Changes in the physicochemical, antioxidant and antibacterial properties of honeydew honey subjected to heat and ultrasound pretreatments

Abstract

The objective of this study was to examine the effect of different treatments on the physicochemical, antioxidant, and antibacterial properties of honeydew honey. Honeydew honey was subjected to heat treatment and 9 different ultrasound treatments. Our results showed that the following parameters were significantly changed: water content, pH, electrical conductivity, diastase activity, HMF content and water activity. The ultrasound resulted in an increase in the total phenol content and the antioxidant capacity (DPPH, FRAP, and ABTS tests) in comparison with the conventional thermal technique. In most cases, the samples subjected to ultrasound improved the antibacterial activity; the heat treatment resulted in a significant reduction of the antibacterial activity, and sample 4 (ultrasound 30 °C, 5 min) showed the best antibacterial activity. The ultrasound treatment, especially at lower temperatures, represents a technique that enables the preservation and improvement of the biological properties of honeydew honey.

Introduction

Honeydew honey is secreted by insects–aphids, whose mouth apparatus is adapted to drilling plant tissue and sucking phloem sap, which is rich in nutrients. This type of honey has specific physicochemical, organoleptic and microscopic properties and shows good antioxidant and antimicrobial activity (Wilczyńska 2010). The antioxidant and antimicrobial activities are important biological properties of honey. The presence of enzymatic (glucose oxidase, catalase) and non-enzymatic antioxidants (flavonoids, ascorbic acid and phenolic acids) has been detected in many types of honey. The antimicrobial activity of honey depends on various factors: osmolarity, pH value, H₂O₂ content, etc. (Leyva-Jimeneza et al. 2019). Honey processing may have a significant impact on its quality. The heat pretreatment is still mainly used for honey, but in spite of many benefits, such as fungal growth inhibition and the delayed crystallisation process (Tosi et al. 2004), it results in significant shortcomings as well: inactivation of enzymes (primarily of diastase), lipid oxidation and protein denaturation, and causes non-enzymatic browning and an increase in the hydroxymethylfurfural (HMF) content. Other methods (ultrasound, high pressure processing, UV light pretreatment, etc.) are therefore intensively examined to eliminate unwanted effects without degrading the nutritional quality of food. Ultrasound did not show any significant effect on physical properties of honey (Majid et al. 2015), but had an impact on its bioactive properties (Quintero-Lira et al. 2017). Taking into consideration all the facts stated above, the objective of this study is to examine the effect of ultrasound pretreatment on physicochemical, antioxidant and antibacterial properties of honeydew honey, and to compare the obtained values to the values obtained through a standard heat pretreatment procedure.

Material and methods

Sample preparation

As the sample for the analysis was used honeydew honey from the territory of the Republic of Srpska, Bosnia and Herzegovina. The untreated honeydew honey was used as the control sample (C). The honeydew honey was subjected to 9 different ultrasound pretreatments (samples 1–9) and to heat pretreatment (sample 10):

1. Ultrasound 30 °C, 1 min

2. Ultrasound 45 °C, 1 min

3. Ultrasound 60 °C, 1 min

4. Ultrasound 30 °C, 5 min

5. Ultrasound 45 °C, 5 min

6. Ultrasound 60 °C, 5 min

7. Ultrasound 30 °C, 10 min

8. Ultrasound 45 °C, 10 min

9. Ultrasound 60 °C, 10 min

10. Pasteurisation at 65 °C, 10 min

Sample processing

Pasteurisation

One hundred grams of honey was put in a beaker, closed with aluminium foil and put into a heated bath with thermoregulation PRECISTERM (J.P. Selecta, Spain) until an inside temperature reached 65 °C. After heating on that temperature for 10 min (Eshete and Eshete 2019), sample 10 was cooled down and kept until the moment of use in the refrigerator at 4 °C. The experiment was carried out in triplicates.

Ultrasound pretreatment

One hundred grams of honey was put in a beaker, closed with aluminium foil and put into an ultrasonic cleaning bath WUC-A03H (Witeg Labortechnik, Germany), which works at the frequency of 40 kHz and the heating power of 172 W, and the honey was treated in accordance with the experimental design. After that, prepared samples (1–9) were cooled down and kept until the moment of use in the refrigerator at 4 °C. All experiments were carried out in triplicates.

Physicochemical analyses

The characteristics and satisfactory quality of all samples were assured through an analysis of the following parameters: water content, diastase activity, HMF content, acidity, reducing sugars, saccharose, electrical conductivity as described by Ordinance on methods for control of honey and other bee products (Official Gazette of BiH no 37/2009). The pH was measured with a pH meter (Hanna Instruments HI-2211). All the chemicals and reagents used were of analytical grade.

Determination of mineral content

The content of different minerals in sample C was determined using atomic emission spectrometry (ICP-OES), by means of an OPTIMA 8000 instrument (Perkin Elmer, USA). The calibration of the instrument was performed using the Instrument Calibration Standard 2 (Perkin Elmer, USA). All the chemicals and reagents used were of analytical grade.

Sample preparation

Ten mL of concentrated HNO₃ was poured over around 5 g of sample C; the mixture was evaporated to dryness and cooled down, after which 10 mL of HNO₃ (1:1) and 10 mL of perchloric acid were added. The mixture was evaporated to dryness again, and the obtained residue was dissolved in 10 mL of HCl (1:3), after which it was filtered and the content was supplemented to the amount of 100 mL. The obtained solutions were used to determine the concentration of the analyzed elements (Na, K, Ca, Mg, P, Fe, Mn, Cu, Zn, Al, Se, Mo, Ni, Cd, Cr, Pb). The concentrations of the elements for the calibration curve were different (for macroelements 15, 10, 5, 2.5 and 1 mg/L; for microelements 5, 2.5, 1, 0.5 and 0.1 mg/L), determined according to preliminary measurements for the specified metals.

Water activity

The water activities of samples were analyzed at 25 ± 0.2 °C by using the instrument LabMaster-aw (Novasina, Switzerland). The detection limit was ± 0.003 a_w. Calibration was performed with saturated salt solutions in the a_w range of 0.2–0.6.

HPLC analysis of phenolics

Chemicals and instruments

The chemicals and reagents used were of analytical grade. All the standards: catechin hydrate, gallic acid, malvidin, rutin, hlorogenic acid, benzoic acid, quercetin, caffeic acid, ferulic acid and naringenin were of purity > 95%. Measurements were performed on HPLC agilent 1260 infinity (USA) equipped with DAD detector.

The HPLC analysis of phenolics was conducted according to Hussein et al. (2011) method with some modifications. Honey samples (10 g) were dissolved in 50 mL of deionized water and filtered through a membrane filter (Liofilchem, Italy) 0.20 μm. Chromatographic conditions: reversed phase column EC-C18, Poroshell-120 (4.6 × 50 mm, particle size 2.7 µm), the mobile phase 0.1% HAC (solvent A) and acetonitrile (solvent B), flow rate 0.5 mL/min, column temperature 25 °C, sample injection volume 2 μL. The chromatograms were evaluated at λ = 280 nm. The following gradient was used for separation: 9% methanol (B) was flowed through the column isocratically with 91% solvent (A) for 0.8 min which was then increased to 55% acetonitrile (B) for 1.5 min, changed to 40% acetonitrile (B) for 2.3 min, to 30% acetonitrile (B) for 6.8 min, and finally 40% acetonitrile (B) for 8 min. The phenolic compounds were identified by comparing the chromatograms and retention times of the analytes with the reference standards.

Antioxidant activity determination

The total phenolic content was determined using the modified method of Folin-Ciocalteu (Wolfe et al. 2003), and the content of flavonoids was determined using the method of Ordenez et al. (2006). The testing of antioxidant activity using the Ferric reducing/Antioxidant power (FRAP) assay was carried out in accordance with Banzie and Strain (1996); the 2,2′-azinobis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS) assay using the modified method of Re et al. (1999) and the 2,2-diphenyl-1-picryl-hydrazyl (DPPH) assay using the method of Brand-Williams et al. (1995) with some modification. A 0.1 mM solution of DPPH (1,1-diphenyl-2-picrylhydrazyl) in methanol was prepared. 1 mL of aqueous honey solution was mixed with 1 mL of DPPH solution. The mixture was left to stand for 30 min in the dark and absorbance was read spectrophotometrically at 517 nm. The results were expressed as µg of gallic acid (GA) /mg honey. All the chemicals and reagents used were of analytical grade.

Antibacterial activity determination

Microorganisms and inoculum preparation

The following bacterial cultures: Escherichia coli ATCC 25922, Staphylococcus aureus ATCC 25923, Pseudomonas aeruginosa ATCC 10145 and Bacillus cereus ATCC 7004 from the collection of the Laboratory of bacteriology, mycology and parasitology of Veterinary Institute of Republic of Srpska "Dr. Vaso Butozan” Banja Luka, Bosnia and Herzegovina were used. The cultures were grown in a nutrient broth and incubated for 24 h at 37assay was carried out in accordance with Banzie and Strain, after which they were inoculated and grown on nutrient agar for the next 24 h.

Antibacterial activity

Agar dilution method was used to determine the minimum inhibitory concentration (MIC) and the minimum bactericidal concentration (MBC) (Balouiri et al. 2016; EUCAST 2000) The density of the cultures was set to 1.5–3 × 10⁶ °Cfu/mL. The experiments were conducted in Petri plates of MHA (Muller–Hinton agar) inoculated with tested bacteria to which various concentrations of the sample had been added: 50, 25, 12.5, 6.25 and 3.125% (w/v). After the 24 h incubation at 37 °C, the inhibition of growth of the bacterial cultures was measured. The highest dilution of the tested honey to inhibit visible growth of bacteria was considered as the MIC value. From the plates showing no visible sign of growth/turbidity in MIC determination, test microorganisms were inoculated onto sterile nutrient agar plates. The plates were then incubated at 37 °C for 24 h. The least concentration that did not show growth of test organisms was considered as the MBC. The results were expressed as the % (w/v) of honey. The antibiotic susceptibility discs Ampicillin (10 μg), Gentamicin (10 μg), Erythomycin (15 μg) and Ciprofloxacin (5 μg) were used as positive and artificial honey as negative control. Artificial honey was prepared by dissolving 39 g fructose, 31 g glucose, 8 g maltose, 3 g sucrose, and 19 g distilled water as described by Majtan and Majtan (2010). All the chemicals and reagents used were of analytical grade.

Statistical analysis

All tests were performed in triplicate (except antibacterial activity) and the results were expressed as means ± standard deviation. Variance analysis (ANOVA) was applied to test significant differences among samples. Tukey’s test was used to identify differences between mean values obtained in samples (p ≤ 0.05). With a view to examining the effect of one ultrasound pretreatment on the characteristic honey quality parameters, a series of experiments in accordance with the experimental design requirements for single-factor categorical parameters were conducted. The experimental design, and the analysis and processing of the obtained results were conducted using the Design–Expert 11 programme (Stat-Ease, Inc., USA).

Results and discussion

Physicochemical analyses

The results of the physicochemical analyses and a_w activities are shown in Table 1.

Table 1.

The results of the physicochemical analyses and a_w activities

Sample	C	1	2	3	4	5	6	7	8	9	10
Water content (%)	18.50a  ± 0.19	17.90b,c  ± 0.06	18.66d,e,f  ± 0.10	18.46a,d,e,f,g,h  ± 0.06	18.14a,b,c,d,g,h  ± 0.02	18.26a,c,d,e,g,h  ± 0.14	18.06a,b,c,g,h  ± 0.18	18.14a,b,c,d,g,h  ± 0.02	18.14a,b,c,d,g,h  ± 0.46	17.98b,c,g,h  ± 0.22	17.86b,c  ± 0.06
pH value	4.33a  ± 0.03	4.19e  ± 0.00	4.21b,f  ± 0.01	4.19e  ± 0.00	4.20b,e,f  ± 0.00	4.17 g,h  ± 0.01	4.15 g  ± 0.00	4.20b,e,f  ± 0.00	4.17 h  ± 0.00	4.19e,f  ± 0.02	4.22b,c  ± 0.01
Acidity (mmol/kg)	50.67a  ± 0.29	46.00b  ± 0.00	45.75b  ± 0.25	47.25c  ± 0.25	48.00d  ± 0.00	46.00b  ± 0.00	47.00c  ± 0.00	46.25b,e  ± 0.25	47.25c  ± 0.25	46.75c,e  ± 0.25	45.75b  ± 0.25
Electrical conductivity (mS/cm)	1.169a  ± 0.001	1.303 h  ± 0.000	1.300i  ± 0.000	1.297j  ± 0.000	1.284 k  ± 0.000	1.259 l  ± 0.000	1.260 m  ± 0.000	1.269n  ± 0.000	1.248p  ± 0.000	1.284 k  ± 0.000	1.261b  ± 0.000
Sucrose content (%)	2.01  ± 0.00	2.18  ± 0.04	2.24  ± 0.20	2.42  ± 0.19	2.37  ± 0.03	2.30  ± 0.08	2.43  ± 0.02	2.60  ± 0.06	2.13  ± 0.06	2.25  ± 0.07	1.80  ± 0.15
Reducing sugar (%)	68.16  ± 4.44	70.51 ± 1.31	67.19  ± 2.55	68.21  ± 0.51	70.68  ± 1.08	68.36  ± 0.25	65.16  ± 5.88	68.42  ± 4.20	70.73  ± 0.95	69.42 ± 4.53	72.08  ± 1.26
HMF content (mg/kg)	5.47a  ± 0.25	23.84i ± 0.14	25.26c,h  ± 0.07	23.05j  ± 0.03	25.50b,c  ± 0.05	23.22j  ± 0.00	25.30c,h  ± 0.17	25.84b  ± 0.09	22.63 k  ± 0.07	24.54f ± 0.18	25.51b,c  ± 0.27
Diastase activity (DN)	47.67a,b  ± 0.69	38.17c,e  ± 1.10	40.03e  ± 0.11	49.52a,d  ± 0.66	43.71f  ± 0.06	46.05b  ± 0.79	50.05d,g  ± 0.83	43.06f  ± 0.38	47.40a,b  ± 0.28	51.69 g  ± 0.46	37.37c  ± 0.38
aw	0.597a  ± 0.000	0.562f  ± 0.001	0.554 g  ± 0.001	0.552 h  ± 0.000	0.568i  ± 0.000	0.556j  ± 0.000	0.567i  ± 0.000	0.558 k  ± 0.001	0.568i  ± 0.001	0.548 l  ± 0.000	0.577b  ± 0.000

^*Means followed by the same small letter in the row do not differ from one another by Tukey test (p ≤ 0.05)

Treatments key: C—Control; 1—Ultrasound 30 °C, 1 min; 2—Ultrasound 45 °C, 1 min; 3—Ultrasound 60 °C, 1 min; 4—Ultrasound 30 °C, 5 min; 5—Ultrasound 45 °C, 5 min; 6—Ultrasound 60 °C, 5 min; 7—Ultrasound 30 °C, 10 min; 8—Ultrasound 45 °C, 10 min; 9—Ultrasound 60 °C, 10 min; 10—Pasteurisation at 65 °C, 10 min

The water content in sample C was 18.5%. The ultrasound pretreatments resulted in a reduction of the water content in all samples, except in sample 2, which showed a slight increase, but the obtained values are higher than the values of the sample subjected to heat pretreatment. The water content affects the preservation of honey, if it is below 20%, it reduces the possibility of fermentation and increases the stability of honey (Chua et al. 2014). The study by Thrasyvoulou (1994) shows that use of ultrasound pretreatments results in a reduction of the water content. Elevation of temperature has a similar effect; the higher the temperature, the higher the degree of the reduction of the water content (Chua et al. 2014).

The pH value of sample C was 4.33. After the pretreatments, it was evident that the pH value was reduced in all samples. The reduction of the pH value could be caused by the release of organic acids from the pollen during the honey processing. A reduction of the pH value due to ultrasound and heat pretreatment was detected by Chaikham et al. (2016).

In the sample C, the acidity was 50.67 mmol/kg, which is slightly higher than the value allowed in Bosnia and Herzegovina (the maximum acid content of 50 mmol/kg). After the pretreatments, it is evident that the acidity in all samples was reduced to 45.75–48.00 mmol/kg. Although the pH value was reduced, the total acidity did not increase, because the pH value of honey is not directly related to acidity, due to the buffering property of phosphates, carbonates and other mineral salts, which are naturally present in honey (Srećković et al. 2019).

The electrical conductivity of sample C was 1.169mS/cm. After the pretreatments, there is an evident increase in all samples. The electrical conductivity reflects the mineral content in honey, and a high level of the ash content is related to high electrical conductivity. Electrical conductivity of honeydew honey is higher in comparison with other types of honey, due to its origin. Trasyvoulou et al. (1994) found that a combination of ultrasound and temperature does not result in any significant changes in the pH value, moisture content and electrical conductivity, which was partly shown in this study as well.

The sucrose content in the control sample was 2.01%, which is in line with the value allowed in Bosnia and Herzegovina (up to 10%). After the ultrasound pretreatment, the sucrose content increased, and the highest content was recorded in sample 7 (2.60%), while it was found that the sucrose content in the pasteurised sample was reduced (1.80%). As regards the reducing sugar content, the situation is different, because the highest content was recorded in the pasteurised sample (72.08%), and the lowest in sample 6 (65.16%). The reducing sugar content in the sample C was 68.16%, which is in line the value allowed in Bosnia and Herzegovina (not less than 45%). The HMF content in the sample C was 5.32 mg/kg, which is indicative of honey freshness, because the internationally recommended quantity of HMF in honeydew honey is 40 mg/kg. After all pretreatments, it is evident that the HMF content significantly increased. The highest value of HMF was recorded in sample 7 (25.84 mg/kg), and the lowest in sample 8 (22.63 mg/kg). HMF content and diastase activity are indicators of the levels of honey freshness and overheating (Tosi et al. 2004). Hidroxymethylfurfural (HMF) is produced by acid catalyzed dehydration of hexoses. There is virtually no HMF in fresh honey, and its formation is influenced by several factors such as temperature and time of heating, storage conditions, pH and floral source from which the honey has been extracted (Rahima 2013). The presence of organic acids and low water activity also favours the production of HMF (Rahima 2013). Effects of thermal treatment on HMF content in honey was studied by Tosi et al. (2004) and they reported that the kinetics of HMF formation did not depend on the initial HMF concentration in honey. They also reported that during thermal processing, the time temperature combination is very crucial for maintaining the HMF level below the maximum permissible limit. Thrasyvoulou et al. (1994) liquefied crystallized honey samples by ultrasonic waves at 23 kHz and by heating at 60 °C for 30 min. They reported that the average increase in HMF content was significantly lower (86%) in samples liquefied by sonication compared to samples liquefied by heating (129%). The increase in the HMF content in the samples subjected to heat and ultrasound pretreatment was confirmed by Chaikham et al. (2016), which is in line with the results shown in our study.

Diastase activity in the sample C was 47.67, which is in line with the the internationally recommended quantity of diastase in honeydew honey—Not lower than 8. The lowest value was recorded in sample 10 (37.37), and the highest in sample 9 (51.69). Diastase is very sensitive to heat, which causes a change in enzyme activity. Trasyvoulou et al. (1994) conclude that both ultrasound and heat pretreatment of honey reduce the diastase activity, which is partly in line with our results, because the diastase activity increased in the samples representing a combination of ultrasound and the highest temperature.

Determination of mineral content

Some elements were found in very small amounts, while others were not detected (Fe, Mn, Se, Mo, Ni, Cd, Cr, Pb). Some heavy metals, such as arsenic, lead, mercury and cadmium are toxic if the maximum limit is exceeded. The most prevalent element was K with the concentration of 6.302 mg/g FW, which is much higher than the content recorded by Vanhanen et al. (2011), which was 3.64 mg/g FW. After potassium, the second most prevalent element was Na (314.76 µg/g FW), followed by Ca (132.4 µg/g FW), Mg (77.29 µg/g FW), P (51.87 µg/g FW), Al (8.79 µg/g FW), Zn (2.47 µg/g FW) and Cu (0.42 g/g FW). The mineral content in honey is dependent on the natural absorption of minerals by plants from the soil and the environment (Vanhanen et al. 2011).

Water activity

The water activity (Table 1) in the sample C was 0.597. After the pretreatments, the water activity values ranged between 0.548 (sample 9) and 0.577 (sample 10). Similar results were presented in the study by Rahima (2013). The a_w limit value for yeasts in honey is around 0.61–0.62 (Rahima 2013), and many types of bacteria will only grow if water activity is 0.94–0.99, while the average water activity in honey ranges between 0.56 and 0.62 (Molan 1992).

HPLC analysis of phenolics

Many studies have shown that honey contains large number of different phenolics and flavonoids; their composition depending on the botanical origin, environmental, seasonal and processing factors (de Oliviera et al. 2017). HPLC–DAD revealed that the phenolic compounds detected in all samples were catechin, malvidin and gallic acid, of which catechin was the most abundant and gallic acid the least represented (Table 2). A significant decrease in the amount of catechin and malvidin relative to the sample C was found in the samples treated with ultrasound for 5 min at 30 °C, while for gallic acid the greatest decrease was observed in the samples exposed to ultrasound for 10 min at 30 °C (same as for 5 min at 45 °C). Probably heat treatment during honey processing as indicated by high HMF content might have caused the complete loss of other phenolic compounds in our samples. It has been postulated that radiation-induced radicals can break glycosidic bonds in honey and form new compounds by releasing phenolics from glycosidic compounds, as well as breaking down larger phenolic compounds into smaller ones (Hussein et al. 2011).

Table 2.

HPLC analysis of phenolics, total phenolic and flavonoid content and antioxidant activity

Sample	C	1	2	3	4	5	6	7	8	9	10
Catechin (mg/100 g)	0.804a	0.886f	1.104b	1.040h	0.694i	1.220j	1.070k	1.108b	1.052l	0.979m	1.012b
Malvidin (mg/100 g)	0.264a	0.090h	0.267a	0.208i	0.062j	0.213k	0.231l	0.194m	0.203n	0.162p	0.222b
Gallic acid (mg/100 g)	0.014a	0.048h	0.013a	0.019i	0.038j	0.008k	0.011b,d	0.008k	0.009b,k	0.016 l	0.010b,c
Total phenolic content (mgGAE/g)	1.493a  ± 0.023	1.469e  ± 0.002	1.505a,f  ± 0.004	1.570 g  ± 0.005	1.453d,e  ± 0.005	1.497a  ± 0.001	1.517f  ± 0.002	1.543h  ± 0.002	1.557h  ± 0.000	1.564g  ± 0.007	1.361b  ± 0.001
Total flavonoid content (mgGAE/g)	0.846a  ± 0.004	0.710b  ± 0.003	0.770c  ± 0.002	0.819f  ± 0.008	0.769c  ± 0.005	0.818f  ± 0.004	0.821f  ± 0.001	0.799g  ± 0.001	0.816f  ± 0.001	0.824f  ± 0.005	0.703b  ± 0.002
DPPH (μgGA/mg)	0.186a  ± 0.001	0.365c  ± 0.002	0.368c  ± 0.002	0.367c  ± 0.002	0.403e  ± 0.003	0.406e  ± 0.001	0.404e  ± 0.000	0.401e  ± 0.005	0.404e  ± 0.003	0.402e  ± 0.001	0.346b  ± 0.002
FRAP (mmol Fe2+/g)	8.006a  ± 0.049	9.592g  ± 0.209	9.589 g  ± 0.024	9.640 g  ± 0.029	8.050a,b  ± 0.000	8.098a,b  ± 0.00	8.147a,b  ± 0.002	8.450d  ± 0.172	8.488d  ± 0.008	8.564d,f  ± 0.034	8.145a,b  ± 0.08
ABTS (g/ml)	0.0044a  ± 0.0006	0.0037d,e  ± 0.0001	0.0034d,e  ± 0.0001	0.0029f,g  ± 0.0000	0.0025g  ± 0.0001	0.0039b,c,d,f  ± 0.0000	0.0038d,e,f  ± 0.0000	0.0033e,f  ± 0.0000	0.0037d,e  ± 0.0001	0.0048a  ± 0.0000	0.0043a,b,c  ± 0.0001

^*Means followed by the same small letter in the row do not differ from one another by Tukey test (p ≤ 0.05)

Treatments key: C —Control; 1—Ultrasound 30 °C, 1 min; 2—Ultrasound 45 °C, 1 min; 3—Ultrasound 60 °C, 1 min; 4—Ultrasound 30 °C, 5 min; 5—Ultrasound 45 °C, 5 min; 6—Ultrasound 60 °C, 5 min; 7—^{Ultrasound 30 °C, 10 min; 8—Ultrasound 45 °C, 10 min; 9—Ultrasound 60 °C, 10 min; 10—Pasteurisation at 65 °C, 10 min}

Antioxidant activity

The results of antioxidant activity are shown in Table 2. The phenolic content in sample C was 1.493 mg GAE/g. After the ultrasound pretreatments, the phenolic content slightly increased in most of the samples. The highest value was recorded in sample 3 (1.570 mg GAE/g), and the lowest in sample 4 (1.453 mg GAE/g). Unlike the ultrasound pretreatments, the heat pretreatment caused a significant reduction of the phenolic content, resulting in the lowest value among all samples (1.361 mg GAE/g). The flavonoid content in sample C was 0.846 mg GAE/g. After the pretreatments, the flavonoid content was reduced in all samples, with the lowest value recorded in sample 10 (0.710 mg GAE/g) and the highest in sample 9 (0.824 mg GAE/g). It is evident that the phenolic and flavonoid content increased within the ultrasound pretreatments with the elevation of temperature, at the same duration of pretreatments. Nadeem et al. (2018) observed that ultrasound has positive effects on the nutritional status of a carrot and grape juice mixture, because it increases the total phenolic and flavonoid content and reducing power, as well as the antioxidant properties of juice, which is in line with our results. Chaikham et al. (2016) were subjected three different honey samples to conventional thermal (90 °C/5 min) and ultrasound (40 and 80% amplitudes/20 kHz/30 min) treatments and found that ultrasonic processing preserved phenols, flavonoids and ascorbic acid as well as FRAP value of honey samples better than the conventional thermal method.On the other hand, after comparing untreated honeys and honeys subjected to ultrasound, Mahnot et al. (2019) found that the phenolic content in honey subjected to ultrasound was slightly reduced.

Obtained value for DPPH assay in the initial untreated sample was 0.186μgGA/mg. After the pretreatments, it was evident that the values for DPPH assay in all samples increased significantly (from 0.346 μgGA/mg in sample 10 to 0.406346 μgGA/mg in sample 5). The obtained value for the FRAP assay in the untreated sample was 8.006 mmol Fe²⁺/g. After the pretreatments, the values for FRAP assay increased in all samples, ranging between 8.098 in sample 5 and 9.640 mmol Fe²⁺/g in sample 3. Apart from that, it is evident that the highest values were obtained for the shortest ultrasound pretreatment, regardless of the temperature. After comparing untreated honeys and honeys subjected to ultrasound, Mahnot et al. (2019) found that there were no major differences. However, the FRAP values for some types did increase or decrease, which is partly in line with our results. The ABTS assay value (IC50) for the untreated honeydew honey was 0.0044 g/ml, and the only sample in which the ABTS value increased in comparison with the untreated sample was sample 9 (0.0048 g/ml). The lowest value was recorded in sample 4 (0.0025 g/ml). Honey contains important components with antioxidant activity, such as: phenolic acids, flavonoids, vitamins and enzymes. Honeydew honey is a dark type of honey, and dark types of honey are known to have stronger antioxidant activity. Apart from that, the improvement of antioxidant properties of honey may be caused by the increase in the availability of phenolic acids and flavonoids, because ultrasound intensifies extraction of bioactive compounds (Pico 2013). Quintero-Lira et al. (2017) showed that all types of honey subjected to ultrasound for five minutes showed insignificant changes in phenolic acids, flavonoids and antioxidant activity, but after 15 min of pretreatment, these parameters increased in some types of honey.

Antibacterial activity

The MIC/MBC values of the samples against the four bacteria are shown in Table 3. The heat pretreatment of honey resulted in an increase in the MIC and MBC values in comparison with the sample C. Compared to sample C only samples 3 and 7 had stronger inhibitory effect on S. aureus and sample 4 on E. coli and P. aeruginosa. Taking everything into consideration, sample 4 showed the best antibacterial activity. Interestingly, G( − ) bacteria were strongly inhibited than G( + ) bacteria which is not in line with the results from many authors (Molan 1992; Bogdanov 1997). The obtained results are interesting because the phenolic content of sample 4 is among the lowest measured, and the antioxidant activity of all samples are similar. Hence it follows that the antimicrobial properties of our samples could be attributed to the individual or synergetic effects of the high sugar osmolarity, the enzymatic generation of hydrogen peroxide or the presence of other minor compounds. Results of artificial honey showed that all bacteria were inhibited at 50% (w/v), except P. aeruginosa [inhibited at 25% (w/v)] with higher MBC values obtained [≥ 50% (w/v)]. The antimicrobial activity of honey is affected by a number of factors: high osmotic pressure, water activity, pH value, production of H₂O₂, methylglyoxal, antimicrobial peptide bee defensin-1, lysozyme, phenolic acids, flavonoids, etc. (Leyva-Jimeneza et al. 2019; Molan 1992). The peroxide activity of honey may be destroyed by heat, light and long-term storage of honey, and the antimicrobial activity of honeydew honey could therefore be explained by the combined effect of the non-peroxide (high concentration of phenolics and flavonoids) and the peroxide antimicrobial activity, with low concentration of free and bound water, a low pH value and high concentration of sugar.

Table 3.

Antibacterial activity of samples, artificial honey (minimum inhibitory—MIC and minimum bactericidal concentration—MBC) and antibiotic discs (diameter of inhibition zone in mm), on bacterial cultures growth

Sample/m.o	S. aureus ATCC 25923		E. coli ATCC 25922		B. cereus ATCC 7004		P. aeruginosa ATCC 10145
	MIC %(w/v)	MBC %(w/v)	MIC %(w/v)	MBC %(w/v)	MIC %(w/v)	MBC %(w/v)	MIC %(w/v)	MBC %(w/v)
C	12.5	25	6.25	25	6.25	12.5	6.25	12.5
1	12.5	50	6.25	25	12.5	12.5	6.25	12.5
2	12.5	50	6.25	25	12.5	25	6.25	12.5
3	6.25	25	6.25	25	12.5	> 50	6.25	6.25
4	12.5	50	< 3.125	25	12.5	25	< 3.125	6.25
5	12.5	50	6.25	25	12.5	12.5	6.25	12.5
6	12.5	50	6.25	25	12.5	25	6.25	6.25
7	< 3.125	25	6.25	25	12.5	> 50	6.25	12.5
8	12.5	25	6.25	25	12.5	> 50	6.25	12.5
9	12.5	25	6.25	25	12.5	50	6.25	12.5
10	25	25–50	25	25–50	12.5	25	12.5	25
Artificial Honey	50	> 50	50	50	50	> 50	25	50
Ampicillin 10 mg (mm)	33.00 ± 2.16		16.00 ± 4.39		10.88 ± 2.53		*NA
Ciprofloxacin 5 mg (mm)	29.25 ± 2.99		37.00 ± 3.37		27.38 ± 3.54		33.33 ± 0.58
Erytromycin 15 mg (mm)	27.75 ± 2.99		10.75 ± 1.50		24.5 ± 2.52		*NA
Gentamicin 10 mg (mm)	29.75 ± 1.90		25.50 ± 1.29		24 ± 2.53		21.33 ± 1.53

^*NA–No activity

Treatments key: C—Control; 1—Ultrasound 30 °C, 1 min; 2—Ultrasound 45 °C, 1 min; 3—Ultrasound 60 °C, 1 min; 4—Ultrasound 30 °C, 5 min; 5—Ultrasound 45 °C, 5 min; 6—Ultrasound 60 °C, 5 min; 7—Ultrasound 30 °C, 10 min; 8—Ultrasound 45 °C, 10 min; 9—Ultrasound 60 °C, 10 min; 10—Pasteurisation at 65 °C, 10 min

The available literature offers various information of the antimicrobial activity of honeydew honey. Fikselova et al. (2014) studied the antimicrobial activity of a number of honeydew honeys and found that the most sensitive bacteria strains were E.coli and P. aeruginosa, while B. cereus was the most resistant, which is in line with our results. On the other hand, in the study by Srećković et al. (2019), honeydew honey was more effective against Gram ( + ) bacteria compared to Gram ( − ). Rahima (2013) tested the antimicrobial activity of rosemary honey-raw, subjected to ultrasound and subjected to heat pretreatment, against several microorganisms, and it was found that the honey subjected to ultrasound showed the best activity, which was explained by a possible synergistic effect of the pretreatment and the substances present in the honey (e.g. glucose oxidase). Nicodim et al. (2014) tested the antibacterial activity of honeys subjected to heat pretreatments and found that the lower temperatures (30 and 50 °C) did not have any effect on its antibacterial properties, while the antibacterial activity was reduced at the temperature of 70 °C, which is in line with our results.

Experimental design

With a view to examining the effect of ultrasound pretreatment on the characteristic honey quality parameters, we conducted a series of experiments in accordance with the experimental design requirements for single-factor categorical parameters. The applied experimental layout and the results of the relevant measurements are presented in Fig. 1.

Fig. 1.

Effect of ultrasound pretreatment (1 min, 30 °C) on all dependent variables

The categorical factor tested is the presence of ultrasound pretreatment (1 min, 30 °C) at two levels, with ultrasound (Yes) and without ultrasound (No), with three repetitions at each level. Table 4 shows that there is some variation in individual values in all dependent variables, except for phenolic content. A specific answer to the question if these variations are statistically significant was obtained after the analysis of the variance of the obtained data.

Table 4.

Experimental design and results of measurements with characteristic parameters of the analysis of variance (ANOVA)

	Factor	Response
	1	1	2	3	4	5	6	7
Run	A:	HMF content	Diastase activity	aw	Phenolic content (Fenoli)	DPPH	FRAP	ABTS
	Pretreatment					value	value	value
		mg/kg	DN		mg/g	μg/mg	mmol Fe2+/g	g/ml
1	Yes	23.7	39.27	0.562	1.467	0.3668	9.3833	0.0036
2	No	5.3193	46.98	0.597	1.4691	0.1875	7.98576	0.0042
3	Yes	23.97	37.07	0.562	1.471	0.3634	9.8007	0.0037
4	Yes	23.835	38.17	0.563	1.469	0.3651	9.592	0.00365
5	No	5.3208	48.36	0.597	1.5142	0.1858	7.97064	0.004
6	No	5.7552	47.64	0.597	1.4942	0.186	8.06143	0.0041
	F-value	12,445.4	160.21	10,816	3.22	25,514.7	164.32	48.6
	p-value	< 0.0001	0.0002	< 0.0001	0.1472	< 0.0001	0.0002	0.0022
	Adjusted R2	0.9996	0.9696	0.9995	0.3074	0.9998	0.9703	0.9049
	Predicted R2	0.9993	0.9452	0.9992	− 0.2467	0.9996	0.9465	0.8289
	Adeq precision	157.768	17.9	147.078	2.537	225.897	18.1295	9.859

HMF content, aw activity and DPPH value have a very high F-value and very low p-values, which implies that the effect of ultrasound pretreatment is statistically very significant. The same conclusions can be drawn for diastase activity, FRAP and ABTS values. The F-value of phenolic content is relatively low (3.22), confirmed by a relatively high p-value, which is higher than the limit value (p = 0.05). Therefore, it cannot be stated that pretreatment has any significant effect on the phenolic content. All above-mentioned claims are supported by the experimental data fitting statistics, according to which the predicted R² for all dependent variables, except for phenolic content, are quite similar to the adjusted values (Adjusted R²), i.e. they only differ by less than 0.2, which is the acceptance limit. Apart from that, the reliability of experimental measurements is quite high, which is expressed by high values of the signal to noise ratio (Adeq precision), and it is higher for all dependent variables, except for phenolic content, than the limit value for the analysis reliability, which is 4. Image 1 shows the charts related to the ultrasound pretreatment effect on all dependent variables for which this effect is significant, as established by previous analyses. The charts show the average values of the dependent variables taking into consideration the relevant level of pretreatment (Yes or No), as well as the relevant LSD intervals, i.e. the intervals of 95% of the least significant differences (Fisher’s LSD test), which is presented by vertical interval lines. As shown in the image, there is no horizontal overlapping of LSD intervals on any chart, which further confirms the significance of ultrasound pretreatment on the numeric values of dependent variables.

Figure 2 shows the chart related to the effect of pretreatment on the phenolic content. As shown on the chart, unlike the previously mentioned charts, there is horizontal overlapping of LSD intervals. Therefore, it cannot be stated with sufficient reliability that ultrasound pretreatment affects the phenolic content.

Fig. 2.

Effect of ultrasound pretreatment (1 min, 30 °C) on the phenolic (fenoli) content

Conclusion

The ultrasound pretreatment, especially at lower temperatures, represents a technique which enables preservation and improvement of the biological properties of honeydew honey. The heat pretreatments of honeydew honey (pasteurisation at 65 °C, 10 min) and 9 different ultrasound pretreatments (a combination of three temperatures: 30, 45 and 60 °C and three different lengths of the process: 1, 5 and 10 min) resulted in a significant change in the following parameters: water content, pH, electrical conductivity, HMF content, diastase and water activities. Ultrasonication yielded higher quality of antioxidant compounds and properties as compared to the conventional thermal technique and noticeably improved the levels of total phenolic content and antioxidant capacity. In most cases, ultrasound improved the antibacterial activity, while the heat pretreatment resulted in a significant reduction of the antibacterial activity.

Author contributions

MS, DC, AS, LT, AV, SP, MŽ, conceived and designed the experiments; performed the experiments; analyzed the data; contributed reagents, materials and analytical tools; wrote the paper.

Compliance with ethical standards

Conflict of interest

The authors declare no conflict of interest.