Journal of Liquid Chromatography & Related Technologies, 38: 855–862, 2015 Copyright # Taylor & Francis Group, LLC ISSN: 1082-6076 print/1520-572X online DOI: 10.1080/10826076.2014.976712
Free Amino Acid Profiles of 17 Turkish Unifloral Honeys İBRAHİM KIVRAK1,2 1 Department of Chemistry and Chemical Treatment Techniques, Muğla Vocational School of Higher Education, Muğla Sıtkı Koçman University, Muğla, Turkey 2 Research Laboratory Center, Food Analysis Laboratory, Muğla Sıtkı Koçman University, Muğla, Turkey
Free amino acids of 58 unifloral honeys of 17 different floras (cedar, eucalyptus, vitex, carob, clover, honeydew, sunflower, citrus, heather, thyme, flower, chestnut, sideritis, acacia, lavender, cotton, and mad honeys) from various regions across Turkey were analyzed using ultra performance liquid chromatography with electrospray ionization tandem mass spectrometry (UPLC-ESI-MS/MS). A method for the determination of free amino acids in honeys was developed and validated. The method comprises sample preparation involving ultrasonic extraction and separation, identification, confirmation, and quantitation. The obtained data have been inquired by principal components analysis (PCA), allowing differentiation of studied honeys by their floral origins. The main amino acids detected in Turkish honeys were phenylalanine (4024.53 mg/kg), proline (1138.18 mg/kg), tyrosine (693.42 mg/kg), and isoleucine (749.81 mg/kg). Thus, the honeys consist of rich free amino acids, and can be accordingly consumed as not only traditional sweeteners, but also supplementary materials for food products especially for foods for children. This method may be used for the determination of individual free amino acids in honeys. Keywords: amino acid analysis, botanical origin, phenylalanine, proline, unifloral honeys, UPLC–ESI–MS/MS
Introduction Honey is highly well known and one of the earliest animal products. It has been appreciated throughout civilization, and has been recently produced widely in the world.[1] The quality of honey is usually evaluated by its sensorial, chemical, physical, and microbiological properties. The physicochemical standards of honey are clearly described by the European Commission (EC) Directive 2001/110.[2] However, specific amino acids are produced by bees and are found in honeys.[3] Considering honeys, the amino acids come from animals and vegetables, and pollen involves the majority. The relative ratios of amino acids (approximately 0.3– 1% w/w) in honeys are related to the source (nectar or honeydew).[4,5] Proline is the major contributor to the total amino acids in honeys.[6] Also, the potential of honey ripeness was associated with the content of proline in honeys in an earlier study.[7] Amino acids, being important constituents of foods, serve as the building blocks for protein biosynthesis, promote the flavor of foods, and are precursors for aroma compounds.[8] In another study, aromas of eucalyptus and lavender honeys displayed an interesting relationship with the composition of amino acids such as phenylalanine and tyrosine present in these honeys, which is much higher than those of other amino acids.[9] Amino acid analysis is used to identify the origin of honeys. [6,10] The European Union food regulations state the ingredients Address correspondence to: İbrahim Kıvrak, Department of Chemistry and Chemical Treatment Techniques, Muğla Vocational School of Higher Education, Muğla Sıtkı Koçman University, Muğla 48000, Turkey. E-mail:
[email protected] Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/ljlc.
and characteristics of honeys. However, only proline amount is specified in the European Union laws, and the proline analysis is the only method for evaluation of the total amount of amino acids.[11] There is no other method stated in laws, which describes how to assess amino acid quantity in honeys. Thus, there is a need for an analytical method for honey authenticity confirmation.[10] There are several techniques concerning amino acid identification, which include multiple steps; sample extraction procedure, derivatization of amino acids, separation, and confirmation/quantitation, and also gas chromatography[12] and liquid chromatography[13,14] methods have been used. Most of the published studies on the determination of amino acids in honeys have used derivatization agents and solid phase extraction (SPE). Although chromatographic separation of amino acids in honeys has been confirmed in the literature, due to derivatization step, peaks appearing in residues and matrixes are still challenging, and frequently complete separation cannot be succeeded. It is considered that mass spectrometric (MS) detection is more selective than ultraviolet-visible (UV-vis) or fluorescence.[10] It is very substantial to pay special attention to use of MS detection methods for amino acid analysis. Thus, in the present study, the preparation of honey samples did not include any derivatization and SPE techniques. However, the separation, identification, and assessment of the free amino acids in honey samples were accomplished by UPLC–ESI–MS/MS. This work aimed to contribute to the limited information related to the constitution of free amino acids in Turkish honeys, and also in this study a -routine analysis method for the food analysis laboratories and food production industry was developed. The discussed method has been applied, and 58 samples of honeys, from 17 different regions of Turkey, were analyzed in this study.
. I . Kıvrak
856 Materials and Methods Reagents, Chemicals, and Solutions All standards (glycine, alanine, serine, proline, valine, threonine, cysteine, leucine, isoleucine, asparagine, aspartic acid, lysine, glutamine, glutamic acid, methionine, histidine, phenylalanine, arginine, tyrosine, tryptophan, cystine, and trans-4-hydroxy proline) were purchased from Sigma–Aldrich Chemie GmbH (Steinheim, Germany). The other reagents were of analytical grade, and purchased from Merck (Darmstadt, Germany). All aqueous solutions were prepared with ultrapure water purified by a Millipore Milli-Q system (Advantage A10 Millipore, Pure Water Systems, France) that involves reverse osmosis, ion exchange, and filtration steps. Amino acid standard solutions were dissolved in methanol for the standard solutions, and diluted with 20% methanol (v/ v) to obtain different concentrations of calibration points within the range of 0–250 ng/mL. Microscopic Analysis of Honey Pollens All the honey samples subjected to the harmonized methods of melissopalynology with slight modifications,[15] first, were homogenized separately. Honey (10 g) was weighted in a 50 mL falcon tube, and distilled water (30 mL) was added, and then the falcon tube was placed in a water bath at 45°C for 15 min in order to dissolve honey completely. The mixture in the tube was mixed using a vortex mixer for 1 min and centrifuged (Eppendorf 5810 R, Germany) at 4000 rpm for 30 min, and the supernatant was discarded. The residue was mixed with 100 µL of glycerine/distilled water (1:1; v/v). Ten microliters of the glycerin jelly was placed on microscope slides. After approximately 12 h, pollen types present in the honey were analyzed using a 40 × ocular microscope (Olympus CX31, Japan). Classification based on pollen counting , and it is specified as dominant, secondary and tertiary percentages.
subsequently centrifuged at 4000 rpm and 4°C, then the supernatant was filtered through 0.20 µm pore diameter polytetrafluoroethylene (PTFE) membranes to remove any solid particles, and added to vials for injection.
UPLC–ESI–MS/MS Analysis The free amino acids, all 21, were identified in honeys without sample cleanup and without derivatization using an easy and fast amino acid method published elsewhere.[16] The analysis was performed using an UPLC–ESI–MS/MS instrument, consisting of a Waters Acquity Ultra Performance LC with a Waters column manager and heater/cooler, binary system manager, sample manager coupled to a Waters Xevo TQ-S triple quadrupole mass spectrometer equipped with electro spray ionization (ESI) (Waters Acquity Ultra Performance LC, Xevo TQ-S MS/ MS, Waters Co., Milford, MA, USA). The mass spectrometry parameters, confirmation and quantification mass transition (m/z), and their collision energies are explained in Table 2. Other parameters related to the evaluation and experimental parameters of the method are described in previous section. Separation operations were accomplished using a C18 column (Acquity UPLC BEH C18 100 mm × 2.1 mm, 1.7 µm particle size). Principal Components Analysis (PCA) PCA was performed using a Partial Least Square (PLS) Toolbox statistical package in the STATISTICA Package program (StatSoft, Inc., 5.0). PCA was applied to honey samples to establish the relationship among the constituents of honey samples and to assess the significant factors responsible for variations. The descriptive statistics skewness and kurtosis test has been evaluated using the StatSoft software package.
Results and Discussion Optimization of UPLC–ESI–MS/MS
Honey Samples and Extraction of Free Amino Acid Compounds Authentic honey samples were supplied from individual beekeepers and The Association of the Beekeeping Organizations of Turkey and the individual city's associations. A total of 58 samples of seventeen different origins of honeys were collected from different regions across Turkey during the harvesting season 2013. The botanical source of samples was indicated by beekeepers and using sensory characteristics, and confirmed by physicochemical analyses and chemometrics. Table 1 summarizes the quantity of the compounds in honey samples of different botanical origins with mean values, standard deviations, limits of detection (LoD), and limits of quantitation (LoQ). The samples were kept at room temperature in the dark up to the time of analysis. To prepare 10% (m/v) water honey solutions, 20% methanol solution (v/v) (20 ml), initially acidified with 0.1% formic acid (v/v), was added to 2.0 g samples of honeys. The resulting mixtures were placed in an ultrasonic bath at 36°C for 10 min to completely mix the extracts of analyzed honey samples and
The UPLC–ESI–MS/MS method previously used by Kıvrak et al.[16] was adapted for the free amino acids in honeys. Optimization work was executed for determining the best performance of ESI–MS/MS. This can be done by achieving the highest signal-to-noise (S/N) ratios of molecular parent ions, and a low number of collision product ions were obtained using the positive ion mode by a multiple reaction monitoring (MRM) mode. MS spectra recorded under S/N conditions were dominated by parent ions, which were subsequently fragmented to establish characteristic fragmentation MS/MS patterns of individual amino acid compounds. These experiments were performed using the infusion with combined mode of a fluidics system. Specific mass spectrometry and chromatographic conditions were described in our previous work. These transition pairs of parent/daughter ions enabled unambiguously distinguishing all amino acid compounds and, therefore, were used for their identification and quantification in analyzed honeys using the MRM analysis mode. Fragmentation patterns in the positive ion mode and pairs of parent and daughter ions obtained for these compounds corresponded to those previously reported in the literature.[16]
857
MV
SD
Gly Ala Ser Pro Val Thr Leu Hyp Ile Asn Asp Lys Gln
MV
MV
SD
Citrus (n ¼ 7)
0.78 189.00 7.51 7.35 166.93 9.73 5.97 117.52 3.32 9.64 1121.16 9.03 7.08 508.54 1.35 3.39 47.78 4.28 8.39 828.59 8.81 3.71 23.72 5.09 6.18 909.67 8.41 7.95 894.68 4.18 9.71 100.67 19.77 9.56 250.18 5.35 5.29 235.70 11.95
SD
Sunflower (n ¼ 5)
56.28 131.94 28.85 1335.97 1080.78 19.82 1607.42 31.75 1834.55 139.65 54.73 192.20 237.25
Amino acids SD
SD
MV
nd 17.75 nd 946.28 72.31 2.46 242.80 2.62 273.45 nd nd 7.69 6.01
4.32 3.61
9.39 5.48 0.30 8.12 0.11 9.52
4.87
SD
Chestnut (n ¼ 3) MV
MV
SD
Mad (n ¼ 3) MV
SD
Vitex (n ¼ 5) MV
SD
Carob (n ¼ 3)
165.55 158.48 33.07 1287.34 180.97 21.44 706.44 16.72 756.56 150.35 27.50 156.61 242.63
MV
8.32 7.61 7.97 24.04 10.72 10.11 7.55 5.63 12.65 7.90 3.22 10.44 6.02
SD
Sideritis (n ¼ 3)
nd 23.24 nd 942.88 65.85 1.51 200.57 2.75 247.73 1.84 nd 2.97 1.49
MV
0.69 0.72
4.82 9.56 0.81 10.78 1.24 8.22 0.76
9.54
SD
Acacia (n ¼ 5)
204.39 176.91 92.67 1117.46 253.21 41.72 893.73 36.24 860.40 105.05 85.65 226.75 209.55
MV
8.37 4.44 8.47 8.25 5.85 3.58 5.90 4.67 16.14 11.79 10.71 5.23 3.04 7.61 6.84 9.96 16.98 8.61 6.35 11.51 3.27
SD
MV
SD
Cotton (n ¼ 3)
207.37 195.48 52.09 1182.39 424.16 40.20 1464.02 43009,00 1317.63 188.80 170.39 181.92 169.44 77.05 65.06 147.50 2568.41 132.09 674.64 70.16 4.00
MV
Clover (n ¼ 4)
4.46 nd Nd 2.89 62.10 4.56 4.43 10.71 1.24 19.05 1274.50 14.02 6.09 399.67 5.43 3.42 20.31 9.63 8.73 1323.99 6.00 7.80 9.36 3.64 6.68 1187.47 8.48 2.79 54.26 9.17 3.14 182.67 5.52 1.09 55.14 8.37 1.77 50.4 7.23
SD
Lavender (n ¼ 4)
17.41 nd nd 54.67 6.46 nd 20.19 94.49 64.43 110.14 8.44 70.34 6.97 7.41 nd nd 41.52 1.85 nd 14.88 1183.90 150.94 1419.63 13.09 1021.84 19.17 15.92 204.61 43.59 416.90 4.40 103.29 6.23 8.38 7.34 2.55 34.53 5.98 3.66 1.09 19.12 318.90 64.52 1188.53 8.43 514.24 9.36 4.04 5.34 2.97 26.30 9.68 10.13 0.3 11.01 283.92 42.66 1130.21 9.25 391.33 8.42 12.40 11.19 1.95 73.05 1.98 2.95 0.2 13.39 1.64 0.61 56.67 2.58 nd 11.34 49.93 10.37 164.34 7.37 68.90 2.6 12.38 67.04 29.04 169.44 7.41 50.98 1.28 16.33 47.67 9.41 205.38 9.39 105.33 5.23 5.87 1.79 0.69 5.93 1.45 2.19 0.24 6.13 772 2.41 59.98 7.05 nd 14.64 1409.19 12.27 13766.06 11.67 3358.89 16.03 9.85 37.67 10.39 42.81 2.29 4.34 0.60 14.36 511.23 46.11 1143.88 14.05 645.50 6.43 4.23 23.34 3.37 268.74 9.21 37.02 13.61 0.61 1.54 0.46 3.29 1.21 nd
SD
Flower (n ¼ 4)
126.75 322.85 62.45 1290.01 483.92 45.75 2033.25 20.13 1438.48 141.75 51.55 141.75 160.25 85.13 38.35 69.02 5299.05 55.24 1047.44 83.65 2.34
nd nd 17.19 11.13 66.60 4.15 nd 2.95 1.86 978.98 7.19 1100.06 10.84 124.20 8.21 141.84 12.57 3.92 4.03 3.62 1.31 289.61 7.79 591.61 15.44 11.87 2.27 14.76 4.82 317.42 3.71 518.61 6.89 3.57 0.37 2.66 0.58 nd 6.05 1.04 14.52 8.11 54.40 7.78 15.67 6.98 56.74 5.98
MV
11.18 7.79 nd 16.23 10.38 0.82 39.72 2.27 19.75 0.69 1.00 9.01 18.37 10.04 6.51 3.95 16.59 2.15 51.62 1.47 0.38
SD
Thyme (n ¼ 5) MV
39.98 205.44 nd 1305.35 151.24 10.75 532.52 15.94 490.68 5.38 7.54 51.42 44.54 78.86 24.42 13.96 3748.54 13.28 612.24 3.84 1.84
MV
Eucalyptus (n ¼ 6)
Heather (n ¼ 4)
Glycine (Gly) nd Alanine (Ala) 31.90 5.93 Serine (Ser) nd Proline (Pro) 897.72 12.85 Valine (Val) 34.00 10.80 Threonine (Thr) 1.72 0.87 Leucine (Leu) 203.91 7.35 trans-4-hydroxy proline (Hyp) 5.47 1.25 Isoleucine (Ile) 191.71 8.79 Asparagine (Asn) 1.69 0.64 Aspartic acid (Asp) nd Lysine (Lys) 1.98 0.05 Glutamine (Gln) nd Glutamic acid (Glu) 8.78 3.32 Methionine (Met) 1.79 0.96 Histidine (His) nd Phenylalanine (Phe) 3276.69 11.09 Arginin (Arg) 1.83 0.64 Tyrosine (Tyr) 568.22 18.41 Tryptophan (Trp) 3.47 0.85 Cystine (Cys) nd
Amino acids
Cedar (n ¼ 3)a
10.06 2.09 7.88 2.75
1.69 19.54 14.45 2.48
0.06 0.09 0.09 0.06 0.03 0.06 0.06 0.06 0.09 0.06 0.06 0.03 0.06 (Continued)
0.02 0.03 0.03 0.02 0.01 0.02 0.02 0.02 0.03 0.02 0.02 0.01 0.02
8.64
SD
1.66 4.74 1.69 19.66 1.40 7.19 4.68
LOD LOQ
nd 26.17 nd 943.58 114.10 3.13 877.80 8.92 597.40 4.89 nd 38.71 24.71 85.08 4.96 nd 3002.02 5.48 664.22 53.37 nd
MV
Honeydew (n ¼ 7)
Table 1. Distribution of amino acid concentrations, Limits of Detection (LoD), and Limits of Quantification (LoQ) for Turkish unifloral honeys. Results are given as mg/kg
. I . Kıvrak 0.06 0.06 0.06 0.06 0.03 0.03 0.03 0.03 0.02 0.02 0.02 0.02 0.01 0.01 0.01 0.01 0.95 44.29 2.59 0.85 38.57 4.21 1.48 42.29 7.70 10.54 3217.97 10.64 8.28 28.14 6.00 15.48 820.69 4.65 11.93 59.09 7.55 0.16 1.67 0.56
Leu Hyp Ile Asn Asp Lys Gln Glu Met His Phe
132.10 132.10 132.20 133.10 134.10 147.00 147.10 148.10 150.20 156.10 166.20
Arg Tyr Trp Cys
175.20 182.16 205.10 241.30
Daughter transition ions (m/z) 30.00, 57.10, 60.00, 43.30, 55.00, 56.10, 102.10 68.11, 69.20, 69.20, 74.00, 74.00, 84.00, 84.10, 84.00, 56.10, 83.10, 77.00, 120.00 60.00, 123.10, 91.00, 74.00,
Collision energies (V)
44.00, 76.00 71.00 88.00 70.10 72.00 74.00, 84.00,
8, 8, 3 8, 8 9, 10 22, 12 18, 10 15, 10, 12, 9
86.08 86.00 86.10 87.13, 115.10 88.00, 116.00 115.00, 130.10 130.10 102.10, 130.20 104.10, 133.20 93.10, 110.19 91.20, 103.10,
14, 12, 8 20, 10 20, 9 15, 10, 10 14, 10, 8 20, 12, 10 16, 10 15, 12, 8 15, 10, 9 22, 20, 15 30, 30, 25, 14
70.00, 116.00 136.10, 165.06 118.10, 188.16 120.00, 152.00
15, 15, 35, 25,
20, 15, 25, 20,
15 9 10 12
number of samples; mean value (MV); standard deviation (SD); not detected (nd).
Sample Extraction Optimization
a
10.75 nd nd 581.30 3.72 138.23 4.84 nd Glu Met His Phe Arg Tyr Trp Cys
112.90 6.07 93.35 2.49 50.55 6.82 87.54 10.15 13.60 2.35 84.06 272.89 8.08 14.83 5.93 8.20 2.27 1.59 0.86 nd 3.12 366.75 7.21 106.02 6.50 nd 10.88 6.26 nd 47.08 4027.75 6.39 1895.67 11.49 2076.35 5.19 5169.91 15.67 1906.46 12.63 1998.83 60.75 1.91 107.67 5.66 3.99 1.20 6.11 2.06 4.15 2.47 47.60 719.25 17.52 446.33 3.96 346.11 21.05 652.15 9.82 442.87 4.31 640.23 206.75 9.71 45.67 9.49 12.23 2.53 236.12 11.75 10.92 2.22 17.36 2.75 1.96 2.95 1.59 nd nd nd 3.10
7.69 1.20 5.97 4.90 6.39 4.36 6.15 2.10
6.74 2.17 5.52 0.46
95.95 2.27 179.75 11114.25 82.75 1714.82 107.75 2.15
76.00 90.00 106.00 116.10 118.10 120.10
1.43
SD MV SD MV
Parent transition ion (m/z)
Gly Ala Ser Pro Val Thr
MV
SD
Table 2. Conditions applied during UPLC-ESI-MS/MS analysis[16] Amino acid
SD MV SD MV SD MV SD MV SD MV SD MV
Sunflower (n ¼ 5) Amino acids
Table 1. Continued
Citrus (n ¼ 7)
Heather (n ¼ 4)
Thyme (n ¼ 5)
Chestnut (n ¼ 3)
Sideritis (n ¼ 3)
Acacia (n ¼ 5)
Lavender (n ¼ 4)
Cotton (n ¼ 3)
LOD LOQ
858
The extraction solvents (methanol:water; 10:90, 20:80, 30:70 (v/v)), solvent volumes (5, 10, 20, 50 mL), and extraction time (1, 5, 10, 20 min) were proved to decide the appropriate extraction efficiency. The optimum extraction procedure was revealed with 2 g of honey sample extracted with 20 mL of 20% methanol solution (v/v) initially acidified with 0.1% formic acid in an ultrasonic bath at 36°C for 10 min. Identification of Free Amino Acids UPLC–ESI–MS/MS ensured the qualification and evaluation of 21 amino acids in the samples analyzed. Free amino acids (phenylalanine, proline, tyrosine, isoleucine, asparagine, leucine, alanine, valine, glutamine, glycine, histidine, and lysine) present in citrus honey and lavender honey are displayed in Figure 1. The summarized data of 58 honey samples for the amino acid content are presented in Table 1. The essential amino acids, which are present in honeys and resulting from pollens, are fundamental for the daily diet of humans.[17] The sulfur-containing amino acid, cysteine, was found in none of the Turkish honeys, and very less amounts of methionine and tryptophan were detected, and also similarly, Hermosín et al.[4] determined methionine and cysteine in less amounts in Spanish honeys. The abundant amino acids revealed for the studied honey samples were phenylalanine, isoleucine, leucine, tyrosine, proline, and valine. Lower but also quite high quantity of aspargine, alanine, and histidine was present. Methionine, cystine
Free Amino Acid Profile
859
Fig. 1. UPLC-ESI-MS/MS total ion chromatograms of some free amino acids present in (a) citrus honey and (b) lavender honey.
(sulfur-containing amino acids), and trans-4-hydroxy proline were found in less amounts, and glycine, histidine, and serine were not detected in several honeys. Pirini and Conte[18] reported that arginine was found in chestnut honey (0.35 mg/100 g), yet not in orange blossom, acacia, rosemary, and lime tree honeys; the authors also determined only tryptophan (0.43 mg/100 g) in acacia honey. On the contrary, Bouseta et al.[9] analyzed eucalyptus and lavender honeys, in which
arginine was found.[4] However, in the current study, arginine and tryptophan were detected in all of the Turkish honeys, and arginine was present in a moderate amount. The amounts of amino acids in Turkish honey samples are in good correlation with data reported for the similar types of European unifloral honeys.[4,5,19] Some specific characteristics related to amino acids as properties of floral origin of honeys can indicate a higher degree of variability in the honey samples studied.
. I . Kıvrak
860 Amino acid analysis reveals that asparagine and glutamine were detected in all honey samples except chestnut and cedar honeys. Acacia, cedar, and carob honeys did not contain glycine, serine, histidine, and cystine. Eucalyptus honey contained all amino acids except serine. Sulfur-containing amino acid, cysteine, was not found in cedar, carob, honeydew, heather, thyme, chestnut, and acacia honeys. Only glycine was found in eucalyptus, flower, vitex, clover, sunflower, citrus, sideritis, and lavender honeys. Lavender and vitex honeys had a higher concentration of phenylalanine and also total amino acid content than other analyzed Turkish honeys. However, acacia honeys had a lowest concentration of phenylalanine. Sunflower and flower honeys had a similar amino acid content and concentration. Several reports underline phenylalanine as the amino acid found in higher quantity in honeys of various botanical species.[4,19,20] In Turkish honey samples, phenylalanine was found as the main amino acid in the range of 499.6 to 15,047.6 mg/kg of honey, on the other hand, phenylalanine is the second highest level amino acid in vitex and acacia honey samples. Three groups of floral varieties can be regarded in terms of phenylalanine content, one with high amount, such as lavender, vitex, thyme, flower, and sunflower, another with moderate amount, such as cedar, eucalyptus, carob, honeydew, and cotton, and the rest with low amount, such as acacia, sideritis, chestnut, heather citrus, mad, and clover. However, high level of proline was found in vitex and acacia Turkish honey samples. When examining that proline in honeys is of bee product-animal origin and that the difference in its amount in unifloral honeys is very high, it is highly difficult to categorize unifloral honeys when only the proline amount is taken into account.[5] On the other hand, tyrosine amounts in the studied honeys displayed three categories according to botanical origin; lavender, flower, vitex, sunflower, and cotton being in a higher content. The high level of proline and phenylalanine contents in the studied honeys is in good correlation with that in Serbian, French, and Estonian honeys.[5,19,21] PCA, a statistical process, was subjected to all honey samples to reveal natural groupings, and to prove that these results displayed significant differences based on the botanical source of the honeys, the data were tested with PCA, which is applied in order to interpret a large data pool. In Table 3, the first three principal components (PCs), and the percent of overall variance determined by each of them are displayed. The first principal component (PC1) accounts for 21.3% of the variance. The cumulative variance for the second component is 40.6%, and the third PC corresponds to 55.5% of the total variability. The statistics produced data considering amino acid constitution and several specific amino acids that discriminate the honeys of different floral sources. However, parameters that can purposely characterize the botanical source of honeys, for advanced analysis of the data, PCA, and cluster analysis, were used to distinguish groups of samples with respect to floral sources of samples. The descriptive statistics was applied to evaluate the medians and variances of 13 amino acids for 17 varieties of honeys (cedar, eucalyptus, flower, mad, vitex, carob, clover, honeydew, sunflower, citrus, heather, thyme, chestnut, sideritis, acacia, lavender, and cotton), and also was applied to examine medians and variances of 21 amino acids for 7 varieties of honeys (flower, vitex, clover, sunflower, citrus, sideritis, and lavender).
Table 3. Component loading matrices for the first three factors and the variance explained by each AA Gly Ala Ser Pro Val Thr Leu Hyp Ile Asn Asp Lys Gln Glu Met His Phe Arg Tyr Trp Cys Eigen value Prp. total Cumulated % variance
PC1
PC2
PC3
0.83472 0.48441 0.74145 0.02508 0.127251 0.69438 0.189508 0.39386 0.372462 0.60773 0.62493 0.76685 0.48413 0.13615 0.040728 0.22189 0.162434 0.60485 0.2032 0.252864 0.06287 4.463065 0.212527 21.3
0.16226 0.065948 0.116973 0.49555 0.69212 0.040764 0.1569 0.57809 0.25927 0.18731 0.022082 0.37979 0.50195 0.6903 0.79923 0.62052 0.30391 0.216455 0.33765 0.82427 0.065147 4.065624 0.193601 40.6
0.09858 0.302178 0.11017 0.026528 0.616942 0.006893 0.635373 0.25053 0.701953 0.1997 0.264699 0.09929 0.13831 0.43386 0.443906 0.345091 0.71244 0.342157 0.58664 0.21429 0.114064 3.124761 0.148798 55.5
Figure 2 displays the 58 samples of the first two PCs. The amino acids most properly identified with the first two PC variables showed that they were in various quantities in at least one of the floral sources. PC variables separated the lavender and vitex honey samples as in positive PC1 and negative PC2, relatively. Although sunflower, citrus, clover, and flower honey samples were present in the positive quarter, one of the citrus and sunflower samples was separated from honeys of similar origins. According to pollen analyses, giving relevant explanation, citrus honey with PC 1 < 0, and sunflower honey with PC 2 < 1 are honeys that contain 58% of citrus pollen, and 53% of sunflower pollen, respectively. There were some other samples, heather and clover honeys, which deviated from each of the groups, yet, this can account for the constitution of specific pollens in honey samples. Although heather honey samples were present in the negative quarter, one of the heather honeys was present in PC 2 > 0. Similarly, clover honey samples were grouped in the positive quarter, though one of them was grouped in PC 1 < 0. However, clover, citrus, and sunflower honeys were grouped together in all positive quarters, though one of the samples of these was present just in the opposite quarters. Besides, cedar, thyme, carop, honeydew, and heather honeys showed up in the negative quarter. Although mad honey and chestnut honey are displayed on the same coordinate grid, each was grouped separately, and this confirmed results of pollen analyses for different botanical origins. Heather, clover, citrus, and sunflower honeys deviated from each of the groups, which may be related to certain pollens in honeys. The abundance of pollen has a crucial effect on the
Free Amino Acid Profile
861
Fig. 2. PCA of 58 unifloral honeys of first two principle component scores.
amino acid amount, the determination of the geographical sources of honeys, and grouping.[11,22] Finally, with statistical evaluation of PCA, amino acid constitution in the studied honeys can be categorized, and some groups displayed similar behavior according to the botanical origin.
sweet taste of honeys, being the crucial source of free amino acids, honeys can also be consumed as supplementary materials for food products especially for –foods for children. Besides, it is now considered that the developed method may serve as an efficient means to designate the botanical origin of Turkish honeys and for the determination of individual free amino acid profiles of honeys.
Method Validation The method was validated for sample preparation, UPLC separation, and MS/MS detection. The method validation procedure consisted of LoD and LoQ (Table 1).
Conclusions The determination of free amino acids consists of sample preparation, chromatographic, and mass spectrometry analyses. The analysis of amino acids was explained by statistical analysis of the data using PCA that grouped and dispersed the varieties of honeys by their botanical origin, though there were apparently similar amino acid profiles. Nevertheless, PCA presented distinguished honeys and amino acids. Lavender, vitex, thyme, and sunflower honeys contain the highest amount of amino acids, followed by acacia, eucalyptus, carop, honeydew, clover, chestnut, and sideritis honeys. Phenylalanine, proline, tyrosine, isoleucine, and leucine were revealed as the main amino acids. Essential amino acids, phenylalanine, threonine, isoleucine, leucine, tryptophan, valine, and lysine (except methionine) were found in apparently high levels in all Turkish honeys. Thus, the Turkish honeys consist of rich free amino acids, which are fundamental for humans and especially for the growth and development of kids. Therefore, apart from the delicious
References 1. Ciulu, M.; Solinas, S.; Floris, I.; Panzanelli, A.; Pilo, M. I.; Piu, P. C.; Spano, N.; Sann, G. RP-HPLC Determination of Water-soluble Vitamins in Honey. Talanta 2011, 83, 924–929. 2. EU. Council Directive 2001/110/CE relating to honey. Off. J. Eur. Communities 2002, L10, 47–52. 3. White, J. W. Composition of Honey. In Honey: A Comprehensive Survey; Crane, E., Ed.; Heinemann Newnes: Oxford, 1976; pp 157–206 4. Hermosín, I.; Chicόn, R. M.; Cabezudo, M. D. Free Amino Acid Composition and Botanical Origin of Honey. Food Chem. 2003, 83, 263–268. 5. Kečkeš, S.; Gašić, U.; Veličković, T. C.; Milojković-Opsenica, D.; Natić, M.; Tešić, Z. The Determination of Phenolic Profiles of Serbian Unifloral Honeys Using Ultra-high-performance Liquid Chromatography/high Resolution Accurate Mass Spectrometry. Food Chem. 2013, 138, 32–40. 6. Paramás, A. M. G.; Bárez, J. A. G.; Marcos, C. C.; García-Villanova, R. J.; Sánchez, J. S. HPLC-fluorimetric Method for Analysis of Amino Acids in Products of the Hive (Honey and Bee-pollen). Food Chem. 2006, 95, 148–156. 7. Ohe, W. V. D.; Oddo, L. P.; Piana, M. L.; Morlot, M.; Martin, P. Harmonized Methods of Melissopalynology. Apidologie 2004, 35, 18–25. 8. Berlitz, H.-D.; Grosch, W.; Schieberle, P. Food Chemistry 4th ed. Springer: Leipzig, 2009.
862 9. Bouseta, A.; Scheirman, V.; Collin, S. Flavor and Free Amino Acid Composition of Lavender and Eucalyptus Honeys. J. Food Sci. 1996, 61 (4), 683–694. 10. Rebane, R.; Herodes, K. A Sensitive Method for Free Amino Acids Analysis by Liquid Chromatography with Ultraviolet and Mass Spectrometric Detection Using Precolumn Derivatization with Diethyl Ethoxymethylenemalonate: Application to the Honey Analysis. Anal. Chim. Acta 2010, 672, 79–84. 11. Anklam, E. A Review of the Analytical Methods to Determine the Geographical and Botanical Origin of Honey. Food Chem. 1998, 63 (4), 549–562. 12. Nozal, M. J.; Bernal, J. L.; Toribio, M. L.; Diego, J. C.; Ruiz, A. Rapid and Sensitive Method for Determining Free Amino Acids in Honey by Gas Chromatography with Flame Ionization or Mass Spectrometric Detection. J. Chromatogr. A 2004, 1047, 137–146. 13. Iglesias, M. T.; de Lorenzo, C.; del Carmen Polo, M.; Martín-Álvarez, P. J.; Pueyo, E. Usefulness of Amino Acid Composition to Discriminate between Honeydew and Floral Honeys. Application to Honeys from a Small Geographic Area. J. Agric. Food Chem. 2004, 52, 84–89. 14. Spano, N.; Piras, I.; Ciulu, M. Reversed-phase Liquid Chromatographic Profile of Free Amino Acids in Strawberry-tree (Arbutus unedo L.) Honey. J. Assoc. Off. Agric. Chem. Int. 2009, 92 (4), 1145–1152. 15. Von der Ohe, W.; Dustmann, J. H.; Von der Ohe, K. Prolinals Kriterium der Reife des Honigs. Deut. Lebensm.-Rundsch. 1991, 87 (12), 383–386.
. I . Kıvrak 16. Kıvrak, İ.; Kıvrak, Ş.; Harmandar, M. Free Amino Acid Profiling in the Giant Puffball Mushroom (Calvatia gigantea) Using UPLC-MS/ MS. Food Chem. 2014, 158, 88–92. 17. Grunfeld, E.; Vincent, C.; Bagnara, D. High-performance Liquid Chromatography Analysis of Nectar and Pollen of Strawberry Flowers. J. Agric. Food Chem. 1989, 37, 290–294. 18. Pirini, A.; Conte, L. S. Capillary Gas Chromatography Determination of Free Amino Acids in Honey As a Mean of Discrimination between Different Botanical Sources. J. High Resolut. Chromatogr. 1992, 15, 165–170. 19. Rebane, R.; Herodes, K. Evaluation of the Botanical Origin of Estonian Uni and Polyfloral Honeys by Amino Acid Content. J. Agric. Food Chem. 2008, 56, 10716–10720. 20. Senyuva, H. Z.; Gilbert, J.; Silici, S.; Charlton, A.; Dal, C.; Gürel, N.; et al. Profiling Turkish Honeys to Determine Authenticity Using Physical and Chemical Characteristics. J. Agric. Food Chem. 2009, 57, 3911–3919. 21. Cotte, J. F.; Casabianca, H.; Giroud, B., Albert, M.; Lheritier, J.; Grenier-Loustalot, M. F. Characterization of Honey Amino Acid Profiles Using High-pressure Liquid Chromatography to Control Authenticity. Anal. Bioanal. Chem. 2004, 378, 1342–1350. 22. Gilbert, J.; Shephard, M. J.; Wallwork, M. A.; Harris, R. G. Determination of the Geographical Origin of Honeys by Multivariate Analysis of Gas Chromatographic Data on Their Free Amino Acid Content. J. Apic. Res. 1981, 20, 125–135.
