2020 Volume 26 Issue 2 Pages 313-317
N-Methyltyramine (N-MeTA) is known as a gastrin-releasing factor in beer. In this study, the agonistic actions of N-MeTA as well as tyramine (TA)/β-phenylethylamine (PEA) and their other N-methylated derivatives were examined to elucidate their structure-activity relationships, using a secreted placental alkaline phosphatase (SEAP)-based reporter assay in HEK-293 cells transiently expressing a G protein-coupled receptor, human trace amine-associated receptor 1 (hTAAR1). We detected the agonistic actions of six test compounds, including N-MeTA (EC50 = 6.78 µM), on hTAAR1. The agonistic actions were reduced depending on the number of N-methyl groups introduced into TA and PEA; the order of potency is PEA > N-methylphenylethylamine > TA ≈ N,N-dimethylphenylethylamine ≥ N-MeTA ≥ N,N-dimethyltyramine. Taken together with our previous study on TA/PEA as agonists for hTAAR1 in the stomach, this finding suggests that hTAAR1 might be the primary target of N-MeTA in the stomach; however, the agonistic potency of N-MeTA is weaker compared to TA and PEA.
Aromatic biogenic amines such as tyramine (TA) and β-phenylethylamine (PEA) (Fig. 1A) are well-known trace amines in the mammalian nervous system. These two amines modulate various neuronal functions via trace amine-associated receptors (TAARs), which belong to the G protein-coupled receptor (GPCR) family (Gainetdinov et al., 2018). TA and PEA are abundantly present in fermented foods such as cheese, wine, chocolate, and traditional fermented food products consumed worldwide (Naila et al., 2010; Kim et al., 2012; Gainetdinov et al., 2018). These amines are also derived from decarboxylation of L-tyrosine and L-phenylalanine during fermentation by specific bacteria, including those belonging to the genera Lactobacillus and Enterococcus (Fernández et al., 2015; Takebe et al., 2016). We recently reported that food-derived TA and PEA are putative agonists for human TAAR 1 (hTAAR1) in the stomach, and especially in the pylorus (Ohta et al., 2017). The pylorus contains G cells, which release gastrin to stimulate gastric fluid secretion. The physiological roles of hTAAR1 and its agonists in peripheral tissues are gaining attention in not only the medicinal field but also the food/nutrition field (Adriaenssens et al., 2015; Raab et al., 2016; Gainetdinov et al., 2018; Batista-Lima et al., 2019).
(A) Structures of tyramine (TA), β-phenylethylamine (PEA), and their N-methylated/N,N-dimethylated compounds, and (B) syntheses of N-methylated TA/PEA derivatives.
A related amine compound, the TA metabolite N-methyltyramine (N-MeTA) (Fig. 1A), is found in various plants, particularly malted barley, with levels as high as 2 mg/g. Thus, an equivalent concentration is present in beer as well (Stohs and Hartman, 2015) and Tsutsumi et al. (2010) reported that the N-MeTA content in beer was 2 mg/L. Administration of N-MeTA has been shown to induce increases in blood pressure, enhancement of cardiac contraction, and inhibition of lipolysis (Stohs and Hartman, 2015). In addition, N-MeTA is known as a gastrin-releasing factor in beer (Yokoo et al., 1999; Stohs and Hartman, 2015). Considering the gastrin-releasing activity of N-MeTA together with our previous work (Ohta et al., 2017), we assumed that N-MeTA acts as an hTAAR1 agonist to stimulate gastrin secretion in the stomach. In this study, the agonistic actions of N-MeTA as well as TA/PEA and their other N-methylated derivatives [N,N-dimethyltyramine (N,N-diMeTA), N-methylphenylethylamine (N-MePEA), and N,N-dimethylphenylethylamine (N,N-diMePEA)] (Fig. 1A) were examined to elucidate their structure-activity relationships, using a secreted placental alkaline phosphatase (SEAP)-based reporter assay system (Ohta et al., 2012) in HEK-293 cells transiently expressing hTAAR1 (Ohta et al., 2017).
Test compounds TA hydrochloride was purchased from Nacalai Tesque, Inc. (Kyoto, Japan). PEA hydrochloride and N,N-diMeTA (hordenine) were obtained from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). N-MePEA was purchased from Sigma-Aldrich Corp. (St. Louis, MO, USA). N,N-diMePEA hydrochloride and N-MeTA hydrochloride were synthesized from the corresponding phenethylbromides and a large excess of N-methylated amines, as shown in Fig. 1B. The products were extracted with ether or chloroform, purified by column chromatography on silica gel (7/3-chloroform/methanol), neutralized by hydrochloric acid, and then isolated as colorless crystals. The NMR data are as follows. N,N-diMePEA hydrochloride: NMR δH(dimethyl sulfoxide-d6): 10.86 (1H, NH+), 7.36–7.24 (5H, aromatic), 3.24 (2H, Ph-CH2-), 3.03 (2H, N-CH2-), 2.78 [6H, N-(CH3)2]. N-MeTA hydrochloride: NMR δH(dimethyl sulfoxide-d6): 9.31 (1H, OH), 8.69 (2H, NH2+), 7.05 (2H, aromatic), 6.73 (2H, aromatic), 3.06 (2H, Ph-CH2-), 2.82 (2H, N-CH2-), 2.56 (3H, N-CH3).
CRE-SEAP reporter assays Prior to the SEAP reporter assays, HEK-293 cells were grown in Dulbecco's modified Eagle's medium (DMEM) (Nacalai Tesque, Inc.) supplemented with 10% fetal bovine serum (FBS; Life Technologies, Carlsbad, CA, USA) at 37 °C and 5% CO2. As shown in our previous study (Ohta et al., 2017), the receptor expression vector pcDNA3-β2N9/hTAAR1 (1 µg) and the reporter vector pCRE-SEAP (1 µg) (Clontech, Mountain View, CA, USA) were transiently co-transfected into HEK-293 cells (6 × 105 cells/35-mm diameter dish) using GeneJuice® Transfection Reagent (Novagen, Gibbstown, NJ, USA). The following cyclic AMP (cAMP) response element (CRE)-SEAP reporter assays were performed according to the procedure described previously (Ohta et al., 2017). TA/PEA and their N-methylated compounds at concentrations ranging from 10−9–10−4 M were added to the transfected cells along with 100 µM 3-isobutyl-1-methylxanthine (IBMX; Nacalai Tesque, Inc.) in FBS-free DMEM. Dimethyl sulfoxide, used to dissolve and dilute the compounds, was included in the FBS-free medium at 1% concentration, without undesirable effects on SEAP activity.
According to the CRE-SEAP assays, TA, N-MeTA, and N,N-diMeTA at concentrations of 10−9–10−4 M showed a dose-dependent increase in SEAP activity in transiently hTAAR1-expressing HEK-293 cells (Fig. 2A), with EC50 values of 4.02 µM [95% confidence limit (CL), 2.85–6.00 µM), 6.78 µM (95% CL, 4.85–10.31 µM), and 10.60 µM (95% CL, 7.65–16.78 µM), respectively. The order of potency was TA ≥ N-MeTA ≥ N,N-diMeTA. With the same experimental conditions, PEA, N-MePEA, and N,N-diMePEA exhibited a typical sigmoidal hTAAR1 agonist curve (Fig. 3A), with EC50 values of 0.56 µM (95% CL, 0.40–0.79 µM), 1.63 µM (95% CL, 1.19–2.26 µM), and 4.31 µM (95% CL, 3.21–5.93 µM), respectively. The relationship between mono/di N-methylation of PEA and agonist potency for hTAAR1 was similar to that of TA; and the order of potency for the three PEA-related compounds was PEA > N-MePEA > N,N-diMePEA. These results indicate that the more methyl groups TA and PEA have at the amino group, the less agonistic potency the TA and PEA derivatives show. Overall, the potency of the six compounds tested can be arranged as: PEA > N-MePEA > TA ≈ N,N-diMePEA ≥ N-MeTA ≥ N,N-diMeTA. As N-MeTA and N,N-diMeTA were the least potent agonists, they did not attain the same maximal SEAP activity at 10−4 M as TA or the PEA compounds (Figs. 2A and 3A). Nonetheless, the respective contents of N-MeTA and N,N-diMeTA (hordenine) in different beer samples are 0.59–4.61 mg/L (3.9–30.5 µM) and 1.05–6.32 mg/L (6.35–38.25 µM) (Sommer et al., 2019), which are comparable with each EC50 value. Therefore, upon direct interaction between the ingested beer and hTAAR1 expressed in the stomach (Ohta et al., 2017), the agonist concentrations in the beer would be sufficient to activate hTAAR1.
Agonist response of TA, N-MeTA, and N,N-diMeTA in (A) transiently β2N9/hTAAR1-expressing HEK-293 cells and (B) HEK-mock cells by CRE-SEAP assay. Each compound at 10−9-10−4 M was added to cells, and the cells were incubated at 37 °Cand 5% CO2 for 1 day. The SEAP activity at the basal level (Ba, SEAP activity with no agonist) was set as 1. HEK cells transiently transfected with the empty pcDNA3 were used as HEK-mock cells for negative control experiments. The adenylate cyclase activator forskolin was used to confirm the negative control experiment. Data represent the mean ± SE for three independent experiments, each performed in duplicate.
Agonist response of PEA, N-MePEA, and N,N-diMePEA in (A) transiently β2N9/hTAAR1-expressing HEK-293 cells and (B) HEK-mock cells by CRE-SEAP assay. Cell cultures and CRE-SEAP assays were performed under identical conditions as described in Fig. 2. The basal SEAP activity (Ba) was set as 1. Data represent the mean ± SE for three independent experiments, each performed in duplicate.
Lindemann et al. (2005) reported that mono N-methyl substitution of TA and PEA had an obscure effect on agonist activity in cAMP assays with HEK-293 cells stably expressing hTAAR1/rat TAAR1 chimera receptors. However, Wainscott et al. (2007) showed that N-methyl and N,N-dimethyl substitution of PEA induced a 2–3-fold and 14-fold decrease in cAMP levels of hTAAR1-expressing GαsAV12-664 cells (Syrian hamster fibroblast cell line stably transfected with rat Gαs protein), respectively. Taken together, it can be concluded that N,N-dimethylation and p-hydroxyl substitution among the test compounds are unfavorable for interaction with the agonist-binding site of hTAAR1.
In negative control experiments where the empty pcDNA3 was introduced into HEK-293 cells, forskolin, as a direct activator of adenylate cyclase, induced a dose-dependent increase in SEAP activity, whereas no SEAP activity was observed for the six compounds tested (Figs. 2B and 3B). This result ensures that all six compounds acted as agonists for hTAAR1 receptors expressed in HEK-293 cells.
In a previous study, N-MeTA was reported to have the property of an α2-adrenergic receptor antagonist (Koda et al., 1999). Most recently, it was shown that N-MeTA, as well as hordenine (N,N-diMeTA), functions as a dopamine D2 receptor agonist (Sommer et al., 2019). Although these receptors might be candidate in vivo targets for N-MeTA as the gastrin-releasing factor in beer (Yokoo et al., 1999), our study revealed that N-MeTA as well as other TA/PEA-related compounds acted as agonists for hTAAR1, with structure-activity relationships as discussed above. This finding is significant in further understanding the molecular mechanism of beer for stimulating gastrin release in the stomach. We are currently studying whether N-MeTA and N,N-diMeTA found in beer secrete gastrin through hTAAR1 at a cellular level.
Acknowledgments Part of this work was conducted at the Kurokami Radioisotope Laboratory of the Institute of Resource Department and Analysis (IRDA) of Kumamoto University. This work was supported in part by a research grant from the SKYLARK Food Science Institute in 2013. We would like to thank Editage (www.editage.jp) for English language editing.
tyramine
PEAβ-phenylethylamine
TAARstrace amine-associated receptors
GPCRG protein-coupled receptor
hTAAR1human TAAR 1
N-MeTAN-methyltyramine
N,N-diMeTAN,N-dimethyltyramine
N-MePEAN-methylphenylethylamine
N,N-diMePEAN,N-dimethylphenylethylamine
SEAPsecreted placental alkaline phosphatase
DMEMDulbecco's modified Eagle's medium
FBSfetal bovine serum
cAMPcyclic AMP
CREcAMP response element
CLconfidence limit
