Biogenic Amine Content of Tarhana: A Traditional Fermented Food

Abstract

Tarhana is a traditional fermented cereal food consumed in nearly all regions of Turkey as a soup having a delicious and desired taste. There are some other products similar to tarhana in Syria, Jordan, Egypt, Iraq, Hungary, Finland, Greece, and Scotland. Fifteen homemade and 5 commercially produced tarhana samples were analyzed for the first time to determine biogenic amine contents. Tyramine was the prevailing biogenic amine. Average tyramine concentrations of homemade and commercially produced tarhana samples were 92.8 and 55.0 mg/kg of tarhana, respectively. Concentrations of biogenic amines were below the permissible limits except for two homemade samples. The pH values of tarhana samples were in the range from 3.43 to 5.03; acidities were from 0.60 to 3.89 g/100 g tarhana (as lactic acid); total dry matters were from 86.42 to 92.32 g/100 g tarhana; and total free amino acid contents were from 0.035 to 1.427 g/100 g tarhana (as leucine).

Keywords:

• Biogenic amines
• Histamine
• Polyamines
• Tyramine
• Tarhana

INTRODUCTION

Recent trends in food security are promoting an increasing search for trace compounds that can affect human health, such as biogenic amines.[1] Biogenic amines are organic bases with aliphatic (putrescine, cadaverine, spermine, and spermidine), aromatic (tyramine and 2-phenylethylamine), or heterocyclic (histamine and tryptamine) structures that may be found in several foods, in which they are mainly produced by microbial decarboxylation of amino acids with the exception of physiological polyamines. Biogenic amines may be of endogenous origin at low concentrations in non-fermented foods, such as fruits, vegetables, meat, milk, and fish. High concentrations have been found in fermented foods as a result of a contaminating microflora exhibiting amino acid decarboxylase activity.[2] Biogenic amines represent a health risk for especially sensitive individuals. Symptoms include nausea, respiratory distress, hot flushes, cold sweat, heart palpitation, headaches, red rash, hypotension, hypertension, etc.[3] The estimation of the biogenic amines, namely histamine, tyramine, 2-phenylethylamine, agmatine, putrescine, and cadaverine, is important not only from the toxicological point of view, but also they can be used as indicators of degree of freshness or spoilage of foods.[1]

The maximum permissible levels of biogenic amines in foods are very difficult to establish because they depend on individual responses and the presence of other amines.[4] Nout[5] proposed acceptable levels for fermented foods of 50–100, 100–800, and 30 mg/kg for histamine, tyramine, and 2-phenylethylamine, respectively. Such levels could be regarded as acceptable also for non-fermented foods. Silla-Santos[2] proposed an acceptable level of 1000 mg/kg for total biogenic amine content. A maximum allowable limit of 100 mg/kg of histamine was reported by Halász et al.[4]

Cereal-based fermented foods play an important role in the nutrition of many people in Asia, Africa, the Middle East, and some parts of Europe. Tarhana, produced with yeast and lactic acid bacteria, is a type of dried powdered food and is consumed as a soup having a delicious and desired taste. Tarhana contains cereal and animal-based nutrients rich with fermentation products.[6] Tarhana is consumed nearly in all regions of Turkey. In general, tarhana is produced in four steps: dough mixing, fermentation, drying, and grinding. Tarhana dough is prepared by mixing wheat flour; yoghurt; yeast (Saccharomyces cerevisiae); vegetables (tomato, onion, green pepper, and paprika); salt; and spices including mint, thyme, dill, tarhana herb, etc.[7] In the central and eastern part of Turkey, one or more ingredients, such as milk, soybean, lentil, chickpea, corn flour, and egg, are also added in order to increase nutritional value. Most of the tarhana consumed in Turkey is homemade. In homemade production of tarhana, vegetables are chopped and added to the dough either raw or after being cooked. Tomato and paprika pastes may be used instead of tomato and paprika. All ingredients are mixed for 5 min to obtain homogenized dough. Dough is fermented at 30–35°C for 1–5 days.8–10 10] Fermentation is usually carried out by yoghurt bacteria (Lactobacillus bulgaricus, Streptococcus thermophilus) and bakers' yeast (Saccharomyces cerevisiae).[7] Both lactic acid bacteria (LAB) and yeast fermentation occur simultaneously during the tarhana production.[7, 11] After fermentation, the mixture is sun-dried and ground to a particle size of <1 mm.[7, 12, 13] However, there is a great commercial potential for the production of tarhana on an industrial scale using modern drying techniques.[7] Basically, four different types of tarhana have been defined by the Turkish Standardization Institute: (a) flour tarhana, (b) göce tarhana, (c) semolina tarhana, and (d) mixed tarhana. Using wheat flour, chopped wheat, and semolina separately or as combinations in the recipe causes some differences.[11]

The composition of 134 tarhana samples collected from different regions of Turkey was determined. It was found that tarhana contained moisture 10.2 g/100 g tarhana, protein 16.0 g/100 g tarhana, carbohydrate 60.9 g/100 g tarhana, fat 5.4 g/100 g tarhana, crude fiber 1.0 g/100 g tarhana, salt 3.8 g/100 g tarhana, and ash 6.2 g/100 g tarhana as mean values.[14] Similar results concerning composition were also reported by other researchers.[9, 12, 15] Because of the low moisture content and low pH, it can be stored for 1–2 years under dry and cold conditions.[16]

There are some other products similar to tarhana, such as kishk (sour milk-wheat mixture with boiled chicken stock) in Syria, Jordan, and Egypt,[17] kushuk (milk-sour dough mixture with turnips) in Iraq,[18] tahonya/talkuna in Hungary and Finland,[19] trahana in Greece, and atole in Scotland.[20] Bacterial genera that are known to have decarboxylating ability include Achromobacter, Aerobacter, Betabacterium, Clostridium, Escherichia, Lactobacillus, Proteus, Pseudomonas, Salmonella, Shigella, Streptecoccus, and Pediococcus. [21, 22] Lactobacillus acidophilus, Lactobacillus casei, and Lactobacillus bulgaricus were mentioned before as biogenic amine-producing bacteria.[23] Microflora of tarhana includes decarboxylase-positive microorganisms. Sengun et al.[24] studied to identify lactic acid bacteria isolated from tarhana. The aims of this study were to enumerate and identify for the first time by molecular biology-based methods predominant LAB isolated during the processing of tarhana. In order to establish the relationship between raw material and the microbiology of tarhana, yoghurt, and wheat flour were also analyzed. A total of 226 Gram-positive and catalase-negative isolates were obtained. Pediococcus acidilactici were found to constitute 27% of the isolates, 19% was identified as Streptococcus thermophilus, 19% as Lactobacillus fermentum, 12% as Enterococcus faecium, 7% as Pediococcus pentosaceus, 5% as Leuconostoc pseudomesenteroides, 4% as Weissella cibaria, 2% as Lactobacillus plantarum, 2% as Lactobacillus delbrueckii spp. bulgaricus, 2% as Leuconostoc citreum, 1% as Lactobacillus paraplantarum, and 0.5% as Lactobacillus casei.

Biogenic amine content of tarhana has not been studied before. Since it is a fermented food containing protein and free amino acids, which might be used by decarboxylase-positive microorganisms, the formation of various biogenic amines might be expected. Consequently, in the authors' laboratory, it was decided to estimate biogenic amine contents of tarhana samples and to determine biogenic amine formation during the processing of tarhana. The aim of the present study was to investigate biogenic amine contents of tarhana samples supplied from different parts of Turkey.

MATERIALS AND METHODS

Materials

Fifteen samples of tarhana produced at homes in different regions of Turkey and five samples of tarhana supplied from markets in Turkey were analyzed to determine biogenic amine contents. The origins and ingredients of homemade tarhana samples were presented in Table 1. Production numbers are available only for commercially produced tarhana changing from 2000 to 3000 t/year depending on temperatures during the winter months.[25] Cadaverine dihydrochloride, tryptamine, 2-phenylethylamine, spermidine trihydrochloride, spermine, histamine dihydrochloride, tyramine, and agmatine sulphate were obtained from Sigma (Steinheim, Germany). Methylamine hydrochloride and 1,7-diaminoheptane (internal standard, IS) were obtained from Merck (Schuchardt, Germany). Putrescine dihydrochloride was obtained from Fluka (Steinheim, Germany). The other reagents: sodium hydroxide, benzoyl chloride, sodium chloride, anhydrous sodium sulphate, trichloroacetic acid (TCA), cadmium acetate dihydrate, ninhydrin, ethanol, acetic acid, L-leucine, and distilled water were supplied from Merck (Darmstadt, Germany); hydrochloric acid from J. T. Baker (Deventer, Holland); methanol, acetonitrile, and diethyl ether (all of them HPLC grade) from Lab-Scan (Dublin, Ireland); sodium acetate trihydrate from Riedel (Germany); and phenolphthalein from Panreac (Barcelona, Spain). The standard solution of biogenic amines and internal standard solution were prepared following the method of Özdestan and Üren.[26]

Table 1 Origins, ingredients, and duration of fermentation of homemade tarhana samples

Sample code	Origin (location)	Ingredients	Duration of fermentation
T1	Izmir	Wheat flour, onion, tomato, strained yoghurt, red pepper, salt	4 days
T2	Bursa	Wheat flour, tomato, red pepper, onion, sour yoghurt, salt, black pepper, olive oil, tarhana herb, yeast	8 days
T3	Uşak	Wheat flour, onion, tomato, red pepper, yoghurt, salt, bread dough, peppermint	25 days
T4	Uşak	Wheat flour, onion, tomato, yoghurt, red pepper, bread dough, peppermint, salt, green pepper	25 days
T5	Uşak	Wheat flour, onion, tomato, yoghurt, red pepper, bread dough, peppermint, salt, green hot pepper	25 days
T6	Balıkesir	Wheat flour, tomato, green pepper, onion, yoghurt, salt, chickpea	10 days
T7	Balıkesir	Wheat flour, tomato, red pepper, onion, chickpea, yoghurt, salt	15 days
T8	Afyon	Wheat flour, rice, chickpea, tomato, garlic, red pepper, yoghurt, peppermint, onion, egg, salt, parsley	15–18 h
T9	Tokat	Wheat flour, yoghurt, egg, salt, chopped wheat	1 h
T10	Gaziantep	Chopped wheat, strained yoghurt, salt	1 h
T11	Kahramanmaraş	Chopped wheat, strained yoghurt, salt, sugar, flour	2 days
T12	Malatya	Chopped wheat, yoghurt, yeast, flour, salt	7 days
T13	Elazığ	Chopped wheat, bread dough, yoghurt, green pepper, red pepper, peppermint, parsley, tomato, sugar beet leaf, salt	4 days
T14	Şanlıurfa	Chopped wheat, strained yoghurt, salt	1 h
T15	Şanlıurfa	Chopped wheat, strained yoghurt, salt	1 h

Methods

Sample preparation

Five grams of tarhana samples were homogenized in 50 ml of 5 g/100 g TCA solution for 2 min. Homogenized samples were mixed for 15 min. The resulting mixture was centrifuged at 2150 g for 15 min. The supernatant was collected, filtered through a Whatman (42) filter paper, and derivatized prior to analysis by HPLC.

Derivatization procedure

Derivatization was achieved following the method of Özdestan and Üren.[26] Benzoyl derivatives of the amines in the tarhana extract were prepared, and then the resulting derivatives were extracted with diethyl ether. Following the evaporation of diethyl ether under a current of nitrogen, the solid residue was dissolved in methanol and injected into HPLC. Chromatograms were obtained for three aliquots of the same tarhana sample that underwent the whole analytical procedure.

Chromatographic conditions

Chromatographic separations of benzoyl derivatives were realized following the method of Yeğin and Üren[27] using a binary gradient elution consisting of methanol and acetate buffer.

Determinations of pH, Acidity, Total Dry Matter, and Total Free Amino Acid Content

The pH values of tarhana samples were measured with a digital pH meter after mixing 6 g of sample with 40 ml of distilled water.[28] The acidities (as lactic acid) of tarhana samples were determined by titration with 0.1 M NaOH solution.[29] The total dry matter was determined by drying the samples at 110°C to a constant weight.[29] The total free amino acid contents (as leucine) of tarhana samples were quantified (as leucine, which has an average formula weight among the amino acids) following the method of Folkertsma and Fox.[30] A 10 g sample of tarhana was homogenized in 50 ml of 0.1 M HCl solution. The homogenized sample was mixed for 15 min. The homogenate was held at 40°C for 1 h and then centrifuged at 2150 g for 30 min. The supernatant was collected and then filtered through a Whatman (42) filter paper. A 0.3 ml portion of the filtrate was diluted to 2 ml with distilled water and treated with 4 ml of Cd-ninhydrin reagent. The mixture was allowed to stand for 5 min at 84°C, cooled to room temperature, centrifuged, and then the absorbance value of the solution was measured at 507 nm.

Apparatus

HPLC separations were performed by using Agilent 1200 liquid chromatograph (Agilent, Santa Clara, CA, USA) equipped with an Agilent 1200 diode array detector (DAD), a gradient elution pump and vacuum degasser, an autosampler system, and a thermostatted column compartment. The chromatographic column was a Hichrom C₁₈ (10 μm particle size, 300 mm × 3.9 mm i.d., Hichrom Ltd., Theale, UK) thermostatted at 20°C. An HI 221 Microprocessor pH meter (Hanna Instruments, Cluj-Napoca, Romania) was used for pH measurements. For the determination of total dry matter contents, an oven (Dedeoğlu, Ankara, Turkey) was used. Absorbance measurements for total free amino acid content were achieved with a Cary 50 UV-vis spectrophotometer (Varian, London, UK).

Statistical Analysis

Throughout the present study, all the experiments were performed in triplicate. The statistical analyses were realized with the SPSS 10.0 statistics package program (IBM Corp., NY, USA). The statistical analyses of the data were achieved by using one-way analysis of variance (ANOVA), the Duncan post-test, and the Pearson correlation test. In all data analyses, a value of P < 0.05 was considered as statistically significant.

RESULTS AND DISCUSSION

The standard addition method and internal standard method were used together to calculate amine contents of tarhana samples. The types and the concentrations of amines detected in 15 homemade tarhana samples from different regions of Turkey are shown in Table 2. The chromatogram for standard biogenic amines and that for a homemade tarhana sample are given in Figs. 1 and 2, respectively. To examine the method performance and the matrix effect of tarhana, recovery rates of amine determinations were estimated. Recovery rates varied from 77.9 to 101.8%; 101.8% recovery was found for methylamine, 96.8% for putrescine, 96.3% for cadaverine, 77.9% for tryptamine, 95.1% for 2-phenylethylamine, 92.3% for spermidine, 88.9% for spermine, 86.3% for histamine, and 94.7% for tyramine. Because of interfering compounds, recovery rate of agmatine could not be calculated. Detection limits of the biogenic amines for the applied method were reported by Özdestan and Üren[26] as 0.5 mg/l or less, except for tyramine, that had a limit of detection value of 2.5 mg/l.

Table 2 Biogenic amine contents (mg/kg) and standard deviation values of tarhana samples.^a

Sample code	Putrescine	Cadaverine	Tryptamine	2-phenylethylamine	Spermidine	Spermine	Histamine	Tyramine	Total concentration
Homemade tarhana samples
T1	27.1^def ± 4.6	23.0^bcd±2.1	22.9^b ± 7.8	10.2^e ± 3.4	45.6 ^a ± 9.3	29.1^bc ± 9.7	103.0^ab ± 6.0	199.1^b ± 30.5	460.0^b
T2	12.3^fg ± 5.2	17.9^bcdef±5.1	7.2^c ± 1.0	17.1^cd ± 5.1	18.6^cde ± 3.6	16.1^cde ± 1.2	42.2^c ± 7.4	146.6^c ± 36.4	278.0^c
T3	243.0 ^a ± 26.8	26.6^b±5.9	67.8 ^a ± 7.7	53.1 ^a ± 2.0	47.7 ^a ± 19.0	50.4 ^a ± 27.6	115.9 ^a ± 10.8	415.4 ^a ± 14.2	1019.9 ^a
T4	48.2^c ± 3.8	28.1^b±12.2	9.3^c ± 3.9	15.6^cde ± 7.4	25.4^bcd ± 5.5	20.2^cde ± 3.0	85.2^b ± 8.4	49.5^d ± 7.8	281.5^c
T5	194.0^b ± 0.2	21.0^bcde±9.1	ND	53.4 ^a ± 5.3	36.3^ab ± 3.0	23.0^bcd ± 0.3	33.1^cd ± 2.8	137.1^c ± 35.3	497.9^b
T6	17.7^efg ± 0.5	13.1^defg±6.6	ND	ND	20.1^cde ± 3.6	ND	ND	40.3^d ± 5.7	91.2^e
T7	9.4^g ± 2.3	12.4^efg±2.1	ND	ND	14.6^de ± 6.2	35.9^b ± 9.1	ND	48.1^d ± 13.9	120.4^de
T8	11.3^fg ± 3.3	13.8^cdefg±9.0	ND	14.4^de ± 5.7	21.0^cde ± 2.3	9.0^de ± 0.1	16.5^de ± 1.6	42.4^d ± 14.6	128.4^de
T9	6.9^g ± 1.2	24.0^bc±3.2	ND	ND	15.9^cde ± 1.2	7.4^de ± 1.6	6.3^e ± 1.4	25.2^d ± 1.1	85.7^e
T10	8.1^g ± 0.7	12.3^efg±2.2	ND	ND	17.0^cde ± 2.6	10.8^de ± 2.6	10.4^e ± 3.9	25.2^d ± 2.5	83.8^e
T11	15.0^efg ± 4.2	40.2 ^a ±0.4	ND	23.5^b ± 3.2	27.6^bc ± 6.9	10.9^de ± 1.6	13.9^de ± 5.5	142.6^c ± 18.7	273.7^c
T12	40.3^cd ± 12.7	7.7^fg±1.5	ND	ND	13.6^de ± 1.9	6.7^e ± 2.4	8.7^e ± 2.0	47.3^d ± 16.5	124.3^de
T13	30.0^de ± 2.2	11.3^efg±2.5	21.0^b ± 5.0	20.7^bc ± 1.6	20.5^cde ± 3.2	20.7^cde ± 4.0	14.0^de ± 2.8	15.7^d ± 2.7	153.9^d
T14	11.1^g ± 4.7	11.9^efg±3.1	19.5^b ± 5.3	4.6^f ± 0.5	15.8^cde ± 2.8	21.5^bcde ± 4.6	48.0^c ± 3.0	35.8^d ± 23.0	168.2^d
T15	15.1^efg ± 7.5	5.6^g±1.4	ND	ND	12.0^e ± 1.7	11.2^de ± 3.9	6.8^e ± 0.7	22.3^d ± 2.1	73.0^e
Commercially produced tarhana samples
T16	11.1^b ± 2.3	11.4^b±1.8	ND	ND	18.4^bc ± 2.3	8.2^b ± 2.2	34.8^b ± 1.5	43.4^b ± 5.6	127.3^b
T17	106.4 ^a ± 13.7	88.8 ^a ±25.4	57.0 ^a ± 6.9	15.1 ^a ± 4.3	14.1^c ± 2.7	9.3^b ± 3.3	48.4 ^a ± 4.2	123.7 ^a ± 37.1	462.8 ^a
T18	15.1^b ± 7.2	15.1^b±5.3	ND	ND	26.1^ab ± 7.3	13.1^ab ± 2.7	21.5^c ± 6.1	23.7^b ± 14.8	114.6^b
T19	11.2^b ± 1.2	14.4^b±1.8	23.9^b ± 7.9	14.8 ^a ± 2.8	26.3^ab ± 2.2	17.0 ^a ± 6.3	33.9^b ± 2.4	50.2^b ± 3.7	191.7^b
T20	16.5^b ± 5.4	15.7^b±4.1	ND	ND	28.8 ^a ± 8.4	13.5^ab ± 2.3	55.1 ^a ± 10.1	33.9^b ± 2.2	163.5^b
^aND = not detected. Different matching letters in a column mean significant differences according to Duncan test (P < 0.05).

Table 2 shows that each of the homemade tarhana samples contained at least four biogenic amines. Putrescine, cadaverine, spermidine, and tyramine were detected in all samples. Among the ten amines under study, tyramine was detected at the highest level. Tyramine contents of samples varied from 15.7 to 415.4 mg/kg of tarhana. Average tyramine concentration of homemade tarhana samples was 92.8 mg/kg. Average putrescine, cadaverine, and spermidine concentrations of homemade tarhana samples were 46.0, 17.9, and 23.5 mg/kg of tarhana, respectively. Total biogenic amine contents of homemade tarhana samples were between 73.0 and 1019.9 mg/kg of tarhana. Concentrations of 2-phenylethylamine for two samples were above the limit value of 30 mg/kg. T3 and T5 samples contained 53.1 and 53.4 mg/kg of 2-phenylethylamine, respectively. Histamine concentration of the T3 sample was 115.9 mg/kg, which was above the limit value of 100 mg/kg. The concentrations of biogenic amines detected in 15 homemade tarhana samples were below the maximum permissible limits except for two samples. The concentrations of 2-phenylethylamine in T3 and T5 samples, and histamine and total amine contents of the T3 sample were above the maximum permissible limits. According to Turkish regulations, the level of histamine in fish must not exceed 200 mg/kg.[31] The European Union requires that nine samples must be taken from each batch of fish species of the following families: Scombridae, Clupeidae, Engraulidae, and Coryphaenidae for histamine analysis. The mean value for histamine in fish must not exceed 100 mg/kg.[32]

Significant differences were detected between concentrations of individual biogenic amines and between total biogenic amine concentrations of homemade tarhana samples (P < 0.05). In Table 2, different matching letters in a column indicate significant differences. Correlations between biogenic amine contents of homemade tarhana samples were tabulated in Table 3. As is seen in this table, statistically significant correlations were obtained between concentrations of various amines (P < 0.05). Based on these powerful correlations between biogenic amines, it may be concluded that all biogenic amines were produced by the same microorganisms. In addition, a significant correlation was obtained between durations of fermentation (Table 1) and total biogenic amine contents for homemade tarhana samples (P < 0.05).

Table 3 Correlations between biogenic amine contents of homemade tarhana samples and biogenic amine contents of commercially produced tarhana samples (P < 0.05). ^a

	Putrescine	Cadaverine	Tryptamine	2-phenylethylamine	Spermidine	Spermine	Histamine	Tyramine	Total biogenic amine
Putrescine		•	•	•	□	□	□	•	•
Cadaverine	□		•	•	□	□	□	•	•
Tryptamine	♦	□		•	□	□	□	•	•
2-Phenylethylamine	♦	♦	♦		□	□	□	•	•
Spermidine	♦	♦	♦	♦		•	□	□	□
Spermine	♦	□	♦	♦	♦		□	□	□
Histamine	♦	♦	♦	♦	♦	♦		□	•
Tyramine	♦	♦	♦	♦	♦	♦	♦		•
Total biogenic amine	♦	♦	♦	♦	♦	♦	♦	♦
^a• indicates a positive correlation between biogenic amines for commercially produced tarhana samples. ♦ indicates a positive correlation between biogenic amines for homemade tarhana samples. □ indicates no correlation.

The types and the concentrations of biogenic amines detected in five commercially produced tarhana samples are shown in Table 2. As is seen in Table 2, putrescine, cadaverine, spermidine, spermine, histamine, and tyramine were detected in all commercially produced tarhana samples. Among the ten amines studied, tyramine was the prevailing biogenic amine. Tyramine contents of samples varied from 23.7 to 123.7 mg/kg of tarhana. The average tyramine concentration was 55.0 mg/kg. Average putrescine, cadaverine, spermidine, spermine, and histamine concentrations of commercially produced tarhana samples were 32.1, 29.1, 22.7, 12.2, and 38.7 mg/kg of tarhana, respectively. Total biogenic amine contents of commercially produced tarhana samples were between 114.6 and 462.8 mg/kg of tarhana. The concentrations of biogenic amines detected in commercially produced tarhana samples were below the maximum permissible limits. Significant differences were detected between concentrations of individual biogenic amines and between total biogenic amine concentrations (P < 0.05). Correlations between biogenic amine contents of commercially produced tarhana samples were tabulated in Table 3. As is seen in this table, statistically significant correlations were obtained between concentrations of various biogenic amines (P < 0.05).

Biogenic amine contents of homemade tarhana samples were compared with those of commercially produced ones, and cadaverine content of commercially produced tarhana samples was found to be greater than that for homemade ones (P < 0.05).

Table 4 pH, acidity, total dry matter, and total free amino acid contents, and standard deviation values of tarhana samples. ^a

Sample code	pH	Acidity (g/100 g tarhana) (as lactic acid)	Total dry matter (g/100 g tarhana)	Total free amino acid (g/100 g tarhana) (as leucine)
Homemade tarhana samples
T1	3.84^h ± 0.01	1.94^d ± 0.02	91.11^b ± 0.13	0.453^cd ± 0.045
T2	4.70^b ± 0.00	1.45^gh ± 0.02	88.51^e ± 0.11	0.477^c ± 0.090
T3	3.83^h ± 0.01	3.89 ^a ± 0.07	90.12^c ± 0.18	0.658^b ± 0.006
T4	3.69^j ± 0.01	3.58^b ± 0.04	88.63^e ± 0.07	0.729 ^a ± 0.028
T5	3.73^i ± 0.01	3.57^b ± 0.07	87.30^g ± 0.13	0.664^b ± 0.011
T6	4.03^f ± 0.01	1.90^d ± 0.08	87.72^f ± 0.03	0.412^d ± 0.053
T7	4.37^d ± 0.02	1.60^efg ± 0.02	87.06^g ± 0.01	0.176^f ± 0.016
T8	4.48^c ± 0.01	1.61^efg ± 0.05	90.03^c ± 0.13	0.285^e ± 0.015
T9	4.29^e ± 0.04	1.85^de ± 0.06	89.48^d ± 0.06	0.086^g ± 0.013
T10	3.87^h ± 0.01	1.54^fgh ± 0.08	90.89^b ± 0.09	0.078^g ± 0.002
T11	3.43^k ± 0.04	2.17^c ± 0.41	90.10^c ± 0.13	0.090^g ± 0.016
T12	3.83^h ± 0.02	1.78^def ± 0.23	91.61 ^a ± 0.10	0.053^g ± 0.007
T13	5.03 ^a ± 0.01	0.60^i ± 0.10	89.51^d ± 0.37	0.238^e ± 0.008
T14	3.96^g ± 0.04	1.38^gh ± 0.12	90.86^b ± 0.43	0.035^g ± 0.005
T15	3.97^g ± 0.01	1.32^h ± 0.12	90.91^b ± 0.13	0.040^g ± 0.003
Commercially produced tarhana samples
T16	3.85^d ± 0.01	2.42^b ± 0.11	90.28^d ± 0.16	0.365^c ± 0.026
T17	3.84^d ± 0.01	1.32^d ± 0.11	86.42^e ± 0.03	0.226^d ± 0.013
T18	4.30^c ± 0.00	2.30^b ± 0.01	91.36^c ± 0.07	0.478^b ± 0.043
T19	4.55 ^a ± 0.01	2.60 ^a ± 0.05	91.57^b ± 0.18	1.427 ^a ± 0.074
T20	4.34^b ± 0.01	1.83^c ± 0.05	92.32 ^a ± 0.05	0.327^c ± 0.032
^aDifferent matching letters in a column mean significant differences according to the Duncan test (P < 0.05).

The acidities, pH, total dry matter contents, and total free amino acid contents of tarhana samples were determined and tabulated in Table 4. The pH values of homemade tarhana samples varied from 3.43 to 5.03, and those of commercially produced ones varied from 3.84 to 4.55. The acidities of homemade tarhana samples varied from 0.60 to 3.89 g/100 g tarhana and those of commercially produced ones varied from 1.32 to 2.60 g/100 g tarhana. According to the Pearson correlation test, negative correlations were obtained between pH and putrescine, cadaverine, spermidine, and total biogenic amine concentrations for homemade tarhana samples (P < 0.05). Statistically significant correlations were obtained between the acidity and putrescine, cadaverine, tryptamine, 2-phenylethylamine, spermidine, spermine, histamine, tyramine, and total biogenic amine concentrations in homemade tarhana samples (P < 0.05). It was seen explicitly that during the fermentation of homemade tarhana the pH of the medium decreased, and acidity and concentrations of biogenic amines increased as a result of activities of microorganisms. A study on the evaluation of free amino acids during fermentation of tarhana was performed by Erbaş et al.[33] and it was reported that the amount of free amino acids increased significantly (P < 0.01) during fermentation. For commercially produced tarhana samples, significant negative correlations were obtained between the pH and putrescine, cadaverine, and tyramine concentrations, but positive correlations between pH and spermidine and spermine concentrations (P < 0.05). In contrast with homemade tarhana samples, there were negative correlations between the acidity and putrescine, cadaverine, tryptamine, histamine, tyramine, and total biogenic amine concentrations in commercially produced ones (P < 0.05). This contrariety between homemade and commercially produced tarhana samples in terms of correlations with pH and acidity might be due to food additives or ingredients, such as milk powder, whey powder, soy flour, and citric acid added to tarhana samples after fermentation for cost reduction or to regulate the quality of the commercially produced tarhana samples.

Since biogenic amine formation occurs by decarboxylation of amino acids, it was thought that free amino acid content of tarhana could effect the biogenic amine formation. Availability of free amino acids contributes to the presence and accumulation of biogenic amines in foods. Consequently, total free amino acid contents of homemade tarhana samples were determined and found between 0.035 and 0.729 g/100 g tarhana, and those of commercially produced ones varied from 0.226 to 1.427 g/100 g tarhana. According to the Pearson correlation test, significant correlations were obtained between the total free amino acid contents and putrescine, cadaverine, tryptamine, 2-phenylethylamine, spermidine, spermine, histamine, tyramine, and total biogenic amine concentrations in homemade tarhana samples (P < 0.05). As is evident, activities of microorganisms and concentrations of biogenic amines were higher when medium contained more substrate, amino acids, for lactic acid bacteria and yeasts. For the commercially produced tarhana samples, no significant correlations were obtained between the total free amino acid contents and biogenic amine concentrations except for spermine (P < 0.05). This contrariety between homemade and commercially produced tarhana samples in terms of correlations between biogenic amines and total free amino acid contents might be due to ingredients added after fermentation. This possibility was supported by the fact that total free amino acid contents of the commercially produced tarhana samples were greater significantly than those for homemade ones (P < 0.05).

The total dry matter contents of homemade tarhana samples varied from 87.06 to 91.61 g/100 g tarhana and those of commercially produced ones varied from 86.42 to 92.32 g/100 g tarhana. No significant correlation was found between total dry matter contents and concentrations of the biogenic amines in homemade tarhana samples (P < 0.05). This result was reasonable since less than 1 g/100 g of the total dry matter consists of amino acids. The remaining part is composed of protein, carbohydrate, fat, crude fiber, and salt. Yeğin and Üren[27] determined biogenic amine content of boza, which is a fermented cereal product. No significant correlation was detected between biogenic amine concentrations and total dry matter contents in boza samples. In commercially produced tarhana samples, a significant correlation was obtained between spermidine concentrations and total dry matter contents (P < 0.05). On the other hand, significant negative correlations were obtained between the total dry matter contents and putrescine, cadaverine, tryptamine, 2-phenylethylamine, tyramine, and total biogenic amine concentrations in commercially produced tarhana samples (P < 0.05). These negative correlations in terms of total dry matter content may be explained by the addition of some ingredients into commercially produced tarhana after fermentation.

Özdestan and Üren[34] found significant correlations between biogenic amine concentrations of shalgam samples and pH values, acidities, total dry matter contents, and total free amino acid contents. Özdestan and Üren[35] reported biogenic amine contents of kefir samples. No significant correlations were detected between biogenic amine concentrations and pHs and total dry matter contents, whereas significant correlations were obtained between biogenic amine concentrations and acidities and total free amino acid contents of kefir samples. Özdestan et al.[36] reported biogenic amine contents of kumru samples. No significant correlations were detected between biogenic amine concentrations and total dry matter contents. Significant correlations were obtained between biogenic amine concentrations and pHs, acidities, and total free amino acid contents of kumru samples.

CONCLUSIONS

Tarhana is one of the most important traditional fermented foods in Turkey, which is produced by lactic acid bacteria and yeast fermentation. Of the ten amines studied, tyramine was the prevailing biogenic amine. The concentrations of biogenic amines in tarhana samples were below the allowable limits except for two homemade ones in which 2-phenylethylamine, histamine, and total biogenic amine contents were above the maximum permissible limits. Consumption of tarhana is less than 40 g per meal and this consumption level appears to be safe for individuals with the exception of two homemade tarhana samples. Statistically significant negative correlations were obtained between biogenic amine concentrations and the pHs, but positive correlations were obtained with the acidities and total free amino acid contents in homemade tarhana samples. These correlations were not observed for commercially produced tarhana samples. This contrariety may be due to the food additives and ingredients added to the commercially produced tarhana after fermentation. Based on the powerful correlations detected between biogenic amine contents, it may be concluded that nearly all of the biogenic amines were produced by the same microorganisms.