Abstract
BACKGROUND:
Honey is a vegetable and animal product which comes from nectar and/or honeydew. It is used in different nutritional and therapeutic fields.
OBJECTIVE:
Melissopalynology and physicochemical analysis of Algerian honeys, determination of their phenolic compounds and authenticity parameters and the evaluation of their antioxidant properties.
METHODS:
Twenty Algerians honey were studied for their physicochemical parameters (moisture, pH, proteins, proline, hydroxymethylfurfural, ash, color, electrical conductivity, and optical rotation), floral origin and phenolic compounds contents. Antioxidant activities were tested too.
RESULTS:
Melissopalynologycal analyses revealed that the studied honeys were twelve multifloral, seven Fabaceae, and one Myrtaceae. All honeys were acidic (3.65≤pH≤4.35) and most of them were low in moisture content. The electrical conductivity varied between 0.29 mS/cm and 1.78 mS/cm. Ash, protein and proline contents results showed that the majority of honeys were in agreement with the legislation and were authentic. The color varied from mimosa yellow to dark brown. The specific rotation was levorotatory in most honey samples and the hydroxymethylfurfural values (from 1.5 mg/kg to 34.73 mg/kg) agreed with the international requirements. Honeys were rich in total phenolic compounds, 22.41 (Honey11) to 96.16 (Honey15) mg gallic acid equivalents/100 g, and flavonoids, 8.90 (Honey11) to 80.02 (Honey02) mg quercetin equivalents/100 g. Honey samples 15, 03, 05, 01, and 06 exerted more than 50% reduction of 1,1-diphenyl-2-picrylhydrazyl and 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonate) radicals and were able to reduce iron while honey samples 12, 18, 19, 14, and 11 chelate efficiently iron. High significant correlations between physicochemical parameters and antioxidant activities were found.
CONCLUSION:
The Algerian honeys analyzed were authentic and variations in their quality parameters and phenolics composition were directly associated with their demonstrated antioxidant properties.
Keywords
Introduction
Honey is the main product of beekeeping. It is produced by Apis mellifera bee from the nectar of flowers, secretions of living parts of plants, or excretions of plant-sucking insects on the living parts of plants. These sugary sources are collected, transformed, combined with specific substances from bees, dehydrated, and stored in the beehives’ combs to mature, resulting in the honey itself [1].
The natural product of honey bees contains ∼200 bioactive substances because of the great variety of plants that provide nectar and pollen. Sugars represent the major component of honey, including around 75% monosaccharides, 10–15% disaccharides, and small amounts of other sugars. Minor compounds include polyphenols, vitamins, enzymes, amino acids, minerals, organic acids, and volatiles. Most of these compounds act together to confer synergistic properties used in different nutritional and therapeutic fieldss [2–4].
The quality, composition, color, aroma, and flavor of honey depend mainly on the flowers, geographical regions, climate, and honeybee species involved in its production and are also affected by weather conditions, processing, manipulation, packaging, and storage time [5, 6].
Honey is a highly symbolic and well-regarded product in the Algerian population. It is a viscous, aromatic, sweet, and flavorful product that comes in the first order of natural products, since it has been used as an alimentary supplement, in medical therapies, and natural food, without the addition of any substance in its elaboration. The chemical composition and health benefits of honey heavily depend on the botanical and geographical origin [7, 8]. Hence, the most important bioactive fraction in honey is made up of secondary metabolites that are present in nectar, and the most important of them are phenolic compounds. These bioactive compounds are used in the treatment of diabetes, dyslipidemia, and skin lesions in addition to having high antibacterial and antifungal effects [9, 10]. Antiproliferative, immunomodulatory, and antioxidant activities have also been reported [11–13]. Variations in several physical properties (acidity, phenolic content, minerals, color, and antioxidant level) of honeys are directly associated with its biological properties, which may be important when assessing and correlating the physical properties and medicinal value of different honey types [14, 15].
Due to its peculiar characteristics, this product has aroused great interest in current functional food research which is a trend in food chemistry. Hence, to contribute more to the knowledge of Algerian honeys and considering the lack of detailed study in the literature, the present research aimed (i) to characterize the botanical origin (blossom and/or honeydew) of honeys, (ii) to determine the physicochemical and the authenticity parameters and (iii) to quantify the amounts of phenolic compounds that are likely responsible for most of the bioactivity in honey and are considered as potential markers for their botanical origin and finally (iv) to evaluate the antioxidant properties with the use of different methods in order to determine the potential functional value of these honeys. This study can contribute to the evaluation of the quality of Algerian honeys, verify their compliance with international standards (Codex Alimentarius Commission, European Commission), and characterize Algerian honey types.
Material and methods
Honey samples
Twenty samples of honey produced in different regions of Algeria of which the majority come from Bejaia were collected from beekeepers. The samples were stored in a refrigerator in airtight plastic containers until analysis. The regions from which the samples of honey were collected were indicated in Table 1.
Location, state and color of Algerian samples honeys
Location, state and color of Algerian samples honeys
Pollen analysis was carried out using the methods previously reported by Louveaux [16].
10 g of honey was dissolved in 20 ml of acidulated water (H2SO4, 5%) and then centrifuged for 10 min at 3000 tours/min. After the supernatant was discarded, 10 ml of acidified water were added to the sediment and then centrifuged again under the same conditions and the supernatant was then removed. The residue obtained was then observed under×40 magnification using a Light optical microscope (Zeiss Axiolab, Göttingen, Germany). Pollen grains were identified by referring to general palynological databases CETAM (Bee Studies Center of Moselle, France) and the existing information about the flora of Algeria and southern regions where honey samples were collected [17].
Physicochemical parameters
Samples were analyzed for moisture content, pH, proline, HMF, electrical conductivity (EC), and specific rotation according the method reported by Bogdanov et al. [18]. Color and protein contents were determined too.
Water content and pH
Water content (moisture) was determined from the refractive index of the honey using ABBE-type refractometre (AR12, SCHMIDT, and HAENSCH). The values were obtained by reference to CHATAWAY table. The pH was assessed in a 10% (w/v) solution of honey in distilled water by mean of HANNA pH meter.
Proline
The proline content was determined by using a color comparison after applying ninhydrin, with a proline standard. The absorbance was determined (510 nm) using Vis-7220G (biotech engineering management GO LTD “UK”) spectrophotometer. Proline concentration in mg/kg of honey was calculated as follows:
where Es is the absorbance of the sample solution; Ea is the absorbance of the proline standard solution (average of 3 readings); E1 is the mg of proline used for the standard solution; E2 is the weight of honey sample in grams; 80 is the dilution factor.
Hydroxymethylfurfural (HMF) content was determined by a spectrophotometric method. A quantity of 5g of honey samples was dissolved in 25 ml of distilled water. A 0.5 ml of Carrez I solution (15%) and 0.5 ml of Carrez II solution (30%) were added. The mixture was adjusted to 50 ml with distilled water. After filtration, the first 10 ml of the filtrate is discarded. Two aliquots of 5 ml are then introduced into two tubes tests, one with 5 ml of distilled water (analysis aliquot) and the other with 5 ml of sodium bisulphate at 2% (reference aliquot). The absorbance was read at 284 nm and 336 nm. The difference between the absorbance of a clear aqueous honey solution (clarified with Carrez I and II solutions) and the same solution after the addition of bisulphite was determined to prevent other components from interfering. The HMF content was calculated as follows:
where A284 is the absorbance at 284 nm; A336 is the absorbance at 336 nm; F is the dilution factor, when the absorbance is greater than 0.6; W is the weight of honey sample in grams; 149.7 is a constant.
The specific rotation was measured in a polarimeter (Polaser-SI) using a clear and filtered aqueous solution of honey. Measurements were taken at 20°C.
Electrical conductivity and ash content
Electrical conductivity was measured in a 20% (w/v) solution of honey in deionized water with low electrical conductivity (< 14μS/cm) using Eutech Con 510 electrical conductivity meter. The EC was calculated as follows:
where 0.032 is the correction factor; T is the temperature of measurement.
The ash content was directly determined according to Piazza et al. [19] as follows:
Honey color intensity was determined according to the method described by Bath and Singh [20]. 1g of honey sample was dissolved in 4ml of distilled water. After homogenization, the absorbance was read at 450 nm.
Proteins content
The content of proteins was determined by the method of Azeredo et al. [21] using an albumin standard solution of bovine serum (BSA). The BSA equivalents for the honey solutions were evaluated. The results were expressed as mg of BSA equivalents per 100g of honey (mg BSAE/100 g) (y = 1.595x; R² = 0.977).
Quantification of antioxidants
Total phenolic compounds
Total phenolics of honey samples were determined spectrophotometrically by the Folin–Ciocalteu method (Ouchemoukh et al. [22]), using gallic acid as the calibration standard (0.04–0.08 mg/ml). The content of total phenolics was expressed as mg of gallic acid equivalents (mg GAE/100 g) (y = 5.650x; R² = 0.998).
Flavonoïds
The content of flavonoïds was determined according to Ouchemoukh et al. [22], using a standard curve with different concentrations of quercetin (0.1–0.5 mg/ml) whose linearity was 0.9968 (R2) (y = 1.2205x). The results were expressed as mg of quercetin equivalents per 100 g of honey (mg QE/100 g).
Antioxidant activities
Reducing power
The reducing power of the honey samples was estimated according to the method described by Beretta et al. [11]. Gallic acid was used as standard to produce the calibration curve (0.0066-0.033 mg/ml). The results were expressed as mg gallic acid equivalents (mg GAE) per 100 g of honey (y = 35.38x; R² = 0.998).
Phosphomolybdenum method
Total antioxidant capacities of the analyzed honeys were evaluated by the phosphomolybdenum method as described by Amessis-Ouchemoukh et al. [23]. Gallic acid was used as standard (0.04–0.08 mg/ml). The results were expressed as mg gallic acid equivalents (mg GAE) per 100 g of honey (y = 3.637x; R² = 0.998).
Ferric reducing antioxidant power
The ferric reducing antioxidant (FRAP) assay was assessed by the method of Benzie and Strain [24], using gallic acid as the calibration standard (0.004-0.02 mg/ml). The results were expressed as mg gallic acid equivalents (mg GAE) per 100 g of honey (y = 45.33x; R² = 0.996).
ABTS radical scavenging assay
The 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonate) (ABTS) radical scavenging assay was evaluated using the procedure described by Amessis-Ouchemoukh et al. [23]. The percentage inhibition was calculated as ABTS radical scavenging activity using the following formula (Eq. 4).
where Ac is the control absorbance; As is the absorbance of the honey sample.
Free radical scavenging capacity for honey samples was also studied as described by Amessis-Ouchemoukh et al. [25], through the evaluation of the free radical scavenging effect on 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical. Results were expressed as a percentage of inhibition of DPPH radical using the given equation 4.
Metal chelating activity
The chelating ability of honey is determined by following the inhibition of the formation of Fe2 +-ferrozine complex after incubation with divalent iron. The ability of honey to chelate iron (II) was assessed by the method of Amessis-Ouchemoukh et al. [26]. The ratio of inhibition of ferrozine-Fe2 + complex formation was calculated with the same formula of Eq. 4.
Statistical analysis
The averages and the standard deviations were calculated with Microsoft Office Excel 2007. The software STATISTICA 5.5 was used to compare the different results by the analysis of variance with one factor (ANOVA). The results were classified by decreasing order. Analyses were performed in triplicate and differences were considered significant at *p < 0.05. The values obtained carrying the same letter do not present any significant difference at p < 0.05.
Results and discussion
Pollen analysis
In the present study, melissopalynologycal analyses was carried out by identifying the most numerous pollen grains and those grains with specific morphologic characteristics. The abundance of each taxa (frequency of pollen appearance) was categorized as follows: Predominant pollen > 45%, secondary pollen: 16–45%; minor pollen: 3–15%, and important minor pollen < 3% [16].
Pollen analysis showed that 60 % of honey samples were multifloral and 40% were monofloral (Table 2), belonging to two different botanical families, Fabaceae sp., which was the most predominant source used by honeybees in Algeria, followed by Myrtaceae sp. These genus were important melliferous plants in the mediterrean area. According to the obtained results, Bejaia region was mainly characterized by multifloral honeys. This result was in agreement with Ouchemoukh et al. [22] which confirmed that the honeys of northern Algeria contain more botanical families than those from other regions; due to the difference in the geographical characteristics of the harvesting sites, the botanical diversity of northern and southern Algeria, and the absence of large-scale monocultures in the studied region. Multifloral honeys were represented with H4-10,13,15,16,19 and H20 samples, and several predominant and secondary pollen types were found in them (Liliaceae, Fabaceae, Apiaceae, Ericaceae, Rhamnaceae, Rutaceae, Myrthaceae, Asteraceae, etc.). Monofloral honeys were represented with H2, H3, H11, H12, H14, H17, and H18 samples where the predominant pollen content was Fabaceae (more than 45%) and H1 sample with the predominant pollen Myrtaceae (50%). In addition, other different minor pollen types exist.
Pollen spectra and pollen percentages of the analyzed honey
Pollen spectra and pollen percentages of the analyzed honey
Physicochemical analyses of honey from different regions of Algeria were summarized in Table 3. Water content, a quality criterion used primarily to estimate the maturity degree of honey, ranged between 14.95% (H20) for Adekar honey (Bejaia) and 22.27% (H08) for Bourdj Bouarreridj honey with an average of 17.64±0.43%. These values were below the maximum limit (20%) set by the European Commission [1] and were different from those reported by Belay et al. [27] (15.87% to 19.35%) on Ethiopian honeys and Escuredo et al. [28] (16.9% to 18%) on Atlantic European area honeys. Honey samples H08, H15, and H10 showed very high humidity which could be mainly explained by the premature extraction, the conditions of elaboration and preservation, and the harvest period of these honeys. As reported by Shantal Rodriguez Flores et al. [29], moisture variation of analyzed honeys may be due to the water content of plants used by worker bees, the strength of bees colonies, the floral origin (nectar and/or honeydew), the climate and the abilities of the beekeeper.
Some physicochemical characteristics of Algerian honeys
Some physicochemical characteristics of Algerian honeys
Values are mean±standard deviation. Means followed by the same letter in each column were not different according to ANOVA (Analysis Of VAriance) (p < 0.05).
The pH results showed that all honeys (multifloral and monofloral) are acidic, having a pH in the range of 3.65–4.35. The acidity of honey is due to the presence of organic acids, particularly the gluconic acid and inorganic ions such as phosphate and chloride [30]. These results agreed with data reported by Silva et al. [31] (3.45 to 4.70) on Portuguese honeys; Serem et al. [32] (3.87 to 5.12) on South African honeys and Özcan and Ölmez [33] (3.61 at 4.67) on Turkish honeys. Though the honeys H04 and H18 were from two different regions, they had the same pH (P < 0.05). The pH is between 3.5 and 4.5 for nectar honeys and between 4.5 and 5.5 for honeydew honeys [27]. Therefore the analyzed honeys are nectar honeys.
The electrical conductivity of honey samples varied between 0.29 mS/cm and 1.78 mS/cm with an average of 0.94±0.01 mS/cm. This average value was higher than that of Italian honeys obtained by Fallico et al. [34] (0.33 mS/cm). The majority of values obtained (except those of H01, H02, and H12) were included in the range of values reported by Belay et al. [27] on Madrid honeys (0.119 to 1.515 mS/cm). The electrical conductivity of samples H05, H17, H13, H19, H16, H07, H08, H15, H06, H03, H12, H01 and H02 exceeded 0.8 mS/cm, suggesting the presence of honeydew in honeys mentioned and the rest samples were nectar honeys. Honeys H08 from Adekar (Bejaia) and H16 from Bouhdjer (El-Taref) had the same conductivity (P < 0.05) which was 0.92 mS/cm. This parameter depends on acidity and the mineral content of the honey, moreover a correlation was found between the latter and the electrical conductivity (r = 1.00, p≤0.001). Indeed, more these parameters are higher, more the corresponding conductivity is high [19].
The ash content of the studied honey samples ranged from 0.17 to 1.02% with a mean of 0.54±0.005%. These results were different from those obtained by Ouchemoukh et al. [22] (0.06 to 0.54%) and Soria et al. [35] (0.003 to 0.99%). The variation of the ash content depends on the botanical origin, the season, and mineral composition of the soil [27]. A correlation was found between ash and color intensity (r = 0.54, p≤0.001). This result was in accordance with Alvarez-Suarez et al. [14] who found that the amber color of honeys corresponded to high ash content, while the lowest values of ash were found in light-colored honeys.
Protein contents of analyzed honeys varied from 34.99 to 137.82 mg BSAE/100 g with an average of 82.42±1.65 mg BSAE/100 g. These results agreed with those obtained by Yücel and Sultanoğlu [36] (13 to 115 mg BSAE/100 g) on Hatay honeys and distinct from those reported by Özcan and Ölmez [33] (60 to 106 mg BSAE/100 g) on Turkish honeys. According to the proteins content, the analyzed honeys were classified in decreasing order: H06 = H01 > H08 > H07 > H03 > H0 = H10 > H18≥H17≥H12 > H09 > H04 > H15 =H02 = H14 > H13 = H16 > H20 > H19 > H11 (P < 0.05). Protein content of Barbacha honey H06 (Bejaia) was higher and differed significantly from other samples. This could be explained by the presence of a high concentration of pollen in the honey and its botanical origin.
The concentrations of proline ranged between 731.27 mg/kg (H11) and 3730.90 mg/kg (H03) with an average of 2139.98±51.26 mg/kg. These rates were widely higher than 180 mg/kg, the minimal limit proposed by Bogdanov et al. [18]. These results indicated the maturity of analyzed honeys and the absence of adulteration. The average obtained was greater than that reported by Meda et al. [37] on Burkina Faso honeys (910±267.9 mg/kg) and it agreed to that found by Khalil et al. [38] on Algerian honeys (2131.47±0.90 mg/kg). The changes in proline levels may be attributed to the strength of bee colonies [39, 40].
The color of the analyzed samples varied from mimosa yellow to dark brown. The results of the color intensity ranged from 0.250 (H11) to 1.778 (H03) with an average of 0.930±0.008. High variability in the color of honeys could be attributed to their floral origin and ash content. The dark color of honey samples H03, H06 and H01 was probably due to the diversity of vegetation in regions of Tavel, Barbacha and Tizi adjissa, respectively, subsequently a high variability in their chemical composition.
The specific rotation was levorotatory in most honey samples. It ranged from –18.46 (H05) to –2.07 (H02). Only sample H12 was dextrorotatory (6.46). Tighremt honey (Bejaia) was dextrorotatory. This could be explained by its high content on trisaccharides (dextrorotatory) compared to fructose (levorotatory). According to Bogdanov et al. [41] natural honey has the property of rotating the plane of polarised light to the left. The overall specific rotation of honey depends on the concentration of the various sugars present in the honey. Chemini honey (Bejaia) was levorotatory over other honeys. This result may be due to the very high ratio fructose/sucrose.
The HMF values of the different analyzed honeys ranged from 1.5 mg/kg (H07) for Bouaiche honey (Bejaia) to 34.73 mg/kg (H08) for Adekar honey (Bejaia) with an average of 11.25±0.45 mg/kg. These results were within the range obtained by Özcani and Ölmez [33] (0.38 mg/kg to 42 mg/kg). Nevertheless, they were distinct from those reported by Belay et al. [27] (0.00 to 1.71 mg/kg). The results of this study showed the good quality of honey samples and they were in agreement with the International Honey Commission (40 mg/kg).
In the present study, the concentrations of total phenolic compounds (TPC) and flavonoïds were determined and presented in Table 4. The total phenolic contents ranged from 22.41 (H11) to 96.16 mg GAE/100 g (H15) with an average of 60.61±2.16 mg GAE/100 g. These results were within the range obtained by Silici et al. [42] on Turkish honeys (0.24 to 141.83 mg GAE/100 g) and Habib et al. [43] on Oriental honeys (30,81 to 132.60±1.94 mg GAE/100 g). However, they were superior to those obtained by Lachman et al. [44] on 40 Czech honey samples (8.36 to 24.25 mg GAE/100 g). Idjissen Honey (Bejaia) and Ilmaten honey (Bejaia) harvested from different regions were the richest in total phenolic compounds, 95.87 and 96.73 mg GAE/100 g, respectively and their rates had a significant difference compared to those of other samples. Nevertheless, the H11 honey recorded the lowest phenolic content (22.41 mg GAE/100 g). The honey samples H13 (Timezrit) and H06 (Barbacha) were not from the same region but contained the same proportion of total phenolic compounds (P < 0.05). The variation in the TPC may be attributed to the botanical origin, the year of the harvest and the storage temperature. Indeed, H16 and H17 honeys harvested in warm regions, El-taref and Guelma, respectively, were less rich in these compounds. Generally, the total phenolic content of clear honeys is less than that of dark honeys [45]. This was confirmed by sample H09 (Bejaia) who had a low TPC and very clear yellow color, while the honey samples (H03) and (H06) had a dark brown color and rich in TPC.
Total phenolic compounds and flavonoids founds in the studied honeys
Total phenolic compounds and flavonoids founds in the studied honeys
Values are mean±standard deviation (n = 3). Means followed by the same letter were not different. (p < 0.05) according to ANOVA (Analysis Of Variance).
Flavonoids (FLA) of honey samples ranged from 8.90 for Tigzirt honey (Tizi-Ouzou) to 80.02 mg QE/100 g for Tissa honey (Bejaia) with an average of 43.14±1.66 mg QE/100 g. These values were different from those reported by Pichichero et al. [46] on Italian honeys (6.73±0.34 to 16.43±0.82 mg QE/100 g) and Habib et al. [43] on Oriental honeys (12.76±0.74 to 109.49±0.99 mg EC/100 g). According to the variance analysis, the studied honeys are ranked in the following decreasing order: H03 = H15 > H10 = H01 > H06≥H16≥H07≥H05 > H12 > H14 = H13 = H04≥H18≥H17≥H02≥H20≥H08 > H19 = H09 > H11 (P < 0.05). Flavonoïd contents of H03, H15 samples were the highest and had a significant difference compared to other honeys. Honey samples H04, H13, H14, H17 and H18 that were harvested in different regions, didn’t register significant differences in the flavonoïds level. According to Lachman et al. [44], the flavonoïds content is influenced by the botanical and geographical origins and climatic conditions.
In this study, several assays were applied to assess the antioxidant activities of honey samples and the obtained results were given in Table 5 and Figs. 1–3.
Ferric reducing power, phosphomolybdenum and FRAP assays of honey samples
Ferric reducing power, phosphomolybdenum and FRAP assays of honey samples
Values are mean±standard deviation. Means followed by the same letter in each column were not different according to ANOVA (Analysis Of Variance).

ABTS radical scavenging activity exerted by honey samples. Vertical bars represent the mean±standard deviation for each data point. Bars marked with the same letter are not different according to ANOVA (Analysis Of Variance).

Radical DPPH scavenging activity of honey samples. Different letter(s) indicate the values are significantly different (*p < 0.05).

Ferrous ion chelating activity showed by samples of honeys. Vertical bars represent the mean±standard deviation for each data point. Bars marked with the same letter are not different according to ANOVA (Analysis Of Variance).
The ABTS radical scavenging activity of honey samples varied between 44.52% (H11) and 97.95% (H15) with an average of 76.11±1.07% (Fig. 1). The average value agreed to that obtained by Wilczynska [47] (79%) on buckwheat monofloral honeys. These results were different from those obtained by Habib et al. [43] on Oriental honeys (65.25 to 80.62%). Honey samples H01, H18 and H06; H02, H07, H17 and H04 had the same scavenger inhibition effect despite they were harvested in different geographical regions. Wilczynska [47] reported that antioxidant activity is high in dark honeys. It is the case of honey samples H15.
The DPPH radical scavenging activity exerted by honey samples (Fig. 2) varied from 33.4 (H11) to 94.50% (H15). The obtained results were different from those reported by Al et al. [48] on Romanian honeys (35.80 to 64.83%). Honey samples H18, H12, H20 and H07, collected in different regions, had the same antiradical activity and differed significantly from other samples. The antioxidant activity of honeys depends on the floral origin, structure, content and nature of the phenolic compounds and the presence of other non-phenolic compounds (catalase, vitamins) as reported in the literature [49, 50]. Very significant correlations were found between DPPH radical scavenging activity and phenolics compouns (Table 6).
Correlation matrix between antioxidants and antioxidant activities
The ferrous ion-chelating activity of honey samples varied between 3.17 (H01) and 77.28% (H12) with an average of 41.03±1.26% (Fig. 3). Honeys (H11, H17), (H04, H08), (H02, H08), (H09, H3), (H10, H016), (H1, H15) and (H20, H7) showed no significant differences, this could be explained by their similar content in iron chelating agents.
The reducing power of all analyzed honeys ranged from 9.71 to 358 mg GAE/100 g with an average of 172.79±2.87 mg GAE/100 g (Table 5). Honey samples M2, M9 and M18 showed no significant difference in their reductive ability (P < 0.05) despite they were harvested in different geographical regions. This could be explained by their similar polyphenols and flavonoïds content and/or other reducing agents present at low concentrations.
In the molybdate assay, the obtained results (Table 5) showed that honey samples had the ability to reduce the molybdenum. Their antioxidant activities ranged from 17180.24 to 25591.69 mg AAE/100 g with an average of 20030.10±272.82 mg AAE/100 g. Honeys H11 and H20; H08 and H12 showed no significant differences. This result may be due to the similarity of their antioxidants components.
The FRAP assay showed a significant reduction variability depending on the honey’s antioxidants. The results presented in Table 5 ranged from 12.10 (H08) to 139.43 mg GAE/100 g (H15) with an average of 42.84±0.41 mg GAE/100 g. These values were different from those obtained by Lachman et al. [44] on Czech honeys (29.5 to 77.6 mg AAE/100 g). The honey samples were classified in decreasing order according to their reduction power as follows: H15 > H01 > H18 > H17 > H12 > H20 > H05 = H06 > H14 > H16 > H03 > H13 > H07 > H02 > H19 > H04 > H10 > H11≥H09≥H08 (P < 0.05). Honeys H05 and H06; H08, H09 and H11 showed no significant differences despite they were collected in different geographical regions.
The correlation matrix between antioxidants and antioxidant activities was presented in Table 6. It indicated a very highly significant correlation between the concentrations of TPC and FLA (r = 0.81, p≤0.001). These results were in agreement with that obtained by Al et al. [48] and Alvarez-Suarez et al. [14]. The rates in antioxidants and antioxidant activities of honey samples showed very highly significant correlations. Indeed, the mean levels of TPC and FLA were highest in honey samples H03, H05 and H15 which correspond to the regions Tavel, Chemini and Ilmaten (Bejaia), respectively and demonstrated a better antioxidant activity. However, honey sample H11 from Tigzirt (Tizi-Ouzou) had a low content of these compounds, therefore a minimal antioxidant activity. Several studies reported that phenolic compounds and flavonoids were directly responsible on the antioxidant activity of honey.
A significant correlations were found between TPC and the following antioxidant assays: DPPH assay (r = 0.77, p≤0.001), ABTS (r = 0.88, p≤0.001), FRAP (r = 0.68, p≤0.001) and RP (r = 0.58, p≤0.001). These results were in agreement with those reported by Aljadi and Kamaruddin [51], Beretta et al. [11], Ferreira et al. [50] and Lachman et al. [44]. Flavonoïds content of the different honey samples showed also very highly significant correlations with DPPH assay (r = 0.50, p≤0.001), ABTS (r = 0.74, p≤0.001), FRAP (r = 0.53, p≤0.001) and RP (r = 0.53, p≤0.001). These results were confirmed by Al et al. [48].
The correlation matrix between physicochemical parameters and antioxidant properties was given in Table 7. The Color registered a very highly significant correlation with antiradical activities ABTS (r = 0.67, p≤0.001) and DPPH (r = 0.68, p≤0.001) and it is linked positively to reducing power (r = 0.53, p≤0.001). The rates of total phenolic compounds and flavonoïds presented a very highly significant correlation with color intensity (Table 7), (r = 0.79, p≤0.001) and (r = 0.76, p≤0.001), respectively. These results agreed with those obtained by Isla et al. [52]. The color intensity is related to the presence of pigments, such as carotenoïds and flavonoïds that are known to have antioxidant properties [38]. Much research demonstrated that honeys with intense color have very high levels of phenolic compounds while the light colored honeys have low concentrations [53–55]. Furthermore, they obtained good correlations between the following parameters: color, total phenolic compounds and antioxidant activities. Also, the color had very highly significant correlations with the protein content (r = 0.73, p≤0.001), proline levels (r = 0.76, p≤0.001) and electrical conductivity (r = 0.54, p≤0.001). It exhibited a very highly significant correlation with pH, this last showed a very highly significant correlation with the anti radical activities (ABTS and FRAP) and reducing power. The matrix correlation of honey samples showed a very highly significant correlation between the biochemical parameters and antioxidant activities. Honeys H03, H05 and H15 which correspond to regions of Tavel, chemini and Ilmaten (Bejaia), respectively were rich in total phenolic compounds, flavonoïds, and ash and showed better antioxidant activity. However, honey H11 from Tigzirt (Tizi Ouzou) contained low concentrations of these constituents and showed a minor antioxidant activity. The amounts of protein and proline gave low correlation coefficients with the antioxidant assays (DPPH and ABTS) which could be explained by their low contribution in antioxidant capacity while no statistically significant correlations were observed between these components and ferric reducing power tests (FRAP and RP).
Correlation matrix between physicochemical parameters and antioxidant activities
Correlation matrix between physicochemical parameters and antioxidant activities
In the present study, 20 honey samples where the majority came from the Wilaya of Bejaia, were studied for their quality criteria and their biological properties. Pollen analysis showed that the botanical origin of the analyzed honeys was mainly multifloral with 12 representative samples. The unifloral samples were represented by Fabaceae (seven honey samples) and Myrtaceae (one honey sample).
Physicochemical characteristics analysis showed that most of the studied samples were low in moisture content and therefore safe from fermentation. The pH results showed that all honeys are acidic (3.65≤pH≤4.35). The electrical conductivity varied between 0.29 mS/cm and 1.78 mS/cm and 13 honey samples were honeydew. Ash and protein contents of the studied honeys varied from 0.17 to 1.02% and from 34.99 to 137.82 mg BSAE/100 g, respectively. The concentrations of proline ranged between 731.27 mg/kg (H11) and 3730.90 mg/kg (H03) indicating the maturity of analyzed honeys and the absence of adulteration. The color varied from mimosa yellow to dark brown according to their floral origin. The specific rotation was levorotatory in most honey samples and the HMF values (from 1.5 mg/kg to 34.73 mg/kg) agreed with the international requirements. For antioxidant metabolites, honeys were found to be rich in total phenolic compounds, 22.41 (H11) to 96.16 (H15) mg GAE/100 g and flavonoids, 8.90 (H11) to 80.02 (H02) mg QE/100 g. So, they present an excellent dietary source of natural antioxidants and can be considered as foods with remarkable benefits for human health. All analyzed honeys exerted antioxidant activities. They were very effective in DPPH and ABTS radicals scavenging activities. On top of that, they showed a very important reducing capacity with iron and molybdenum or metal chelating activities. The matrix correlation of honey samples showed a very highly significant correlation between the biochemical parameters and antioxidant activities.
These results prove that Algerian honeys are an excellent source of antioxidants that can serve the industries and could be exploited in pharmaceutic and cosmetic fields. There is a need for further study of its composition and the identification of active ingredients to enable the clinical application of honey as an alternative treatment in medical practice.
Footnotes
Acknowledgments
We thank the Algerian Ministry of High Education and Scientific Research for their funding and encouragement for scientific research.
Author contributions
Nadia Amessis-Ouchemoukh: Conceptualization, Methodology, Investigation, Data curation, Writing the original draft; Nacera Maouche: Methodology, Investigation; Amar Otmani: Methodology, Investigation; Anass Terrab: Investigation; Khodir Madani: Investigation; Salim Ouchemoukh: Review and editing supervision.
Conflict of interest
The authors declare no conflict of interest for this paper.
Funding
This work was supported by the Ministry of Higher Education and Scientific Research, Algeria Research Projects University Training PRFU project N°: D01N01UN060120180005.
