001). The quantitative data for the statistical analysis were calculated as IC₅₀ values with standard deviation (±SD) mean using confidence intervals (95% CI) with sample replicates (n; n = 3).

3. Results and Discussion

3.1. Extraction Yields and Determination of TPC and TFC

The yield percentage calculated based on dry matter weight (w/w) demonstrated that the PGHE extraction process achieved a yield of 6.16%. Extraction of bioactive compounds using the PGHE method (70% methanol) showed a moderate yield. Methanol as a solvent for extraction is usually favored when targeting broad phytochemicals like flavonoids, phenols, and tannins due to its polarity.

The TPC present in the extract was expressed in terms of gallic acid equivalent (y = 3.2793x − 0.0004; r ² = 0.9938), determined as 147.2 ± 0.98 mg GAE/g, while the TFC expressed in quercetin equivalent (y = 16.699x + 0.0036; r ² = 0.9975) was determined as 70 ± 1.67 mg QUE/g, respectively. Polyphenolic compounds are naturally occurring phytochemicals that are characteristic of comprising multiple phenol units and are categorized as phenolic acid, flavonoids, tannins, and so forth. The high TPC and TFC of 147.2 ± 0.98 mg GAE/g and 70 ± 1.67 mg QUE/g suggest richness in polyphenolic constituents in the extract. Methanolic leaf extracts are known to show high yields of polyphenols [22, 23]. The high correlation coefficients (r ² > 0.99) achieved for both standards show the versatility of the analytical method.

3.2. Untargeted Metabolite Profiling Using HRLCMS–ESI–QTOF–MS/MS

An untargeted phytometabolites profiling was done by a Triple quadrupole LC‐MS/MS instrument using a 1290 Infinity HPLC coupled with a 6545 QTOF (Agilent Technologies) equipped with dual ESI source. A total of 555 tentative metabolites were identified using LC–ESI–QTOF–MS/MS (positive and negative mode). The total ion chromatograms for both the mode of ionizations and the tentative identification of some potential phytometabolites present in the PGHE using LC–ESI–QTOF–MS/MS are provided in Figure 1, respectively. A total of 66 metabolites, including 15 polyphenols and flavonoids, 10 fatty acids and lipids, 7 terpenoids and saponins, 6 alkaloids and amines, 3 steroids, 3 sugar/sugar conjugates, 3 nucleosides, 2 anthocyanins, and others (Table S1) were identified with high confidence after filtering the score limit ≥ 99 (positive mode) and ≥ 92 (negative mode).

FIGURE 1.

UPLC–QTOF–MS/MS total ion chromatogram for PGHE (A) and (B) represent positive and negative ionization modes, respectively.

HRLC–MS is a vital tool for phytochemical profiling, enabling the identification of bioactive compounds in plant extracts for potential medical and agricultural applications. Many conventional drugs, including antidiabetic, anticancer, and antibiotic agents, have been developed from such plant‐derived bioactives [24]. The present study also reports on the identification of seven phytometabolites reported to have the glucose‐lowering potential, namely, myricetin 3‐galactoside eluting at RT 8.688, myricetin (RT‐8.704), ellagic acid (RT‐8.955), quercetin (RT‐9.55), quercetagetin/ 6‐hydroxyquercetin (RT‐10.479), epigallocatechin (RT‐4.149), and gallic acid eluting at RT‐1.335, respectively. Seven compounds belonging to three different chemical categories, like flavonoids, benzopyrans, and phenolics, were characterized. A flavanol, namely, myricetin 3‐galactoside, was detected at retention time (RT) 8.68 min, showing molecular ion [M−H]⁻ at m/z values at 481.0978 [25]. Myricetin, a flavonoid derivative at 8.95 min, produces molecular ion [M−H]⁻ at m/z 319.0451 [25]. Another flavonoid derivative eluting at 9.55 and 10.47 min produces molecular ion [M−H]⁻ at m/z 303.0499 and 319.0445, respectively [25, 26]. Ellagic acid, a phenolic class of compound, was detected at RT 8.95, producing molecular ions at 303.0133 [27]. In the negative mode of ionization, gallic acid, a phenolic, was detected at RT 1.33 min, producing molecular ion [M−H]⁻ at m/z 169.015 [25]. Another phenolic derivative, epigallocatechin, eluted at 4.14 min, producing molecular ion [M−H]⁻ at m/z 305.0661 [28] (Table 1). Myricetin, a flavonol, is widely distributed in plant species within the Myrtaceae family, particularly in the fruits, leaves, and seeds. Myricetin has been well documented for its protective role against oxidative stress, anti‐inflammation, and antidiabetic [29, 30]. Niisato and Marunaka [31] have reported on the multifunctional role of myricetin, including treatment of T2DM. Emblica officinalis was reported to contain ellagic acid, which is reported for its antidiabetic properties through the action on the β cells of the pancreas and antioxidant activities [32]. Quercetin, a class of flavonoid, is reported to show antidiabetic effects in streptozotocin‐induced diabetic rats, possibly by reducing blood glucose levels as well as reducing oxidative stress related to diabetic complications [33]. Adil et al. [23] have reported on the identification of quercetin in the methanol extract of Euphorbia parviflora that showed the highest antioxidant activity as compared to other solvents used for extraction.

TABLE 1.

HRLCMS–ESI–QTOF–MS/MS profiles of PGHE (analysis in both negative and positive mode).

Source	RT (min)	Compound name	Chemical formula	Acquired (m/z)	Theoretical (m/z)	Literature references
Positive mode	8.688	Myricetin 3‐galactoside	C21H20O13	481.0977	481.0978	[25]
	8.704	Myricetin	C15H10O8	319.0451	319.0448	[25]
	8.955	Ellagic acid	C14H6O8	303.0133	300.9997	[27]
	9.55	Quercetin	C15H10O7	303.0499	303.0499	[25]
	10.479	Quercetagetin/6‐hydroxyquercetin	C15H10O8	319.0445	319.0466	[26]
Negative mode	1.335	Gallic acid	C7H6O5	169.015	169.0131	[25]
	4.149	Epigallocatechin	C15H14O7	305.0661	305.0663	[28]

3.3. Evaluation of Free Radical Scavenging Potential by In Vitro Methods

DPPH and ABTS activity assays are the most utilized methods for evaluating the antioxidant potential of compounds to neutralize free radicals. DPPH and ABTS are capable of donating hydrogen and hence are considered to be effective antioxidants. According to the DPPH method, the IC₅₀ values for AA and PGHE were 10.18 ± 0.5 and 33.23 ± 0.54 µg/mL, respectively. AA and PGHE had ABTS radical scavenging activity IC₅₀ values of 12.34 ± 0.15 and 77.62 ± 0.04 µg/mL, respectively (Figure 2). PGHE exhibited lower antioxidant potential as compared to the standard ascorbic acid in both assays; however, it still demonstrated considerable radical scavenging ability. In the present study, the higher content of polyphenolics could be attributed to the extract's antioxidant effectiveness. Although DPPH showed a marginal increase in antioxidant potential over ABTS, this could, however, be due to the extract's components' greater affinity for DPPH radicals or their improved solubility in the assay medium. The findings also further suggest the effectiveness of the extraction method that is able to isolate the plants' most bioactive components and also maintain the integrity and concentration of key antioxidant compounds.

FIGURE 2.

A nonlinear dose–response radical scavenging effects with different concentrations of standard antioxidant (ascorbic acid) and PGHE. (A) ABTS radical scavenging assay, (B) DPPH radical scavenging assay. Statistical analysis for IC₅₀ and Hill slope was done by sum‐of‐squares F‐test, where the comparison strategy included the global curve‐fitting, and the specific model is the two‐parameter logistic model, and the testing level (α = 0.05) and p values (p < 0.0001). The quantitative data for the statistical analysis were calculated as IC50 values using confidence intervals (95% CI) and n = 3.

3.4. Assessment of Glucose‐Lowering Potential by In Vitro Methods

The inhibition of α‐amylase and α‐glucosidase is a widely used therapeutic strategy to manage postprandial hyperglycemia for T2DM. While α‐amylase breaks down complex carbohydrates into simple sugars or smaller sugars, α‐glucosidase is known to break the smaller sugars into glucose monomers, thereby facilitating their absorption. Being critical for carbohydrate digestion, their inhibition thereby leads to slow glucose absorption and lower levels of glucose in the blood. In the current study, the glucose‐lowering potential of PGHE was assessed through in vitro enzyme inhibition assays, with ACR, a clinically used α‐glucosidase inhibitor, serving as the reference standard. In comparison to the standard drug ACR, PGHE showed a comparative inhibitory action against both enzymes. The IC₅₀ values for the α‐glucosidase inhibitory activity of PGHE and ACR (standard inhibitor) were found to be 53.32 ± 0.25 and 19.08 ± 4.2 µg/mL, respectively. The IC₅₀ values for α‐amylase inhibition were 79.28 ± 1.1 and 23.63 ± 0.81 µg/mL, respectively (Figure 3). PGHE exhibited considerable α‐glucosidase inhibitory activity as compared to ACR, while the α‐amylase inhibitory activity of PGHE was also evident, although less pronounced than α‐glucosidase inhibitory activity. Foods containing high polyphenolics have been associated with lowering risk of many epidemiological diseases such as cardiovascular diseases, diabetes, antibacterial, inflammatory, and others [34, 35]. In the present study compounds such as quercetin, myricetin, ellagic acid, and gallic acid, identified in through LC–ESI–QTOF–MS/MS analysis, have been reported to result in the inhibition of carbohydrate‐hydrolyzing enzymes through competitive or non‐competitive mechanisms, often via binding to the active site or allosteric regions of the enzymes [31, 32, 36, 37]. Therefore, the report underscores the therapeutic potential of PGHE as a natural antidiabetic agent. The balanced inhibition profile of a moderate α‐glucosidase and milder α‐amylase inhibitory activity of PGHE supplemented by its lower side effects could thereby serve as a potential therapy for glycemic management against the synthetic drugs.

FIGURE 3.

A nonlinear dose–response enzyme inhibitory activity with different concentrations of standard (acarbose) and PGHE. (A) α‐Glucosidase inhibitory activity, (B) α‐amylase inhibitory activity. Statistical analysis for IC₅₀ and Hill slope was done by sum‐of‐squares F‐test, where the comparison strategy included the global curve‐fitting, and the specific model is the two‐parameter logistic model, and the testing level (α = 0.05) and p values (p < 0.0001). The quantitative data for the statistical analysis were calculated as IC₅₀ values using confidence intervals (95% CI) and n = 3.

3.5. Targeted Metabolite Detection and Quantitation Using HPTLC

HPTLC profiling of PGHE showed the presence of gallic acid, quercetin, and ellagic acid—with corresponding R _f values of 0.592, 0.745, and 0.503, respectively (Figure 4a, 4b, 4c). The mobile phase was found to produce a compact spot for the three phytometabolites. The solvent system could effectively resolve these polyphenolic markers, indicating an optimum polarity balance aiding their separation. The developed plate was scanned at two wavelengths, λ ₂₅₄ and λ _366 nm, using a Camag TLC Scanner 3. The image of the scanned plate was captured by CAMAG Reprostar 3 (Figure 5). A good linear precision relationship between the peak areas and concentration, with the correlation coefficient (r ²) > 0.97, was obtained. The phytometabolites showed a concentration of 0.093%, 0.047%, and 0.028% for quercetin, ellagic acid, and gallic acid, respectively (Figure 5a–c). Lawag et al. [38] have reported on the detection of phenolic and flavonoid compounds due to their conjugated aromatic systems at a 254 nm wavelength, which enabled the clear visualization of bands with good intensity. Gallic acid, ellagic acid, and quercetin are potent radical scavengers and carbohydrate‐hydrolyzing enzymes [32, 36, 39, 40]; their existence, therefore, supports the bioactivity profile seen in PGHE. In both traditional and modern phytomedicine, standardizing herbal extracts using bioactive markers is essential for ensuring quality and therapeutic consistency. The effective application of TLC in identifying and characterizing these markers in PGHE confirms its value as a reliable tool for routine phytochemical screening, particularly in resource‐limited settings lacking advanced chromatographic equipment.

FIGURE 4a.

Densitogram at λ _254 nm showing (A) standard gallic acid, (B) PGHE (R _f 0.592) in PGHE (0.028%).

FIGURE 4b.

Densitogram at λ _366 nm showing (A) standard quercetin, (B) PGHE (R _f 0.745) in PGHE (0.093%).

FIGURE 4c.

Densitogram at λ _254 nm showing (A) standard ellagic acid, (B) PGHE (R _f 0.503) in PGHE (0.047%).

FIGURE 5.

HPTLC profile of Psidium guajava extract taken at (A) 254 nm (B) 366 nm showing the presence of standard quercetin, gallic acid, and ellagic acid.

3.6. Molecular Docking Studies Against α‐Amylase In Silico

Computational‐aided tools are used to study the interaction between two molecules and are widely used in drug discovery [41], particularly for identifying potent inhibitors against carbohydrate‐hydrolyzing enzymes such as α‐amylase and α‐glucosidase, key targets in diabetes mellitus management. In the current study, to find potential inhibitors against α‐amylase enzyme, we screened from the data obtained from LC–ESI–QTOF–MS/MS analysis. Of the 67 compounds identified tentatively from LC–ESI–QTOF–MS/MS analysis 07 phytometabolites were selected. All the filtered compounds were promoted to docking along with the standard ACR to human salivary amylase (PDB ID: 1SMD). Table 2 details the interaction of molecular docking studies of the phytoconstituents on human salivary amylase. Docking studies of the phytocompounds against human salivary amylase (PDB ID: 1SMD) showed good docking scores in the range of −5.1 to −7.6 (Table 2). The docking results revealed that ellagic acid (−7.6 kcal/mol) exhibited the strongest binding affinity—on par with ACR (−6.7 kcal/mol). Other flavonoids such as quercetagetin (−6.9 kcal/mol), myricetin and its 3‑galactoside (both −6.7 kcal/mol), epigallocatechin (−6.6 kcal/mol), and quercetin (−6.7 kcal/mol) demonstrated moderately strong interactions, while gallic acid showed weaker affinity (−5.1 kcal/mol). Ellagic acid and quercetin were among the top‐performing phytochemicals, achieving binding energies up to −7.6 kcal/mol—comparable to ACR—in a library of 257 compounds targeting α‑amylase [42, 43]. Gull et al. [39] report on the interaction of ellagic acid with α‑amylase, with binding energies around −8.3 kcal/mol, indicating consistent strength across different docking models. The crucial amino acids within the active site that are mostly found to interact with the ligand molecules are ARG195, ASN298, and ARG337 amino acid residues; these include the enzyme catalytic site [44]. Also, other amino acid residues such as TYR, ASP, ALA, GLU, GLN, and LEU are known as key residues. The docking score of the standard inhibitor (ACR) was −6.7, respectively. Other than gallic acid, which showed less conventional H‐bond with only interaction at LYS227 and ILE230. Ellagic acid, with a stronger docking score of −7.6, exhibited conventional H‐bonds with SER4, THR6, ARG10, ARG252, PRO332, and ARG398, and π‐interactions with PHE335. The compound quercetagetin, with a docking score of −6.9, showed a conventional H‐bond interaction with SER4, THR6, SER289, PRO332, GLY334, and ASP402. ACR, the standard compound, was found to interact with SER3, SER4, ASN5, THR6, GLY9, ARG10, ARG252, and SER289 via a conventional H‐bond and a van der Waals interaction with GLY7. While the enzyme catalytic site of human salivary amylase and pancreatic amylase includes ARG195, ASN298, and ARG337, however, in our study, we have observed that the amino acids SER4, THR6, ARG10, ARG252, SER289, PRO332, and GLY334 are important for a stable interaction. Visvanathan et al. [45] have reported on resveratrol tetramers, that is, (−)‐hopeaphenol comprising many hydroxyl groups near the enzyme surface, which form bonds with specific parts of the enzyme, forming H‐bonds with ARG421, GLY403, GLY334, SER289, and SER4. The compound (−)‐hopeaphenol also displayed moderate inhibitory activity against human salivary amylase, while it showed a mild effect against pancreatic amylase [32]. The H‐bonds formed between the amino acid residues in the human pancreatic amylase and hydroxyl groups in the polyphenolic compounds showed interaction with amino acids such as ASP, HIS, GLU, GLN, and GLY, due to their capacity to give and receive hydrogen bonds [46]. In the current study, the amino acids at position SER4, THR6, ARG252, SER289, GY334, PRO332, and ASP402, which suggests the interaction between the enzyme and ligand is through the interacting donation and acceptance of the atoms between these residues. In the present study, ellagic acid stands out in its bioactivity as a compelling α‑amylase inhibitor, with binding potential stronger than ACR, while quercetagetin, myricetin derivatives, quercetin, and epigallocatechin exhibit promising but moderately strong binding potential, and gallic acid shows lower docking affinity. The bioactivity for inhibition of human salivary amylase, as analyzed by the in silico studies, shows polyphenols as the most active phytometabolites conferring the effect. The study confirms the potential of guava phytochemicals that act as competitive enzyme inhibitors and strengthens their relevance for natural glycemic control strategies. While molecular docking provides valuable initial insights into ligand–receptor interactions and helps prioritize compounds efficiently, it does have limitations such as assuming rigid protein structures and lacking confirmation of binding stability or biological activity, which need to be validated by molecular dynamics simulation studies. Despite these constraints, docking, however, it remains a powerful, cost‐effective screening tool, especially in the early stages of drug discovery.

TABLE 2.

Details of molecular docking studies of the phytoconstituents on human salivary amylase (PDB ID: 1SMD), which include docking score, conventional hydrogen bonds, π‐interactions, and active site pocket residues.

Phytocomponents	Docking score (kcal/mol)	Conventional H‐bond	van der Waals	π‐Interaction
Ellagic acid 3D_CID_5281855	−7.6	ARG10, THR6, SER4, ARG252, ARG398, PRO332	THR11	PHE335
Epigallocatechin 3D_CID_72277	−6.6	SER4, GLN7, THR6, GLY334, PRO332	—	PHE335
Gallic acid 3D_CID_370	−5.1	LYS227, ILE230	ARG252 GLY251 ASN250 PRO228	—
Myricetin 3‐galactoside 3D_CID_5491408	−6.7	SER3, THR6, GLN7, ARG252, SER289	—	—
Myricetin3D_CID_5281672	−6.7	THR6, ARG10, ARG252, SER289, PRO332	SER54	PHE335
Quercetagetin 3D_CID_5281680 (6)	−6.9	SER4, THR6, SER289, PRO332, GLY334, ASP402	—	—
Quercetin3D_CID_5280343	−6.7	SER4, THR6, SER289, PRO332, GLY334	—	—

As most of the standard drugs, like ACR, used for the management of diabetes, come with various side effects exerted due to their strong inhibition, prompting the ingested carbohydrate reaching the colon unaltered, thereby leading to various gastrointestinal issues, thus natural compounds, although less potent, have multifaceted bioactive effects leading to their increased interest in this area (Figure 6).

FIGURE 6.

2D depiction showing various interactions of the phytocompounds with human salivary amylase (PDB ID: 1SMD). 1—Ellagic acid, 2—Epigallocatechin, 3—Gallic acid, 4—Myricetin 3‐galactoside, 4—Myricetin, 6—Quercetagetin, 7—Quercetin, 8—Acarbose.

3.7. Statistical Analysis

Nonlinear regression analysis of dose–response curves was performed to determine the IC₅₀ values of PGHE compared with the standard references (ascorbic acid for antioxidant assays and ACR for enzyme inhibition assays). In the ABTS radical scavenging assay, PGHE exhibited a significantly higher IC₅₀ (77.62 µg/mL, 95% CI: 77.14–78.11) than ascorbic acid (12.34 µg/mL, 95% CI: 12.25–12.44), with global curve comparison confirming a highly significant difference (F(2,44) = 73,159, p < 0.0001). Similarly, in the DPPH assay, PGHE showed weaker activity (IC₅₀ = 33.23 µg/mL, 95% CI: 32.31–34.18) relative to ascorbic acid (IC₅₀ = 10.18 µg/mL, 95% CI: 7.263–13.66), and the difference was statistically significant (F(2,38) = 98.96, p < 0.0001). For α‐glucosidase inhibition, PGHE demonstrated weaker inhibitory potential (IC₅₀ = 53.32 µg/mL, 95% CI: 52.87–53.76) compared to ACR (IC₅₀ = 19.08 µg/mL, 95% CI: 18.04–20.14), with global curve comparison confirming significance (F(2,47) = 1218, p < 0.0001). Similarly, for α‐amylase inhibition, PGHE showed weaker inhibitory potential (IC₅₀ = 79.28 µg/mL, 95% CI: 75.89–82.81) compared to ACR (IC₅₀ = 23.63 µg/mL, 95% CI: 23.38–23.88), with global curve comparison again confirming significance (F(2,41) = 1300, p < 0.0001).

A global model comparison encompassing all four assays revealed significant variability in PGHE's activity (F(6, 94) = 436.0, p < 0.0001). Among the tested bioactivities, PGHE was most potent in DPPH radical scavenging (IC₅₀ ≈ 33 µg/mL) and least potent against α‐amylase (IC₅₀ ≈ 79 µg/mL) (Figure S1).

4. Conclusion

The phytochemical profiling and biological activities of PGHE reveal a promising therapeutic potential, particularly in managing oxidative stress and T2DM. Extraction with 70% hydro alcohol yielded a moderate yield (6.16%) rich in phenolic and flavonoid contents at 147.2 ± 0.98 mg GAE/g and 70 ± 1.67 mg QUE/g, respectively. LC–ESI–QTOF–MS/MS analysis, identified 555 metabolites, with key compounds such as quercetin, ellagic acid, myricetin, and gallic acid that are known to possess potent antioxidant‐ and enzyme‐inhibiting activities. PGHE showed good in vitro radical scavenging with DPPH (IC₅₀ = 34.27 ± 0.54 µg/mL) and ABTS (IC₅₀ = 78.5 ± 0.04 µg/mL), indicating its effectiveness in reducing oxidative stress, while it showed significant enzyme inhibition against α‐glucosidase and a moderate α‐amylase inhibitory activity. HPTLC confirmed the presence of ellagic acid, quercetin, and gallic acid, which supports the extract's bioactivity. Molecular docking studies corroborated the in vitro findings, which showed strong interactions with key amino acids specific to the bioactivity in salivary amylase. The study links chemical diversity with functional efficacy and molecular interaction, offering a detailed in vitro and in silico validation of guava's glucose‐lowering potential, more than what conventional studies have provided. In conclusion, PGHE exhibits significant antioxidant and glucose‐lowering potential due to its rich polyphenolic composition, bioactivity, and stable molecular interactions, thereby highlighting its possible utilization as an alternative or complementary therapeutic agent for glycemic control and oxidative stress‐related disorders. Although PGHE demonstrates strong antioxidant and antidiabetic actions in vitro and in silico, in vivo studies are essential to confirm its efficacy, safety, bioavailability, and therapeutic relevance in living systems, bridging the gap between laboratory findings and potential practical application.

Author Contributions

Neeta Pathaw: writing – original draft, validation, methodology, investigation, formal analysis, data curation. Keithellakpam Ojit Singh: investigation, data curation, writing – review and editing. Laikangbam Lamyanba: HPTLC analysis. Tonjam Bidyasana Chanu: investigation and data curation. Shweta Rai: investigation and data curation. Oinam Shajan Singh: sample collection and processing. Sunil Thorat: investigation, data analysis, molecular docking. Nanaocha Sharma: writing – review and editing, supervision, resources, project administration, conceptualization.

Funding

The authors are thankful to the Department of Biotechnology (DBT), Government of India, for funding grant (BT/PR45281/ NER/95/1934/2022).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: cbdv70778‐sup‐0001‐SuppMat.docx

Data Availability Statement

The original contributions presented in the study are included in the article/Supporting Information, further inquiries can be directed to the corresponding author.