Effects of Soaking and Fermentation Time on Biogenic Amines Content of Maesil (Prunus Mume) Extract

Maesil extract, a fruit-juice concentrate derived from Prunus mume prepared by fermenting with sugar, is widely used with increasing popularity in Korea. Biogenic amines in maesil extract were extracted with 0.4 M perchloric acid, derivatized with dansyl chloride, and detected using high-performance liquid chromatography. Among 18 home-made maesil extracts collected from different regions, total biogenic amine content varied from 2.53 to 241.73 mg/L. To elucidate the effects of soaking and fermentation time on biogenic amine content in maesil extract, maesil was soaked in brown sugar for 90 days and the liquid obtained was further fermented for 180 days at 15 and 25 °C, respectively. The main biogenic amines extracted were putrescine and spermidine and the total biogenic amine content was higher at 25 °C than at 15 °C. Soaking at 15 and 25 °C increased the total biogenic amines content from 14.14 to 34.98 mg/L and 37.33 to 69.05 mg/L, respectively, whereas a 180 day fermentation decreased the content from 31.66 to 13.59 mg/L and 116.82 to 57.05 mg/L, respectively. Biogenic amine content was correlated with total amino acid content (particularly, arginine content). Based on these results, we have considered that biogenic amine synthesis can be reduced during maesil extract production by controlling temperature and fermentation time.


Introduction
Maesil (Prunus mume) known as Japanese Ume has been used not only as a food but also as a medicine on account of its various functionalities [1][2][3]. As the seed of maesil has a toxic substance called amygdalin [4], maesil has been processed into various products such as alcoholic beverage, juice, pickle or extract rather than eaten raw [5]. Maesil extract is a fruit-juice concentrate produced by the fermentation of maesil and sugar. Recently, it has been increasingly used as a seasoning to impart sweetness and a unique flavor to foods [5][6][7][8]. Traditionally, maesil extract is soaked for a long period (90 days) at room temperature and fermented naturally under different conditions in individual households. Therefore, uncontrolled fermentation can lead to the formation of biogenic amines, which are produced by molds and bacteria.
As biogenic amines are mainly produced by the microbial decarboxylation of free amino acids, they are easily found in fermented foods [9,10]. These biogenic amines have been reported to be abundant and they have been found in a wide range of food products, including fish products, soy sauce, Chunjang (traditional fermented soybean paste in Korea and China), and agricultural products [11][12][13][14][15]. As a high intake of biogenic amines can cause various detrimental effects such as migraine and gastrointestinal problems, their ingestion needs to be restricted [16,17]. Indeed, their content is currently regulated in certain food products. For example, the histamine content in fish

Preparation of Food Samples
During the period from 2010 to 2014, we collected samples 18 maesil extracts from different households in Korea for analysis of biogenic amines content. We also prepared our own maesil extract, following the method of Choi and Koh (2016) [5], and the process of preparation is shown in Figure 1. Maesil fruits obtained from a local market were washed with pure water, and drained at room temperature (23 ± 1 • C). To the 400 g of maesil, we added 400 g of brown sugar and the mixture was then placed in 1 L clear plastic jars, which were maintained in incubators set at 15 • C and 25 • C, respectively. After the maesil fruits were taken out from the jar after 90 days of soaking, the obtained liquid (490 mL at 15 • C and 476 mL at 25 • C) was further fermented for the next 180 days in the same jar. Biogenic amines were analyzed at 30, 45, 75, and 90 days of soaking period and 30, 60, 120, 150, and 180 days of fermentation.

pH Measurement
To measure the pH, maesil extract (10 g) was mixed with 10 mL deionized water for 3 min and then filtered through Whatman paper No.2 filter paper (Advantec, Tokyo, Japan). The pH was measured using a pH meter (Beckman Coulter, FL, USA) following the method of Shukla et al. [21].

Amino Acids Analysis
Amino acids in the maesil extract were analyzed using an HPLC system (Dionex Ultimate 3000, Thermo Fisher Scientific, Waltham, MA, USA), equipped with a 1260 Infinity fluorescence detector (Agilent Technologies, Waldbronn, Germany), following the method described by Jajic et al. (2013) [22] with slight modifications. The samples were derivatized with OPA and FMOC via a programmed autosampler. After derivatization, samples (0.5 µL) were injected into an Inno-C 18 column (4.6 × 50 mm, 5 µm, Youngin Biochrom, Korea) at 40 • C. The fluorescence was detected at excitation and emission wavelengths of 340 and 450 mm, respectively for OPA, and at 266 and 305 nm, respectively, for FMOC. Primary and secondary amino acids were analyzed based on the OPA and FMOC derivatives, respectively. The mobile phase solvent A was 40 mM sodium phosphate (pH 7), and solvent B was a 10:45:45 (v/v) mixture of distilled water, acetonitrile, and methanol. The gradient program was run at a flow rate of 1.0 mL/min as follows: 5% B for 3 min; followed by elution with 5% to 55% B in 24 min; 55% to 90% B in 25 min; maintained 90% of B for next 6 min; and 90% to 5% B for 3.5 min, maintained for 0.5 min.

Extraction of Biogenic Amines
Biogenic amines were extracted from the maesil extract using the method of Shukla et al. (2014) [21] with slight modifications. Briefly, 10 mL of 0.4 M perchloric acid solution was mixed with 5 g maesil extract, homogenized for 3 min, and then centrifuged at 3000× g for 10 min at 4 • C. The residue was re-extracted with 0.4 M perchloric acid solution (10 mL). After the supernatants were combined and 0.4 M perchloric acid solution was added to adjust the final volume to 50 mL. After filtering through Whatman filter paper No.1 (11 µm, Adventec, Tokyo, Japan), 1 mL of the extract was used for derivatization with dansyl chloride.

Derivatization of Biogenic Amines
Biogenic amines were derivatized following the methods described by Shukla et al. (2010) [23] and Frias et al. (2007) [24]. An extract sample (1 mL) or standard solution mixture (1 mL) was mixed with 200 µL 2 M sodium hydroxide; next, 300 µL of sodium hydrogen carbonate solution was added to saturate the solution. To the mixture, 1 mL of a dansyl chloride solution (10 mg/mL in acetone) was added and kept for 45 min at 40 • C. To stop the reaction, 100 µL of 25% ammonium hydroxide was added to the mixture and reacted for 30 min at 25 • C. Then, the derivatized biogenic amines were extracted twice with 1 mL of diethyl ether. Subsequent to drying in a nitrogen stream, the extract was redissolved in acetonitrile (1 mL) and filtered through a 0.22 µm polyvinylidene fluoride (PVDF) filter (Millipore Co., Bedford, MA, USA) for injection into the HPLC system.

Method Validation
The HPLC method for biogenic amines analysis was validated for linearity, limits of detection (LOD) and limits of quantification (LOQ), accuracy, and precision [22]. The linearity was evaluated using five concentrations (0.5, 1, 2, 5, and 10 mg/L) of each the biogenic amine standards (PUT, CAD, HIS, TRP, 2-PHE, TYR, SPD, and SPM) by constructing a calibration curve. The LOD and LOQ values were calculated using the following equations: LOD = 3.3 × (standard deviation (SD)/slope of calibration curve) and LOQ = 10 × (SD/slope of calibration curve). The accuracy of the method was verified by triplicate analysis of spiked samples at two different levels (5 and 10 mg/L) and expressed as % recovery. The recoveries were calculated by contrasting the peak area of measured concentration with the peak area of the spiked concentrations. To evaluate the precision, repeatability, inter-day, and intra-day were performed and expressed as the percentage relative standard deviation (RSD) of the peak area measurements. Repeatability was estimated by analysis of six consecutively injected samples. The inter-day precision was determined at two different levels, 5 and 10 mg/L, and the analyses were performed over a period of three consecutive days. The intra-day precision was determined by spiking five blank samples at concentrations levels of 5 and 10 mg/mL and the evaluation was based on the results obtained using the method operating over a single day under the same conditions.

Statistical Analysis
Quantitative data are expressed as the means ± SD of at least three measurements. Statistical analysis was performed using a one-way analysis of variance (ANOVA) and Duncan's multiple range test by SAS software, version 8.0 for Windows (SAS Institute, Cary, NC, USA). The probability value of p < 0.05 was considered statistically significant.

Method Validation
The results obtained from the different method validations are presented in Table 1. Standard curves for biogenic amines were constructed from triplicate analyses of five concentrations in the range 0.5-10 mg/L. With the exception of spermine (correlation coefficient (R 2 ) > 0.998), the linearity of the calibration curves for each biogenic amine was >0.999. The precision, expressed as %RSD, of inter-day variation was between 0.17% and 5.20%, and the RSD values for intra-day variation were between 0.07% and 6.46%. The LOD and LOQ of the biogenic amines ranged from 0.01 to 0.20 mg/L and 0.02 to 0.61 mg/L, respectively. The accuracy of the method with regard to recovery was between 89.4% and 110.8%.

Content of Biogenic Amines in Home-Made Maesil Extract
Among the 18 home-made maesil extracts analyzed, the total content of biogenic amines ranged from 2.5 to 241.7 mg/L, the major individual biogenic amines were putrescine (not detectable (ND)-80.82 mg/L) and spermidine (ND-219.20 mg/L), followed by tryptamine (Table 2). Putrescine, histamine, tyramine, cadaverine, 2-phenylethylamine, spermidine, spermine, agmatine, and tryptamine are the main biogenic amines in wine [27]. Among these amines, putrescine has been reported to be generated from the raw material or by microbial decarboxylation [28]. In the case of wine, putrescine content has been found to be influenced by geographical region and grape variety [29]. Histamine and spermine detected in wine [27,29,30] are known to have toxicity or play a role in enhancing toxicity [11,31]. However, we were unable to detect either of these two amines in the 18 maesil extracts examined in the present study. These results imply that the amount and composition of biogenic amines may differ widely among different fruit-derived products and that these differences could be attributed to differences in manufacturing practice and fruit material.   The content of biogenic amines is known to be affected by fermentation conditions, including temperature, microorganisms, and the synthetic pathways of the biogenic amine formation [32][33][34][35]. In wine, cadaverine, histamine, putrescine, and tyramine are mainly detected, the content of which can vary depending on fermentation factors, storage, microbial decarboxylase activity, and vinification [27,30,36]. Marcobal et al. (2006) have reported that the content of biogenic amines in wine ranged from ND to 54.02 mg/L [31]. Garai et al. (2006) found that the main biogenic amine in commercial apple ciders was putrescine and that the total biogenic amine content ranged from ND to 23.26 mg/L [12]. In comparison, the results of this study indicate that the biogenic amine content in home-made maesil extracts is considerably higher than that reported in wine or apple ciders [12,27,30], thereby emphasizing the necessity to control biogenic amines productions during the fermentation of maesil extract.

Content of Biogenic Amines During Soaking and Fermentation
During the 90 day soaking of maesil examined in the present study, we found that the total biogenic amines content increased from 14.1 to 35.0 mg/L and 37.3 to 69.1 mg/L at 15 and 25 • C, respectively, indicating that the content was higher at the latter temperature throughout the soaking period ( Figure 1a). Previous studies have reported that biogenic amines are generated via the catalytic activity of decarboxylase enzymes produced during the growth of microorganisms such as lactic acid bacteria [37], and thus, the increase in biogenic amines during the soaking period might be caused by microbial decarboxylase activity. At both incubation temperatures we assessed, the predominant biogenic amines detected in maesil extract were putrescine and spermidine (Figure 2b,c), and the latter comprised approximately 80% of the total biogenic amines. The content of biogenic amines is known to be affected by fermentation conditions, including temperature, microorganisms, and the synthetic pathways of the biogenic amine formation [32][33][34][35]. In wine, cadaverine, histamine, putrescine, and tyramine are mainly detected, the content of which can vary depending on fermentation factors, storage, microbial decarboxylase activity, and vinification [27,30,36]. Marcobal et al. (2006) have reported that the content of biogenic amines in wine ranged from ND to 54.02 mg/L [31]. Garai et al. (2006) found that the main biogenic amine in commercial apple ciders was putrescine and that the total biogenic amine content ranged from ND to 23.26 mg/L [12]. In comparison, the results of this study indicate that the biogenic amine content in home-made maesil extracts is considerably higher than that reported in wine or apple ciders [12,27,30], thereby emphasizing the necessity to control biogenic amines productions during the fermentation of maesil extract.

Content of Biogenic Amines During Soaking and Fermentation
During the 90 day soaking of maesil examined in the present study, we found that the total biogenic amines content increased from 14.1 to 35.0 mg/L and 37.3 to 69.1 mg/L at 15 and 25 °C, respectively, indicating that the content was higher at the latter temperature throughout the soaking period (Figure 1a). Previous studies have reported that biogenic amines are generated via the catalytic activity of decarboxylase enzymes produced during the growth of microorganisms such as lactic acid bacteria [37], and thus, the increase in biogenic amines during the soaking period might be caused by microbial decarboxylase activity. At both incubation temperatures we assessed, the predominant biogenic amines detected in maesil extract were putrescine and spermidine (Figure 2b,c), and the latter comprised approximately 80% of the total biogenic amines.  After removing the maesil fruit from the sample jars at the end of the soaking period, the residual liquid was subsequently fermented for 180 days, during which, the content of biogenic amines decreased from 31.7 to 13.6 mg/L and 116.8 to 57.1 mg/L at 15 and 25 • C, respectively (Table 3). Generally, the extracts fermented at 25 • C exhibited biogenic amines content that was twice as high as that obtained at 15 • C. Moreover, at the end of the fermentation period, the total biogenic amines content at 15 • C was 23.8% of that at 25 • C. In addition to putrescine and spermidine, tryptamine was also detected at 0.33 mg/L when maesil was fermented at the higher temperature for 30 days. In this regard, Chong et al. (2011) have reported that temperature was the most important factor affecting biogenic amines formation [38], and Pinho et al. (2001) reported a higher increase in biogenic amines at a storage temperature of 21 • C than at 4 • C [35]. In addition, Kim et al. (2002) found that 25 • C was the optimum temperature for histamine production in fish muscles [39]. Maesil extract is typically produced by natural fermentation without controlling the temperature or starter culture. Moreover, it is sometimes consumed immediately after a 90 day soaking without subsequent fermentation. On the basis of the results obtained in this study, we recommend that, to yield a product with lower levels of biogenic amines, maesil extract should be fermented at a relatively low temperature and for a long period of time.

Effects of Processing Factors on Biogenic Amines Formation
The pathways implicated in the synthesis of biogenic amines can vary depending on the temperature, sugar content, precursors, and microorganisms involved in the fermentation of various food items [16,32,40]. Generally, putrescine is derived from the decarboxylation of arginine and ornithine or is already present in raw materials [28,32], whereas, spermidine is produced from arginine and ornithine or is converted from putrescine by spermidine synthase [16,31]. Poveda (2019) and Bardocz (1995) reported that most putrescine is either converted to spermidine or spermine, or is catabolized to succinate and other amino acids via succinate [28,31].
In the present study, to determine the effects of the factors influencing biogenic amines formation, we performed Pearson's correlation. Among these factors, we detected no significant correlation between biogenic amines content and pH, which had a narrow range (pH 2.9-3.3) during the fermentation. Arena et al. (2008) reported a negative correlation between biogenic amines and sugar concentration and found that; the additions of glucose and fructose at 5 and 20 g/L reduced biogenic amines production by 82%-93% and 61%-99%, respectively [40]. Cid et al. (2008) reported that lower glucose concentration is associated with a high activity of ornithine-decarboxylase produced by Lactobacillus [41]. In our study, we found that, although there was a negative correlation between sugar content and biogenic amines content (Figure 3a), the relationship was not significant, which could be attributable to the narrow range of sugar content (61 to 81 • Brix) during fermentation.  The total biogenic amines content showed a positive correlation with the total amounts of amino acids (R = 0.6581, p < 0.05), which could be explained by the fact that amino acids are precursors of biogenic amines [42]. We also detected a strong positive correlation between the amounts of putrescine and spermidine (R = 0.9277, p < 0.01; data not shown), consistent with the findings of Bardocz (1995) [31] and Nuriez et al. (2016) [16], which indicates that putrescine is a precursor of spermidine. However, apart from a positive correlation between arginine and total biogenic amines content (R = 0.6910, p < 0.05), we detected no correlation between individual biogenic amines and their respective precursor amino acid, which is consistent with the findings reported by Soufleros et al. (1998) [43]. Gezginc et al. (2013) reported that arginine serves as a precursor of putrescine, which can, in turn, be converted to spermidine [32]. Furthermore, it has been found that, in plants and some microorganisms, there are alternative pathways in which putrescine is generated from arginine via agmatine [33]. These results indicate that the fermentation of maesil extract at low temperature could reduce the production of biogenic amines. In addition, biogenic amine formation in maesil extract could be affected by the origin of maesil, the number of amino acids as well as the content of biogenic amine precursors.

Conclusions
The present study was conducted to evaluate the changes in biogenic amines formation and the relationship between biogenic amines and amino acids in maesil extract during the fermentation of this product. Although the consumption of maesil extract is currently increasing, there has, to date, been a lack of studies on the changes that biogenic amines undergo during maesil extract fermentation. The results of this study showed that the biogenic amines content in maesil extract is affected by both the inherent amino acids content and fermentation temperature and time. Moreover, the content of some biogenic amines may also be affected by the presence of other biogenic amines. We found that both amino acids and biogenic amines content was lower during fermentation at 15 • C than at 25 • C and decreased with increasing fermentation time. Accordingly, these observations indicate that employing protracted low-temperature fermentations could be an effective approach for reducing the production of biogenic amines in maesil extract. In further research, it will be necessary to study the types of microorganisms and formation on biogenic amines in maesil extract.