Application of the Precolumn Derivatization Reagent CIM-C₂-NH₂ for Labeling Carboxyl Groups in LC-MS/MS Analysis of Primary Organic Acids in Japanese Sake

Abstract

Japanese sake, a traditional alcoholic beverage, contains several organic acids that may contribute to its sour taste. To identify these, a precolumn derivatization reagent, benzyl 5-(2-aminoethyl)-3-methyl-4-oxoimidazolidine-1-carboxylate (CIM-C₂-NH₂), developed for labeling carboxyl groups, was synthesized and applied to liquid chromatography–tandem mass spectrometry (LC-MS/MS) analysis of organic acids in six commercial sake samples. The majority primarily contained lactic acid (LA), and dicarboxylic acids, such as succinic acid (SA), malic acid (MA), and citramalic acid (CMA). The organic acid concentrations and compositions in the sake differed among brands. Notably, both l- and d-forms of LA were detected in all samples, while only d-CMA was present. To estimate the total acidic content, neutralization titration with sodium hydroxide was performed. In four of the six samples, titration results closely matched LC-MS/MS data, suggesting that l-LA, d-LA, SA, MA, and d-CMA were the primary contributors for the sour taste in these sakes. The discrepancy between titration and LC-MS/MS data for the other samples was attributed to the presence of other organic acids, which will be investigated in future studies.

1. Introduction

Japanese sake, namely Japanese rice wine, is a traditional alcoholic beverage in Japan and is now consumed worldwide. Japanese sake generally contains several organic acids [1], which are produced during its preparation process and may contribute to a sour taste. A previous study reported that a number of organic acids are present in sake, with the representative ones being lactic acid (LA), succinic acid (SA), and malic acid (MA) [2]. The composition of organic acids has been reported to vary across different brands of sake [1,3], and may depend on the type of yeast, rice, kouji, shubo, moromi, and the region where the sake is produced owing to differences in traditional preparation methods. Therefore, the differing organic acid compositions in each brand of sake may contribute to the variation in sourness.

Chromatography techniques can be used to analyze the composition of organic acids in complex matrices such as food and beverages because of their ability to separate each organic acid both quantitatively and qualitatively. Furthermore, both enantiomers of LA are present in sake [4,5]. Therefore, an enantiomer separation technique is preferable for investigating the organic acid components in sake.

We have recently reported a precolumn derivatization reagent, benzyl 5-(2-aminoethyl)-3-methyl-4-oxoimidazolidine-1-carboxylate (CIM-C₂-NH₂; Figure S1), which was developed for labeling carboxyl groups, and can be applied to liquid chromatography–tandem mass spectrometry (LC-MS/MS) analysis of organic acids [6,7]. This method allowed the separation of organic acid enantiomers in commercial wine [6,7].

Thus, in the current study, the main organic acids in Japanese sake were investigated by LC-MS/MS using CIM-C₂-NH₂. Derivatization with CIM-C₂-NH₂ enhances MS/MS detection because CIM-C₂-NH₂ can tag carboxyl groups (Figure S1), generating a prominent fragment ion at m/z 91 for monocarboxylates, while m/z (M + H-277) is observed for dicarboxylates [6]. Therefore, the use of CIM-C₂-NH₂ is suitable for the quantitative determination of a compound having carboxyl groups such as organic acids in food and beverages. However, in our previous study [6], the synthesis route for CIM-C₂-NH₂ was suboptimal owing to a low final yield.

Thus, we attempted to synthesize CIM-C₂-NH₂ via an alternative synthesis route to overcome the limitation of low yield.

Subsequently, the organic acids in sake were investigated by LC-MS/MS using a precolumn derivatization with CIM-C₂-NH₂. Finally, a portion of the sake sample was subjected to neutralization titration with a standard solution of NaOH, and the correlation between the titration results (which reflect total acidity) and the organic acid concentrations was investigated.

2. Materials and Methods

2.1. Chemicals

(R)-4-Amino-2-(((benzyloxy)carbonyl)amino)butanoic acid was purchased from Watanabe chemical industries, Ltd. (Hiroshima, Japan). Phthalic anhydride, 40% methylamine in H₂O, (4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium chloride (DMT-MM), paraformaldehyde, hydrazine hydrate, triphenylphosphine (TPP), and 2,2′-dipyridyl disulfide (DPDS) were purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan).

Sodium dl-lactate, d-lactate, and l-lactate for LA, and dl-citramalate and d-citramalate for CMA were purchased from Sigma-Aldrich Inc. (St. Louis, MO, USA).

Standard solutions of 0.01 and 0.02 mol/L NaOH in H₂O (factor: 1.000), first-grade acetic acid, sodium bicarbonate, anhydrous sodium sulfate, methanol, concentrated hydrochloric acid, potassium hydroxide, ethyl acetate, hexane, chloroform, special grade p-toluenesulfonic acid monohydrate, 28% ammonia, toluene, SA, l-malic acid (MA), high-performance liquid chromatography (HPLC)-grade and liquid chromatography–mass spectrometry (LC-MS)-grade CH₃CN, and HPLC-grade formic acid were purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan).

A packed SepaFlash^TM Standard Series ultra-pure irregular silica column with a particle size of 40–63 µm was obtained from Santai Science Inc. (Montreal, QC, Canada).

The water used was purified using a Milli-Q Lab system (Nihon Millipore; Tokyo, Japan). Sodium d-lactate ¹³C₃ (98%, d-LA¹³C₃), used as an internal standard (IS), was purchased from Cambridge Isotope Laboratories, Inc. (Tewksbury, MA, USA).

Six brands of Japanese sake were purchased from different local markets.

2.2. Apparatus

NMR spectra were recorded using a spectrometer (JMS-ECS 400, JEOL Ltd.; Tokyo, Japan). Mass spectra were recorded on a time-of-flight mass spectrometer (JMS-100LP AccuTOF LC-plus; JEOL Ltd., Tokyo, Japan) for high-resolution mass spectrometry (HRMS). The LC-MS/MS system used in this study was a tandem quadrupole mass spectrometer (LCMS-8040, Shimadzu Corporation; Kyoto, Japan), consisting of an autosampler (SIL-20AC), dual intelligent pumps (LC-20AD), PC software (LabSolutions ver. 5.80, Shimadzu Corporation), and column oven (CTO-20A), set at 40 °C.

2.3. Synthesis of CIM-C₂-NH₂

2.3.1. (R)-2-(((benzyloxy)carbonyl)amino)-4-(1,3-dioxoisoindolin-2-yl)butanoic Acid

Phthalic anhydride (10.1 mmol) and (R)-4-amino-2-(((benzyloxy)carbonyl)amino)butanoic acid (9.70 mmol) were dissolved in acetic acid (20 mL) and the resulting solution was stirred at reflux temperature for 1 h. The product was then extracted with ethyl acetate (100 mL) and washed with H₂O (2 × 100 mL) and saturated sodium bicarbonate (4 × 50 mL). The organic layer was dried with sodium sulfate, filtered, and concentrated under vacuum. The residue was purified by flash chromatography (silica gel, hexane/ethyl acetate (1:1)) to obtain the title compound (8.37 mmol, 86%). m/z [M + H] 383.1236 (calculated for C₂₀H₁₉N₂O₆ = 383.1243). ¹H-NMR (400 MHz, Chloroform-D) δ 9.17 (s, 1H, CO₂H), 7.79–7.76 (m, 2H, PhthH), 7.66–7.62 (m, 2H, PhthH), 7.34–7.29 (m, 5H, PhH), 5.77 (d, J = 8.5 Hz, 1H, NH), 5.07 (d, J = 12.9 Hz, 2H, PhCH₂), 4.46–4.41 (m, 1H, α-CH), 3.78 (t, J = 6.9 Hz, 2H, γ-CH₂), 2.27–2.14 (m, 2H, β-CH₂), ¹³C-NMR (101 MHz, Chloroform-D) δ 175.5, 168.2, 156.1, 136.0, 134.0, 131.7, 128.4, 128.1, 128.0, 123.3, 67.2, 51.5, 34.1, 30.3.

2.3.2. Benzyl (R)-(4-(1,3-dioxoisoindolin-2-yl)-1-(methylamino)-1-oxobutan-2-yl)carbamate

(R)-2-(((benzyloxy)carbonyl)amino)-4-(1,3-dioxoisoindolin-2-yl)butanoic acid (8.38 mmol) and 40% methylamine in H₂O (720 µL, 8.64 mmol) were dissolved in methanol (50 mL), then DMT-MM (10.7 mmol) was added to the solution and stirred for 1 h. The product was extracted with ethyl acetate (200 mL) and washed with 1 M hydrochloric acid (2 × 100 mL) and saturated sodium bicarbonate (2 × 100 mL). The organic layer was dried with sodium sulfate, then filtered and concentrated under vacuum. The residue was purified by recrystallization with aqueous ethanol to obtain the title compound as a white solid (6.56 mmol, 78%). m/z [M + H] 396.1541 (calculated for C₂₁H₂₂N₃O₅ = 396.1560). ¹H-NMR (400 MHz, Chloroform-D) δ 7.83–7.79 (m, 2H, ArH), 7.71–7.67 (m, 2H, ArH), 7.35–7.28 (m, 5H, ArH), 6.94 (d, J = 4.5 Hz, 1H, NH), 5.98 (d, J = 8.4 Hz, 1H, NH), 5.07 (d, J = 26.7 Hz, 2H, PhCH₂), 4.22 (d, J = 7.8 Hz, 1H, α-CH), 3.83–3.69 (m, 2H, γ-CH₂), 2.75 (d, J = 4.8 Hz, 3H, NCH₃), 2.15–2.09 (m, 2H, β-CH₂), ¹³C-NMR (101 MHz, Chloroform-D) δ 171.2, 168.5, 155.9, 134.0, 131.8, 128.4, 128.0, 127.8, 123.2, 123.2, 66.8, 52.4, 34.4, 31.5, 26.2.

2.3.3. Benzyl (R)-5-(2-(1,3-dioxoisoindolin-2-yl)ethyl)-3-methyl-4-oxoimidazolidine-1-carboxylate

Benzyl (R)-(4-(1,3-dioxoisoindolin-2-yl)-1-(methylamino)-1-oxobutan-2-yl)carbamate (5.45 mmol), p-toluenesulfonic acid monohydrate, and paraformaldehyde (7.5 mmol) were dissolved/suspended in toluene (100 mL), then heated at 120 °C for 45 min, at 150 °C for 15 min, and then, after the addition of paraformaldehyde (8.2 mmol), at 150 °C for another 30 min. The product was extracted with ethyl acetate (100 mL) and washed with saturated sodium bicarbonate (2 × 100 mL). The organic layer was dried with sodium sulfate, then filtered and concentrated under vacuum. The residue was purified by flash chromatography (silica gel, hexane/ethyl acetate (1:1)) to obtain the title compound (2.86 mmol, 52%). m/z [M + H] 408.1540 (calculated for C₂₂H₂₂N₃O₅ = 408.1560). ¹H-NMR (400 MHz, Chloroform-D) δ 7.81–7.76 (m, 2H, ArH), 7.68 (d, J = 2.4 Hz, 2H, ArH), 7.32 (d, J = 3.5 Hz, 5H, ArH), 5.13–4.94 (m, 2H, PhCH₂), 4.76–4.70 (m, 2H, γ-CH₂), 4.23 (d, J = 11.2 Hz, 1H, α-CH), 3.87–3.81 (m, 1H, NCH₂O), 3.68–3.62 (m, 1H, NCH₂O), 2.71 (d, J = 24.4 Hz, 3H, NCH₃), 2.55–2.36 (m, 2H, β-CH₂), ¹³C-NMR (101 MHz, Chloroform-D) δ 168.5, 168.4, 167.7, 153.6, 152.9, 135.5, 135.4, 133.6, 131.7, 128.3, 128.0, 127.8, 127.7, 122.7, 67.3, 67.0, 63.4, 63.1, 56.6, 56.4, 33.0, 28.2, 27.1, 27.0, 26.9.

2.3.4. Benzyl (R)-5-(2-aminoethyl)-3-methyl-4-oxoimidazolidine-1-carboxylate, (R)-CIM-C₂-NH₂

Benzyl (R)-5-(2-(1,3-dioxoisoindolin-2-yl)ethyl)-3-methyl-4-oxoimidazolidine-1-carboxylate (2.86 mmol) and hydrazine hydrate (3.09 mmol) were dissolved in methanol and stirred at room temperature for 18 h. The product was transferred to a separating funnel with ethyl acetate (100 mL) and extracted in 1 M hydrochloric acid (3 × 100 mL). The aqueous layer was adjusted to pH 11 using potassium hydroxide. The organic layer was extracted with chloroform (3 × 100 mL) and dried with sodium sulfate, then filtered and concentrated under vacuum. The residue was purified by flash chromatography (ULTRAPACK^® C18 column, Yamazen Corp.; Osaka, Japan) with an eluent of H₂O/methanol/28% ammonia (50/50/0.1)) to give (R)-CIM-C₂-NH₂ (1.87 mmol, 66%). m/z [M + H] 278.1490 (calculated for C₂₀H₁₉N₂O₆ = 278.1499). ¹H-NMR (400 MHz, Chloroform-D) δ 7.40–7.32 (m, 5H, PhH), 5.17 (q, J = 11.9 Hz, 2H, PhCH₂), 4.85 (dd, J = 20.6, 5.6 Hz, 1H, NCH₂O), 4.63 (dd, J = 12.2, 5.1 Hz, 1H, NCH₂O), 4.29–4.23 (m, 1H, α-CH), 2.92–2.64 (m, 5H, NCH₃ and γ-CH₂), 2.17–1.98 (m, 2H, β-CH₂), 1.32 (s, 2H, NH₂), ¹³C-NMR (101 MHz, Chloroform-D) δ 170.6, 153.9, 153.3, 135.8, 128.6, 128.4, 128.2, 67.6, 67.5, 63.8, 63.5, 56.9, 37.8, 37.5, 34.6, 34.3, 27.6.

2.4. Sample Preparation

Japanese sake was sampled from three different bottles of each brand. An aliquot of each sake sample was diluted 10-fold with H₂O. To 10 μL of the diluted solution, 10 μL of an IS (100 μM of d-LA¹³C₃ in H₂O) was added, followed by 10 μL of the derivatization reagent CIM-C₂-NH₂ [6] (25 mM) in the presence of condensing agents TPP (250 mM, 10 μL) and DPDS (250 mM, 10 μL) in CH₃CN. The resulting mixture was vortexed for 1 min and heated at 60 °C for 30 min. Subsequently, 50 μL of 0.05% formic acid in CH₃CN/H₂O (20/80) was added, vortexed for 1 min, and diluted 100-fold with the same solution. From the final solution, 5.0 μL was injected into an LC-MS/MS system as described in Section 2.5.

2.5. LC-MS/MS Parameters

The analytical column used for LC-MS/MS was a reversed-phase C18 column, InertSustain^® C18 (150 × 2.1 mm i.d., 3 µm; GL Sciences Ltd.; Tokyo, Japan) with a guard column (10 × 1.5 mm i.d., 3 µm). The mobile phases were 0.05% formic acid in H₂O (A) and 0.05% formic acid in CH₃CN (B). A time-programmed gradient elution was employed with a flow rate of 0.3 mL/min as follows: 0–30 min, B% = 15; 30.01–40 min, B% = 20; 40.01–75 min, B% = 30; 75.01–85 min, B% = 100; 85.01–95 min, B% = 15. The temperatures of the desolvation line and heat block were set at 250 and 400 °C, respectively. The flow rates of the nebulizing and drying gases were 3.0 and 15 L/min, respectively. The ion spray voltage and collision-induced dissociation (CID) gas pressure were 4.5 kV, and 230 kPa, respectively. Ions were detected in multiple reaction monitoring mode and quantified using MS/MS detection (positive ion mode). Other MS/MS parameters for each organic acid derivative labeled with CIM-C₂-NH₂ are shown in

Table 1. MS/MS parameters for each organic acid derivative labeled with CIM-C₂-NH₂.

Organic Acid	m/z		Q1 Prebias (V)	Collision Energy (CE)	Q3 Prebias (V)
	Precursor	Product
LA	349.9	91.15	−14	−28	−18
LA-IS	353.45	91.10	−17	−39	−18
MA	653.1	91.05	−32	−54	−17
SA	637.1	91.15	−24	−45	−18
CMA	667.1	91.05	−24	−48	−18

.

2.6. Linearity, Precision, and Accuracy

Linearity was evaluated by constructing calibration curves for each organic acid. Each calibration curve was constructed by plotting the peak area ratio of the IS to each organic acid against the concentration (n = 4). The concentration ranges of l-LA, d-LA, and MA were 6.25–800 and 12.5–800 μM, respectively, while those of SA and d-CMA were 6.25–650 and 1.25–100 μM, respectively. A 10-fold diluted sample of sake A with H₂O (10 μL) was spiked with 10 μL of H₂O (0 μM) or standard organic acid solution, and 10 μL of IS, then vigorously mixed. Next, 10 μL of the precolumn reagent CIM-C₂-NH₂ was added and treated in a manner similar to that described in Section 2.4 (n = 3). Precision was expressed as relative standard deviation (RSD;%), and accuracy was expressed as relative mean error (RME;%), as in our previous study [8].

2.7. Neutralization Titration

2.7.1. Actual Titration Volume

Sake samples (20 mL) were pipetted from each bottle into a conical beaker using a hole pipette, and then 20 mL of purified water was added and mixed. Subsequently, two drops of a pH indicator (1% phenolphthalein in ethanol solution) were added and mixed in the conical beaker.

A standard solution of 0.01 or 0.02 mol/L NaOH in H₂O (factor: 1.000) was used to titrate the sample using a 50 mL glass burette. The titration was stopped when the solution turned slightly pink, and the volume of NaOH solution used was regarded as the equivalence point.

2.7.2. Predicted Titration Volume

The moles of l-LA and d-LA in 20 mL of sake were calculated from their concentrations. Accordingly, the volume of 0.01 or 0.02 mol/L NaOH standard solution required for neutralization was predicted using Equation (1):

Volume of 0.01 or 0.02 mol/L NaOH standard solution (mL) = [(mole amount of l-LA and d-LA)/(0.01 or 0.02)] × 10³

(1)

For sakes A–E, 0.01 mol/L NaOH standard solution was used, while 0.02 mol/L NaOH standard solution was used only for sake F, due to its relatively high organic acid content.

Similarly, the mole amounts of SA, MA, and d-CMA in 20 mL of sake were calculated from their concentrations, and the required volume of NaOH standard solution for neutralization was estimated using Equation (2):

Volume of 0.01 or 0.02 mol/L NaOH standard solution (mL) = [(mole amount of SA, MA, and d-CMA)/(0.01 or 0.02)] × 2 × 10³

(2)

The sum of the volumes (mL) calculated from Equations (1) and (2) was regarded as the predicted volume of NaOH standard solution required for complete neutralization of the organic acids.

3. Results and Discussion

3.1. Synthesis of CIM-C₂-NH₂

As mentioned in Section 1, we have previously reported a precolumn derivatization reagent, CIM-C₂-NH₂, whose precursor was the chiral compound Cbz-l-Asp(OtBu)OH (Figure S2). This compound was converted into CIM-C₂-NH₂ via a five-step synthetic route including the Curtius rearrangement reaction [9,10]. However, the overall yield was quite low, necessitating the development of an alternative route. Therefore, in the current study, the synthetic route for CIM-C₂-NH₂ was modified, as shown in Figure 1, to address the issue of low yield. The precursor was the chiral compound (R)-4-amino-2-(((benzyloxy)carbonyl)amino)butanoic acid, whose γ-NH group was subsequently protected with anhydrous phthalic acid to produce phthalimide [11,12] in the first reaction step. This was followed by amidation of the carboxyl group with methylamine, cyclization with formaldehyde to construct imidazolidinone ring, and deprotection of phthalic acid by hydrazine. Each reaction step in Figure 1 proceeded satisfactorily. In the final purification step, flash chromatography under reversed-phase (C₁₈) conditions was effective in isolating CIM-C₂-NH₂ with minimal product loss. As a result, CIM-C₂-NH₂ was successfully synthesized via a four-step reaction, and the overall yield increased to 23%, compared with 0.7% obtained using the previous method.

3.2. Chromatographic Detection of Organic Acids in Sake

Organic acids present in sake were investigated using an LC-MS/MS following precolumn derivatization with the proposed reagent, CIM-C₂-NH₂ [6]. In this study, six commercial sake samples of different brands were selected. By using CIM-C₂-NH₂ derivatization followed by LC-MS/MS analysis, peaks corresponding to several organic acids, such as LA, SA, MA, and CMA derivatives, tagged with CIM-C₂-NH₂ were clearly detected in the sake samples (Figure 2). Regarding the absolute configuration of LA, both l-LA and d-LA were detected. Notably, most sake samples contained approximately equal amounts of d- and l-LA, with sake F containing a particularly high amount of d-LA. Previously, Kodama et al. also reported the presence of both d- and l-LA in sake using capillary electrophoresis with chiral 2-hydroxypropyl β-cyclodextrin as an additive in the electrolyte [4]. d-LA is generally present in fermentation products such as wine [6,13] and LA beverages [13]. Given that sake is one of the representative fermentation products in Japan [14], the detection of d-LA in sake is plausible.

CMA has recently gained attention as a metabolite of interest in human biological fluids such as urine, owing to reported alterations in its levels under disease conditions in recent metabolomics studies [15,16,17]. However, the configuration of CMA in human biological fluids remains unclear. In the current study, only d-CMA was detected in all sake samples examined. d-CMA is enzymatically produced from achiral pyruvate and acetyl coenzyme A by CMA synthase [18,19]. The detected d-CMA likely originated from yeast because d-CMA is produced during the beer production process [20].

In our previous study, we found that both the fruit and peel of three apple varieties contained only d-CMA; however, one study reports that l-CMA is also present in the fruit [21].

As a result of investigating six commercial sake products, we have determined that most yeast strains used for Japanese sake production predominantly produce d-CMA rather than l-CMA.

Collectively, these findings suggest that if CMA in human body fluids is derived entirely from exogenous sources, the configuration of CMA may primarily exist in the d-form. However, further studies are necessary to confirm this.

3.3. Organic Acid Concentrations

In the next experiment, calibration curves for each organic acid were constructed using standard solutions. As shown in Figure S3, the calibration curves exhibited linearity, with correlation coefficients in the range of 0.994 to 0.999. The obtained RSD and RME values were within 10% (Table S1), indicating acceptable accuracy and precision for quantifying organic acids in sake.

Using these calibration curves, the concentrations of organic acids in the sake samples analyzed in this study were determined and are summarized in Figure 3. The concentrations and composition of organic acids varied among sake samples A–F, although the SA concentration remained relatively consistent across all samples. As previously reported, the current results confirmed that LA, SA, and MA were present at high concentrations (mM order) in sake. In sakes A and E, the concentrations of d-LA and l-LA were nearly equal, whereas in sakes D and F, the d-LA concentrations were approximately 1.5- and 10-fold higher than those of l-LA, respectively. This finding suggests that yeasts capable of producing higher levels of d-LA than l-LA may have been predominantly used in the production process of sakes D and F.

The concentration of d-CMA was approximately 1/20 that of the other organic acids in sakes A–F. Although the presence of CMA in alcoholic beverages such as beer [19] and wine [6] has been reported, d-CMA in sake is presumed to have contributed minimally to the sour taste compared with LA, SA, and MA.

3.4. Acidic Components Determined by Neutralization Titration

Neutralization titration is typically used to estimate the total acidic components in sake [2]. As described above, previous studies have reported that the primary organic acids in sake include LA, SA, and MA [1], all of which were also detected in this study. To assess the contribution of these organic acids to sake acidity, neutralization titration was performed on sakes A–F using a standard NaOH solution (0.01 or 0.02 mol/L) and phenolphthalein as the pH indicator. The samples were titrated to the equivalence point, indicated by a slight shift to a pink color.

Figure 4 presents a correlation plot comparing the actual titration volumes with the predicted volumes calculated from the sum of organic acid (l-LA, d-LA, SA, MA, and d-CMA) concentrations determined by LC-MS/MS in commercial sakes A–F.

In general, the actual titration volumes were greater than the calculated predicted values for sakes A–F, suggesting the presence of additional acidic species. However, for sakes A–D, the actual titration volumes were in close agreement with the predicted values based on LC-MS/MS, implying that the acidic components responsible for sour taste in these sakes may be primarily attributed to l-LA, d-LA, SA, MA, and d-CMA.

In contrast, the actual titration volumes of sakes E and F were significantly higher than those predicted by LC-MS/MS, indicating the presence of other organic acids with significant acidity contributions.

A limitation of the present study is that the precolumn derivatization reagent, CIM-C₂-NH₂, selectively reacts with mono- and dicarboxylic acid, but not with tricarboxylic acids such as citric acid. Therefore, the presence and concentration of citric acid and other tricarboxylic acids in sake could not be determined. The identification of these tricarboxylic acids and other acids should be explored in future research.

In conclusion, for four of the six commercial sake samples, the titration data were consistent with the values determined by LC-MS/MS.

4. Conclusions

This study demonstrated that precolumn derivatization with CIM-C₂-NH₂, followed by LC-MS/MS analysis, was an effective method for the determination of l-LA, d-LA, SA, MA, and d-CMA in commercial sake. In sakes A–D, these organic acids primarily contribute to the acidic component responsible for sour taste. The developed LC-MS/MS method may also be applicable to the analysis of organic acids in other alcoholic beverages.

In sake samples E and F, the titrated volumes exceeded those predicted by LC-MS/MS. This indicates that organic acids apart from those identified above were present. The identities of these acids will be investigated in future research.