Since the exact composition of the natural mixtures used as starting material is unknown, some average properties of EOs or propolis components were used to estimate the appropriate amount of each reagent. The number of moles of reacting molecules was estimated considering 150 Da as the average molecular weight of EOs, 300 Da as the average molecular weight of propolis extract components, and one, as the average number of reacting group per molecule. Finally, 20 mol of hydrazine monohydrate and 1.1 mol of Selectfluor® per “estimated” mol of starting mixture components were used for the reactions.

Typical procedure for preparation of NFEOs: A solution of EO1 (100 mg, 0.66 mmol taking an average MW of 150 Da) and hydrazine monohydrate (645 μL, 13.32 mmol) in ethanol (5 mL) was stirred for 20 h under reflux. The reaction solution volume was reduced approximately to 1/3 under reduced pressure, water was added (5 mL), and the resulting solution was extracted with dichloromethane (DCM) (3 × 5 mL). The DCM fractions were dried (anh. Na2SO4), filtered, and evaporated at reduced pressure. The obtained crude (71.9 mg, 0.48 mmol) was dissolved in ethanol (5 mL), Selectfluor® was added (186.8 mg, 0.53 mmol), and the solution was stirred at room temperature for 20 h. Water was added (5 mL) and the aqueous solution was extracted with DCM (2 × 5 mL). The DCM fractions were dried (anh. Na2SO4), filtered, and evaporated at reduced pressure.

Typical procedure for preparation of NFPEs: A solution of PE1 (100 mg, 0.33 mmol taking an average MW of 300 Da) and hydrazine monohydrate (322 μL, 6.66 mmol) in ethanol (5 mL) was stirred for 20 h under reflux. The reaction solution volume was reduced approximately to 1/3 under reduced pressure, water was added (5 mL), and the resulting solution was extracted with DCM (3 × 5 mL). The DCM fractions were dried (anh. Na2SO4), filtered, and evaporated at reduced pressure. The obtained crude (51.5 mg, 0.17 mmol) was dissolved in ethanol (5 mL), Selectfluor® was added (66.90 mg, 0.19 mmol) and the solution was stirred at room temperature for 20 h. Water was added (5 mL) and the aqueous solution was extracted with DCM (2 × 5 mL). The DCM fractions were dried (anh. Na2SO4), filtered, and evaporated at reduced pressure.

The NFPE1 was chromatographed in MPLC-UV (Elldex-Alltech). 50.0 mg were directly loaded on a Latek model M2 (2 cm Id × 33 cm length) glass column filled with Silica gel 60 RP-18 (15–25 μm, LiChroprep RP-18, Merck). Mobile phase: 1% Formic acid solution and methanol. Method: 0–60 min 60% methanol, 100 min 100% methanol. Injection solvent: methanol. Flow: 4 mL/min. The effluent of the column was monitored at 254 nm. Fractions were automatically collected every two minutes to obtain fifty fractions (F1 − F50). The fractions F13-15 contained 1.5 mg of pure of pure fluorinated pyrazole 3 (3.0% final yield).

Hydrazine monohydrate (390 µL, 7.8 mmol) was added to a solution of chrysin (100 mg, 0.393 mmol) in absolute ethanol (7 mL), and the mixture was stirred for 20 h under reflux. After that, the reaction solution volume was reduced approximately to 1/3 under reduced pressure, water was added (7 mL), and the resulting solution was extracted with DCM (3 × 7 mL). The DCM fractions were dried (anh. Na2SO4), filtered, and evaporated at reduced pressure. The obtained crude (87.0 mg, 0.34 mmoles) was dissolved in ethanol (7 mL), Selectfluor® was added (132.49 mg, 0.37 mmol) and the solution was stirred at room temperature for 20 h. Water was added (7 mL) and the aqueous solution was extracted with DCM (2 × 7 mL). The DCM fractions were dried (anh. Na2SO4), filtered, and evaporated at reduced pressure. The mixture was purified by MPLC-UV on reversed-phase silica gel using methanol/H2O gradients to obtain pyrazole 2 (22.8 mg, 28.5% yield) and pyrazole 3 (12.2 mg, 15.2% yield).

Pyrazole 2. Mp: 168–170 °C. 1H NMR (300 MHz, (CD3)2CO): δ = 7.83–7.79 (m, 2H, Ar–H); 7.49–7.44 (m, 3H, Ar–H and Pyr-H); 7.39–7.37 (m, 1H, Ar–H); 6.96 (t, 1H, J = 8.10 Hz, Ar–H); 6.46 (d, 2H, J = 8.10 Hz, Ar–H). 13C NMR (75 MHz, (CD3)2CO): δ C1 = C3 = 157.99; C3’ = 150.57; C5’ = 143.70; C1’’ = 130.45; C3’’ = C5’’ = 129.87; C4’’ = 129.31; C5 = 129.27; C2’’ = C6’’ = 126.35; C4 = C6 = 107.99; C2 = 106.10; C4’ = 104.76. IR (neat) ν = 3401, 1701, 1626, 1454, 1015, 768, 694 cm−1. HRMS: found m/z = 275.0783, calculated m/z for C15H12N2O2Na [M + Na]+ = 275.0791 (0.8 mDa error).

Pyrazole 3. Mp: 133–134 °C. 1H NMR (300 MHz, (CD3)2CO: δ = 10.65 (1H, HO-Ar); 10.12 (1H, HO-Ar); 7.86 (m, 2H, Ar–H); 7.55 (s, 1H, Pyr-H); 7.52 (m, 2H, Ar–H); 7.43 (m, 1H, Ar–H); 6.94 (dd, 1H, J1 = 8.26 Hz, J2 = 4.08 Hz, Ar–H); 6.44 (dd, 1H, JHH = 8.9 Hz, JHF = 10.5 Hz, H-Ar). 13C NMR (75 MHz, (CD3)2CO: δ C1 = 153.51; C3’ = 150.40; C4 = 146.53 (d, 1JC-F = 228.89 Hz); C3 = 145.40 (d, 2JC-F = 15.46 Hz); C5’ = 144.00; C1’’ = 130.31; C3’’ = C5’’ = 130.10; C4’’ = 129.65; C2’’ = C6’’ = 126. 61; C5 = 115.35 (d, 2JC-F = 19.66 Hz); C2 = 107.88 (d, 3JC-F = 1.90 Hz); C6 = 106.48 (d, 3JC-F = 6.75 Hz); C4’ = 104.92. 19F NMR (282 MHz, (CD3)2CO: δ = − 150.67 (1F, m). IR (neat) ν = 3415, 1703, 1634, 1494, 1462, 1026, 982, 854, 766, 695 cm−1. HRMS: found m/z = 271.0881, calculated m/z for C15H11FN2O2 [M + H]+ = 293.0877 (0.4 mDa error).

The hydrolysis of p-nitrophenyl-α-O-D-glucopyranoside (α-pNPG) was continuously measured in a 96-well microplate using a method similar to that applied by Arnaldos et al. [57]. Wells were filled in triplicate with α-Glc (yeast) in 0.1 M, pH 7 phosphate buffer (0.088 U/mL end concentration per well), α-cyclodextrin, the same buffer solution (1.22 mM end concentration per well) and 10 µL of test compound in dimethylsulfoxide (DMSO) solution. Wells containing the corresponding volume of DMSO without an inhibitor were used as the references of maximum enzymatic rates. The final volume per well was 270 µL. The enzymatic reaction was initiated by adding α-pNPG (1.63 mM end concentration per well). The plate was shaken for 2 s and the increase in absorbance at 405 nm was monitored at 37 °C for 10 min.

DMSO solutions (0.084 mg/mL) of doubly chemically modified extracts were employed for α-Glc inhibition percentage determination.

For IC50 determination, ten serial dilutions of the compounds were prepared in DMSO, following equally spaced points on a neperian logarithm scale, starting at 64.7 mM and finishing at 0.00647 mM (end concentration per well: 2397 to 0.2514 µM). IC50 calculated using Prism V5.01 (GraphPad Software Inc., La Jolla, CA, USA) applying a non linear regression curve fit for a log[inhibitor] vs. normalized answer model with variable slope. Standard drug acarbose was used as enzyme inhibition control.

The α-Glc inhibition properties of the mixtures were surveyed by TLC autography using reported protocols [47]. Briefly, a Silica gel-TLC plate (64 cm2) was sprayed with the 2-naphthyl-α-d-glucopyranoside: Fast Blue B salt (1:1, V/V) solution using a glass reagent sprayer operated with compressed air. Then, the plate was dried under air current at room temperature. 80 mg of agar was dissolved at 80 °C in 9.4 mL phosphate buffer (100 mM, pH 7.0), the solution was allowed to cool down (40 °C) and 188 μL of α-Glc solution (12.5 U/mL) was added. The obtained solution was mixed by inversion and distributed over the TLC plate. After cooling and solidification, the plate was incubated at 37 °C (10 min) in a stove.

DAD-HPLC measurements were performed on a Hewlett Packard HP 1050 series, coupled to a G1306AX DAD. The samples were directly loaded on a Phenomenex Gemini C18 column (5 µm × 150 mm × 4.6 mm). Mobile phase: A: Methanol with formic acid (5%); B water/methanol (39/61) with formic acid (5%). Method: 0 min, B 100%, 3 min B 100%, 5 min B 90%, 8 min B 90%, 15 min B 82%, 16 min B 82%, 17 min B 70%, 20 min B 60%, 22 min B 60%, 23 min B 10%, 25 min B 10%. Injected volume: 3 µl. Flow: 0.3 mL/min. Column temperature: 30.0 °C. Detection at 254 nm. Solutions of 0.15 mg/mL (pure compounds) or 10 mg/mL (for mixture) were injected.