Physicochemical, Antimicrobial Properties and Mineral Content of Several Commercially Available Honey Samples

Abstract

Ensuring food safety and protecting consumers are major aspects for commercialized products. Honey, the most prominent in the class of bee products, requires special regulations due to its origin as a natural product. Mislabeling, imitation, and adulteration represent a source of risks for human health. Specific determinations and analyses are essential for controlling the sector and preventing unfair competition. To compare and establish the correct labeling of several different honeys, melissopalynological, physicochemical, mineral content, and microbiological analyses were carried out on 18 samples commercially available in different countries, namely Türkiye, Romania, Bulgaria, and Northern Cyprus. The honey labels were in accordance with the determined pollen content. The physiochemical parameters showed high variability: 4.07–5.25 (pH), 79.95–83.45 (°Brix), 0.262–1.452 µS/cm (electrical conductivity), and 14.6–18.4% (moisture). The samples were quantitatively high in K, P, Na, and Ca, with the highest cumulative mineral content being found for honeys containing Fagaceae pollen. Additionally, the antimicrobial potential of the various honey samples was evaluated against selected bacteria, employing the disk diffusion and serial dilution methods. Results revealed that the honey samples exhibited increased antibacterial activity against Gram-negative bacteria, with notable activity against S. typhimurium, and moderate activity against Gram-positive S. aureus.

1. Introduction

There is an increasing global interest in the consumption of high-quality foods with additional beneficial effects. Honey, representing one of the oldest foods acknowledged for its naturalness, typically commands a high price for its geographical origin and stated botanical source, as well as for its superior organoleptic or bioactive properties. Due to this fact, consumers have a particular interest in the accurate labeling and traceability of honey and the lack of its adulteration [1]. During food adulteration, valuable food ingredients may be added, removed, or replaced with relatively cheaper substances for unfair economic gain. This process not only lowers the quality of honey, but it also poses several risks due to the potential toxicity [2,3]. Honey fraud poses a real economic threat to the producers as well, and it is estimated to occur with a prevalence of 10 to 30% worldwide, with even higher incidence in some regions [4].

Due to the unknown content of honey, it can be marketed under several different labels, and the botanical origin of honey represents a very active sector of modern apicultural research [5,6,7,8]. For this reason, correct labeling of honey is necessary, and it should aim to achieve the following points: the health protection of customers (by avoiding misleading claims regarding its beneficial properties and quality), the quality parameters required legally (by meeting specific criteria and being in the acceptable quality range), and the market integrity (by ensuring the trust of customers) [9,10,11].

The determinations in the field of honey quality highlight the necessity to make botanical origin analysis compulsory for unifloral honey, to prevent unfair competition, since these varieties are sold at a higher price compared to other polyfloral kinds of honey. It was determined that 33% of honey samples collected as chestnut honey in Türkiye did not have the minimum pollen content reported in the national food codex [12]. For honey originating from Poland, it was concluded that 48% of the samples were misclassified in comparison to the information declared on the label [7]. In Hungary, adulteration of acacia honey with sugar syrups had negative effects on some physicochemical parameters, such as moisture, sugar content, electrical conductivity, color, and pH [13]. In another study, it was concluded that 79.42% of honey samples originating in eastern Romania complied with the quality regulations. Moreover, these quality parameters are dependent on certain factors, and the associated risks can be mitigated through different strategies [14]. Indeed, producers can use different health claims and beneficial effects while using incorrect details regarding the botanical origin of the pollen. Therefore, regular monitoring and testing of honey samples can help detect any potential fraud or contamination. Implementing strict quality control measures can help maintain the integrity of the honey industry and ensure consumer trust in the product.

According to this pervasive potential for fraud and mislabeling of honey, it becomes evident that exhaustive and elaborate data on this matter are required for researchers around the world. Thus, the aim of the present study was to investigate commercially available samples purchased from four different countries (Türkiye, Romania, Bulgaria, and Northern Cyprus) in terms of their labeling, which was accomplished by assessing the correctness of the honey type (by determining the botanical origin of the present pollen grains), as well as through the determination of physicochemical parameters, mineral content, and antimicrobial potential. The limits of quality parameters have been compared to the available local food regulations and, where possible, to the existing scientific literature data.

2. Materials and Methods

2.1. Honey Samples

A total of 18 honey samples (of which 16 were monofloral and 2 were polyfloral) were used for the analysis, including the assessment of different quality parameters, and are represented in Figure 1. The honey was produced and marketed in different countries, namely Bulgaria, Romania, Türkiye, and the Turkish Republic of Northern Cyprus (Lefke), during 2023. After their purchase, the samples were stored in a dark environment at room temperature (20–24 °C) until further analysis. Detailed information about the honey samples is given in

Table 1. Detailed information regarding the considered honey samples.

ID	Code	Label Information and Botanical Origin		Honey Type	Country of Origin
H1	B1	Linden, Organic honey	Tilia tomentosa	Flower	Bulgaria
H2	B3	Mixture	Polyfloral	Flower	Bulgarian
H3	B2	Acacia	Robinia pseudoacacia	Flower	Bulgarian
H4	K2	Mountain	Mountain honey	Flower	Cyprus
H5	K1	Citrus	Citrus limon	Flower	Cyprus
H6	R1	Raspberry	Rubus idaeus	Flower	Romania
H7	R2	Knotweed	Fallopia aubertii	Flower	Romania
H8	R3	Linden	Tilia tomentosa	Flower	Romania
H9	R4	Oak, Honeydew honey	Quercus robur	Secretion	Romania
H10	4296	Thyme	Thymus sp.	Flower	Türkiye
H11	7480	Linden	Tilia sp.	Flower	Türkiye
H12	1374	Linden	Tilia sp.	Flower	Türkiye
H13	7694	Chestnut	Castanea sativa	Flower	Türkiye
H14	19701-1149	Chestnut	Castanea sativa	Flower	Türkiye
H15	6326	Chestnut	Castanea sativa	Flower	Türkiye
H16	1373	Chestnut	Castanea sativa	Flower	Türkiye
H17	3184	Chestnut	Castanea sativa	Flower	Türkiye
H18	T1	Oak, Honeydew honey	Quercus sp.	Secretion	Türkiye

Note: The designation of samples (ID) that are presented throughout the manuscript.

. According to the products’ labels, some of the honeys were described using specific terms, such as “polyfloral” and “mountain” honey (describing pollen originating from mountainous areas) [15]. The term “acacia honey” is used for the designation of honey produced from the flowers of Robinia pseudacacia [16].

2.2. Chemicals

Ultrapure water obtained from a purification system (Purelab CLASSIC UV, ELGA, Wycombe, UK) was used in all dilutions. Nitric acid (65%) and hydrogen peroxide (30%) were purchased from Merck (Darmstadt, Germany). The element standard solutions were prepared by adequate dilution of a multielement standard of 1000 mg/L Al, B, Ca, Cr, Cu, Fe, K, Mg, Mn, Na, Ni, P, and Zn and were obtained from Merck. The chemicals used in the analysis were of analytical-grade purity.

2.3. Melissopalynological Analysis

Melissopalynological determinations were applied to determine and confirm the botanical origin of the honey samples. The slides were prepared according to Louveaux et al.’s original method, from 1978 [17]. In this method, 10 g of honey was weighed, dissolved in 20 mL of water at 40 °C, centrifuged, and the sediment was spread across a microscope slide. The analysis was carried out by counting 1000 pollen grains present on the slide. The pollen grains were examined under a light microscope (ZEISS Axioscope 5, Oberkochen, Germany) using different magnifications: 40× for counting and 100× for identification of botanical origin. Only the percentages of pollen contained in each honey sample were reported, which was necessary for the validation of the correct labeling of the product that was commercially available.

2.4. Physicochemical Analyses

The physicochemical analysis was applied using several methods, namely the determination of honey moisture, pH, electrical conductivity, Brix values, color analysis, and water activity.

Honey moisture analyses were performed according to a previously described method [18], using a digital refractometer (Atago RX-9000α, Tokyo, Japan). After the refractive index of the refractometer was determined at 20 °C, the water content was calculated using a standard table. The percentage of water was used to express the results.

pH analysis was carried out with a pH meter (Thermo Scientific Orion Star A 215, Waltham, MA, USA) on a solution containing a fixed 10% (w/v) concentration, obtained by dissolving 10 g of honey in 100 mL of distilled water [18].

The determination of electrical conductivity was carried out at 20 °C on a solution containing a fixed 20% (w/v) concentration, obtained by dissolving 20 g of honey in 100 mL of CO₂-free and deionized distilled water, using a portable conductivity meter (Thermo Scientific Orion Star A 215, Waltham, MA, USA). Values were expressed as μS/cm [18].

Brix values in honey were determined by refractometry (Atago RX-9000α, Tokyo, Japan) following the method 31.119 described by AOAC International [19]. Water activity was measured at 25 °C using a water activity meter (ACQUALAB 4TE, Pullman, WA, USA) [20].

For the color analysis, the color values (L, a, b) of honey samples were determined using a colorimeter (CR-400 Chroma Meter, Konica Minolta, Tokyo, Japan). The color values were calculated according to the criteria determined by the International Commission of Illumination (CIE) based on three-dimensional color measurement. L*, a*, b*, color intensity, and shade were expressed accordingly: L* represents the sample’s lightness (100) to darkness (0), a* stands for the red (positive) to green (negative) degrees, whereas b* represents the yellow (positive) to blue (negative) degrees [21].

2.5. Mineral Content

The mineral content was determined using an inductively coupled plasma optical emission spectroscopy method (ICP-OES), after the preliminary preparation of the samples. Approximately 0.3 g of honey samples was weighed into the Teflon cups of a microwave acid digestion device (MARS 2 model, CEM, Matthews, NC, USA). To each sample, 6 mL of nitric acid and 1.5 mL of hydrogen peroxide were added and then burned in the microwave incinerator. Blind digestion was also performed without samples [22].

For the elemental analysis of the samples, the ICP-OES spectroscope (OPTIMA 2000 DV, Perkin Elmer Inc., Hebron, KY, USA) was used, presenting a quartz nebulizer gasifier, cyclonic spray chamber, and an integrated auto-sampler. A wash solution containing 3% nitric acid was prepared using 18.2 MΩ·cm ultrapure water obtained from a purification system (Purelab CLASSIC UV, ELGA, Wycombe, UK). For the preparation of ICP-OES calibration solutions, commercially available elemental standards were diluted with 1% nitric acid in ultradistilled water. In addition, ICP-OES calibration was performed before each measurement.

With the help of a peristaltic pump, the samples were sent to the cyclonic spray chamber with an argon gas flow. ICP-OES instrument software WinLab32, version 5.5 was used to control the instrument, including tuning, interferences, data acquisition, and data analysis. High-purity nitrogen gas was used in addition to argon gas to avoid interference.

2.6. Antibacterial Activity

For the determination of antimicrobial activity, the honey samples were preliminarily diluted in sterile deionized water to achieve different concentrations, representing 100, 75, 50, 25, 12.5, and 6.25% (expressed as w/v). The pathogenic bacterial strains included three Gram-negative bacteria, Escherichia coli (Migula) (ATCC 8739), Klebsiella aerogenes (ATCC 13048), and S. typhimurium (ATCC 14028), as well as three Gram-positive bacteria, Staphylococcus aureus subsp. aureus (ATCC 25923), Streptococcus pyogenes (ATCC 19615), and Listeria monocytogenes (ATCC 13932).

The studied bacteria were grown on nutrient agar at a temperature of 37 ± 1 °C for a total of 24 h. The bacterial inoculum for each species was standardized to 1.0 × 10⁸ CFU/mL in physiological saline using a 0.5 McFarland’s standard.

The antibacterial activity of all the honey samples against each of the bacterial organisms was examined using the agar well diffusion method [23]. For this, a volume of 200 µL of inoculum from each microorganism was mixed in Mueller–Hinton agar plates and solidified. After, four 8 mm diameter wells were opened in each Petri dish, and the analysis was carried out in two parallels. The prepared wells were filled with the honey sample itself and the previously diluted honey solutions. The dishes were incubated for 18–24 h at 37 ± 1 °C for bacterial growth. The zone of inhibition (in millimeters) was measured using a digital caliper.

The approximate minimum inhibitory concentrations (MICs) were determined by the agar well diffusion method [24]. Each honey sample was used to determine the lowest honey concentration (expressed as %) that provides an inhibition zone with a diameter greater than 8 mm.

2.7. Statistical Analysis

The analysis of variance (one-way ANOVA) was used to examine the collected data. Post hoc tests were conducted to identify any significant differences between the means after rejecting the null hypothesis. Tukey’s HSD test was employed to analyze the means and detect statistically significant differences between them at p < 0.05. The statistical analyses were assessed using SPSS software (version 19, SPSS Inc., Chicago, IL, USA). The data are represented as means ± standard error. Principal component analysis (PCA) was performed using the Facto Miner factoextra package [25]. Heatmaps and dendrograms were generated using the Euclidean distance with complete linkage to provide a visual representation to emphasize the similarities and differences between the color characteristics and to assess the effects between the botanical origin and the accumulation of minerals depending on the collection region of the honey samples using the following packages: Cluster R (version 4.0.5) [26] and ggplot packages (R software version 4.0.5) [27].

3. Results and Discussion

3.1. Melissopalynological Analysis

Pollen analysis of honey, also known as melissopalynology, is crucial for quality control. Honey consistently contains numerous pollen grains and honeydew elements, which together provide a detailed fingerprint of the environment from which the honey originates. Therefore, pollen analysis can be useful for determining and verifying the geographical and botanical origins of honey. However, sensory and physicochemical analyses are also necessary for an accurate identification of its botanical origin [28].

The melissopalynological analysis proved the label declaration of the botanical origin of the honey samples taken in the study (

Table 2. The results of melissopalynological analyses, compared to the described label of the commercial samples and the actual content of predominant pollen.

ID	Declared Honey Type	Actual Pollen Identified Confirming the Sample Labels (%)	Other Types of Pollen Identified in the Samples
H1	Linden, Organic	Malvaceae: Tilia spp. 83.65%	Astragalus, Asteraceae, Apiaceae, Fabaceae, Robinia pseudoacacia
H2	Mixture	Brassicaceae: Brassica 20%, Asteraceae 20%, Rosaceae 15%, Fabaceae 10%, Apiaceae 2%
H3	Acacia	Fabaceae: Robinia pseudoacacia 21.0%	Rosaceae, Apiaceae, Brassicaceae, Lamiaceae
H4	Mountain Honey	Brassicaceae: Brassica sp. 15%, Fabaceae 25%, Rosaceae 5%, Asteraceae 8%, Lamiaceae 10%, Plantaginaceae 2%, Linaceae 1%, Apiaceae 4%, Solanaceae 2%, Scrophulariaceae: Verbascum phlomoides 3%
H5	Citrus	Rutaceae: Citrus 57.60%
H6	Raspberry	Rosaceae: Rubus idaeus 26%
H7	Knotweed	Poligonaceae: Fallopia sp. 13.86%	Tiliaceae, Asteraceae
H8	Linden	Malvaceae: Tilia sp. 2.10%
H9	Oak Honey (Honeydew)	Fagaceae: Quercus sp.
H10	Thyme	Lamiaceae: Thymus sp. 8.10%
H11	Linden	Malvaceae: Tilia 5.40%
H12	Linden	Malvaceae: Tilia 5.10%
H13	Chestnut	Fagaceae: Castanea sativa 81.80%
H14	Chestnut	Fagaceae: Castanea sativa 75.30%
H15	Chestnut	Fagaceae: Castanea sativa 81.30%
H16	Chestnut	Fagaceae: Castanea sativa 79.10%
H17	Chestnut	Fagaceae: Castanea sativa 71.40%
H18	Oak Honey (Honeydew)	Fagaceae: Quercus sp.

).

It is reported that the Tilia sp. pollen content in linden honey varies widely across the world, with the highest being 50–99% in the Russian Far East and northeast China [29], while in Hungarian samples, Tilia sp. pollen was found to have 5.3–66.4% [6]. According to the analysis of 50 acacia honey samples in Ukraine, it was reported that the pollen rate varied from 5–25% [30]. In Croatia, it was found to be 11–71% in 200 acacia honey types [31]. In Türkiye, it was reported that the pollen rate was 90% or above in 43 chestnut honey samples collected in the province of Sinop [32].

Der Ohe et al. specified that samples of chestnut honey must contain over 86% chestnut pollen because Castanea sativa is a plant with over-represented pollen [28]. In some cases, underrepresented pollen could originate from species in genera such as Robinia (7–60%), Thymus (13–68%), and Tilia (1–56%), and the underrepresented pollen normally belongs to Citrus species, with limits ranging from 2 to 42%. The regulations of Türkiye specify the following pollen lower limits: 70% for chestnut honey, 10% for citrus and thyme, and 15% for acacia honey [33]. Fallopia honey is a new type that is being studied, and the limits of the pollen content are still being refined [34]. The pollen interval for Rubus sp. honey presents a high variability, as also noticed by Escuredo et al. who determined an interval between 46.4% and 91.3% [35]. These aspects are highly variable when comparing to the Romanian legislation standard for quality requirements at delivery by producers (SR 784-1:2009) which specifies a minimum of 20% pollen content, with 20% for acacia and 25% for linden honeys [36].

3.2. Physicochemical Characteristics

Several different physicochemical parameters have been determined using standard methods [37], and the results are summarized in

Table 3. Results of physicochemical characterization of honey samples. Results are expressed as ± standard deviation of three different analyses (n = 3).

ID	pH	Brix	L*	a*	b*	H°	EC (μS/cm)	Water Activity	Moisture Content
H1	5.12 ± 0.15 a	80.93 ± 0.19 a	92.9 ± 5.43 a	−3.01 ± 0.03 d	45.42 ± 0.04 c	−86.21 ± 0.02 d	616.3 ± 5.4 c	0.5672 ± 0.01 a	17.5 ± 0.21 a
H2	4.07 ± 0.26 b	82.29 ± 0.05 a	95.83 ± 3.89 a	−1.62 ± 0.01 d	21.55 ± 0.02 d	−85.71 ± 0.06 d	443.28 ± 0.99 d	0.5679 ± 0.00 a	17.5 ± 0.61 a
H3	4.28 ± 0.11 b	80.74 ± 0.02 b	86.12 ± 5.23 a	6.53 ± 0.05 bc	82.3 ± 0.04 a	85.46 ± 0.00 a	118.48 ± 0.17 f	0.5389 ± 0.05 a	16.0 ± 0.81 b
H4	4.08 ± 0.10 b	81.31 ± 0.07 a	81.12 ± 2.03 a	8.69 ± 0.01 b	68.61 ± 0.05 b	82.79 ± 0.04 a	419.4 ± 0.14 d	0.5591 ± 0.01 a	17.0 ± 0.55 ab
H5	3.31 ± 0.13 c	81.99 ± 0.03 a	74.95 ± 8.02 b	1.68 ± 0.04 c	39.17 ± 0.04 c	87.54 ± 0.05 a	262.53 ± 0.11 e	0.5455 ± 0.07 a	16.2 ± 0.62 b
H6	4.55 ± 0.21 ab	82.51 ± 0.03 a	81.85 ± 4.56 a	7.93 ± 0.01 b	72.21 ± 0.08 ab	83.74 ± 0.03 a	703.25 ± 1.37 c	0.5588 ± 0.01 a	15.6 ± 0.98 b
H7	4.09 ± 0.05 c	80.86 ± 0.02 b	25.72 ± 2.85 d	31.22 ± 0.41 a	44.09 ± 0.07 c	54.7 ± 0.02 c	654.48 ± 2.1 c	0.5779 ± 0.01 a	17.2 ± 0.28 ab
H8	4.51 ± 0.07 b	81.21 ± 0.02 b	91.04 ± 3.31 a	0.02 ± 0.00 c	58.85 ± 0.02 b	89.98 ± 0.00 a	490.9 ± 0.76 d	0.5669 ± 0.03 a	17.0 ± 0.45 ab
H9	4.84 ± 0.03 ab	83.45 ± 0.17 a	80.64 ± 6.51 b	9.36 ± 0.01 b	76.94 ± 0.01 ab	83.07 ± 0.04 a	901.68 ± 4.92 b	0.5466 ± 0.01 a	14.6 ± 0.49 b
H10	4.19 ± 0.03 bc	82.7 ± 0.04 a	75.86 ± 4.59 b	12.55 ± 0.01 b	68.15 ± 0.03 b	79.57 ± 0.01 ab	373.53 ± 0.12 de	0.5467 ± 0.01 a	15.6 ± 0.31 b
H11	4.22 ± 0.01 bc	81.03 ± 0.01 b	83.23 ± 5.69 ab	6.7 ± 0.01 b	72.17 ± 0.04 ab	84.7 ± 0.01 a	428.03 ± 0.29 d	0.5796 ± 0.04 a	17.2 ± 0.28 ab
H12	4.72 ± 0.02 ab	80.15 ± 0.02 c	78.65 ± 6.71 b	12.35 ± 0.06 b	84.03 ± 0.03 a	81.64 ± 0.05 a	855.6 ± 4.52 b	0.6043 ± 0.02 a	18.2 ± 0.45 a
H13	4.59 ± 0.11 ab	80.13 ± 0.01 c	56.98 ± 5.42 bc	37.72 ± 0.51 a	94.45 ± 0.07 a	68.23 ± 0.02 b	1097 ± 5.00 b	0.5936 ± 0.01 a	18.1 ± 0.76 a
H14	4.85 ± 0.20 a	79.95 ± 0.01 c	66.51 ± 3.29 b	23.25 ± 0.08 a	90.4 ± 0.01 a	75.58 ± 0.03 ab	1076 ± 8.90 b	0.6007 ± 0.02 a	18.4 ± 0.28 a
H15	4.47 ± 0.17 ab	81.32 ± 0.03 b	66.91 ± 7.18 b	26.21 ± 0.10 a	97.73 ± 0.01 a	74.99 ± 0.07 ab	949.48 ± 0.95 b	0.5811 ± 0.01 a	17.0 ± 0.31 ab
H16	5.25 ± 0.18 a	79.98 ± 0.01 c	68.68 ± 9.54 b	25.44 ± 0.13 a	100.05 ± 0.02 a	75.74 ± 0.01 ab	1408.25 ± 1.5 a	0.5962 ± 0.01 a	18.3 ± 0.49 a
H17	5.13 ± 0.04 a	80.86 ± 0.02 b	16.33 ± 8.87 d	27.8 ± 0.09 a	28.14 ± 0.03 cd	45.35 ± 0.02 c	651.38 ± 0.26 c	0.5749 ± 0.01 a	17.4 ± 0.44 a
H18	4.91 ± 0.03 a	81.55 ± 0.01 a	52.5 ± 6.32 bc	35.1 ± 0.21 a	86.16 ± 0.01 a	67.84 ± 0.01 b	1452 ± 0.82 a	0.5932 ± 0.02 a	16.7 ± 0.32 ab

Notes: EC—electrical conductivity; L*, a*, b*—color intensity and shade. Data are presented as means ± standard error (n = 3). Different lowercase letters in a column indicate significant differences (p < 0.05).

.

The pH values ranged from 4.07 ± 0.26 (H2) to 5.25 ± 0.18 (H16), showing the typical slightly acidic character of honey samples. This variability is frequently obtained in pH determinations and is believed to be due to the difference in the botanical origin of the honey [38]. These values also prevent microbial growth, which contributes to the antimicrobial properties of honey [39]. The most acidic samples were represented by honey with pollen of mixed botanical origin, while the least acidic were chestnut and linden honeys. These results are similar to previous determinations, and pH values up to 5.398 were found for linden honey, yet smaller values were found for multifloral honey samples originating in eastern Romania [14]. Moreover, a range of 4.46–5.19 was identified for Turkish chestnut honey [32].

The values of Brix degrees are used to estimate the total soluble sugars in honey, where one °Brix represents the percentage of solubilized sugars [14]. In our samples, the values ranged from 79.95 (H14) to 83.45 (H9) °Brix, which is in line with previous determinations in Romanian honey [14,40].

The color of the honey is an indicator of pigments, which can be derived from various sources such as nectar, pollen, or propolis. The color can range from light yellow to dark amber, depending on the type of flowers visited by the bees. Different colors may also indicate different flavors and nutritional properties in honeys. The difference in honey color can be explained based on the variation in mineral content, yet it can also be used for the identification of botanical origin and the determination of purity [41,42]. Color analyses were accomplished using the parameters L*, a*, b*, and H° (hue), and details regarding the variation in color characteristics are visualized in Table 3. Notable variations in terms of luminance L* include the highest values in H2 (95.83 ± 3.89), H1 (92.9 ± 5.43), and H8 (91.04 ± 3.31), while the lowest include darker honeys (i.e., H17 and H7) with values of 16.33 ± 8.87 and 25.72 ± 2.85, respectively. The redness a* ranged between −3.01 ± 0.03 (H1) and 37.72 ± 0.51 (H13), denoting the tendency of red coloration observed in samples with predominant Castanea sp. (H13–H17) and Fallopia sp. (H7) and the honeydew honey from Türkiye (H18). The lowest and negative a* values were recorded in the lighter-colored honey samples (i.e., H1 and H2) with −3.01 ± 0.03 and −1.62 ± 0.01, respectively. Lower values but positive ones were observed in samples H5 (1.68 ± 0.04), H8 (0.02 ± 0.00), and H11 (6.70 ± 0.01). The samples with darker tones presented elevated levels in b*, in particular the honey types with predominant Castanea sp. from Türkiye. The other lighter-colored honey types presented lower b* values. In accordance with the results obtained, other studies also reported increased a* and lower L* coordinates to be associated with darker and reddish heather honeys, whereas honeydew honeys tend to be lighter in color with red to yellow hues and higher a* and b* coordinates [43]. The color of honey is associated with the consumers’ preference; for instance, in Europe, darker honeys with amber hues and stronger flavors are preferred [44].

The electrical conductivity (EC) of honey, determined using a 20% solution in water, is another very frequent analysis that can help in the assessment of purity and botanical origin. This parameter is variable due to the different concentrations of solubilized salts and organic compounds, such as proteins and organic acids [38,45]. In the analyzed samples, the EC values were in the range of 0.262 (H5) to 1.452 (H18) mS/cm. Previous determinations of Serbian honey reported EC values between 0.08 and 1.99 mS/cm [45]. In Lithuanian honey, the identified variation in EC was between 0.27 and 0.89 mS/cm, while rapeseed honey was found to have the lowest EC [46]. For chestnut honey produced in the province of Sinop in Türkiye, EC values were found to be above 0.8 mS/cm [47].

With sufficiently low water activity (a_w), honey can be microbially stable. However, a critical limit of a_w is around 0.6, above which yeasts can grow, and for this reason a_w acts as a critical quality parameter [48]. Moreover, it was determined that a_w presents a linear relationship with the moisture content [49]. In a study conducted on Slovenian honey, a statistically significant linear correlation was found between a_w and water content. At the same water content, a_w in sweet sap (secretion) honey is higher than in flower honey [50]. In Turkish samples, the a_w and moisture values of flower honey were determined to be between 0.470 and 0.563 and between 15.0% and 20.4%, respectively. The a_w and moisture values of pine honey samples were determined between 0.492 and 0.589 with 15.1% and 20.4%, respectively [20]. In the other eight honey samples originating from Mexico, a_w values were determined between 0.569 and 0.613 [51].

The relevance of the water content of honey lies in the fact that it affects the physical, microbiological, and sensorial properties, as well as the commercial value of honey [52,53]. For the present study, the moisture values of the samples were 14.6–18.4%, which is a typical pattern. It was reported that moisture higher than 20% could be problematic due to the possible initiation of fermentative processes [54], meaning that the commercial samples were acceptable. In a study conducted on Hungarian honey, it was reported that the moisture content was below 20% [6], while commercial Indian honey showed moisture of 17 to 22.6% [37].

3.3. Mineral Content

The mineral content of the honey samples was determined using an ICP-OES method, and the results are presented in

Table 4. Results of mineral content determinations. Results are expressed as mg of specific mineral per kg of sample (ppm) ± standard deviation of two different analyses (n = 2).

No.	Al	B	Ca	Cr	Cu	Fe	K	Mg	Mn	Na	Ni	P	Zn
H1	0.41± 0.07 cd	2.42 ± 0.89 bc	47.26 ± 7.83 f	0.02 ± 0.02 a	0.11 ± 0.06 c	0.14 ± 0.12 d	1502.5 ± 340.12 cd	18.46 ± 3.61 d	0.4 ± 0.1 d	119.23 ± 97.97 b	ND	52.84 ± 13.25 e	ND
H2	0.51 ± 0.01 cd	1.54 ± 0.19 c	23.93 ± 5.13 g	0.03 ± 0.01 a	0.02 ± 0.00 c	0.42 ± 0.03 cd	132.7 ± 1.98 g	6.33 ± 0.57 e	0.04 ± 0.0 e	2.07 ± 0.03 f	0.04 ± 0.01 c	31.07 ± 1.3 f	ND
H3	0.58 ± 0.13 cd	4.97 ± 0.24 a	112.3 ± 4.81 bc	0.01 ± 0.01 a	0.16 ± 0.07 bc	0.78 ± 0.2 cd	830.15 ± 61.59 f	24.91 ± 1.1 d	0.59 ± 0.0 d	62.99 ± 33.81 d	0.07 ± 0.01 c	79.97 ± 10.93 d	ND
H4	0.9 ± 0.04 c	3.34 ± 0.03 b	63.98 ± 7.35 e	ND	0.09 ± 0.01 c	0.94 ± 0.1 c	610.1 ± 11.88 f	15.79 ± 1.12 de	0.2 ± 0.01 e	26.03 ± 5.24 e	ND	59.34 ± 3.9 de	ND
H5	0.54 ± 0.1 cd	3.91 ± 0.1 a	54.24 ± 0.72 ef	0.02 ± 0.02 a	0.08 ± 0.00	0.97 ± 0.14 c	309.1 ± 11.6 g	12.5 ± 0.15 e	0.12 ± 0.0 e	19.51 ± 3.47 e	ND	55.65 ± 10.61 de	ND
H6	5.7 ± 0.02 b	4.07 ± 0.16 a	50.58 ± 12.66 ef	0.02 ± 0.01 a	0.46 ± 0.01 a	1.44 ± 0.09 b	1368 ± 42.43 d	59.57 ± 2.68 b	5.88 ± 0.04 b	ND	0.35 ± 0.02 b	150 ± 6.65 b	ND
H7	0.6 ± 0.04 cd	6.15 ± 0.19 a	119.3 ± 0.14 b	0.03 ± 0.01 a	0.29 ± 0.00 b	1.24 ± 0.2 bc	1009.25 ± 46.32 e	50.93 ± 2.06 b	0.59 ± 0.02 d	5.59 ± 7.91 f	0.09 ± 0.00 c	144.7 ± 15.7 b	ND
H8	0.28 ± 0.18 d	2.79 ± 0.88 bc	119.39 ± 27.45 b	0.02 ± 0.00 a	0.06 ± 0.00 c	0.78 ± 1.07 cd	1247.9 ± 391.88 d	46.74 ± 8.76 b	0.35 ± 0.03 d	158.7 ± 32.95 b	0.03 ± 0.02 c	134.3 ± 35.21 bc	ND
H9	4.03 ± 0.04 b	3.53 ± 0.21 b	35.41 ± 2.52 fg	0.03 ± 0.00 a	0.65 ± 0.02 a	1.56 ± 0.06 b	1961 ± 45.25 b	68.33 ± 4.05 b	5.04 ± 0.07 b	76 ± 23.33 c	0.73 ± 0.03 a	187.25 ± 24.68 b	0.18 ± 0.26 b
H10	0.62 ± 0.13 cd	2.35 ± 0.01 bc	84.7 ± 16.91 d	0.05 ± 0.02 a	0.18 ± 0.02 b	1.16 ± 0.1 b	707.75 ± 127.21 f	21.86 ± 4.36 d	0.16 ± 0.02 e	109.36 ± 20.56 bc	0.02 ± 0.02 c	138 ± 18.81 bc	0.49 ± 0.17 b
H11	0.99 ± 0.01 c	2.11 ± 0.00 c	76.57 ± 3.0 d	0.02 ± 0.02 a	0.12 ± 0.01 bc	1.6 ± 0.13 b	772.6 ± 22.49 f	18.98 ± 0.04 d	0.45 ± 0.03 d	83.97 ± 2.55 c	0.02 ± 0.0 c	126 ± 5.66 bc	ND
H12	1.7 ± 0.14 c	1.61 ± 0.03 c	93.84 ± 2.84 c	0.03 ± 0.03 a	0.23 ± 0.01 b	1.54 ± 0.03 b	1906 ± 84.85 c	36.83 ± 1.16 c	1.3 ± 0.01 c	100.48 ± 3.85 bc	0.03 ± 0.01 c	154.1 ± 2.12 b	ND
H13	1.39 ± 0.04 c	1.12 ± 0.17 cd	165.95 ± 6.43 b	0.02 ± 0.02 a	0.2 ± 0.02 b	1.32 ± 0.08 b	2193.5 ± 26.16 b	42.29 ± 0.51 bc	4.86 ± 0.09 b	84.67 ± 16.1 c	ND	117.5 ± 18.24 c	ND
H14	2.52 ± 0.12 c	2.17 ± 0.07 c	121.85 ± 3.89 b	0.02 ± 0.01 a	0.39 ± 0.03 ab	3.27 ± 0.05 a	2328.5 ± 57.28 b	44.69 ± 1.55 b	2.35 ± 0.04 c	90.32 ± 2.51 c	0.06 ± 0.01 c	159.4 ± 2.55 b	ND
H15	11.73 ± 0.35 a	3.4 ± 0.01 b	133.95 ± 1.48 b	0.02 ± 0.01 a	0.42 ± 0.00 a	2.4 ± 0.05 b	1968.5 ± 68.59 b	55.55 ± 0.25 b	1.61 ± 0.02 c	96.28 ± 3.61 c	0.14 ± 0.01 bc	181.4 ± 4.24 b	ND
H16	1.1 ± 0.01 c	0.09 ± 0.05 d	98.1 ± 15.85 b	0.01 ± 0.01 a	0.07 ± 0.01 c	1.36 ± 0.04 b	2421 ± 82.02 b	25.61 ± 3.64 d	3.1 ± 0.05 bc	253 ± 15.7 a	ND	98.68 ± 0.74 c	1.86 ± 0.35 a
H17	0.27 ± 0.04 d	ND	58.37 ± 2.46 e	ND	ND	0.3 ± 0.02 cd	964.05 ± 46.32 e	45.98 ± 1.87 b	0.03 ± 0.00 e	311.75 ± 12.66 a	ND	70.35 ± 0.33 d	ND
H18	2.09 ± 0.38 c	2.66 ± 0.03 bc	104.81 ± 8.05 a	0.02 ± 0.00 a	0.76 ± 0.00 a	2.04 ± 0.09 bc	3303 ± 72.12 a	107.3 ± 2.12 a	8.83 ± 0.03 a	91.68 ± 31.28 c	0.11 ± 0.02 bc	257.1 ± 14.71 a	ND

Notes: The following minerals are represented in the table: aluminum (Al), boron (B), calcium (Ca), chromium (Cr), copper (Cu), iron (Fe), potassium (K), magnesium (Mg), manganese (Mn), sodium (Na), nickel (Ni), phosphorus (P), and zinc (Zn). ND—not detected or no data available. Data are presented as means ± standard error (n = 3). Different lowercase letters in a column indicate significant differences (p < 0.05).

. As a general observation, the highest identified content was in potassium (K) and phosphorus (P). The values for these minerals were 132.7–3303 mg/kg (average of 1418.6 mg/kg) and 31.07–257.1 mg/kg (average of 122.1 mg/kg), respectively, which shows a really similar profile to a previous study on honey samples from Mexico [51]. Moreover, it was previously highlighted that K, Na, and Ca represent the most abundant minerals in Malaysian honey [55]. In our samples, Na was the third most prevalent mineral, with values ranging from 2.07–311.75 mg/kg (average 105.7 mg/kg), followed by Ca, with 23.93–165.95 mg/kg (average of 86.9 mg/kg), and Mg, with 6.33–107.3 mg/kg (average 39.0 mg/kg). The other elements were underrepresented, with B, Al, Mn, and Fe showing average amounts higher than 1.0 mg/kg. The Zn content was low, except for sample H16, which indicates there is no heavy metal contamination in the other samples [56].

When comparing the samples in terms of cumulative mineral content, it was observed that the highest amounts were found in the samples H18, H16, H14, and H13, all of them being representative of pollen originating from the Fagaceae species. On the other hand, the samples H2, H5, H4, and H10 showed the lowest content. Ultimately, the differences in elemental composition are due to the geographical and botanical origin of the pollen, as well as the environmental pollution [57,58].

The mineral content of the Fallopia sp. honey sample is similar to previous reports highlighting an accumulation of macroelements, particularly Ca, K, and Mg, and trace elements (i.e., B), emphasizing the importance of consuming this type of honey for its potential health benefits [59]. Research has been carried out to fully understand how these macroelements contribute to the optimal functioning of various biological systems, including cardiac, nervous, and musculoskeletal systems [60]. Furthermore, B, found in significant amounts in knotweed honey, is an essential trace element for the nervous system and bone health, and a main ingredient (i.e., boromycin) in the treatment of arthritis [61]. Chestnut honey is another source of essential nutrients, including Zn, an anti-inflammatory agent involved in numerous biological functions, such as DNA replication and damage repair, collagen, and protein synthesis associated with wound healing [62]. Additionally, chestnut honeys accumulate significant values of Fe, an essential nutrient responsible for appropriate functionality in the human body, including oxygen transport and normal DNA synthesis [63]. Therefore, chestnut honey can be safely consumed by individuals with iron deficiency (i.e., especially pregnant women and the elderly), which manifests as anemia [64].

3.4. Antibacterial Activity

Details regarding the antimicrobial potential of the various honey samples against selected Gram-positive bacteria (S. pyogenes, S. aureus, and L. monocytogenes) and Gram-negative bacteria (E. coli, K. aerogenes, and S. typhimurium) can be observed in

Table 5. Inhibition zones of honey samples against different microorganisms (in mm), expressed as average ± standard deviation (n = 2).

ID	EC	KA	ST	SP	SA	LM
H1	11.7 ± 0.7 ab	10.0 ± 0.0 d	16.1 ± 0.1 a	26.8 ± 0.1 a	21.3 ± 0.4 a	12.7 ± 0.5 b
H2	12.2 ± 0.8 ab	15.5 ± 0.8 a	15.6 ± 0.2 a	23.1 ± 1.0 b	22.3 ± 0.8 a	15.1 ± 0.3 a
H3	10.6 ± 0.0 cd	10.5 ± 0.5 d	11.6 ± 0.1 c	17.4 ± 0.5 c	16.9 ± 0.2 d	11.5 ± 0.1 b
H4	10.9 ± 0.4 c	14.9 ± 1.4 a	15.6 ± 0.8 a	21.5 ± 3.2 b	20.6 ± 1.0 b	14.5 ± 0.1 a
H5	9.9 ± 0.3 d	13.9 ± 0.3 b	12.8 ± 1.1 c	18.2 ± 0.6 c	16.4 ± 0.4 d	12.6 ± 0.1 b
H6	10.8 ± 0.1 c	14.0 ± 0.0 b	12.5 ± 0.8 c	15.4 ± 0.4 c	18.3 ± 1.6 c	13.6 ± 0.4 ab
H7	10.7 ± 0.1 c	12.5 ± 1.8 c	13.5 ± 1.2 b	19.6 ± 0.0 bc	18.9 ± 1.3 bc	14.9 ± 0.1 a
H8	10.7 ± 0.3 c	11.9 ± 0.1 c	10.2 ± 0.1 c	ND	ND	ND
H9	11.5 ± 0 b	13.5 ± 0.8 b	13.9 ± 0.8 b	22.0 ± 0.5 b	22.2 ± 0.9 a	15.3 ± 0.4 a
H10	12 ± 0.8 ab	13.7 ± 1.1 b	16.1 ± 0 a	20.5 ± 1.1 b	11.76 ± 1.3 d	12.5 ± 0.2 b
H11	10.5 ± 1.0 cd	14.4 ± 0.8 ab	11.9 ± 0.0 c	11.0 ± 0.1 d	17.9 ± 0.4 c	ND
H12	10.8 ± 0.3 c	14.3 ± 1.9 ab	13.2 ± 0.1 b	21.7 ± 0.6 b	23.2 ± 1.2 a	13.2 ± 0.2 ab
H13	10.6 ± 0.2 cd	13.0 ± 0.8 bc	11.2 ± 1.6 c	9.9 ± 0.4 d	14.2 ± 1.1 d	ND
H14	10.7 ± 0.6 c	11.8 ± 1.5 cd	14.3 ± 1.8 b	17.9 ± 0.2 c	18.4 ± 0.0 c	ND
H15	10.8 ± 0.1 c	9.6 ± 0.0 d	ND	19 ± 0.3 c	20.0 ± 0.5 c	11.2 ± 0.0 b
H16	10.9 ± 0.1 c	10.9 ± 0.3 d	10.6 ± 0.6 c	ND	12.8 ± 0.7 d	ND
H17	10.5 ± 0.1 cd	9.8 ± 0.1 d	10.0 ± 0.1 c	ND	ND	ND
H18	13.9 ± 0.8 a	13 ± 0.2 bc	16.2 ± 1.7 a	22.9 ± 1.6 b	22.7 ± 0.4 a	16.3 ± 0.5 a

Notes: EC—Escherichia coli; KA—Klebsiella aerogenes; ST—Salmonella typhimurium; SP—Streptococcus pyogenes; SA—Staphylococcus aureus; LM—Listeria monocytogenes. ND—not detected or no data available (no inhibition zone). Values in the same column annotated with lowercase characters denote statistically significant variations in inhibitory activity against the evaluated microorganisms (established by post hoc Tukey HSD test, p < 0.05).

. The MIC values were determined using the agar well diffusion method (Figure 2).

3.4.1. Inhibition Zones of Honey Extracts against Different Microorganisms

The disk diffusion method revealed that honey samples exhibited increased antibacterial activity against Gram-negative bacteria, particularly S. typhimurium, and moderate activity against Gram-positive bacteria, with notable activity against S. aureus. Regarding the Gram-positive strains, the growth inhibition ranged between 23.2 ± 1.2 (H12) and 11.76 ± 1.3 (H10) against S. aureus. The Bulgarian samples (i.e., H1 and H2) manifested the capability to inhibit S. pyogenes. No activity against all Gram-positive bacteria has been observed in the case of H8 and H18 samples. Significant activity manifested against the Gram-negative bacteria, particularly S. typhimurium, which displayed an inhibition diameter ranging from 16.1 ± 0.1 (H1) to 10.0 ± 0.1 (H17), with no inhibition observed in H15. Overall, out of the evaluated honey types, the Bulgarian samples (i.e., H1 and H2), predominantly linden, polyfloral type, and honeydew honeys (H9 and H18) proved to have a significant inhibition potential against both Gram-positive and Gram-negative bacteria (Figure 3).

3.4.2. Minimum Inhibitory Concentration (MIC) of Honey Extracts against Different Microorganisms

Determining the minimum inhibitory concentration (MIC) was performed alongside the disk diffusion method in order to guarantee reliability and precision and evaluate the antibacterial capacity of the honey samples. Figure 3 illustrates the minimum inhibitory concentration (MIC) of the honey types required to completely prevent the microbiological development of the tested strains and the assessed relationship between bacterial strains and the evaluated honey types (Supplementary Table S1). Sugar solutions (i.e., fructose, glucose, and sucrose) were assessed to highlight the absence of microbial growth.

The present analysis was carried out to guarantee that the honey’s inherent antibacterial components may be responsible for the antimicrobial activity reported in the samples, instead of osmotic reactions (details of control solutions were excluded from Figure 3). Results of MIC revealed that the honey samples exhibited a targeted antimicrobial effect according to their botanical origin. Thus, high MIC values were recorded for the Gram-positive strains, particularly S. aureus and S. pyogenes, whereas the lowest MIC values were observed for the Gram-negative K. aerogenes. The most susceptible were the honey samples from Bulgaria with predominant Tilia sp. (H1) and polyfloral types (H2), but also the honey samples from Türkiye with predominant Thymus sp. (H10) and honeydew honey samples from both Romania (H9) and Türkiye (H18). The most effective against Gram-negative bacteria, particularly E. coli, were several chestnut honey samples from Türkiye (i.e., H13, H16, and H17). Other reports describe the antimicrobial potential of honey in association with its botanical origin and nutritional and phenolic composition. Thus, as evidenced in the present report, thyme honey shows significant inhibitory activity against P. aeruginosa and S. epidermis, compared with manuka honey [65]. In the present report, Fallopia sp. honey exhibited moderate inhibitory activity against the evaluated microorganisms, including Gram-negative E. coli and Gram-positive L. monocytogenes. Romanian honey samples with predominant Fallopia sp. pollen presented similar activity, with higher inhibitory potential against S. aureus and moderate activity against E. coli [66]. Coniferous honey from Greece exhibited the highest inhibition against Gram-positive S. aureus and Gram-negative E. coli and S. typhimurium., followed by significant activity of thyme honey against Gram-positive S. aureus and S. pyogenes [67]. A different report revealed that E. coli and Enterococcus faecalis proved to be the most susceptible to chestnut honey compared with manuka honey [68].

3.5. Statistical Analysis

Employing PCA modeling (Figure 4), trends associated with the relationships between the nutritional profile, color characteristics, and melissopalynological analysis of various honey types were shown.

The first two components accounted for 47.7% of the overall variance and revealed specific color patterns and possible connections between nutrients and botanical origin by displaying certain aggregations. The refractive index, °Brix, and L values were negatively correlated with moisture, electrical conductivity, and the coordinates of a* and b*, respectively. Regarding honey samples, the first quadrant distinguishes the samples with predominant Castanea sp. pollen types and a positive correlation with a* and b* values, moisture content, and electrical conductivity and a negative correlation between these characteristics and the L* and refractive index levels. A similar pattern has been observed in samples H7, H15, H17, and H18, with predominant Castanea sp. and Fallopia sp. pollen types. In the second quadrant, the samples with predominant pollen from Citrus sp., Robinia sp., and Rubus sp. were positively associated with °Brix and refractive index values and negatively associated with moisture content, electrical conductivity, and pH. The subsequent quadrant highlights the samples with predominant Brassica sp. (H2) and Tilia sp. (H1) pollen types as outliers due to their relatively similar composition; these Bulgarian samples were positively associated with lightness and negatively associated with a*, b*, and hue values. These findings suggest a potential correlation between specific pollen types and certain quality parameters in the honey samples.

A heatmap summarizing the mineral profile pattern in honey samples according to their color characteristics and botanical origin is visualized in Figure 5. Following the cluster analysis and importance score, the upper main cluster highlights the samples from Bulgaria and Cyprus differentiated from the other honey types mainly due to their botanical origin with a predominance of pollen from Brassica, Citrus, and Robinia genera and Fabaceae and higher luminosity (L*), but also due to their elevated levels of Cr and B. The following cluster underlines the honeydew honey samples (collected from Türkiye and Romania) and Rubus sp. and Fallopia honey types from Romania (i.e., H6 and H7, respectively) which, according to their position in the heatmap, outline the importance of the botanical origin that influences the mineral composition. These honey types presented significant levels of Al, Mg, P, Cu, Mn, and Ni. The outlier sample with predominant Fallopia (H7) is imperative to be mentioned as it accumulated the highest level in B and has the most vivid red hue (a*) compared with all the honey types under study. Subsequently, the honey samples with a predominance of Castanea sp. presented noteworthy accumulations of both Fe and Zn, along with significant levels of Ca and K. Therefore, the botanical origin of honey plays a crucial role in the determination of mineral composition, as evidenced by the significant levels of various elements found in different honey types [69,70]. As previously highlighted, the outlier sample from Fallopia (H7) stands out for its high levels of B and vibrant red hue, while the honey samples with Castanea sp. as the predominant botanical origin show notable accumulations of Fe, Ca, K, and Zn.

4. Conclusions

Suitable labeling of commercial honey is necessary, and it should be imposed legally. However, producers can use different health claims and beneficial effects while using incorrect details regarding the botanical origin of the pollen. This study aimed to determine the mislabeling of several different honey samples from four countries. The melissopalynological analyses determined the correct labeling of pollen origin, and the physicochemical parameters were in the ranges described by established methods in the field of honey research. In the case of samples collected from Türkiye, they proved to have an accurate labeling according to the Turkish regulations which state that for chestnut honey the pollen content is expected to reach 70% to be categorized as Castanea sp. In terms of mineral content, there were several differences between the samples, with K, P, Na, and Ca being the most represented elements. The highest cumulative content was determined for the Fagaceae-pollen-containing samples.

The statistical analysis highlighted the association between the honey samples’ botanical origin, color characteristics, and mineral composition. The samples with predominant Brassica, Citrus, and Fabaceae were revealed to have increased levels of Cr and B. The honeydew samples and those with predominant Rubus sp. and Fallopia sp. accumulated elevated levels of Mg, P, Cu, and B. The honey samples from Türkiye with Castanea sp. as the predominant pollen type are noteworthy due to their accumulation of Fe, Zn, Ca, and K. The honey samples exhibited a targeted antimicrobial effect, according to their botanical origin. Thus, significant inhibitory activities were recorded for the Gram-positive strains, particularly S. aureus and S. pyogenes, whereas the most effective against Gram-negative bacteria, particularly E. coli, were several chestnut honey samples from Türkiye.