Development of a colorimetric enzymatic assay method for aromatic biogenic monoamine-producing decarboxylases

Abstract

Biogenic amines (BAs) produced by the action of bacterial amino acid decarboxylases in fermented foods cause various health problems in human. Despite the importance, detailed characterizations of the BA-producing decarboxylases are relatively less progressed than the studies on BA-producing bacteria, due to the time-consuming chromatography-based assay method. In this study, a simple and general colorimetric assay for aromatic amino acid decarboxylases coupled with an amine oxidase from Arthrobacter aurescens (AMAO) and horseradish peroxidase was developed using a tyrosine decarboxylase from Enterococcus faecium DSM20477 (EfmTDC) as a model enzyme. The activity profiles over pH and temperature and the kinetic analysis for EfmTDC revealed that the results by the colorimetric assay are compatible with those by the chromatographic assay. In addition, due to the broad substrate specificity of AMAO for histamine and 2-phenylethylamine, the colorimetric assay would be applicable to the characterization of other aromatic amino acid decarboxylases including histidine decarboxylases.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10068-021-00938-4.

Introduction

Biogenic amines (BAs) in foods are considered as food poisons causing health problems in human (Ruiz-Capillas and Herrero, 2019), such as migraine and hypertensive crisis known as cheese-reaction by tyramine (Price and Smith, 1971) and the allergy-like poisoning known as scombroid poisoning by histamine (Hungerford, 2010). As BAs are found in various fermented foods including fermented dairy products, wine, meat, soybean pastes, and so on (Coton et al., 2010; Linares et al., 2012; Mah et al., 2019; Ruiz-Capillas and JimÉNez-Colmenero, 2005), various studies regarding BAs in foods have been extensively progressed; either culture- or polymerase chain reaction-based BA-producing bacteria detection (Chang and Chang, 2012; Landete et al., 2007), development of BA-degrading enzyme and bacteria (Barbieri et al., 2019; García-Ruiz et al., 2011; Jiang et al., 2020), BA-negative bacteria as starters (Heo et al., 2020; Xiong et al., 2020), inhibitor screening against the decarboxylases (Kang et al., 2017; Kim et al., 2014; Wendakoon and Sakaguchi, 1995) and the processes control of BA-producing bacteria contamination (Gardini et al., 2016; Naila et al., 2010).

BAs are produced by the action of bacterial decarboxylases using free amino acids as the substrates. So far, there is much information regarding BA-producing decarboxylases at the molecular level, whereas detailed enzymatic characterization is still less in progress. The general assay method for the BA-producing decarboxylases is chromatographic analysis of the products derivatized to introduce an UV-active functional group using high pressure liquid chromatography (HPLC) with the high resolution and sensitivity (Munir and Badri, 2020). For the facile analysis of BA-producing decarboxylases, a simple and rapid quantitative enzyme assay methods is required. As a rapid assay, CO₂ detection using a selective CO₂ electrode has been applied, but the detection limit was relatively low compared to the chromatographic methods (Moreno-Arribas and Lonvaud-Funel, 2001).

Amine oxidases (EC1.4.3.21) catalyze oxidative deamination toward primary amines. An amine oxidase from Arthrobacter aurescens (AMAO) exhibited a broad substrate specificity for aromatic monobiogenic amines including tryptamine, 2-phenyethylamine, tyramine and histamine, producing the corresponding aldehydes, ammonia, and hydrogen peroxide (Lee and Kim, 2013). Through a coupled reaction using AMAO and horse radish peroxidase (HRP), enzymatic colorimetric determination of total monobiogenic amines in cheeses and fermented soybean pastes has been developed (Lee and Kim, 2013; Lee et al., 2013b). Therefore, this enzymatic colorimetric BAs determination would be potential assay method to measure the activities of aromatic amino acid decarboxylases.

In this study, a recombinant tyrosine decarboxylase (TDC) from Enterococcus faecium DSM20477 (EfmTDC) in Escherichia coli was used as a model enzyme for assay methodology for aromatic monobiogenic amine-producing decarboxylases coupling with AMAO and HRP (Fig. 1). Our results suggest that this enzymatic colorimetric assay exhibited sensitive performance compatible with HPLC-based assay.

Fig. 1.

Scheme of colorimetric assay for tyrosine decarboxylases

Materials and Methods

Bacterial strains

Enterococcus faecium DSM20477, which purchased from Korean Collection for Type Cultures (Jeongeup, Korea), used as the source of tyrosine decarboxylase gene. E. coli TOP10 and E. coli BL21 (DE3) were purchased from Thermo Fischer Scientific (Waltham, MA, USA) were used as a host for DNA manipulation and protein overexpression, respectively.

Construction of a tyrosine decarboxylase

The DNA fragment containing the gene for a tyrosine decarboxylase was amplified from the genomic DNA of E. faecium DSM20477 with two synthetic primers (EfmTDC-Nc-F, 5’-TCACCATGGGTGAATCATTGTCGAAAG-3’ and EfmTDC-Xh-R, 5’-CCTCTCGAGTTT TGCTTCGCTTGCCAA-3’ containing NcoI and XhoI restriction sites) which were designed based on the genome sequence of E. faecium DO (GenBank assembly accession: CP003583). The amplified DNA fragment treated with NcoI and XhoI was ligated with pET28a digested by the corresponding enzymes to create pET28-EfmTDC6xH. Analysis of the recombinant DNA sequences was carried out by Cosmogenetech (Daejeon, Korea).

Purification of the recombinant enzymes in E. coli

Overexpression and purification using Ni-affinity chromatography of 6xHis-Tagged EfmTDC and AMAO have been followed by the previous studies (Lee and Kim, 2013). The purified AMAO was dialyzed against 100 mM potassium phosphate buffer (pH 7.0). EfmTDC was stored in Ni–NTA elution buffer at 4 °C, and 150-fold diluted EfmTDC solution using 100 mM sodium acetate buffer (pH 5.5) was prepared freshly. HRP was purchased from Sigma (St. Louis, MO, USA).

Colorimetric assay of EfmTDC coupling with AMAO and HRP

Twenty-microliter of the diluted enzyme was mixed with 180 μL of 100 mM sodium acetate buffer (pH 5.5) containing 0.2 mM pyridoxal 5-phosphate (PLP) and 5.5 mM either L-tyrosine or L-phenylalanine and then incubated for 10 min. The reaction mixtures were boiled for 10 min to terminate the reaction. Then, 50 μL of the reaction mixture was transferred to a 96-well plate and mixed with 130 μL of reaction mixture (final concentrations of prepared solution; 9 mM sodium hydroxybenzoate, 0.9 mM 4-aminoantipyrine, 0.15 U/mL HRP, 100 mM potassium phosphate/pH 7.0). After prewarming 5 min at 37 °C, the reaction was initiated by addition of AMAO (20 μL of 2 U/mL). The absorbance of the mixture was measured with a microplate reader (VERSAmax™, Molecular device, Sunnyvale, USA) at wavelength of 498 nm until absorbance was saturated. For determination of the optimal temperature, EfmTDC was incubated in 100 mM sodium acetate buffer (pH 5.5) containing 0.2 mM PLP and 5.5 mM L-tyrosine at temperature ranging from 15 to 70 °C. For optimal pH, the enzyme reaction was carried out at following buffers; 100 mM sodium acetate buffer (pH 3.0–6.0) and 100 mM potassium phosphate buffer (pH 6.5–8.0) containing 0.2 mM PLP and 5.5 mM L -tyrosine at the optimum temperature.

Kinetic analysis of EfmTDC

The reaction mixtures (180 μL) were prepared in 100 mM sodium acetate buffer (pH 5.5) containing 0.2 mM PLP and 0 to 5.5 mM L-tyrosine and L-phenylalanine. After adding 20 μL of the diluted EfmTDC, they were incubated for 10 min at 35 °C. The aliquot of the reaction mixtures taken in time intervals (3, 6, and 10 min) was boiled for 5 min for the reaction termination. The color development was conducted as described above. The reaction rates were obtained from the slops of the linear regression plots for the produced amine concentrations over reaction time. The kinetic parameters were calculated by fitting the data to the Michalis-Menten equation using SigmaPlot program (Systat Software, Inc., San Jose, CA, USA).

Tyramine and phenylethylamine measurement using HPLC

To determine tyramine and β-phenylethylamine contents by HPLC, 100 μL of EfmTDC reaction mixtures used in the EfmTDC assay was diluted up to 200 μL with distilled water. The dansylation of the tyramine in the mixtures and HPLC analysis were conducted as described previously (Lee et al., 2013b).

Results and Discussion

Expression of EfmTDC in E. coli

Upon expressing the gene for EfmTDC in E. coli TOP10 using a constitutive promoter system, the recombinant protein formed inclusion body (Data not shown). By contrast, the T7 promoter-based expression system in E. coli BL21(DE3) successively produced soluble EfmTDC at 25°C for 24 h after 0.05 mM IPTG induction. The band intensities of the proteins corresponding to EfmTDC in the total protein fraction and the soluble extraction were almost same, suggesting that most EfmTDCb was correctly folded during the low temperature fermentation (Fig S1 in supplementary information). After Ni–NTA affinity chromatography, generally 648 mg of EfmTDC was obtained from 1 L culture.

Determination of condition for colorimetric assay

As the substrate concentration was 5.5 mM of tyrosine, the maximum initial rate for EfmTDC in the assay condition should be lower than 0.55 mM tyramine production for 10 min of the reaction time. To determine the amount of HRP as the enzyme for the coupled reaction, the reactions catalyzed by different amount of HRP was monitored using 0.7 mM hydrogen peroxide (higher than 0.55 mM) as the substate. Upon employing 0.15 U/mL of HRP, the reactions were completed within less than 3 min (Fig. 2A). Therefore, HRP rate by 0.15 U/mL of HRP was chosen as the HRP dose for the colorimetric assay.

Fig. 2.

(A) Effects of HRP on the absorbance of the reaction mixtures using H₂O₂ as the substrate, (B) Effect of AMAO on the absorbance of the reaction mixtures supplemented with HRP (0.15 U/mL) using tyramine as the substrate. (C) Standard curve of the colorimetric assay mixtures supplemented with AMAO (2.5 U/mL) and HRP (0.15 U/mL) using tyramine as the standards. The slopes of the standard curves were calculated by the fitting of the absorbance was measured at 5 min to a linear functional equation with linear regression

In the case of AMAO, the reactions were conducted by varied amount of AMAO ranging from 1.5 to 5 U/mL with supplemented 0.15 U/mL HRP using 0,15 mM of tyramine as the substrate (Fig. 2B). As expected, the lower dose of AMAO requested the longer reaction time to complete the reaction. To reduce assay time, the high dose of AMAO (final concentration of 5 U/mL) was chosen. Interestingly, the absorbance after the maximum point gradually decreased over time (Fig. S2A). The reason for such absorbance decline is unclear, but the standard curves prepared using the absorbance at a certain time point showed great linearity up to 0.7 mM tyramine; R² values of the standard curves using data measured at 5, 10, 15, and 20 min were higher than 0.999 (Fig. 2C and Fig. S2). These results suggested that the absorbance which are collected at a certain time point after the absorbance reaches the maximum, would be acceptable for the standard curve for tyramine. More importantly, the data collection time of the TDC assay reactions during the colour development should be same as that for the standard curve. From this point of view, the time point for measuring the absorbance was determined as 5 min for the TDC assay (Fig. 2C).

Given the substate specificity of AMAO (Lee et al., 2013a), another aromatic monobiogenic amines, including 2-phenylethylamine and histamine could be used as the substrates. Upon employing the selected dose of AMAO and HRP, the reactions using 2-phenylethylamine as the standards were completed within 5 min (Fig. 3A). Due to the less preference of AMAO for histamine, threefold AMAO was requested for the absorbance saturation within 5 min (Fig. 3A). Indeed, the standard curves for 2-phenylethylamine and histamine prepared using absorbance measured at 5 min showed great linearity with R² values of 0.9999 and 0.9993, respectively (Fig. 3B and C). Therefore, the colorimetric assay in this study could be used for TDC as well as phenylalanine decarboxylases and histidine decarboxylase using the corresponding amino acids as the substrates.

Fig. 3.

(A) Time course analysis of the reaction mixtures supplemented with HRP using tyramine, 2-phenylethylamine, histamine as the substrates. Standard curves of the colorimetric assay using 2-phenylethylamine (B) and histamine (C) as the standards. The slopes of the standard curves were calculated by the fitting of the absorbance was measured at 5 min to a linear functional equation with linear regression

Comparison of colorimetric and chromatographic assay for EfmTDC

To verify the accuracy of the colorimetric assay, the pH and temperature profiles for EfmTDC was determined by the colorimetric assay, followed by comparison of those by the chromatographic assay using HPLC. The effect of pH on the EfmTDC activity was investigated at varied pH from pH 3.0 to 8.0 at 25 °C which is the optimum temperature for a TDC from E. faecium R615Z1 (Liu et al., 2014). EfmTDC showed its maximum activity at pH 5.5 (Fig. 4A), which was consistent with those of other bacterial TDCs ranging from pH 5.0–pH 5.5 (Liu et al., 2014; Maini Rekda et al., 2019; Zhang et al., 2011; Zhu et al., 2016) At fixed pH at 5.5, the maximum activity for EfmTDC was observed at 50 °C (Fig. 4B), which was quite different from the previously reported value by a TDC from E. faecium R615Z1 (Liu et al., 2014) with 100% matched amino acid sequence with EfmTDC. However, the profiles for pH and temperature determined by the colorimetric assay were fully matched with those by the HPLC assay (Fig. 4).

Fig. 4.

Effects of pH and temperature on EfmTDC activity. (A) pH and (B) temperature profiles determined by the enzyme assay coupling with AMAO and HRP (black circles) and HPLC (white circles) in varied range of pH and temperature, respectively. Mean values and standard deviations from three independent experiments are shown.

Kinetic analysis of EfmTDC

According to literature, several bacterial TDC showed activities for not only L-tyramine but also L-DOPA and L-phenylalanine (Liu et al., 2014; Zhang and Ni, 2014). Unfortunately, dopamine, the product of decarboxylation of L-DOPA, was not the substrate of AMAO (Data not shown). The relative activities of EfmTDC for L-tyrosine and L-phenylethylamine were 100: 7.5 (Fig. S3), whereas L-histidine was not decarboxylated by EfmDC. To investigate the substrate specificity of EfmTDC, kinetic analysis was conducted by the colorimetric and HPLC assay (Fig. S4, Table 1). Upon employing L-tyrosine as the substrate, the k_cat and K_M values by the colorimetric assay were 29.0 ± 0.68 s⁻¹ and 1.71 ± 0.10 mM, respectively, and those by the HPLC assay were 29.0 ± 0.48 s⁻¹ and 1.76 ± 0.07 mM, respectively, which were consistent with those for E. faecium W54 (van Kessel et al., 2019). In the case of L-phenylalanine by the colorimetric and HPLC assay determined k_cat of 3.43 ± 0.31 s⁻¹ and 3.58 ± 0.29 s⁻¹, respectively, and the K_M values were 14.1 ± 1.64 mM and 15.7 ± 1.61 mM, respectively. Therefore, the relative substrate specificities of EfmTDC based on the catalytic efficiencies (k_cat/K_M) of 100:1.44 were determined for L-tyrosine and L-phenylalanine, respectively.

Table 1.

Kinetic parameters of EfmTDC and other bacterial TDCs

Substrates	Bacteria	Assay	kcat	KM	kcat/KM	Condition	References
			(s−1)	(mM)	(s−1·mM−1)
L-Tyrosine	E. faecium DSM20477	Colorimetric	29.0 ± 0.68	1.71 ± 0.10	17.0 ± 0.96	pH 5.5/35 °C	This study
		HPLC	29.0 ± 0.48	1.76 ± 0.07	16.5 ± 0.64
	E. faecium W54	HPLC	36.7	1.5	24.5	pH 5.0	(van Kessel et al., 2019)
	E. faecalis strain EnGen0310	HPLC	63.6	0.315	201.9	pH 5.5/RT	(Zhu et al., 2016)
	E. faecalis ATCC 700,802	HPLC	116.1	0.6	193.5	pH 5.0	(Maini Rekda et al., 2019)
	Lactobacillus brevisCGMCC 1.2028	HPLC	343.1	0.59	581.5	pH 5.0/40 °C	(Zhang and Ni, 2014)
			124.8	0.6	216
L-Phenylalanine	E. faecium DSM20477	Colorimetric	3.43 ± 0.31	14.1 ± 1.64	(24.5 ± 3.14) × 10–2	pH 5.5/35 °C	This study
		HPLC	3.58 ± 0.29	15.7 ± 1.61	(22.9 ± 2.61) × 10–2

In conclusion, the AMOA-based colorimetric assay for EfmTDC yielded almost identical results to the HPLC assay. To our best knowledge, there are commercialized colorimetric kits for determination of histamine (Kikkoman Co., Tokyo, Japan) and putrescine (BioVision Inc., San Francisco, USA) which would be applicable to the decarboxylase assays producing the corresponding BAs. By contrast, as the kinetic analysis revealed, the colorimetric assay demonstrated the performance for analysis of not only for TDC but also phenylalanine decarboxylase activities. Therefore, the colorimetric assay in this study would be a general assay method for aromatic amino acid decarboxylases.

Supplementary Information

Below is the link to the electronic supplementary material.

Declarations

Conflict of interest

The authors declare no conflicts of interest.