Collection of H. sabdariffa calyx samples, chemicals, and reagents

The dried calyces of H. sabdariffa were taken from a local distributor in Damanhur, Al-Bohaira, Egypt that were freshly obtained from an Experimental Farm at the Faculty of Agriculture, Tanta University, Tanta, Egypt [25]. The seeds (Sabahia 17, dark-red color) of these calyces were previously identified by the Herbarium of the Medicinal and Aromatic Plants Research Farm, El-Kanater El-Khairia, El-Kalubia Governorate, Ministry of Agriculture, ARC, Giza, Egypt. Furthermore, we obtained an experimental approval from the Botany and Microbiology Department, Faculty of Science, Menoufia University, Shebin El-Kom, Menoufia, Egypt. The seeds of Roselle can readily cultivate in well-drained and/or poor soils (the Middle and the Upper Egypt clayey soils) with ∼ 65% humidity under drought and high temperatures (˃ 35 °C). In the cultivation process of Roselle, the organic fertilizers widely use to improve the growth rate, maturity time, plant yield and its phytochemical components, and the seed capabilities. The ideal NPK fertilizers include 68 Kg N, 32 kg P2O5, 24 kg K2O, and 4 L humic acid per fed. The organic fertilizers as 500 kg ammonium sulfate (100 kg N), 150 kg calcium superphosphate (22.5 kg P2O5), 50 kg potassium sulfate (24 kg K2O) as well as azotobacterine and phosphorein (biofertilizers) per fed are also used. The seeds of Roselle were cultivated in the second week of May and harvested in the third week of October in both seasons of 2022 and 2023, which were irrigated with 100 cubic meters per acre per 6–7 days to improve the phytochemical components of Roselle [26]. In the present study, HPLC-grade methanol and ethanol, dimethyl sulfoxide (DMSO), quercetin (95%), gallic acid (79.5%), Folin-Ciocalteu’s phenol reagent, sodium carbonate (NaCO3), aluminum chloride (AlCl3), ascorbic acid (vitamin C), and 1,1-diphenyl-2-picrylhydrazyl (≥ 95% DPPH) were supplied by Sigma-Aldrich (Merck, St. Louis, MO, USA). Other materials, chemicals, and reagents were an analytical grade (Merck) used as received. The ultrapure deionized water was produced by the Milli-Q synthesis system (Millipore Corp., Billerica, MA, USA). All solutions and buffers in this study were prepared using ultrapure deionized water.

Extraction of calyx H. sabdariffa phytochemicals by microwave-mediated process

Prior to the in vitro study, the dried H. sabdariffa calyx samples were re-dried using a shade-drying process for two weeks at room temperature (25 °C) on a laboratory bench to reduce their moisture content. The re-dried Roselle samples were blended into a powdered form using a Panasonic Blender PANA-MX-801 S HG, Panasonic, Malaysia. The powdered form was preserved in a dark, air-tight polyethylene container and stored in a refrigerator (4 °C) until further usage. First, 100 g of the powdered form of H. sabdariffa calyces was soaked in a flask that contained 1000 mL of an aqueous ethanolic solution (1:10 w/v; 60–70% ethanol v/v), following a previously described method with minor modifications [16]. Subsequently, this flask was placed in the microwave cavity for 120 min at 70–90 °C and 400-W microwave power. After the extraction process was completed, the Roselle mixture was filtered using Whatman qualitative No.4 filter paper. The filtrated form was sterilized using a 22 μm pore size filter (Millipore, Molsheim, France). Then, this sterilized filtrate was evaporated and concentrated using a rotary vacuum evaporator (N-1000; Eyela, Tokyo, Japan) at 120 rpm, 70 °C, and 500 mbar. This was followed by the freeze-drying process using a lyophilizer (LGJ-12MC, Shin Lab Co., Korea). The extraction yield was 8.82 g of the hydroethanolic extract of Roselle per 100 g of the powdered form of H. sabdariffa dried calyces.

Identification of calyx H. sabdariffa volatile aromatic components using GC-MS analysis

Separation and identification of volatiles of the hydroethanolic extract of H. sabdariffa calyces were done using Direct DB-1/5MS and TG-5MS fused silica capillary separating columns (30 m length, 0.25 mm inner diameter, and 0.25 μm film thickness) along with a trace GC1310-Ultra/ISQ mass spectrometer (Thermo Scientific, Austin, TX, USA). Roselle sample was initially injected on a 1:50 split mode at 250 °C. The oven temperature of the separating column was firstly maintained at 50 °C for 2 min and raised at an increasing rate of 5 °C/min to 230 °C held for 2 min. Moreover, a rate of 30 °C/min was added to the final temperature of 290 °C held for 2 min. Helium (He) was employed as a carrier gas with a constant flow-rate of 1.0 mL/min. The temperature of the ion-source M+ injector and the MS transfer line were maintained at 200 °C and 260 °C, respectively, for a total run time of 48 min. Using the Splitless mode of the GC and the Autosampler AS1300, a diluted sample of 1.0 µL was automatically injected for 30 s. The mass spectra profile was obtained using a full scan electron impact mode, specifically an electron ionization system, with ionization energy voltages of 70 eV. The scan range was from 10 to 400 m/z. The Roselle GC volatile aromatic fractionations were determined by comparing the Roselle volatile components’ retention time (RT) and mass spectra (m/z) to those in the WILEY 2009 and NIST 2011 v.2.3 mass spectral commercial libraries database. The relative retention times (RTs) of a series of n-alkanes (C7-C36) were used to calculate the relative retention indices (RIs) for all identified Roselle volatile compounds. All analyses were replicated twice. The GC–MS analyses were performed using a Perkin Elmer Clarus 500 gas chromatography (Shelton, CT, USA) coupled with a Perkin Elmer Clarus 500 MSD spectrometer (Shelton, CT, USA) [21].

Phytochemical screening and antioxidant capacity of the hydroethanolic extract of H. sabdariffa calycesTotal anthocyanin content

The total anthocyanin content (TAC) of the hydroethanolic extract of Roselle calyces was determined using a V-5600 UV-Vis spectrophotometer and pH differential spectrophotometry. The pH differential method was utilized [27,28,29], with slight modifications. After diluting 0.5 mL Roselle samples with 4.5 mL buffer solutions of pH 1.0 and pH 4.5 to 5 mL, the absorbance was measured at 520 and 700 nm, respectively. Delphinidin-3-O-sambubioside is considered the most abundant anthocyanin in Roselle calyces. It has a molecular weight (MW) of 597.5 g/mol and a molar adsorption coefficient (ε) value of 23,800 L.mol− 1.cm− 1. In contrast, cyanidine-3-O-glucoside is considered the least abundant anthocyanin pigment in Roselle calyces, with a MW of 449.4 g/mol and a ε value of 26,900 L.mol− 1.cm− 1. All measurements were performed in triplicate. The results were expressed as mg delphinidin-3-O-sambubioside or cyanidin-3-O-glucoside equivalents of anthocyanins according to the following equation:

$$TAC = \left( \begin\left[ } - \,}} \right)}_}\, - \,} - \,}} \right)}_}} \right] \hfill \\*\,MW*\,DF\,*\,2 \hfill \\ \end \right)/\varepsilon *\,I$$

(1)

Where TAC represents mg cyanidine-3-O-glucoside equivalents of anthocyanins/g of the dried form of the hydroethanolic extract of Roselle calyces, MW is the anthocyanin’s molecular weight, ε is the molar adsorption coefficient, DF is the dilution factor, and I represents the quartz cuvette path length (0.64 cm).

Total flavonoid content

The total flavonoid content (TFC) of the hydroethanolic extract of H. sabdariffa calyces was evaluated using the AlCl3 colorimetric method [28, 30, 31], with minor modifications. Quercetin was used as a reference standard flavonoid at concentrations of 0.1, 0.5, 1, 2, 4, 6, 8, 10, 20, and 50 mg/mL. In a test tube, 0.5 mL of the hydroethanolic extract of H. sabdariffa calyces and 0.5 mL of 10% AlCl3 were thoroughly mixed and incubated in a dark place at room temperature (25 °C) for 60 min. The absorbance was recorded at 400–420 nm using a V-5600 UV-Vis spectrophotometer to measure the formation of a golden yellow color. All measurements were performed in triplicate. The TFC was estimated from the standard calibration curve of quercetin (Y = 1.231x + 0.015, R2 = 0.978). The result was expressed as mg quercetin equivalents (QUE) of flavonoid/g of freeze-dried hydroethanolic extract of H. sabdariffa calyces using the following equation:

Where Conc represents the flavonoid concentration in the Roselle samples from the standard quercetin calibration curve (mg/mL).

Total phenolic content

The total phenolic content (TPC) of the hydroethanolic extract of H. sabdariffa calyces was spectrophotometrically determined using the Folin-Ciocalteu method [32,33,34], with slight modifications. Gallic acid was used as a reference standard phenol at concentrations of 0.1, 0.5, 1, 2, 4, 6, 8, 10, 20, and 50 mg/mL. In a test tube, 0.1 mL of diluted Roselle extract (1:10) and 0.2 mL of diluted Folin-Ciocalteu’s phenol reagent (1:20) were mixed and vortexed. To develop a greenish-blue color, 0.5 mL of 20% Na2CO3 solution was added, and the mixture was then incubated under complete darkness conditions. After 120 min, the absorbance was determined at 765 nm against the control negative sample using a V-5600 UV-Vis spectrophotometer. All measurements were performed in triplicate. Using the standard calibration curve of gallic acid (Y = 0.116x + 0.087, R2 = 0.975), the TPC was estimated as mg gallic acid equivalents (GAE) of phenol/g of freeze-dried Roselle hydroethanolic extract and expressed using the following equation:

$$TPC = Conc*250*1000/25$$

(3)

Where Conc represents the phenol concentration in the Roselle samples, as determined by the standard gallic acid calibration curve (mg/mL).

DPPH radical scavenging activity

The antioxidant activity of the hydroethanolic extract of H. sabdariffa calyces was determined using the DPPH free radical scavenging assay [35, 36], with slight modifications. In the presence of antioxidants, the stable purple-colored DPPH free radicals are rapidly neutralized and reduced to a yellow-colored and reduced DPPH form [7]. The changing of the optical density was recorded using a V-5600 UV-Vis spectrophotometer at 517 nm. The hydroethanolic extract of H. sabdariffa calyces was prepared at ten concentrations: 1, 5, 10, 20, 40, 60, 80, 100, 150, and 200 mg/mL. In a test tube, the total volume of the mixture was 500 ?L of Roselle samples, 125 ?L of DPPH solution (4 mg DPPH dissolved in 100 mL of 50% methanol), and 375 ?L of 50% methanol that mixed, vortexed, and incubated at 25 °C for 30 min. As a positive control, ascorbic acid as a reference standard antioxidant was used instead of Roselle samples. As a negative control, the total volume of the mixture was 125 ?L of DPPH solution and 875 ?L of 50% methanol. The change of color was determined at 517 nm using a V-5600 UV-Vis spectrophotometer. All measurements were performed in triplicate. The DPPH radical scavenging activity was expressed using the following equation:

$$\beginDPPH}radical}scavenging}activity\left( \% \right) = \, \hfill \\\,\,\,\,\,\,\,\,\,\,\,Ab} - \left( }\,/\,Ab}} \right)\,*\,100 \hfill \\ \end $$

(4)

Where Abscontrol is the absorbance of negative control, and Abssample is the absorbance of Roselle samples or positive control (ascorbic acid) [35].

The antioxidant capacity of Roselle was expressed as a DPPH radical IC50 value (mg/mL) compared to ascorbic acid. Based on the linear regression equation, the DPPH radical IC50 value represented the concentration of Roselle (mg/mL) that inhibited 50% of DPPH free radicals compared to ascorbic acid (Fig. S6, Tables S2 and S3).

Antibacterial activities of the hydroethanolic extract of H. sabdariffa calyces against selective MDR clinical bacterial isolatesCollection, isolation, and purification of MDR clinical bacterial isolates

Clinical isolates of the urinary tract (urine), wounds (pus), and respiratory tract (sputum) infections were extensively administered for selective pathogenic bacteria. The selected isolates were carefully collected, preliminary identified, isolated, and purified from patients that correlated to Alexandria University Hospitals, Faculty of Medicine, Alexandria, Egypt, from January to August 2022. The selective pathogenic Gram-negative (GN) bacteria were A. baumanii, E. coli, K. pneumoniae, and P. aeruginosa. Before processing, the purified specimens were quickly transported under aseptic conditions to the Bacteriology Lab, Damanhur Regional Joint P.H. Lab, Health Affairs Directorate, Damanhur, Al-Bohaira governorate, Egypt. This study was approved by the Ethics Committee and an Institutional Review Board (IRB) of Alexandria University Hospitals, Alexandria, Egypt.

Bacterial identification and antimicrobial susceptibility testing

Colony samples from primary cultures were subjected to cultural, morphological, and biochemical tests to verify the identification and characterization of specific clinical bacterial isolates [37]. The purified bacterial colonies were identified using an accurate and rapid direct inoculation method. The method was performed using the full automated colorimetric VITEK®-2 GN Cassette and N-280/AST panel Compact Systems according to the manufacturer’s recommendations (Biomérieux, Marcy-l’E´ toile, France). The VITEK®-2 test cards were employed to identify Gram-negative rods (GNRs) through using the Gram-negative Cassette [38]. The probability was 99% A. baumanii, 96% E. coli, 98% K. pneumoniae, and 99% P. aeruginosa. The antimicrobial susceptibility testing (AST) was conducted on selective MDR clinical bacterial isolates using multiple standard antibiotics (Oxoid Ltd., UK). The testing was performed on plate, nutrient or Mueller–Hinton agar (MHA) media using the agar-disc diffusion method. The interpretation criteria based on the guidelines that provided by the Clinical and Laboratory Standards Institute (CLSI) as bacterial susceptibility (S), moderate susceptibility (I), and bacterial resistance (R) (Table S1, Fig. S4A-D) [39]. According to the in vitro antimicrobial susceptibility assay, the MDR bacteria are identified according to their resistance (non-susceptibility) for at least one standard antibacterial drug from ≥ 3 potent antibiotic classes. While, the XDR strains are termed resistant to all standard antibiotics [3, 4].

Antibacterial activity assay

The clinical bacterial isolates were cultivated in nutrient or Mueller-Hinton broth media for 24 h at a temperature of 37 °C until reaching a turbidity level of 0.5 on the McFarland standards scale. The inoculum size used was 1.5 × 108 colony-forming units (CFU)/mL. In order to perform the agar-disc diffusion method, these inoculated suspensions were carefully swabbed onto the surface of MHA (Oxoid Ltd, UK) plates using sterile cotton swabs and allowed to dry. Sterile filter paper discs (5–6 mm; Difco, Detroit, MI, USA) were saturated in triplicates with 25 µL of serial dilution of tigecycline (TGC) as a positive control (4, 2, 1.2, 0.6, 0.2, 0.04, and 0.02 mg/mL). Alternatively, 50 µL of Roselle calyx samples (10, 6, 4, 2, 1, 0.5, and 0.1 mg/mL) were placed on the surface of each inoculated plate for each MDR bacterial isolate. As a negative control, 50 µL of 10% DMSO was used. Subsequently, the labeled cultured petri plates were incubated at 37 °C for 24 h. The antibacterial effectiveness of the hydroethanolic extract of H. sabdariffa calyces was assessed by measuring the zone diameters of inhibition bacterial growth (mm), minimum inhibitory concentrations (MICs), minimum bactericidal concentrations (MBCs), and MBC/MIC ratios. These measurements were compared to those of TGC that served as a positive control, following the CLSI guidelines [39]. The measurements were conducted three times. The zones of inhibition observed in bacterial growth indicate the absence of bacterial growth due to the inhibitory properties of Roselle concentrations that diffused into semisolid culture agar media under the Roselle-saturated discs. The study conducted by Abu Eleneen et al. assessed the diameter of inhibition zones for bacterial growth using the hydroethanolic extract of H. sabdariffa calyces. The criteria for evaluation were as follows: zone ≤ 9 mm (bacterial resistance), 10–15 mm (moderate resistance), and ≥ 16 mm (bacterial susceptibility) [4]. MIC refers to the lowest concentration of the diluted hydroethanolic extract of H. sabdariffa calyces that inhibited any visible growth of the tested bacterial isolates after incubation at 37 °C for 24 h [2, 40]. In contrast, the MBC of the diluted hydroethanolic extract of H. sabdariffa calyces was found to be the lowest concentration that completely inhibited the growth of the tested bacterial isolates after incubation at 37 °C for 48 h [37, 41, 42]. The MBC/MIC ratio is ˂ 4, demonstrating the bactericidal characteristics, while it is ≥ 4, reflecting the bacteriostatic activities of the selective potential antimicrobial drugs [43, 44].

Scanning electron microscopy (SEM) examination

The scanning electron microscope (SEM) was used to evaluate the morphological changes that presented on the surface of the treated clinical isolates of A. baumanii, E. coli, K. pneumoniae, and P. aeruginosa with the hydroethanolic extract of H. sabdariffa L. calyces compared with the untreated forms. The selected samples were coated with gold and dried using a series of ethyl alcohol. Then, the preserved specimens were examined under SEM (JSM 1400 PLUS-JEOL, Tokyo, Japan).

Molecular docking simulations analysis

Molecular docking is a significant computational-based method used to generate the conformations and orientations of ligands into the active-binding sites of their potential targets. Ranked poses were generated by search algorithms based on their scoring functions [45, 46]. In order to evaluate the effectiveness of selective promising volatile aromatic components of the hydroethanolic extract of H. sabdariffa calyces as potent antibacterials, the molecular docking simulations analysis was carried out between (E)-10-Octadecenoic acid methyl ester (MF: C19H36O2, MW:296.5 g/mol, CID:5364425), Butanedioic acid, 3-hydroxy-2,2-dimethyl-, diethyl ester (MF: C10H18O5, MW:218.25 g/mol, CID:589302), or Diethyl succinate (MF: C8H14O4, MW:174.19 g/mol, CID:31249) with the X-ray crystallographical structures of E. coli K-12 1,4-Dihydroxy-2-naphthoyl-CoA synthase (MenB, lyase, EC: 4.1.3.36, PDB DOI: 3T88, 2.00Å) [47] and E. coli K-12 DNA Gyrase-DNA binding and cleavage domain in State1 without TOPRIM insertion (DNA gyrA and B subunits, isomerase, EC: 5.6.2.2, PDB DOI: 6RKS, 4.00Å) [48].

The protein structure files (PDB-file) were obtained from the RCSB-PDB database server, accessed on 5 March 2024 (http://www.rcsb.org/). The canonical SMILES of (E)-10-Octadecenoic acid methyl ester, Butanedioic acid, 3-hydroxy-2,2-dimethyl-, diethyl ester, and Diethyl succinate were retrieved from the PubChem database server (https://pubchem.ncbi.nlm.nih.gov/). Then, their 2D-chemical structures were generated and cleaned using ACD/ChemSketch software as.mol format files. The cleaned 2D-chemical structures of Roselle main volatile components were converted to.pdb format files using OpenBabel v2.3.2 software. The Swiss-PDBViewer v4.1.0 program was utilized to minimize the energy of target proteins. The protein models’ quality was validated using the Ramachandran plot study through the PROCHECK/PDBsum database server (https://www.ebi.ac.uk/thornton-srv/software/PROCHECK/) [49]. The Ramachandran plots and their statistics for the selected antibacterial-correlated targets are demonstrated in Fig. 1. The Ramachandran plot statistics indicate that 92.9% and 81.8% of the amino acid residues in the MenB lyase and DNA gyrase targets, respectively, are in the most favored regions, as shown in Fig. 1.

Fig. 1

The Ramachandran plots and their statistics represent the good-quality stabilized models for selective promising antibacterial-related survival proteins (https://www.ebi.ac.uk/thornton-srv/software/PROCHECK/). Most favored regions (A, B, L) as red color; Additional allowed regions (a, b, l, p) as brown color; Generously allowed regions (∼ a, ∼b, ∼l, ∼p) as bright-yellow color; Disallowed regions (XX) as light-yellow color. The selected antibacterial-related survival targets were demonstrated as the following: E. coli K-12 1,4-Dihydroxy-2-naphthoyl-CoA synthase (MenB, lyase, 3T88) and E. coli K-12 DNA Gyrase-DNA binding and cleavage domain in State1 without TOPRIM insertion (DNA gyrA and B subunits, isomerase, 6RKS)

The molecular docking simulations tool Autodock 4.2.6 was used to investigate the estimated ligand-protein free energy of binding (kcal/mol), estimated inhibition constant (Ki), and the stability root-mean-square deviation-tolerance (RMSD-tol) values. The grid box properties were 0.375Å spacing, -27.419X-, 36.819Y-, and − 24.954Z-center, and 200 as a number of points in X-, Y-, and Z-dimensions. Free energy of binding was estimated to be less than − 5 kcal/mol, indicating that the protein target exhibits a specific binding affinity towards the lead [50,51,52,53]. The estimated free energy of binding for the docked ligands is positively correlated with their binding affinities and docking properties toward selective promising targets. The main Roselle volatile components with the highest binding energies/the lowest negative values toward E. coli MenB lyase and DNA gyrase were chosen after conducting molecular docking simulations. After molecular docking simulations, the main Roselle volatile components with the highest binding energies/the lowest negative values toward E. coli MenB lyase and DNA gyrase were selected to visualize their docked forms using the BIOVIA Drug Discovery Studio Visualizer software (Figs. 6-9).

Fig. 2

(A) The GC-MS chromatogram of twenty-seven fractionations/volatile organic compounds of the hydroethanolic extract of H. sabdariffa calyces/flowers. (B) The GC-MS analysis plot, mass spectrum, and 2D-molecular structure of (E)-10-Octadecenoic acid methyl ester (25.20 RT; 59.23%) that represented the most abundant fraction and the major active ingredient in the hydroethanolic extract of H. sabdariffa flowers

Fig. 3

(A) The 2D- and 3D-molecular structures of the major volatile organic compounds in the hydroethanolic extract of H. sabdariffa calyces were (E)-10-Octadecenoic acid methyl ester (25.20 RT and 59.23%; C19H36O2), 8,11-Octadecadienoic acid, methyl ester (28.71 RT and 4.27%, 29.28 RT and 4.17%, and 25.58 RT and 3.06% fractionations; C19H34O2), Butanedioic acid, 3-hydroxy-2,2-dimethyl-, diethyl ester (31.96 RT and 6.22%; C10H18O5), Diethyl succinate/Butanedioic acid, diethyl ester (24.70 RT and 1.68% and 20.93 RT and 0.67% fractionations; C8H14O4), and Heptadecanoic acid, 16-methyl-, methyl ester/Methyl isostearate (26.44 RT and 2.31%; C19H38O2). Zone diameters of inhibition growth (mm) of 10 (500 µg/disc) and 6 (300 µg/disc) mg/mL hydroethanolic extract of H. sabdariffa calyces toward clinical A. baumanii (B), E. coli (C), K. pneumoniae (D), and P. aeruginosa (E) isolates compared to 10% DMSO as a negative control

Fig. 4

Phytochemical screening (TPC, TFC, and TAC) (A), DPPH radical scavenging activity (%) (B), DPPH radical IC50 value (C), zone diameters of inhibition growth (mm) (D), ClustVis heatmap distribution analysis (E), MIC (F), and MBC (G) of the hydroethanolic extract of H. sabdariffa calyces toward clinical A. baumanii, E. coli, K. pneumoniae, and P. aeruginosa isolates. Ascorbic acid and TGC were used as a reference standard antioxidant compound and the most effective antibiotic against these clinical MDR bacterial isolates, respectively. Data values are expressed as means ± S.D. (n = 3). * indicates statistically significant differences (p < 0.05) compared to ascorbic acid as a standard antioxidant or TGC as a reference standard antibiotic. ZIG, Zones of inhibition bacterial growth (mm)

Fig. 5

The surface morphological characteristics (SEM) of the untreated (A, C, E, and G) and treated (B, D, F, and H) A. baumanii, E. coli, K. pneumoniae, and P. aeruginosa clinical isolates, respectively with the hydroethanolic extract of H. sabdariffa calyces were demonstrated. Magnifications = 15,000x. Scale bar = 1 µM

Statistical analysis

All data are statistically expressed as means ± standard deviations of the means (means ± S.D.) for in vitro study. The paired-sample test analysis of SPSS Windows Version 19.0 (SPSS, Inc., Chicago, IL, USA) was utilized to assess the statistical significance (p < 0.05) of the differences [54]. Compared to ascorbic acid, the linear regression analysis revealed that the roselle hydroethanolic extract exhibited a 50% inhibition (IC50) against DPPH free radicals. The ClustVis web server applications are freely available at http://biit.cs.ut.ee/clustvis/ server and were utilized to summarize and visualize the data values in an ideal heatmap distribution form [55].