Gas chromatography–mass spectrometry-based quantitative method using tert-butyldimethylsilyl derivatization for plasma levels of free amino acids and related metabolites in Japanese Black cattle

2.1 Plasma of Japanese Black cattle

Plasma during the early fattening (12 months of age), middle fattening (18 and 22 months of age), and late fattening phases (26 and 28 months of age) in 20 Japanese Black cattle (two samples from each age) were taken. The plasma of each age group was mixed in the same proportion, dispensed, and stored at −80°C until analysis as the quality control (QC) sample. This animal experiment was performed at the Hyogo Prefectural Technology Center for Agriculture, Forestry and Fisheries. All procedures were approved by “The Guideline for the Care and Use of Experimental Animals in Hyogo Prefectural Institute of Agriculture, Forestry and Fisheries” (approval no.: H2019-02).

2.2 External and internal standards

The external standards used in this study are listed in Table S1. The internal standards for amino acids were selected according to the method by Makino et al. (2021). The internal standard for other metabolites was 2-isopropylmalic acid because it has been used in untargeted metabolomic analysis (Takemoto et al., 2017; Tomonaga et al., 2018; Ueda et al., 2019). These internal standards used in this study are listed in Table S2.

2.3 Amino acid and related metabolite standards analysis by GC/MS

First, we established the analytical conditions for amino acids and related metabolites using their external and internal standards, according to previous studies (Jiménez-Martín et al., 2012; Okahashi et al., 2019) with some modifications. A 250 μL of standard solution was combined with an equal volume of acetonitrile. The mixture was vortexed and centrifuged at 20,000 × g for 5 min at 4°C. A 400 μL of the solution was evaporated after freezing it at −80°C. For oximation, 50 μL of 40 mg/mL methoxyamine hydrochloride (MPBio, Tokyo, Japan) dissolved in pyridine was mixed with the sample, before being shaken at 800 rpm for 60 min at 30°C. Next, for TBDMS derivatization, 50 μL of N-tert-butyldimethylsilyl-N-methyltrifluoroacetamide (Thermo Scientific, Waltham, MA, USA) was added and incubated at 800 rpm for 60 min at 60°C. The supernatant was then subjected to GC/MS. GC/MS analysis was performed using a GCMS-QP2010 Ultra (Shimadzu, Kyoto, Japan). A DB-5 column (30 m × 0.25 mm, i.d.; film thickness 1.00 μm; Agilent, Tokyo, Japan) was used for GC. The GC column temperature was programmed to maintain an initial temperature of 100°C for 4 min, then increase to 320°C at a rate of 11°C/min, and finally remain at 320°C for 11 min, producing a total GC run time of 37 min. The inlet temperature was maintained at 280°C, and helium was used as the carrier gas at a constant flow rate of 39.0 cm per second. A sample of 1.0 μL was injected in splitless mode, and the mass conditions were as follows: ionization voltage, 70 eV; ion source temperature, 200°C; full scan mode in the m/z range 50–600 at an interval of 0.3 s. Chromatogram acquisition, mass spectral peak detection, and waveform processing were performed using Shimadzu GC/MS solution software (Shimadzu). Based on the mass spectrum, quantitative and qualitative ions for each metabolite were selected based on their specificity and analytical conditions for the specific ion monitoring (SIM) mode. In addition, the retention index was calculated based on the retention time of n-alkanes (hydrocarbon mixed solution; GL Sciences Inc., Tokyo, Japan) injected in the same run. This retention index is useful for identifying metabolites.

2.4 Confirmation of plasma free amino acid and its metabolite levels in Japanese Black cattle

To confirm the detection of free amino acids and their metabolites in beef cattle plasma, we checked the degree of chromatographic separation of the selected ions, the intensity ratio of quantitative and qualitative ions, and the linearity of the calibration curves for each metabolite. A 20-μL QC sample or a standard solution of amino acid and related metabolite were mixed with 180 μL of phosphate-buffered saline (PBS, FUJIFILM Wako Pure Chemical Corporation), 50 μL of internal standards mixture (stable isotope-labeled internal standards for amino acids and 2-isopropylmalic acid mixture for other metabolites; these internal standard concentrations are in Table S3), and 250-μL acetonitrile. This was vortexed before being centrifuged at 20,000 × g for 5 min at 4°C. The subsequent pre-treatment and GC/MS analytical conditions were the same as those for the SIM method described in the previous section. In addition, for the detected metabolites, the types of derivatization were annotated by referring to a previous study (Ohie et al., 2000), the NIST Chemistry WebBook (Linstrom & Mallard, 2001), and/or HMDB 5.0 (Wishart et al., 2022).

2.5 Intra-day repeatability, inter-day reproducibility of assay, and spike recovery rate

For the metabolites detected in plasma, the intra-day repeatability and inter-day reproducibility of the assay were investigated. The QC samples were analyzed thrice a day on three separate days. The pretreatment and GC/MS analytical conditions were the same as those described in the previous section. For quantification, the peak area of each ion was calculated and normalized to that of the internal standard. In this study, we determined that the acceptable relative standard deviation was less than 20%, based on a previous study on untargeted metabolomics using the GC/MS TMS derivatization method in Japanese Black cattle (Ueda et al., 2019).

Spike recovery rates were also investigated. Based on the quantified levels, the spiked metabolites were divided into three groups: low, medium, and high. For the analysis, a 20-μL plasma sample with 160 μL of PBS (FUJIFILM Wako Pure Chemical Corporation), 20 μL of the spiked mixture of standards, and 50 μL of internal standards mixture (stable isotope-labeled amino acids and 2-isopropylmalic acid mixture; Table S3) were used. The subsequent pre-treatment and GC/MS analytical conditions were the same as those used for the SIM method described in the previous paragraph. Acceptable recovery ratios in this study ranged from 80% to 120%.

2.6 Comparison with TMS derivatization method

To evaluate the analytical quality of the present method from a different perspective, a comparison was made with the TMS derivatization method, which has been used for untargeted metabolomics (Putri et al., 2022). The intra-day repeatabilities of the two derivatization methods were compared. The relationship between the analytical results of these two methods in Japanese Black cattle was investigated using 20 Japanese Black cattle plasma samples. The TMS derivatization method was based on the method described by Tomonaga et al. (2018) with minor modifications. A 50-μL plasma sample was mixed with 250 μL of a liquid mixture (MeOH:H₂O:CHCl₃, 2.5:1:1), after adding 5 μL of 1 mg/mL 2-isopropylmalic acid (Sigma-Aldrich, Tokyo, Japan) dissolved in distilled water, following which the solution was vortexed and shaken at 1200 rpm for 30 min at 37°C, before being centrifuged at 16,000 × g for 5 min at 4°C. A 200-μL volume of the supernatant was transferred to a clean tube containing 200 μL of distilled water. After being vortexed, the solution was centrifuged at 16,000 × g for 5 min at 4°C, and a 250-μL volume of the supernatant was transferred to a clean tube before being lyophilized using a freeze dryer. For oximation, 40 μL of 20 mg/mL methoxyamine hydrochloride (MPBio, Tokyo, Japan) dissolved in pyridine was mixed with the lyophilized sample, before being shaken at 1200 rpm for 90 min at 30°C. Next, 20 μL of N-methyl-N-trimethylsilyl-trifluoroacetamide (Thermo Scientific, Yokohama, Japan) was added for derivatization and incubated at 1200 rpm for 45 min at 37°C. The supernatant was then subjected to GC/MS. GC/MS analysis was performed using a GCMS-QP2010 Ultra (Shimadzu, Kyoto, Japan). The GC conditions were the same as those used for the TBDMS method. The mass conditions were as follows: ionization voltage, 70 eV; ion source temperature, 200°C; full scan mode in the m/z range 45–600 at an interval of 0.3 s. Chromatogram acquisition, mass spectral peak detection, and waveform processing were performed using Shimadzu GC/MS solution software (Shimadzu). The metabolites were identified using the Smart Metabolites Database (Shimadzu), which includes the retention index and qualitative and quantitative ions of each metabolite. For semi-quantification, the peak area of each quantified ion was calculated and normalized to the peak area of 2-isopropylmalic acid (internal standard). For the metabolites in TMS derivatization, the types of derivatization and the quantitative and qualitative ions used are shown in Table S4.

2.7 Statistical analysis

All the statistical analyses were performed using Microsoft MS-Excel (ver. 2306). Relationship between TMS and TBDMS derivatization data in each metabolite was judged by simple regression analysis. Statistical significance was set at p < 0.05.

3.1 Amino acid and related metabolite standards analysis by GC/MS

Using the TBDMS derivatization method, the retention time and mass spectra of 51 external standards and 25 internal standards were confirmed, and their quantitative and qualitative ions were determined (Table S5). In addition, based on the retention time of the n-alkanes analyzed in the same run, the retention index for each metabolite was calculated (Table S5).

3.2 Confirmation of plasma metabolites in Japanese Black cattle and linearity of calibration curve for each metabolite

Using the SIM mode of the TBDMS derivatization method, 34 metabolites, including 27 amino acids, one amine, three organic acids, three branched-chain keto acids, and 22 internal standards, were detected and quantified in bovine plasma. Their MS conditions were selected from Table S5 and shown in Table 1. The types of derivatization were annotated for these quantified metabolites (Table S6). The linearity of the calibration curves for these metabolites was acceptable (R² > 0.97; Table S7). On the other hand, as metabolites that could not be quantified, there are eight (norvaline, 4-aminobutyric acid, 3-hydroxyanthranilic acid, homocysteine, melatonin, spermidine, carnosine, and anserine) that were undetectable in bovine plasma, three (taurine, taurine-¹³C₂, O-phosphoserine) that could not be detected possibly due to overlapping peaks with the phosphorus of PBS as a solvent, and five (L-arginine, L-arginine-¹⁵N₄, 3-hydroxyisovaleric acid, L-cysteine, and L-citrulline) that could not make stable calibration curve.

3.3 Intra-day repeatability and inter-day reproducibility of assay

Intra-day repeatability and inter-day reproducibility were calculated for the 34 metabolites quantified in plasma (Table 2). All metabolites in bovine plasma were acceptable for both intra-day repeatability and inter-day reproducibility. In contrast, in the TMS derivatization method, of the same 34 metabolites, five (3-aminoisobutyric acid, pipecolic acid, serotonin, Nτ-methylhistidine, and 2-ketoisovaleric acid) were not detected, and five (asparagine, glutamine, histidine, kynurenine, and 2-ketoisocaproic acid) were not acceptable. The intra-day repeatability of the TBDMS method was better than that of the TMS method (Table 2).

Acceptable recovery ratios in this study ranged from 80% to 120%. Of the 34 metabolites, 24 were stable and acceptable at all spiked levels. Aspartic acid and Nτ-methylhistidine were partly acceptable. Eight were unacceptable for all spiked levels (glycine, β-alanine, 3-aminoisobutyric acid, 2-aminoadipic acid, 3-hydroxybutyric acid, ethylmalonic acid, glutaric acid, and pipecolic acid) (Table 3).

3.4 Relationship of analytical results between TMS and TBDMS derivatizations

The relationships between the analytical results of TMS and TBDMS derivatization were investigated using simple regression analysis (Table 4). Simple regression analysis was conducted on 29 of the 34 metabolites that could be compared. The relationships were significant but not high in 18 metabolites (0.50 < R² < 0.65: lysine, glycine, and glutamic acid; R² < 0.50; 2-aminoadipic acid, 2-aminobutyric acid, 3-hydroxybutyric acid, alanine, asparagine, ethylmalonic acid, glutamine, histidine, isoleucine, methionine, ornithine, proline, threonine, tyrosine, and β-alanine). No significant relationships were observed for the remaining 11 compounds (aspartic acid, glutaric acid, hydroxyproline, kynurenine, leucine, phenylalanine, serine, tryptophan, valine, 2-ketoisocaproic acid, and 2-keto-3-methylvaleric acid).