Physico-chemical, rheological and sugar profile of different unifloral honeys from Kashmir valley of India

2.1 Chemicals for estimation of sugar profile

Standards of sugars: maltose, isomaltose, trehalose, raffinose, melezitose, glucose, maltotriose, maltotetraose, sucrose, fructose, xylose, ultrapure water and organic solvents (HPLC grade) (Sigma–Aldrich Co., Mumbai, India).

2.2 Sample preparation and pollen analysis

Thirty seven samples from four varieties of raw and fresh honey (saffron honey, apple honey, cherry honey and P. rugosus honey) were collected from local bee-keepers during March 2013 to September 2014 and packed in glass bottles before storing at 4 °C under refrigeration condition. The origins of each honey sample were confirmed by melissopalynology. Honey samples were classified according to their botanical origin using the method described by Von der Ohe et al. (2004). The following terms were used for frequency classes: predominant pollen (>45% of pollen grains counted), secondary pollen (16–45%), important minor pollen (3–15%) and minor pollen (<3%).

2.3 Physico-chemical and enzymatic analysis

2.4 Determination of different sugars by HPLC–Refractive Index

The sugar profile was determined using high-performance liquid chromatography. 0.5 g of honey was weighed into polypropylene tubes and mixed with 10 ml of 60% methanol. Afterward, a milliliter of the dissolution was filtered through a 0.45 μm filter prior to HPLC analysis. The determination of sugar was performed with HPLC system equipped with refractive index (RI) detector (Waters isocratic (USA)). The separation was performed by using Waters X-bridge Amide HPLC Column, 5 μm (250 × 4.66 mm), maintained at 85 °C throughout the analysis. The injection volume of the sample was 20 μl, with a flow rate of 0.6 ml/min, using as mobile phase prepared by dissolving 80% of acetonitrile in ultrapure water. The freshly prepared mobile phase (ACN to water proportion, 80:20) was filtered and degassed by vacuum filtration through a 0.45 μm PTFE membrane filter. Before injection onto the column all syringes were fitted with syringe filters to remove particle/impurities from a sample so that the column should not be blocked. The separated carbohydrate peaks were identified by comparing the retention times obtained from standards. The honey samples were also spiked with standards in order to verify the identity of the chromatographic peaks. Duplicate injections were performed and average peak areas were used for the peak quantification (Mora and Marioli, 2001).

2.5 Sugar quantification by FTIR–ATR

Spectra from all samples were recorded in the one-bounce ATR mode using an Agilent Technologies Cary 660 FTIR spectrometer equipped with a Universal ATR accessory. Samples were placed on Diamond/ZnSe crystal plate (Perkin–Elmer) and scanned between 4000 cm⁻¹ and 400 cm⁻¹ representing the average of 64 scans. All spectra were recorded at room temperature (25 °C) under ambient conditions (Anjos et al. (2015)).

2.6 Rheological measurement

The dynamic rheological characteristics of honey samples were obtained by using Modular Compact Rheometer (MCR-102, M/s. Anton Paar, Austria), equipped with parallel plate system (50 mm diameter) at a gap of 0.5 mm at different temperatures (0, 5, 10, 20, and 30 °C). In order to determine the rheological data, frequency sweeps over the range of 0.63–63 rad/s at 3% strain were performed. A Rheoplus data analysis software (32 V3.40.) was used to obtain the experimental data to calculate storage modulus (G′) and loss modulus (G″). Results were obtained as an average of two measurements. The honey samples were heated to 50 °C for 1 h in a water bath to dissolve crystals, and then kept in 30 °C to remove air bubbles (Yoo, 2004).

4.1 Pollen analysis

The percentages of pollen spectra are related to pollens of nectar producing plants. Apple honey (M. domestica) contained 54–60% pollen of M. domestica sp., Cherry honey (P. avium) possessed 46–57% pollen of P. avium, P. rugosus contained 42–54% of Plectranthus sp. pollens while saffron honey (C. sativus) contained 47–49% pollen of C. sativus.

4.2 Physico-chemical and enzymatic analysis

The moisture content of all analyzed samples was within the range of 17.5–19.1% recommended by Codex Alimentarius (<20%) as shown in Table 1. Chirife et al. (2006) stated that moisture content was affected by climate, season, and moisture content of original plant nectar and was considered unripened at the moisture content higher than 20%. All the honey samples tested were found to be acidic in nature (3.01–4.35) and statistical significant difference (P < 0.05) was observed among the tested varieties (Table 1). In general, honey is acidic in nature irrespective of its variable geographical origin. The pH values were in agreement with the results of Algerian, Brazilian, Spanish and Turkish honeys (Azeredo et al., 2003; Ouchemoukh et al., 2007; Ozcan and Olmez, 2014).

Hydroxymethylfurfural (HMF) content is widely used as an indicator of honey freshness. All the tested samples showed lower HMF level than the limit (40 mg/kg), recommended by the Codex Alimentarius and statistical analysis revealed significant difference among all the four varieties (P < 0.05). The maximum content of HMF was found in P. rugosus (22.66 mg/kg) followed by apple (10.37 mg/kg), cherry (8.62 mg/kg) and saffron honey (7.13 mg/kg) as shown in Table 1. These values suggested that the samples were raw and unprocessed. Our results were in agreement with the values reported by Ozcan and Olmez, (2014).

The water activity is an important factor, which governs the food stability by preventing or limiting microbial growth. The water activity of all the honey samples varied from 0.514 to 0.597 (Table 1). Our results were quite similar to those of Greek honeys for which the a_w values ranged from 0.530 to 0.670 and Indian honeys whose a_w values ranged from 0.570 to 0.700 (Lazaridou et al., 2004; Saxena et al., 2010). A high positive correlation exhibited between water activity and water content (r = 0.93) indicated high water activity due to high moisture content.

The proline content of honey is also a criterion for its ripeness as well as adulteration with sugar (Von der Ohe et al., 1991). A minimum value of 180 mg/kg for genuine honey has been accepted in codex standards. The proline content values were significantly different among all the samples (P < 0.05) and the proline content ranged from 205.3 to 551.74 mg/kg (Table 1). Therefore, all the honey samples analyzed contain higher proline limit than the standard limit and thus can be considered as ripened and unadulterated ones. Our results were consistent with Algerian honey, Malaysian honey and Indian honey (Saxena et al., 2010; Ouchemoukh et al., 2007).

4.3 Enzymatic activity

Diastase activity is one of the main criteria utilized in the determination of the intensity of heating of honey during processing and storage. The diastase activity of the tested samples ranged from 12.18 DN (saffron honey) to 22.52 DN (apple honey) (Table 1), and all the samples showed the values within the Codex Standard (>8 DN) which indicated that all the samples were unprocessed and properly stored. Our results were in agreement with those reported by Saric et al. (2008) in Croatian honeys. By comparing our results with those reported by Cantarelli et al. (2008), it can be concluded that the Kashmir valley being a moderate temperature region of India; thus, the honey available in valley will have high diastase activity. Invertase number (IN) indicator of honey freshness was found within the Codex standard (>4 IN) and it ranged from 9.71 IN (saffron) to 19.69 IN (apple). Both IN > 4 and DN > 8 determined that all honey samples were fresh and unprocessed. Our results were consistent with the reporting of Saric et al. (2008).

4.4 Sugar analysis

Sugars are the main components of honey which depend mostly on floral and geographical origins and less on seasonal, processing and storage conditions (Dobre et al., 2012; Ouchemoukh et al., 2010). Sugar composition has been used to discriminate honey samples on the basis of floral as well as geographical origin (Gomez-Barez et al., 2000). However, many authors have concluded that sugar composition alone is not enough to discriminate honeys (Foldhazi, 1994). In this study, eleven types of sugars were identified and quantified, including three monosaccharides, four disaccharides and four trisaccharides (Table 1). Fig. 1 shows the graphical representation of typical HPLC chromatographic profile of sugars in saffron honey. The HPLC-RI detection proved to be very responsive as reflected from the values of average recovery (94–98%), limits of detection (LOD) (0.05–0.19 g/100 g) and limits of quantification (LOQ) (0.08–0.23 g/100 g). The total sugar content of honey samples varies between 73.89% and 78.45%. The result of total sugar content of our tested unifloral honeys is in agreement with that reported by Bentabol Manzanares et al. (2011) in blossom honey. The results showed that the monosaccharide glucose and fructose are the main sugars in all samples which confirmed that all honey varieties are genuine honeys. Statistical differences were observed in identified sugars and moisture content from all honey types (P < 0.05). The fructose values of all the samples ranged between 36.42% and 40.61% (Table 1) with highest percentage in apple and lowest in cherry. The range of amounts quantified in our study is same as reported for some Spanish unifloral honeys (De La Fuente et al., 2011), but higher ranges for fructose 41.3–43.30% (acacia honey) and 39.70–49.10% (black locust honey) were reported by Escuredo et al. (2014) and Primorac et al. (2011), respectively.

Close

Figure 1

HPLC chromatographic profile of sugars in Saffron honey (1: Fructose, 2: Glucose, 3: Xylose, 4: Maltose, 5: Isomaltose, 6: Trehalose, 7: Sucrose, 8: Raffinose, 9: Maltotriose, 10: Melezitose, 11: Maltotetraose).

Export to PPT

The glucose content of all analyzed samples from four varieties ranged from 30.56% to 33.98% (Table 1). Manikis and Thrasivoulou (2001), showed the quicker crystallization of honey samples with glucose content more than 30%. The glucose values obtained in present study were similar to lime honey as reported by Escuredo et al. (2014) in their study on honey from Northwest of Spain and different regions of Romania. All the honeys presented a value of glucose plus fructose higher than 60 g/100 g, which is the value, required for all the kinds of honey in the European and Codex standards. In accordance with Soria et al. (2004) the sum of glucose plus fructose was a discriminatory variable used to distinguish between blossom and suspected honeydew honeys. Xylose is one of the rare reducing sugars found in honeys. The xylose content estimated in four different honey varieties ranged between 0.16% and 0.29% which was lower as compared to Croatians unifloral honeys (0.1–0.7%) reported by Primorac et al. (2011), which may be due to different geographical origins or different climatic variations. All the honey samples had sucrose content lower than the limit (5 g/100 g), prescribed by European Community Directive (European Economic Community, 2002) which confirmed that these honeys are at an advanced stage of ripening.

4.5 FTIR–ATR spectroscopy

ATR–FTIR spectroscopy was used to compare honey samples based on their spectral differences in the 4000 cm⁻¹ to 400 cm⁻¹ spectral region. A representative ATR-FTIR spectrum of saffron honey is shown in Fig. 2. The band from about 1400 to 750 cm⁻¹ (Fig. 2) corresponds to the most sensitive absorption region of the honey sugars. The characteristic bands of major monosaccharides absorption (glucose, fructose) and disaccharide (sucrose) were present in the region between 1400 and 900 cm⁻¹ and major infrared bands of water were located at 3330 cm⁻¹ for O–H stretching and 1645 cm⁻¹ for H–O–H bending vibrations (Fig. 2). According to some authors, peak at 918 cm⁻¹, 1043 cm⁻¹ and 1254 cm⁻¹ corresponds to the C–H bending of the carbohydrate, C–O stretch in the C–OH group as well as the C–C stretch in the carbohydrate structure, respectively (Subari et al., 2012; Tewari and Irudayaraj, 2005). Similarly, Anjos et al. (2015) reported that spectral region located between 1500 and 750 cm⁻¹ was characteristic of the carbohydrate configuration. As shown in Fig. 2, the small peak located at 1110 cm⁻¹ corresponds to stretching of the C–O band of the C–O–C linkage while the peak that appeared at 1321 cm⁻¹ was due to O–H bending of the C–OH group. The broad band located between 3200 and 3000 corresponds to the C–H stretching of carboxylic acids and NH₃⁺ stretching band of free amino acids (Fig. 2). The similar spectral band corresponding to carboxylic acids and NH₃⁺ was reported by Zhbankov et al. (1998), Tewari and Irudayaraj (2004), Bureau et al. (2009), Gallardo-Velazquez et al. (2009), Gok et al. (2015).

Close

Figure 2

Representative FTIR–ATR spectrum of saffron honey in the 4000–400 cm⁻¹ spectral region.

Export to PPT

4.6 Rheology

Dynamic frequency sweep tests were carried out to determine the rheological parameters (G′ and G″) and their dependence on angular frequency. The frequency dependent magnitude obtained for loss modulus (G″) was much greater and significant than storage modulus (G′) which determined that all honey samples show more viscous behavior than elastic (G″ ≫ G′). Similar behaviors for different types of honey were predicted by Yoo (2004), Gomez-Diaz et al. (2009) and Witczak et al. (2011). Both magnitudes G′ and G″ increased with increase in angular frequency. Fig. 3a–d shows the typical diagram of loss modulus (G″) as a function of angular frequency (ω) for all honey samples at various temperatures from 0 °C to 30 °C. Therefore in this study the effect on rheological properties of honey varieties by temperature was characterized by G″ only. As mentioned earlier G″ is much informative than G′; thus, the data obtained for both parameters at different temperatures (0–30 °C) were also subjected to linear regression, as suggested by Rao and Cooley (1992), using the power law equation (Eqs. (1 and 2)) as follows:

Close

Figure 3

(a–d) Loss modulus (G″) as a function of angular frequency (ω) for honey samples of (a) apple, (b) cherry, (c) Plectranthus rugosus, and (d) saffron at various temperatures from 0 °C to 30 °C.

Export to PPT

Close

Figure 4

(a–d) Viscosity as a function of angular frequency of different unifloral honeys: (a) saffron, (b) Plectranthus rugosus, (c) apple and (d) cherry at different temperatures (0, 5, 10, 10, 20 and 30 °C).

Export to PPT

Close

Figure 5

Effect of temperature on viscosity of different honey types (S = saffron; C = cherry; P = Plectranthus rugosus; A = apple).

Export to PPT

Close

Figure 6

Loss modulus (G″) at 20 °C in dynamic oscillatory procedure for all honey samples (S = saffron; C = cherry; P = Plectranthus rugosus; A = apple).

Export to PPT

4.7 Effect of temperature on honey viscosity and loss modulus (G″)

The effect of temperature on viscosity and loss modulus (G″) was described using the Arrhenius model as follows: $$\mu ={\mu }_{\text{o}}\text{exp}[E_{\text{a}}/\mathit{RT}]$$ where μ is viscosity (Pa s), μ_o is constant, E_a is activation energy (kJ/mol), R is gas constant (8.314 kJ/mol/K) and T is absolute temperature in Kelvin. The pre-exponential factor (μ_o) of Arrhenius model and activation energies (E_a) were estimated, respectively, from the intercept and slope of straight lines obtained by the least square regression. The significant effect of temperature was found to be more up to 20 °C. However, at temperatures beyond 20 °C, the differences in viscosity are very small in all honey samples (Table 3). The values of activation energy of the analyzed honey samples ranged from 77.18 kJ/mol in P. rugosus to 85.59 kJ/mol in saffron honey (Table 4). The sensitivity of the viscosity to the temperature change is indicated by activation energy. Higher E_a value means that the viscosity of honey is relatively more sensitive to temperature change. Table 4 shows that saffron honey is the most sensitive (highest E_a value), whereas P. rugosus is the least sensitive (least E_a value) among the honey varieties analyzed. Fig. 7 shows the linear relationship of ln(η) vs (1/T) obtained from the linearization of Arrhenius equation and fitted to experimental data. Similar values for activation energy were reported for Greek honeys (Lazaridou et al., 2004) and Polish honeys (Juszczak and Fortuna, 2006). Honey with less moisture content possessed less activation energy, so the highest one was achieved by saffron honey and the smallest one by P. rugosus. According to Al-Malah et al. (2001), the material constant (pre-exponential factor in the Arrhenius equation), represents viscosity at a temperature approaching infinity. Higher the values of material constant, higher the viscosity of the samples. Thus P. rugosus was more viscous and saffron honey least viscous of all samples analyzed. A similar effect of temperature on loss modulus (G″) was described using the Arrhenius model and the values of activation energy of the analyzed honey samples ranged from 77.80 kJ/mol in P. rugosus to 86.88 kJ/mol in saffron honey (Table 4) which was quite similar to the values reported by Oroian et al. (2014) in Spanish honeys.

Close

Figure 7

Arrhenius model fit for different types of unifloral honey (S = saffron; C = cherry; P = Plectranthus rugosus; A = apple).

Export to PPT