Comprehensive Profiling of Free Proteinogenic and Non-Proteinogenic Amino Acids in Common Legumes Using LC-QToF: Targeted and Non-Targeted Approaches

Abstract

Legumes, a dietary staple for centuries, have seen an influx of conventional and unconventional varieties to cater to human care conscious consumers. These legumes often undergo pretreatments like baking, soaking, or boiling to mitigate the presence of non-proteinogenic amino acids (NPAAs) and reduce associated health risks. The recent tara flour health scare, linked to the NPAA baikiain, emphasizes the need for robust analytical methods to ensure the safety and quality of both traditional and novel plant-based protein alternatives. While traditional techniques provide insights into protein and non-proteinogenic amino acid profiles, modern liquid chromatography-mass spectrometry (LC-MS) offers superior sensitivity and specificity for NPAA detection. This study employed an LC-QToF method with MS/MS analysis to comprehensively map the distribution of free NPAAs and proteinogenic amino acids (PAAs) in various legume samples. A total of 47 NPAAs and 20 PAAs were identified across the legume samples, with at least 7–14 NPAAs detected in each sample. Sulfur-containing NPAAs, such as S-methyl-L-cysteine, γ-glutamyl-S-methyl cysteine, and S-methyl homoglutathione, were predominantly found in Phaseolus and Vigna species. Cysteine and methionine were the sulfur-containing PAAs identified. Gel electrophoresis and soluble protein quantification were also conducted to understand legume protein composition holistically. This orthogonal approach provides a valuable tool for ensuring the overall quality of plant-based proteins and may aid in investigating food poisoning or outbreaks related to such products.

1. Introduction

Fabaceae, the legume family, is one of the largest plant families, with six sub-families cultivated globally for millennia as a staple food and significant livestock feed. Common examples include Phaseolus vulgaris L. (common bean), Lens culinaris L. (lentils), Cicer arietinum (chickpea), Cajanus cajan L. (pigeon pea), Vigna unguiculata (cow pea), Pisum sativum L. (green pea), as well as Glycine max L. (soy), Arachis hypogaea L. (peanut), and Medicago sativa (alfalfa). With their nitrogen-fixing ability, legumes are rich in alkaloids, amines, cyanogenic glucosides, and proteins consisting of common proteinogenic amino acids (PAAs) and non-proteinogenic amino acids (NPAAs). Other secondary metabolites such as phenolics, polyketides, and terpenoids are also part of legume chemical composition [1].

Legumes, a major source of non-animal protein worldwide, typically contain all 20 proteinogenic amino acids [2]. However, they also include NPAAs, which differ from PAAs in their inability to be incorporated into proteins. Certain NPAAs specific to particular legume species can serve as valuable chemical markers of product authenticity and the presence of adulterants [3]. Several NPAAs, such as L-Dopa, γ-aminobutyric acid (GABA), L-theanine, and taurine offer potential health benefits when consumed in appropriate amounts. However, it is crucial to note that the safety of these amino acids can vary depending on individual intake. Excessive consumption may lead to adverse or even toxic effects. Additionally, some NPAAs, like furosine, are used to assess the nutritional quality of thermally processed foods [4].

While NPAAs play a role in plant defense, they can be harmful or toxic to humans [5]. When inadvertently incorporated into proteins, they can cause misfolding and aggregation, leading to health problems. Examples of such harmful NPAAs include homoarginine in legumes, hypoglycin A in unripe ackee fruit, L-canavanine in alfalfa, and more recently, baikiain in tara flour [6,7]. Common household methods for reducing NPAAs in legumes include soaking, rinsing, and cooking. Fermentation and processing are often employed on a larger scale, but their effectiveness can vary depending on the specific NPAA and legume.

Tara flour, a plant protein linked to adverse health effects, was implicated in 2022 outbreak of adverse effects involving Daily Harvest’s French Lentil & Leek Crumbles [7,8]. Hundreds of consumers experienced gastrointestinal distress and liver problems. The FDA identified tara flour as the likely culprit, leading to product recalls [9]. Similar issues arose with Revive Superfoods smoothies, where tara protein flour was one of the newly introduced ingredients. Research suggests that the NPAA baikiain in tara flour may be responsible since studies in mice showed that it negatively impacts liver function [7]. Hence, in the event of an outbreak of adverse effects related to food, the rapid identification and characterization of NPAAs are essential for addressing the safety of plant-based protein products.

Recently, a shift towards plant-based proteins has emerged, with various options now available to consumers. Health benefits, sustainability concerns, and the desire for animal-free alternatives drive this trend. However, the safety of such protein alternatives is largely unknown, specifically the potential for population-specific adverse health effects. These effects can vary based on genetic, dietary, and environmental factors. For example, certain populations might be more susceptible to the toxic effects of specific NPAAs due to genetic predispositions or dietary habits. Additionally, some NPAAs can interfere with nutrient absorption and metabolism, leading to health issues that might be more pronounced in certain demographic groups [10].

Early studies of NPAAs relied upon one-dimensional and two-dimensional thin-layer chromatography (TLC) [11] and paper electrophoresis [12,13] and were limited by detection limits that exceeded the normally encountered trace levels of NPAAs. High-performance liquid chromatography (HPLC) [14,15,16], gas chromatography (GC) [17,18], ion exchange chromatography [19], and capillary electrophoresis (CE) [20] are among the methods currently used for the detection of NPAAs. HPLC methods often rely upon the derivatization of the NPAAs to form a chromophore for detection by photometric detectors. The addition of mass spectrometry as a detector for HPLC allows discrimination between analytes not only by retention time but also by their mass-to-charge ratio (m/z).

To comprehensively analyze the distribution of amino acids, specifically NPAAs in various legume species, a liquid chromatography-mass spectrometry (LC-MS) method is essential. Tandem mass spectrometry (MS/MS) analysis of NPAAs and common amino acid reference standards will enable their confident identification. In comparison to traditional amino acid profiling, the proposed method of analysis offers comprehensive qualitative profiling on an array of non-proteinogenic amino acids in addition to proteinogenic amino acids. In addition, the developed method provides extended sensitivity towards NPAA and PAA identification with minimal sample volume and time of analysis. Once validated, this method can be applied to a large cohort of legume seed samples to assess quality or safety issues. Moreover, orthogonal methodologies such as gel electrophoresis and protein quantifications of total proteins in legume samples are expected to provide critical data on these macromolecules.

2. Materials and Methods

2.1. Plant Materials

A total of 54 dry seed samples were obtained from various sources (Table 1, Figure 1). These samples included Phaseolus (P. vulgaris, P. lunatus, P. coccineus, and P. acutifolius), Lens culinaris, Pisum sativum, Cicer arietinum, Vigna (V. radiata, V. mungo, V. unguiculata, V. angularis, V. umbellata, and V. aconitifolia), Vicia faba, Cajanus cajan, Lathyrus sativus, Mucuna pruriens, Mimosa pudica, Medicago sativa, Caesalpinia (C. spinosa and C. bonducella), and Glycine max. In this study, whole or split seed samples were used. For each seed sample, two extraction methods for free PAAs and NPAAs were used: in the first method (refer to Method 1 below for additional details), the seed samples were soaked for 15–16 h at room temperature, and in the other extraction method (refer to Method 2 below), dried seeds were directly powdered and extracted with water. In both cases, the clear supernatant solution was analyzed in duplicate. The detailed classification of legume samples used in this study is depicted in Figure S1 based on their family, subfamily, genus, and species.

Table 1.

Code #, scientific name, common name, and source information of seeds from different Genera.

#	NCNPR Code #	Plant Name	Common Name	Plant Part	Source Information
1	25277	Phaseolus vulgaris	Anasazi bean	Seed	Missouri Botanical Garden, St. Louis, MO, USA
2	PV-01	Phaseolus vulgaris	Kidney bean	Seed	Local store
3	PV-03	Phaseolus vulgaris	Black turtle bean	Seed	Amazon.com
4	PV-04	Phaseolus vulgaris	Pinto bean	Seed	Local store
5	25309	Phaseolus vulgaris	Cranberry bean	Seed	Missouri Botanical Garden, St. Louis, MO, USA
6	25322	Phaseolus vulgaris	Cannellini bean	Seed
7	25283	Phaseolus coccineus	Black Runner/Ayocote Pinto	Seed
8	25281	Phaseolus coccineus	Scarlet Runner	Seed
9	25289	Phaseolus coccineus	Corona bean	Seed
10	25298	Phaseolus coccineus	Ayocote Morado bean	Seed
11	25292	Phaseolus acutifolius	White tepary bean	Seed
12	25290	Phaseolus acutifolius	Brown tepary bean	Seed
13	25302	Phaseolus acutifolius	Black tepary bean	Seed
14	25279	Phaseolus lunatus	Christmas speckled lima bean	Seed
15	25317	Phaseolus lunatus	Large lima bean	Seed
16	PL-01	Phaseolus lunatus	Lima bean	Seed	Local store
17	22446	Vigna unguiculata	Black-eyed pea or cowpea	Seed	Missouri Botanical Garden, St. Louis, MO, USA
18	VU-01	Vigna unguiculata		Seed	Medicinal Plant Garden, Oxford, MS
19	22454	Vigna mungo	Urdbean or black gram	Seed	Missouri Botanical Garden, St. Louis, MO, USA
20	1308	Vigna mungo		Seed	Local store
21	22465	Vigna aconitifolia	Mat bean, moth bean, matki, or dew bean	Seed	Missouri Botanical Garden, St. Louis, MO, USA
22	25282	Vigna angularis	Adzuki bean azuki bean, aduki bean, red bean, or red mung bean	Seed
23	VR-01	Vigna radiata	Green gram or moong bean	Seed	Local store
24	8222	Vigna radiata		Seed
25	5537	Vigna umbellata previously Phaseolus calcaratus	Ricebean or rice bean	Seed
26	22970	Vicia faba	Broad bean, fava bean, or faba bean	Seed	Missouri Botanical Garden, St. Louis, MO, USA
27	12757	Vicia faba		Seed
28	12550	Vicia faba		Seed
29	5471	Pisum sativum	Pea	Seed	Local store
30	22459	Pisum sativum		Seed	Missouri Botanical Garden, St. Louis, MO, USA
31	5587	Lens culinaris	Brown lentil	Seed	Local store
32	5591	Lens culinaris	Dark brown lentil	Seed
33	8220	Lens culinaris	Brown lentil	Seed	Commercial source
34	8764	Lens culinaris	Masoor lentil or red lentil	Seed	Missouri Botanical Garden, St. Louis, MO, USA
35	22467	Lens culinaris	Masoor lentil or red lentil	Seed
36	25315	Lens culinaris	Beluga: black, bead-like lentil	Seed
37	12553	Lens culinaris	Dark brown lentil	Seed
38	CA-1	Cicer arietinum	Chickpea or garbanzo bean	Seed	Local store
39	CA-01	Cicer arietinum	Green chickpea	Seed
40	CA-02	Cicer arietinum	Brown chickpea	Seed
41	10102	Mucuna pruriens	Monkey tamarind, velvet bean	Seed	Commercial source
42	24108	Mucuna pruriens		Seed	Missouri Botanical Garden, St. Louis, MO, USA
43	12565	Medicago sativa	Alfalfa	Seed
44	22379	Mimosa pudica	Lajwanti seeds, Chui Mui seeds	Seed powder *
45	25291	Glycine max	Golden soybean, soybean, or soya bean	Seed
46	5881	Glycine max		Seed
47	25299	Glycine max	Black soybean	Seed
48	25300	Lathyrus sativus	Grass pea, cicerchia, blue sweet pea, chickling pea, chickling vetch, Indian pea, white pea, and white vetch	Seed
49	25313	Cajanus cajan	Pigeon pea	Seed
50	7418	Cajanus cajan		Seed
51	22450	Cajanus cajan		Seed
52	15573	Caesalpinia spinosa	Tara (Quechua), Peruvian carob, or spiny holdback-	Seed
53	2112PR	Caesalpinia spinosa		Seed powder *	Commercial source
54	7488	Caesalpinia bonducella	Grey nicker, nicker bean, fever nut or knicker nut, or ivy gourd	Seed

* indicates the sample is in powder form.

Figure 1.

Representative seeds/beans from 22 species belonging to 13 genera.

2.2. Chemicals and Reagents

Methanol, acetonitrile, formic acid, and ultrapure water were purchased from Fisher Chemicals (ThermoFisher, Waltham, MA, USA). The solvents used for extraction and characterization were of analytical and LC-MS grade, respectively. An AccQ-Tag Ultra derivatization kit was purchased from Waters Corporation (Milford, MA, USA). A Pierce™ Bradford Protein Assay Kit was purchased from ThermoFisher Scientific (Waltham, MA, USA), and SDS-PAGE gels and the electrophoresis system were purchased from BioRad (Hercules, California, USA). The BLUeye Pre-stained Protein Ladder was purchased from Millipore Sigma (St. Louis, MO, USA).

2.2.1. Free Proteinogenic Amino Acids (PAAs)

All twenty amino acid standards (L-histidine, L-isoleucine, L-leucine, L-lysine, L-methionine, L-phenylalanine, L-threonine, L-tryptophan, L-valine, L-arginine, L-cysteine, L-glutamine, L-glycine, L-proline, L-tyrosine, L-alanine, L-aspartic acid, L-asparagine, L-glutamic acid, and L-serine) were purchased from Sigma-Aldrich (St. Louis, MO, USA).

2.2.2. Free Non-Proteinogenic Amino Acids (NPAAs)

β-N-oxalyl-L-α, β-diaminopropionic acid (β-ODAP) or BOAA, 3-hydroxyproline, S-methyl-L-cysteine, α-aminoadipic acid, L-3,4-dihydroxyphenylalanine (L-dopa), L-homoserine, L-canavanine, azetidine-2-carboylic acid (AZE), taurine, γ-aminobutyric acid (GABA), 2,3-diaminopropionic acid (DAPA), L-2,4-diaminobutyric acid (DABA), L-theanine, L-citrulline, L-norvaline, L-norleucine, 5-hydroxy-L-tryptophan, L-homoserine, N-(2-aminoethyl) glycine (AEG), 4-amino-L-phenylalanine, L-2-amino- adipic acid, O-acetyl-L-serine, S-methyl-L-cysteine, cystine, taurine, 4-hydroxy-L-isoleucine, γ-glutamylphenylalanine, 3-hydroxy norvaline, 2,6 diaminopimelic acid (DMP), 4-amino-3-hydroxy butyric acid, L-mimosine, N⁵-methyl-L-glutamine, L-2-amino butyric acid, and DL-hydroxylysine were purchased from Sigma-Aldrich (St. Louis, MO, USA). Hypoglycin A was purchased from Santa Cruz Biotech Inc. (Dallas, TX, USA), 4-hydroxypipecolic acid from BOC Sciences (Shirley, NY, USA), 4-carboxy-phenylalanine from Chem-Impex Intl Inc (Wood Dale, IL, USA), γ-glutamyl-leucine from AA Blocks (San Diego, CA, USA), and γ-glutamyl-tyrosine from MedChem Express (Monmouth Junction, NJ, USA). (S)-Willardiine was purchased from ChemBlock (Hayward, CA, USA), and homoarginine, tranexamic acid, L-ornithine, canaline, β-cyanoalanine, ibotenic acid, and N-acetyl-L-ornithine were purchased from Cayman Chemicals (Ann Arbor, MI, USA). Pipecolic acid and 5-hydroxy pipecolic acid were purchased from 1PlusChem LLC (San Diego, CA, USA). Baikiain (4,5-dehydropipecolic acid), baikiain isomer, 3-hydroxy-methyl-phenylalanine, and 3-hydroxy methyl tyrosine were isolated at the National Center for Natural Products Research (NCNPR), University of Mississippi, University, MS, USA. The purity of each of these compounds was greater than 95%. The identity and purity of these compounds were confirmed by chromatographic and high-resolution mass spectrometry (HRMS) data.

2.3. Seed Extract Preparation for Soluble Protein Analysis

Twenty-two legume seed samples were ground in a Retsch MM400 mixer mill using two metal balls. One gram of the finely powdered material was resuspended in 10 mL of 1X PBS (Phosphate Buffer Saline), pH 7.2, at the concentration of 1:10 (w/v) in 50 mL round bottom tubes. The samples were vortexed and mixed overnight at 4 °C on a tube rotator. The soluble protein suspensions were then centrifuged at 12,000 rpm for 30 min at 4 °C. The supernatant containing soluble total proteins was used as a crude extract and stored at −20 °C.

2.4. Bradford’s Assay

The soluble protein content in the crude extracts of different legume samples was quantified using Bradford’s method [21]. The Pierce™ Bradford Protein Assay Kit was used, and protein concentrations were measured using an Ultrospec 2100 pro spectrophotometer (Amersham, England, UK). The BSA (Bovine Serum Albumin) standard curve was used for soluble protein estimation.

2.5. SDS-PAGE Analysis

Sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was performed according to the method described by Laemmli (1970) [22]. The 2X reducing Laemmli buffer containing 0.5 M Tris buffer–HCl (pH 6.8), glycerol, 10% SDS, bromophenol blue, and 0.01 M β-mercaptoethanol was diluted with crude extract at a 1:2 concentration. The samples were then boiled for 10 min, cooled, and centrifuged at 3000× g for 1 min.

For soluble protein profile analysis, 12% Criterion gels (Bio-Rad, Hercules, CA, USA) were used. A 10 µL supernatant of each protein sample was loaded onto the gel, and proteins/ peptides were separated using a Midi vertical electrophoresis unit. The BLUeye Prestained Protein Ladder served as the molecular weight standard. Electrophoresis was performed at a constant voltage of 80 V for 30 min, followed by 120 V for 45 min. The gels were stained overnight with Coomassie Brilliant Blue R-250 on a shaker and then de-stained with 40% (v/v) ethanol and 10% (v/v) acetic acid until a clear background was obtained. Finally, the gels were visualized on a GelDoc imaging system (Bio-Rad) after de-staining.

2.6. Sample Preparation for Free Protein and Non-Proteinogenic Amino Acids

2.6.1. Preparation of Standard Solutions

A stock solution of the free proteinogenic amino acids, and the non-proteinogenic amino acids were prepared to obtain a 10 μg/mL final concentration for each analyte. The stock solution was stored at −20 °C. PAAs and NPAAs were dissolved in water.

2.6.2. Seed Material and Extraction Procedure

All the seed samples selected for analysis are used for human consumption except for M. sativa, which is cultivated as livestock feed.

• Method 1:

Cleaned seed samples of approximately 1–2 g was transferred into a 25 mL beaker, covered with twice as much water as the volume of the seeds, and soaked for about 15 h. After 15 h, the soaked water was centrifuged at 10,000 rpm, and the clear supernatant solution was used for analysis.

• Method 2:

About 1 g of powdered seed sample was weighed, then sonicated in 3.5 mL of water for 30 min, followed by centrifugation for 15 min at 10,000 rpm. The supernatant was transferred to a 10 mL volumetric flask. The extraction procedure was repeated three times, and the respective supernatants were combined. The final volume was adjusted to 10 mL with water, mixed thoroughly, and centrifuged.

Finally, the AccQ-Tag Derivatization Kit from Waters (Milford, MA, USA) was applied to analyze amino acids. The clear solutions from the above two methods were mixed separately with AccQ-Tag buffer to adjust pH and AccQ-Tag derivatization reagent and heated to 55 °C to expedite the reaction rate. Briefly, a 10 μL sample was mixed with 20 μL AccQ-tag Ultra derivatization reagent and with 70 μL borate buffer (pH 8.8), mixed well, and kept in a tightly sealed vial in a heating block for 10 min at 55 °C, and similarly, a derivatization blank was prepared without sample [23]. After cooling to room temperature, 2 μL of the sample was injected.

2.7. Analysis Using LC-QTOF-MS/MS

The liquid chromatographic system was an Agilent Series 1290 rapid resolution (Santa Clara, CA, USA) equipped with a binary pump, and an autosampler using an Agilent Poroshell 120 Aq C18 column (150 mm × 2.1 mm, 2.7 μm particle size). The mobile phase consisted of water with 0.1% formic acid (A) and acetonitrile with 0.1% formic acid (B) at a flow rate of 0.21 mL/min. The analysis was performed using the following gradient elution: 1% B to 25% B in 25 min, then to 100% B in the next 5 min. Each run was followed by a 3 min wash with 100% B and an equilibration period of 5 min with 99% A/1% B. Two microliters of the sample were injected. The column temperature was 45 °C.

The system was coupled to a 6530 Agilent Accurate-Mass Q-TOF LC/MS (Palo Alto, CA, USA) equipped with an ESI interface. Data acquisition (2 GHz) in the profile mode was governed via MassHunter Workstation software (Agilent Technologies). The spectra were acquired in the positive ionization mode, over a mass-to-charge (m/z) range from 50 to 1000. The detection window was set to 10 ppm. An ESI source with Jet Stream technology was used with the following parameters: drying gas (N₂) flow rate, 13 L/min; drying gas temperature, 300 °C; nebulizer pressure, 30 psig; sheath gas temperature, 300 °C; sheath gas flow, 11 L/min; capillary voltage, 3500 V; nozzle voltage, 0 V; skimmer, 60 V; Oct RF V, 750 V; and fragmentor voltage, 125 V. All the operations and the acquisition and analysis of data were controlled by Agilent MassHunter Acquisition Software Ver. A.10.1 and processed with MassHunter Qualitative Analysis software Ver. B.07.00. Accurate mass measurements were obtained employing ion correction techniques using reference masses at m/z 121.0509 (protonated purine) and 922.0098 [protonated hexakis (1H, 1H, 3H-tetrafluoropropoxy) phosphazine or HP-921] in the positive ion mode. The samples were analyzed in the all-ion MS-MS mode, where experiment 1 was carried out with a collision energy of zero and experiment 2 with a fixed collision energy of 45 eV. For the ESI-MS-MS CID experiments, precursor ions of interest were mass-selected by the quadrupole mass filter. The selected ions were then subjected to collision with nitrogen in a high-pressure collision cell. The collision energy was optimized to yield good product ion signals, which were subsequently analyzed with the ToF mass spectrometer. Analysis was performed in reflection mode with a resolving power of about 20,000 at m/z 922.0098. The instrument was set to an extended dynamic range (up to 10⁵ with lower resolving power). MS/MS spectra were recorded simultaneously at a rate of 2.0 spectra s⁻¹. An isolation window of 4.0 m/z was set for the quadrupole to filter selected precursor ions for MS/MS.

2.8. Statistical Analysis

GraphPad Prism v10.4.1 was used to display the distribution of each PAA and NPAA across all 22 legume variety samples. Data were analyzed by a two-way analysis of variance (ANOVA) followed by Tukey multiple comparison testing. Results are reported in terms of the significance of the p value (* p = 0.01–0.05, ** p = 0.001–0.009, *** p = 0.0001–0.0009, and **** p = <0.0001).

3. Results and Discussions

3.1. Soluble Protein Analysis

Total soluble seed proteins were extracted from 38 legume samples belonging to 20 distinct species within the Fabaceae family. An SDS-PAGE analysis was conducted to profile the protein polypeptides in each species, estimate their molecular weight, and assess their relative abundances. All the tested species, except C. bonducella, exhibited protein bands on the SDS-PAGE gel. The absence of bands in C. bonducella is likely attributed to the presence of tannins, which can interfere with protein extraction, as confirmed by a negative Bradford protein assay.

The remaining 19 species displayed various protein bands ranging from 5 to 250 kDa. Each species exhibited a unique protein profile characterized by band intensity and position variation, reflecting the heterogeneity of their seed proteins [24]. However, different varieties or samples of the same species, such as P. vulgaris, P. coccineus, V. radiata, V. unguiculata, L. culinaris, C. arietinum, and V. faba, exhibited comparable band patterns.

Globulins constitute legumes’ primary seed storage proteins, while albumins are present in smaller quantities [25]. The Bradford protein assays revealed a range of protein concentrations from 15 to 110 mg/g in the crude dry legume samples. Phaseolus species demonstrated the highest protein concentrations, while V. faba exhibited the lowest (Table 2, Figure 2). However, the small sample size and limited replication (1–3 replicates) for most samples raise concerns about the generalizability of these trends. Further investigation is needed to determine whether protein analysis varies based on seed variety, age, or geographic location.

Table 2.

The soluble protein concentrations in different bean samples were estimated using Bradford assay. The protein concentrations vary in different samples, and no protein was obtained from C. bonducella, in congruence with SDS-PAGE analysis.

#	NCNPR #	Species	Soluble Protein Concentration (mg/g Dry Weight)
1	25277	Phaseolus vulgaris	110
2	25283	Phaseolus coccineus	87
3	PL-01	Phaseolus lunatus	111
4	25292	Phaseolus acutifolius	69
5	VR-01	Vigna radiata	45
6	22454	Vigna mungo	39
7	22446	Vigna unguiculata	31
8	22465	Vigna aconitifolia	29
9	25282	Vigna angularis	23
10	5537	Vigna umbellata	71
11	25315	Lens culinaris	57
12	CA-1	Cicer arietinum	21
13	22970	Vicia faba	15
14	25300	Lathyrus sativus	30
15	25313	Cajanus cajan	26
16	10102	Mucuna puriens	98
17	25291	Glycine max	94
18	5471	Pisum sativum	52
19	2112PR	Caesalpinia spinosa	45
20	7488	Caesalpinia bonducella	-

Note: Medicago sativa and Mimosa pudica samples not analyzed due to presence of excess mucilage.

Figure 2.

SDS-PAGE profile of crude extracts of different pulse proteins from family Fabaceae. ‘M’ indicates protein molecular standards, and kDa indicates the molecular weight of kilodaltons.

3.2. Soaking

Seed extracts and powdered extracts showed similar patterns of free protein and non-proteinogenic amino acids, although the powdered samples generally displayed higher peak intensities. Soaking was found to reduce the levels of free protein and non-proteinogenic amino acids compared to the original seeds. About 1–2 g of the sample was added to a beaker containing 5 mL of HPLC grade water. The samples were soaked for approximately 15 h at room temperature then placed on filter paper to eliminate surface moisture. These soaked samples were weighed to ±0.0001 g to determine moisture gain. The samples were dried in an oven at 70 °C for 24 h, cooled to room temperature, and weighed to obtain a post-soaking weight.

Soaking is a common preparation method for legumes, facilitating the extraction and elimination of certain NPAAs as well as enabling a reduced cooking time [26]. Our analysis revealed that most legumes exhibit significant weight gain due to water absorption during soaking, indicating effective hydration. However, Caesalpinia species showed no change in weight. Despite this, most legumes experienced a notable reduction in dry weight after soaking, suggesting that soluble compounds were extracted. These findings confirm the efficacy of water extraction for obtaining PAAs and NPAAs from legumes and support the conventional practice of soaking them before cooking. Compared to the un-soaked seeds, the soaked seed samples were larger and around double in weight. The dried soaked seed weights were close to the original seed weight. The differences between the dry weight of the pre-soaked, soaked, and post-soaking dry seeds are depicted in Figure 3.

Figure 3.

The average weight of seeds before soaking, after soaking, and dried post-soaking of 21 samples after a soaking time of 15 h.

To determine the qualitative differences in the free proteinogenic and non-proteinogenic amino acid profiles from edible seeds, the seeds (Table 1) were subjected to whole seed soaking (Method 1 under Section 2.6.2), with the dried seeds then ground into fine powder and subjected to an exhaustive extraction method (Method 2 under Section 2.6.2) using water as the solvent. Upon comparison there is no qualitative difference observed in the proteinogenic and non-proteinogenic amino acids profiles. We considered the optimal peak intensities of analytes observed in the samples obtained through Method 1 in comparison to Method 2, for the further derivatization of samples using AccQ. Tag reagent was used to generate fluorescent labelled amino acids with greater sensitivity in terms of detection via mass spectrometric detectors coupled with liquid chromatography. Although grinding un-soaked seed samples produced adequate peak intensities, to maintain detection sensitivity, minimize saturation, and sustain repeatability in retention times, the soaked seed samples were chosen to represent qualitative profiles.

Non-proteinogenic amino acids have been studied to a lesser extent than PAAs but have been shown to have an important role in the quality and safety of food. The chemical and physical diversities of metabolites make them difficult to identify based solely on MS data. In the present study, the analyte identification in seed samples was achieved primarily through an accurate mass-based search for molecular formula followed by a search of published literature search engines (Google, PubMed) and databases (SciFinder). Reference compounds of these putative identifications were then subjected to a tandem MS/MS experiment side-by-side with the sample. By comparing the MS/MS spectra and retention times of the reference compounds with the molecules of interest in the sample, the identities of the molecules were confirmed.

A sensitive LC–QToF–MS system was employed to analyze the free form of PAAs and NPAAs that are expected to be at relatively low concentrations in legume samples. Direct analysis (un-derivatized) and derivatization approaches were used for the analysis of free amino acids. Direct analysis is rapid and simple but was found to be less sensitive due to matrix complexity and interferences. To increase the sensitivity and selectivity of amino acid analysis in a complex matrix, the derivatization approach was used to improve chromatographic separation. A readily available 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC) reagent was utilized to effectively derivatize the amino acids and enhance the selectivity and sensitivity. Both primary and secondary amines are reported to form stable urea adducts. Excess AQC reagent reacts with water to produce 6-aminoquinoline (AMQ), which is chromatographically resolved or further reacts with another AQC molecule to form a stable bis-aminoquinoline urea [23]. The resulting undesired side products do not often interfere with separating and identifying analytes of interest.

Upon the derivatization of both extracts with AQC, the derivatized amino acids (PAAs and NPAAs) were separated by liquid chromatography and detected by mass-specific detection using [M+H]⁺ ions in the positive ion mode with extractive ion monitoring (EIM). This method detected and confirmed mono-derivatized and bis-derivatized AQC-amino acid adducts. Most amino acids produced a mono-derivatized product with AQC, while cystine, lysine, 2, 3 diamino propionic acid, 2, 4-aminobutyric acid, L-ornithine, AEG, β-methylamino-L-alanine (L-BMAA), DMP, L-canaline, and hydroxylysine were bis-derivatized. A similar phenomenon has been reported in skin-related amino acids analysis [27]. The major advantage of this method is that no sample solvent evaporation is necessary. The samples were prepared at a 1:10 dilution in the reagents, which may have hindered the identification of low-abundance compounds.

3.3. Fragmentation Patterns PAAs and NPAAs Using LC-QToF-MS Analysis

This study examined fragmentation patterns of protonated amino acids using LC-QToF-MS/MS with collision-induced dissociation (CID) in the mass range of 50 to 1100 Da. The fragmentation patterns of PAAs and NPAAs were analyzed, excluding glycine (75 Da) and alanine (89 Da). For each derivatized amino acid, fragmentation at the 6-aminoquinoline carbonyl group produced a high-intensity common fragment ion at m/z 171.0553, along with 6-aminoquinoline (AMQ, m/z 145.076 C₉H₈N₂), and a fragment corresponding to the amino acid’s original mass. The observed fragment ions of protonated amino acids are summarized in Table 3.

Table 3.

Extracted high-resolution masses (m/z) of mono-derivatized and bis-derivatized charged ions of AQC-derivatized pure amino acids.

#	Compound Name	Chemical Formula	Rt (min)	Mass	[M+H]+	AQC Tag Mass	Protonated Fragment Ions (m/z) of Native Compounds
1	Histidine (H)	C6H9N3O2	7.16	155.0695	156.0768	326.1251 (326.1248) **	110.0711 [M+H−H2O−CO]+, 93.0446 [M+H−H2O−CO−NH3]+
2	Isoleucine (I)	C6H13NO2	21.87	131.0946	132.1019	302.1504 (302.1499)	86.0964 [M+H−H2O−CO]+, 69.0699 [M+H−H2O−CO−NH3]+
3	Leucine (L)	C6H13NO2	22.34	131.0946	132.1019	302.1503 (302.1499)	86.0961 [M+H−H2O−CO]+
4	Lysine (K)	C6H14N2O2	16.45	146.1055	147.1128	487.2091 * (487.2088)	130.0861 [M+H−NH3]+, 84.0806 [M+H−NH3−H2O−CO]+
5	Methionine (M)	C5H11NO2S	17.92	149.051	150.0583	320.1071 (320.1063)	133.0315 [M+H−NH3]+, 104.05275 [M+H−H2O−CO]+
6	Phenylalanine (F)	C9H11NO2	23.39	165.079	166.0863	336.1348 (336.1343)	120.0806 [M+H−H2O−CO]+, 103.0541 [M+H−H2O−CO−NH3]+, 93.0698, 91.0541
7	Threonine (T)	C4H9NO3	11.33	119.0582	120.0655	290.1141 (290.1135)	102.0549 [M+H−H2O]+, 74.0600 [M+H−H2O−CO]+
8	Tryptophan (W)	C11H12N2O2	24.35	204.0899	205.0972	375.1460 (375.1452)	188.0703 [M+H−NH3]+, 170.05975 [M+H−NH3−H2O]+, 146.0597 [M+H−CH2CO]+, 143.0727 [M+H−NH3−H2O−CO2]+, 118.0649 [M+H−CH2CO−CO]+, 91.0541 [M+H−CH2CO−CO−HCN]+
9	Valine (V)	C5H11NO2	18.15	117.079	118.0863	288.1349 (288.1343)	72.0808 [M+H−H2O−CO]+, 55.0543 [M+H−H2O−CO−NH3]+
10	Arginine (R)	C6H14N4O2	8.39	174.1117	175.119	345.1679 (345.167)	116.0704 [M+H−CH5N3 (guanidine group)]+ 70.0655 [M+H−CH5N3−H2O−CO]+
11	Cysteine (C)	C3H7NO2S	13.88	121.0197	122.027	292.0754 (292.0750)	88.0392 [M+H−H2S]+(minor), 76.0215 [M+H−H2O−CO]+ 58.9952 [M+H−H2O−CO−NH3]+
12	Glutamine (Q)	C5H10N2O3	9.02	146.0691	147.0764	317.1249 (317.1244)	130.0497 [M+H−NH3]+, 101.0549 [M+H−H2O−CO]+, 84.0443 [M+H−NH3−H2O−CO]+, 56.0497 [M+H−NH3−H2O−2CO]+
13	Glycine (G)	C2H5NO2	9.61	75.032	76.0393	246.0875 (246.0873)	−
14	Proline (P)	C5H9NO2	13.36	115.0633	116.0706	286.1191 (286.1186)	70.0651 [M+H−H2O−CO]+
15	Tyrosine (Y)	C9H11NO3	17.61	181.0739	182.0812	352.1300 (352.1292)	165.0543 [M+H−NH3]+, 147.0439 [M+H−NH3−H2O]+, 136.0754 [M+H−H2O−CO]+, 123.0439 [M+H−NH3−H2O−CH2CO]+, 119.0491 [M+H−H2O−CO−NH3]+, 95.0490 [M+H−NH3−H2O−CH2CO−CO]+
16	Alanine (A)	C3H7NO2	12.22	89.0477	90.055	260.1036 (260.1030)	−
17	Aspartic acid (D)	C4H7NO4	10.29	133.0375	134.0448	304.0928 (304.0928)	116.0340 [M+H−H2O]+, 88.0391 [M+H−H2O−CO]+, 74.0236 [M+H−H2O−CH2CO]+
18	Asparagine (N)	C4H8N2O3	8.11	132.0535	133.0608	303.1093 (303.1088)	87.0552 [M+H−H2O−CO]+, 70.0287 [M+H−H2O−CO−NH3]+
19	Glutamic acid (E)	C5H9NO4	10.88	147.0532	148.0604	318.1089 (318.1084)	130.0497 [M+H−H2O]+, 102.0548 [M+H−H2O−CO]+, 84.0443 [M+H−2H2O−CO]+, 56.0497 [M+H−2H2O−2CO]+
20	Serine (S)	C3H7NO3	9.03	105.0426	106.0499	276.0983 (276.0979)	88.0393 [M+H−H2O]+, 70.0287 [M+H−2H2O]+
21	Azetidine-2-carboxylic acid (Aze)	C4H7NO2	10.97	101.0477	102.055	272.1035 (272.103)	84.0449 [M+H−H2O]+, 56.0500 [M+H−H2O−CO]+
22	Hypoglycin A	C7H11NO2	21.59	141.0790	142.0863	312.1348 (312.1343)	124.0762 [M+H−H2O]+, 96.0813 [M+H−H2O−CO]+, 74.0240, 67.0548
23	γ-Aminobutyric acid (GABA)	C4H9NO2	12.55	103.0633	104.0706	274.1192 (274.1186)	86.0601 [M+H−H2O]+, 87.0440 [M+H−NH3]+
24	L-3,4-Dihydroxyphenylalanine (L-DOPA)	C9H11NO4	15.68	197.0688	198.0761	368.1250 (368.1241)	181.0502 [M+H−NH3]+, 152.0709 [M+H−H2O−CO]+, 163.0399 [M+H−H2O−NH3]+, 135 [M+H−H2O−NH3−CO]+, 107.0492 [[M+H−H2O−NH3−2CO]+
25	L-Canavanine	C5H12N4O3	8.11	176.0909	177.0982	347.1465 (347.1462)	160.0726 [M+H−NH3]+, 118.0850, 102.0513
26	5-Hydroxy-pipecolic acid	C6H11NO3	11.86	145.0739	146.0812	316.1296 (316.1292)	128.0706 [M+H−H2O]+, 100.0759[M+H−H2O−CO]+, 82.0654 [M+H−2H2O−CO]+, 55.0547
27	Baikiain isomer	C6H9NO2	16.87	127.0633	128.0706	298.1194 (298.1186)	110.0595[M+H−H2O]+, 82.0658[M+H−H2O−CO]+, 80.0496, 74.0243, 65.0387, 55.0542
28	Baikiain	C6H9NO2	17.21	127.0633	128.0706	298.1192 (298.1186)	110.0595[M+H−H2O]+, 82.0655[M+H−H2O−CO]+, 80.0491, 74.0242, 65.0381, 55.054
29	Pipecolic acid	C6H11NO2	17.79	129.0790	130.0863	300.1345 (300.1343)	112.0756 [M+H−H2O]+, 84.0811(C4H6NO) [M+H−H2O−CO]+, 56.0502 (C2H2NO)
30	3-Hydroxymethyl tyrosine	C10H13NO4	16.17	211.0845	212.0917	382.1402 (382.1397)	194.0810 [M+H−H2O]+, 177.0525 [M+H−H2O−NH3]+, 148.0739, 91.0528, 77.0372, 55.0168
31	3-Hydroxymethyl phenylalanine	C10H13NO3	18.59	195.0895	196.0968	366.1452 (366.1448)	178.0847 [M+H−H2O]+, 103.0536 [M+H−NH3−H2O−CO]+, 91.0543, 77.0389
32	2,3 Diaminopropionic acid (DAPA)	C3H8N2O2	13.33	104.0586	105.0659	275.1151 (275.1139) 445.1622 * (445.1619)	88.0394 [M+H−NH3]+, 87.0553 [M−H−H2O]+
33	2,4-Diaminobutyric acid (DABA)	C4H10N2O2	13.66	118.0742	119.0815	289.1312 (289.1295) 459.1783 * (459.1775)	101.0714 [M+H−H2O]+, 102.0586 [M+H−NH3]+
34	Homoarginine	C7H16N4O2	10.1	188.1273	189.1346	359.1836 (359.1826)	172.1089 [M+H−NH3]+, 144.1138 [M+H−NH3−CO]+, 84.0801, 67.0542, 56.0499, 51.0243
35	4-Aminobenzoic acid (PABA)	C7H7NO2	20.4	137.0477	138.055	308.1035 (308.1030)	120.0445 [M+H−H2O]+, 94.0660, 92.0494 [M+H−H2O−CO]+, 75.0241 [M+H−H2O−CO−NH3]+, 77.0310, 65.0393
36	Theanine	C7H14N2O3	12.4	174.1004	175.1077	345.1555 (345.1557)	158.0826 [M+H−NH3]+, 130.0499 [M+H−NH2CH2CH3]+, 84.0445, 56.0485
37	L-Citrulline	C6H13N3O3	10.4	175.0957	176.103	346.1513 (346.1510)	159.0770 [M+H−NH3]+, 113.0706 [M+H−H2O−CO−NH3]+, 70.0656
38	L-Norvaline	C5H11NO2	18.8	117.0790	118.0863	288.1346 (288.1343)	110.0285 [M+H−H2O]+, 72.0799, 55.0538
39	Tranexamic acid	C8H15NO2	19.3	157.1103	158.1176	328.1657 (328.1656)	141.0916 [M+H−NH3]+, 140.1075 [M+H−H2O]+, 123.0810,112.1126, 95.0861, 77.0389, 67.0541, 55.0540
40	L-Norleucine	C6H13NO2	23.1	131.0946	132.1019	302.1499 (302.1499)	114.0916 [M+H−H2O]+, 86.0969, 77.0389, 69.0696
41	5-Hydroxy-L-tryptophan	C11H12N2O3	17.1	220.0848	221.0921	391.1403 (391.1401)	204.0668 [M+H−NH3]+, 186.0560, 162.0563, 116.0502, 107.0501, 77.0384
42	L-Homoserine	C4H9NO3	9.3	119.0582	120.0655	290.1138 (290.1135)	102.0534 [M+H−H2O]+, 74.0581, 56.0479
43	L-Ornithine	C5H12N2O2	14.8	132.0899	133.0972	473.1938 * (473.1932)	116.0705 [M+H−NH3]+, 115.0865 [M+H−H2O]+, 70.0651 [M+H−NH3−H2O−CO]+
44	trans-4-Hydroxy-L-proline	C5H9NO3	7.2	131.0582	132.0655	302.1136 (302.1135)	114.0563 [M+H−H2O]+, 86.0600, 68.0500, 58.0652
45	N-(2-Aminoethyl) glycine (AEG))	C4H10N2O2	13.5	118.0742	119.0815	459.1777 * (459.1775)	102.0533 [M+H−H2O]+, 56.0477
46	β-Methylamino-L-alanine (L-BMAA)	C4H10N2O2	13.3	118.0742	119.0815	459.1776 * (459.1775)	102.0554 [M+H−H2O]+, 58.0646
47	L-Canaline	C4H10N2O3	14.1	134.0691	135.0764	475.1727 * (475.1724)	118.0508 [M+H−NH3]+, 89.0709, 72.0444, 56.0495
48	β-ODAP	C5H8N2O5	8.5	176.0433	177.0506	347.0989 (347.0986)	160.0254 [M+H−NH3]+, 133.0607, 131.0448, 116.0341, 105.0655, 87.0551, 67.0289, 59.0602, 57.0444
49	4-Amino-L-phenylalanine	C9H12N2O2	10.7	180.0899	181.0972	351.1455 (351.1452)	164.0677 [M+H−NH3]+, 135.0886, 118.0612, 106.0620, 94.0620, 91.0504, 77.0350
50	β-Cyanoalanine	C4H6N2O2	10.5	114.0429	115.0502	285.0986 (285.0982)	89.0471 [M+H−CN]+, 74.0237, 69.0447 [M+H−H2O−CO]+ (CN=26.0031)
51	Ibotenic acid	C5H6N2O4	10.2	158.0328	159.04	329.0885 (329.0880)	113.0345 [M+H−H2O−CO]+, 94.0406, 67.0291, 68.0139, 56.0132, 52.0181
52	Muscinol	C4H6N2O2	12.05	114.0429	115.0502	285.0986 (285.0982)	98.0237 [M+H−NH3]+, 67.0216
53	L-2-Amino-adipic acid	C6H11NO4	13.25	161.0688	162.0761	332.1245 (332.1241)	144.0658 [M+H−H2O]+, 127.0411 [M+H−H2O−NH3]+, 116.0705 [M+H−H2O−CO]+, 55.0173
54	O-Acetyl-L-Serine	C5H9NO4	13.12	147.0532	148.0604	318.1087 (318.1084)	130.0498 [M+H−H2O]+, 131.0347 [M+H−NH3]+, 106.0457, 88.0392, 70.0295, 60.0450
55	S-Methyl-L-cysteine	C4H9NO2S	15.35	135.0354	136.0427	306.0908 (306.0907)	119.0124 [M+H−NH3]+, 77.0040, 73.0232
56	Cystine (L-cysteine derivative and a non-proteinogenic L-alpha-amino acid)	C6H12N2O4S2	16.1	240.0238	241.0311	581.1271 * (581.1271)	224.0046 [M+H−NH3]+, 205.9940 [M+H−NH3−H2O]+, 195.0256 [M+H−H2O−CO]+, 177.9991 [M+H−H2O−CO−NH3]+, 151.9835 [M+H−C3H7NO2]+, 122.027 [M+H−C3H5NO2S], 74.0236 [M+H−C3H7NO2−CH2S2]+
57	4-Carboxy-phenylalanine	C10H11NO4	18.83	209.0688	210.0761	380.1241 (380.1241)	164.0701, 146.0596, 103.0544, 91.0542, 77.0389
58	Taurine	C2H7NO3S	8.75	125.0147	126.0219	296.0702 (296.0699)	80.9645 [SO3]
59	4-Hydroxy-L-isoleucine	C6H13NO3	15.44	147.0895	148.0968	318.1456 (318.1448)	132.1015, 86.0964 [M+H−H2O−CO]+, 69.0699 [M+H−H2O−CO−NH3]+
60	γ-Glutamyl phenylalanine	C14H18N2O5	20.45	294.1216	295.1288	465.1770 (465.1768)	166.0852, 120.0797 [M+H−H2O−CO]+, 103.0535 [M+H−H2O−CO−NH3]+, 93.0694, 91.0537
61	3-Hydroxy norvaline	C5H11NO3	14.6	133.0739	134.0812	304.1299 (304.1292)	116.0700, 88,0758, 70.0651
62	2,6 Diaminopimelic acid (DAP)	C7H14N2O4	12.99	190.0954	191.1026	361.1510 (531.1986)	128.0731, 82.0671
63	4-Amino-3-hydroxy butyric acid	C4H9NO3	10.17	119.0582	120.0655	290.1141 (290.1135)	102.0541, 84.0440
64	N-Acetyl-L-ornithine	C7H14N2O3	12.19	174.1004	175.1077	345.1560 (345.1557)	158.0819, 157.0967, 133.0970, 115.0862, 116.0735, 70.0651
65	γ-Glutamyl-leucine	C11H20N2O5	19.32	260.1372	261.1445	431.1931 (431.1925)	244.1180, 198.1126, 132.1017, 130.0494, 86.0958, 84.0442, 56.0494
66	γ-Glutamyl-tyrosine	C14H18N2O6	15.28	310.1165	311.1238	481.1718 (481.1718)	294.0975, 248.0915, 202.0855, 182.0804, 165.0539, 147.0429, 136.0750, 130.0495, 123.0426, 119.0488, 95.0487, 91.0529, 84.0435
67	L-Mimosine	C8H10N2O4	8.98	198.0641	199.0713	369.1198 (369.1193)	112.0395 (Dihydroxypyridine), 94.0304, 79.0218
68	N5-Methylglutamine	C6H12N2O3	10.63	160.0848	161.0921	331.1405 (331.1401)	144.0652 [M+H−NH3], 130.0502 [M+H−CH3NH2], 115.0869 [M+H−H2O−CO], 84.0448 [M+H−CH3NH2−H2O−CO], 56.0497 [M+H−CH3NH2−H2O−2CO]
69	(S)-Willardiine	C7H9N3O4	9.56	199.0593	200.0666	370.1148 (370.1146)	191.1031, 175.1081, 161.0920, 154.0617, 144.0654, 130.0500, 115.0869, 84.0448
70	4-Hydroxy-pipecolic acid	C6H11NO3	9.51	145.0739	146.0812	316.1297 (316.1292)	128.0701 [M+H−H2O]+, 100.0755[M+H−H2O−CO]+, 82.0654 [M+H−2H2O−CO]+, 56.0508, 55.0551
71	DL-5-Hydroxylysine	C6H14N2O3	12.2	162.1004	163.1077	503.2038 * (503.2037)	145.0969 [M+H−H2O]+, 128.0702 [M+H−H2O−NH3]+, 100.0758 [M+H−H2O−NH3−CO]+, 82.0652, 67.0439, 56.0494, 55.0542
72	α-Aminobutyric acid (AABA)	C4H9NO2	14.7	103.0633	104.0706	274.1187 (274.1186)	86.0590 [M+H−H2O]+
73	β-Aminobutyric acid (BABA)	C4H9NO2	12.53	103.0633	104.0706	274.1185 (274.1186)	58.0645 [M+H−H2O−CO]+
74	β-Aminoisobutyric acid (BIABA)	C4H9NO2	13.56	103.0633	104.0706	274.1180 (274.1186)	58.0639 [M+H−H2O−CO]+
75	Tryptamine	C10H12N2	27.1	160.1000	161.1073	331.1558 (331.1553)	144.0814 [M+H−NH3]+ (C10H10N), 127.0546 [M+H−2NH3]+ (C10H8), 117.0689 (C7H5N2), 115.0580 (C7H3N2), 91.0580 (C5H3N2)
76	PEA (phenethylamine)	C8H11N	26.2	121.0891	122.0964	292.1452 (292.1444)	105.0706 [M+H−NH3]+, 77.0377, 51.0224

Note: * = double derivatization; ** theoretical accurate mass; and in all samples analyzed, two common fragment ions detected were 6-aminoquinoline carbonyl group, m/z 171.0558 [M + H]⁺ C₁₀H₇N₂O and 6-aminoquinoline (AMQ), m/z 145.076 [M + H]⁺ C₉H₉N₂.

Derivatized amino acids showed predominantly the fragment ions (m/z 171.0553, 145.0760) associated with the derivatization agent while also showing ions similar to those seen with underivatized or native amino acid fragment ions, i.e., fragment ions of the amino acid backbone (Table 3). Amino acids containing sulfur, such as methionine and cysteine, exhibited losses of ammonia (17 Da, NH₃), water (18 Da, H₂O), carbon monoxide (CO), ethanimine (C₂H₅N), and methyl thiol (CH₃SH). Branched-chain amino acids or hydrophobic compounds like leucine, isoleucine, and valine showed sequential losses of H₂O, CO, and NH₃. Similarly, protonated aromatic amino acids, such as tryptophan, phenylalanine, and tyrosine, demonstrated losses of CH₂CO, CO₂, NH₃, H2O+CO, HCN, CH₃, H₂O, and CO. Polar uncharged amino acids, including glutamine, asparagine, threonine, and serine, showed losses of NH₃, H₂O+CO, and H₂O. The observed fragment losses for the negatively charged side chains, such as those in glutamic and aspartic acid, included H₂O and H₂O+CO. Positively charged amino acids, including histidine, arginine, lysine, and ornithine, showed losses of H₂O+CO, NH₃, CO₂, or combinations thereof. While most amino acids fragmented uniquely, some structurally similar amino acids produced common fragment ions [28,29]. Similar patterns of fragmentation were observed for NPAAs (Table 3). Extracted ion chromatograms (EIC) for each analyte with mass spectrometry information in MS and MS/MS modes for free PAAs and NPAAs are presented in Figure S2.

3.4. Proteinogenic Amino Acids (PAAs) Content

The 20 free proteinogenic amino acids were detected in most seed and seed powder samples (Table 4 and Figure 4). The PAAs mostly found in these samples were leucine, lysine, alanine, aspartic acid, and glutamic acid. The analysis of P. sativum, L. culinaris, C. arietinum, and V. faba analysis revealed lysine, proline, leucine, and arginine as major common amino acids with lower levels of methionine, cysteine, histidine, and tryptophan. Methionine and tryptophan were identified at minimal levels in lentils.

Table 4.

Screening of free proteinogenic and non-proteinogenic amino acids in 22 distinct legume water-soaked samples.

#	Compound Name	RT	AQC Tag Mass	1	2	3	4	5	6	7	8	9	10	11	12	13 *	14	15	16	17	18	19 *	20	21 *	22 *	Detected Samples/ Total Number
Essential Free Protein Amino Acids
1	Histidine	7.1	326.1248	L	L	L	L	L	M	L	L	L	L	L	L	M	L	M	L	L	L	M	L	L	-	21/22
2	Isoleucine	21.87	302.1499	M	M	M	M	M	H	H	M	M	M	M	M	M	M	M	M	M	M	M	M	H	H	22/22
3	Leucine	22.34	302.1499	H	M	H	H	H	H	H	H	M	M	H	H	M	H	H	H	H	H	M	H	H	H	22/22
4	Lysine	16.43	487.2088	H	L	M	H	H	H	H	H	M	M	H	H	L	H	H	H	H	H	M	M	H	-	21/22
5	Methionine	17.91	320.1063	L	L	L	L	L	L	L	L	L	L	L	L	L	L	L	L	L	L	L	L	L	L	22/22
6	Phenylalanine	23.4	336.1343	M	M	M	H	M	H	H	H	M	M	H	M	M	L	M	M	M	H	M	M	M	M	22/22
7	Threonine	11.24	290.1135	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	22/22
8	Tryptophan	24.37	375.1452	H	L	M	L	M	L	L	L	L	L	L	L	M	M	L	M	L	L	L	M	L	L	22/22
9	Valine	18.11	288.1343	H	H	H	M	M	H	H	M	L	M	H	M	H	M	M	M	H	M	M	M	M	M	22/22
Conditionally Essential Free Protein Amino Acids
10	Arginine	8.4	345.167	M	M	M	H	H	H	H	H	H	H	M	H	H	H	H	H	M	M	M	H	H	L	22/22
11	Cysteine	13.90	292.075	L	L	L	L	L	L	L	L	L	L	L	L	L	L	L	L	-	L	L	L	-	-	19/22
12	Glutamine	9.01	317.1244	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	H	M	M	-	21/22
13	Glycine	9.59	246.0873	M	H	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	H	22/22
14	Proline	13.34	286.1186	M	H	M	M	M	M	M	M	H	H	M	H	H	H	M	M	H	M	H	H	M	M	22/22
15	Tyrosine	17.61	352.1292	L	M	M	M	L	M	M	M	L	L	L	M	M	M	M	M	M	M	M	M	H	M	22/22
Non-Essential Free Protein Amino Acids
16	Alanine	12.21	260.103	H	M	H	M	M	M	M	M	H	H	M	H	H	M	M	M	M	M	H	M	M	M	22/22
17	Aspartic acid	10.27	304.0928	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	22/22
18	Asparagine	8.10	303.1088	M	M	M	M	M	M	M	M	M	M	M	M	H	M	M	M	M	M	L	M	M	-	21/22
19	Glutamic acid	10.87	318.1084	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	22/22
20	Serine	9.01	276.0979	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	M	H	M	M	M	M	M	22/22
Non-Proteinogenic Amino Acids (NPAAs)
21	Pipecolic acid	17.75	300.1343	H	H	H	H	-	L	M	H	T	L	H	H	M	-	L	L	L	L	T	T	-	-	18/22
22	γ-Glutamyl-leucine	19.33	431.1925	M	H	M	L	L	M	M	H	H	L	L	M	-	L	T	-	L	-	L	-	-	-	16/22
23	γ-Acetyl-L-ornithine	12.19	345.1557	L	L	L	L	L	M	L	L	M	T	L	M	M	M	-	-	L	-	-	-	-	-	15/22
24	γ-Aminobutyric acid (GABA)	12.55	274.1192	H	H	H	H	H	H	H	H	H	H	H	H	L	H	H	H	M	H	L	H	M	T	22/22
25	5-Hydroxynorvaline	11.4	304.1292	M	M	L	L	-	-	-	-	-	-	-	L	-	-	-	-	-	M	-	-	-	-	6/22
26	γ-Glutamyl-phenyl-L-alanine	20.45	465.1768	L	L	L	L	L	L	L	M	M	L	T	H	-	L	H	-	-	H	L	M	L	L	19/22
27	Homoserine	9.3	290.1135	M	L	L	H	H	-	-	-	-	-	-	-	T	H	H	H	M	M	-	-	-	-	11/22
28	α-Aminoadipic acid	13.2	332.1241	L	L	L	L	M	M	L	L	L	L	L	M	L	L	M	M	L	L	L	L	L	H	22/22
29	N-Methylglutamine	10.63	331.1401	-	-	-	-	M	-	-	-	-	-	-	-	-	M	-	-	-	M	-	-	-	-	3/22
30	α-Amino-β-acetylaminopropionic acid	9.8	317.1244	-	-	-	-	M	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	1/22
31	β-Aminoisobutyric acid (BAIBA)	13.58	274.1186	-	-	-	-	-	-	-	-	-	-	-	M	-	-	-	-	-	-	-	-	-	-	1/22
32	α-Aminobutyric acid (Homoalanine) (AABA)	14.87	274.1186	T	T	T	T	-	-	-	-	-	-	-	T	-	-	-	-	-	-	-	-	-	-	5/22
33	γ-Glutamyl-tyrosine	15.28	481.1718	-	-	-	-	-	-	-	-	-	-	-	M	-	-	H	T	-	H	-	M	-	-	5/22
34	Baikiain	17.18	298.1186	-	-	-	-	-	-	-	-	-	-	-	-	H	-	-	-	-	-	-	-	-	-	1/22
35	Baikiain isomer	16.89	298.1186	-	-	-	-	-	-	-	-	-	-	-	-	M	-	-	-	-	-	-	-	-	-	1/22
36	Hydroxy-L-proline	7.2	302.1135	-	-	-	-	-	-	-	-	-	-	-	-	T	-	-	-	-	-	-	T	-	-	2/22
37	4-Hydroxypipecolic acid	9.49	316.1292	-	-	-	-	-	-	-	-	-	-	-	-	M	-	-	-	-	-	-	-	-	-	1/22
38	5-Hydroxypipecolic acid	11.86	316.1292	-	-	-	-	-	-	-	-	-	-	-	-	L	-	-	-	-	-	-	-	-	-	1/22
39	3-Hydroxymethyl phenylalanine	18.57	366.1448	-	-	-	-	-	-	-	-	-	-	-	-	M	-	-	-	-	-	-	-	-	-	1/22
40	3-Hydroxymethyl tyrosine	16.2	382.1402	-	-	-	-	-	-	-	-	-	-	-	-	H	-	-	-	-	-	-	-	-	-	1/22
41	m-Carboxy-phenylalanine	19.84	380.1241	-	-	-	-	-	-	-	-	-	-	-	-	M	-	-	-	-	-	-	-	-	-	1/22
42	γ-Hydroxyornithine	6.86	319.1401	-	-	-	-	-	-	-	-	-	-	-	-	-	M	-	-	-	-	-	-	-	-	1/22
43	Homoarginine	10.1	359.1826	-	-	-	-	-	-	-	-	-	-	-	-	-	L	-	-	H	L	-	-	-	-	3/22
44	γ-Hydroxyarginine	7.03/7.27	361.1619	-	-	-	-	-	-	-	-	-	-	-	-	-	M	-	T	-	-	-	-	-	-	2/22
45	γ-Hydroxy-L-homoarginine	7.8	375.1775	-	-	-	-	-	-	-	-	-	-	-	-	-	M	-	-	M	-	-	-	-	-	2/22
46	l-3,4-Dihydroxy-phenylalanine (L-DOPA)	15.68	368.1241	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	H	-	-	L	-	H	-	3/22
47	Deoxyfructose L-dopa (L-DOPA hexose isomer)	14.52	530.1769	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	H	-	-	-	-	-	-	1/22
48	β-N-Oxalyl-L-R, β-diaminopropionic acid (β-ODAP)	8.5	347.0986	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	H	-	-	-	-	-	1/22
49	α-N-Oxalyl-L-R, β-diaminopropionic acid (α-ODAP, less toxic than β-ODAP)	9.22	347.0986	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	H	-	-	-	-	-	1/22
50	γ-N-Oxalyl-L-α, γ-diaminobutyric acid (γ-ODAB)	8.8	361.1142	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	L	-	-	-	-	-	1/22
51	2,3 Diaminopropionic acid (DAPA)	13.3	445.1619	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	L	-	-	-	-	1/22
52	2,4-Diaminobutyric acid (DABA)	13.66	459.1775	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	L	-	-	-	-	1/22
53	L-Canavanine	8.11	347.1462	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	H	-	-	-	-	1/22
54	Mimosine	8.98	369.1193	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	H	-	-	-	1/22
55	3-Carboxysalsolinol/N-acetyl-L-tyrosine	15.1/19.4	394.1397	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	M	-	1/22
56	L-3-Carboxy-6,7-dihydroxy-1,2,3,4-tetrahydroisoquinoline	16.78	380.1241	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	M	-	1/22
57	N-Methyl-L-DOPA	18.51	382.1397	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	M	-	-	-	-	M	-	2/22
58	γ-Ethylidene glutamic acid	16.08	344.1241	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	M	1/22
59	γ-Methylene glutamic acid (L-MDAM)	13.65	330.1084	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	M	1/22
60	Hydroxy-methyl glutamic acid	9.48	348.119	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	L	1/22
61	γ-Ethyl glutamic acid	15.74	346.1397	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	M	1/22
62	Taurine	8.7	296.0677	-	-	-	-	-	-	-	-	-	-	-	-	-	-	-	L	-	-	L	-	-	-	2/22
63	S-Methyl-L-cysteine	15.38	306.0907	H	H	H	H	L	H	H	H	H	H	T	L	-	L	L	L	L	-	-	L	-	-	17/22
64	γ-Glutamyl-S-methylcysteine (GMC)	13.75	435.1333	M	M	H	H	-	H	L	L	H	H	L	T	-	L	-	-	L	-	-	L	-	-	14/22
65	S-Methylhomoglutathione	13.89	506.1704	L	M	L	M	-	M	L	M	-	L	T	-	-	L	-	-	-	-	-	-	-	-	10/22
66	L-Homoglutathione	12.72	492.1547	L	T	L	-	-	L	L	M	-	M	L	M	-	-	-	-	-	-	-	-	-	-	9/22
67	γ-Glutamylmethionine (GGM)	15.75	449.1489	L	L	M	L	-	-	-	-	-	-	-	T	-	-	L	-	-	-	-	-	-	-	6/22

Note: (1) 1. Phaseolus vulgaris; 2. P. coccineus; 3. P. acutifolius; 4. P. lunatus; 5. Pisum sativum; 6. Vigna radiata; 7. V. mungo; 8. V. umbellata; 9. V. unguiculata; 10. V. aconitifolia; 11. V. angularis; 12. Cajanus Cajan; 13. Caesalpinia spinosa; 14. Lens culinaris; 15. Cicer arietinum; 16. Vicia faba; 17. Lathyrus sativus; 18. Medicago sativa; 19. Mimosa pudica; 20. Glycine max; 21. Mucuna pruriens; 22. Caesalpinia bonducella; (2) H = high (ToF-MS signal intensity >7 × 10⁶ cps); M = medium (signal intensity between 7 × 10⁵ to 7 × 10⁶ cps); L = low (signal intensity between 5 × 10⁴ to 7 × 10⁵ cps); and T = trace (signal intensity < 5 × 10⁴ cps). H, M, L, and T values were based on the analysis of compounds within the sample but not between the samples. (3) * Data shown for the powdered sample; [C. spinosa and C bonducella did not absorb water upon soaking]. (4) ’-‘ = not detected.

Figure 4.

Relative distribution of free proteinogenic and non-proteinogenic amino acids across different Leguminosae seeds in water under soaking conditions.

Similarly, M. pruriens showed arginine, isoleucine, leucine, lysine, and tyrosine as major PAAs, whereas methionine, cystine, tryptophan, and histidine were present in lower amounts. The P. coccineus analysis showed proline, glycine, and valine as major compounds. V. unguiculata and V. aconitifolia were rich in alanine, arginine, and proline. V. umbellata, M. sativa, and P. lunatus showed arginine, leucine, lysine, and phenylalanine as major PAAs. The samples of P. acutifolius and P. vulgaris showed alanine, lysine, valine, and leucine as major compounds with lower amounts of methionine, histidine, cysteine, and tyrosine. V. angularis was high in lysine, leucine, phenylalanine, and valine, whereas lysine, valine, isoleucine, leucine, phenylalanine, and alanine contents were high in V. radiata and V. mungo. L. sativus showed higher valine, leucine, lysine, proline, and serine levels.

In the C. bonducella powder samples, the major compounds were aspartic acid, glycine, leucine, histidine, and isoleucine, but at lower amounts were methionine, tryptophan, phenylalanine, cysteine, and valine. Glutamic and aspartic acids were major components in all samples except in C. cajan, where alanine has relatively higher levels of lysine, valine, leucine, and glutamine. In most legume samples, the sulfur-containing amino acids methionine and cysteine were the limiting essential amino acids [30]. In tara powder from C. spinosa, lysine is the limiting amino acid, with leucine, arginine, glycine, and alanine as major ones. Asparagine is the most important PAA used in the biosynthesis of proteins and was found in all the samples.

The amino acids in the seed extracts showed more variability than they did in the powder extracts. It is plausible that the extractability of such PAAs may be more efficient with powdered samples compared to intact, soaked legume seeds. While most PAAs were not detected, trace amounts of phenylalanine, valine, alanine, glutamic acid, and serine were found. Notably, C. spinosa extracts exhibited higher amino acid levels than C. bonducella. No significant weight changes were observed in soaked seeds of either Caesalpinia species.

3.5. Free Non-Proteinogenic Amino Acid (NPAAs) Content

NPAAs are primarily found in free form, but some can be conjugated to carbohydrates or linked to glutamyl residues [31]. Many NPAAs have structural similarities to PAAs and often contain hydroxylated aromatic or heterocyclic rings. Heterocyclic NPAAs may include oxygen, nitrogen, and sulfur atoms in their structures [32]. These compounds are generally considered more toxic than PAAs and can deter herbivores through direct toxicity or by making plants unpalatable. Combinations of NPAAs often have a stronger deterrent effect than individual compounds [33]. A previous study observed changes in NPAA levels during the germination of legume seeds [34].

PAAs and NPAAs do contribute taste and flavor to food products especially in terms of being savory or bitter. These factors vary based on chemical composition along with protein breakdown during digestion as well as the concentrations of PAAs and NPAAs. Specific NPAAs may have an impact on taste and flavor profiles [35]. According to Kim et al., proline, asparagine, tyrosine, alanine, leucine, and glutamate can contribute to the bitterness of food products. In addition, the hydrophobic and electronic properties of amino acids and the critical spatial structure of peptides influence the intensity of bitterness [36]. Traditional processing like soaking before cooking increases the nutritional value and lowers the amount of harmful components and antinutrients, such as NPAAs, phytic acid, tannins, lectin, and trypsin inhibitor. Considering the above factors, it is evident that the analyzed and identified PAAs and NPAAs likely impact the taste and flavor profiles of respective legume seeds, but, because of the complexities in PAAs and NPAAs composition, investigations to establish specificity in taste related to non-proteinogenic amino acids will be challenging.

Our analysis revealed qualitative differences in NPAAs among the seeds. Several diverse chemical structures derived from nicotinic acid or common proteinogenic amino acids were detected in these seeds and could serve as chemotaxonomic markers. Trigonelline (a nicotinic acid derivative) and γ-aminobutyric acid (GABA) were detected in all the seeds. Some of the NPAAs, such as α-aminoadipic acid, homoserine, pipecolic acid, hydroxy norvaline, acetyl ornithine, O-oxalyl homoserine, hydroxy arginine, α-ODAP_, or L-dopa, were detected in these samples with variable concentrations and are dependent upon plant species. While their effects on humans are largely unknown, toxicity cannot be ruled out. For instance, L-canavanine in M. sativa is toxic to herbivores and interferes with protein synthesis [37]. Lathyrism, a neurodegenerative disease, can be caused by the excessive consumption of L. sativus seeds due to the presence of β-ODAP [38]. Conversely, L-dopa (L-3,4-dihydroxyphenylalanine), extracted from M. pruriens seeds, is claimed to treat Parkinson’s disease [39].

As reported earlier [40,41], our analysis revealed distinct chemical profiles among the examined Vigna and other Faboideae species. Notably, Vigna species were characterized by the absence of pipecolic acid, a compound found in other genera. Furthermore, we observed significant differences in the types and concentrations of NPAAs between Vicia and Lathyrus seeds. Vicia species preferentially accumulated C6 guanidino amino acids, such as arginine and γ-hydroxy arginine, while Lathyrus species contained higher levels of C7 guanidino amino acids, including homoarginine, γ-hydroxy homoarginine, and lathyrine. These unique chemical markers provide a reliable method for distinguishing between species within these genera.

The analysis of M. pruriens revealed the presence of several non-proteinogenic amino/imino acids, including L-dopa, L-3-carboxy-6,7-dihydroxy-1,2,3,4-tetrahydroisoquinoline, and 1-methyl-3-carboxy-6,7-dihydroxy-1,2,3,4-tetrahydro-isoquinoline. A more diverse NPAA profile was observed with L. culinaris varieties [42], where detected components included γ-hydroxyarginine (γ-OH-Arg), γ-hydroxyornithine (γ-OH-Orn), homoarginine (Har), γ-hydroxy-L-homoarginine, γ-acetylornithine, γ-glutamyl-phenylalanine, homoserine, methylglutamine, and γ-glutamyl leucine. Several other NPAAs, such as γ-hydroxynorvaline, δ-hydroxynorvaline, homoserine, and δ-acetylornithine, were found in varying distributions.

The analysis of G. max seeds resulted in the identification of eight NPAAs including GABA, γ-glutamyl-phenyl-L-alanine, γ-glutamyl tyrosine, hydroxy-L-proline, 5-methyl-L-cysteine, and γ-glutamyl-S-methyl-L-cysteine (GMC). These are dipeptides belonging to the family of N-acyl-α-amino acids and derivatives. According to Shibata et al., these dipeptides impact the taste of food products such as Kokumi [43]. Additionally, two other NPAAs were identified along with dipeptides, i.e., pipecolic acid and α-aminoadipic acid in very low amounts.

In the Caesalpinioideae subfamily, C. bonducella seeds contained seven NPAAs, including γ-ethylidene glutamic acid, γ-aminobutyric acid, γ-ethylidene glutamic acid, γ-methylene glutamic acid, hydroxy-methyl glutamic acid, and γ-ethyl glutamic acid. C. spinosa seeds, on the other hand, contained 13 NPAAs, including γ-aminobutyric acid (GABA), homoserine, baikiain, baikiain isomer, pipecolic acid (PA), 3-hydroxyproline, 5-hydroxypipecolic acid (5-HPA), 4-hydroxypipecolic acid (4-HPA), diaminopimelic acid (DAP), 3-hydroxymethyl phenylalanine (3-HMP), 3-hydroxymethyl tyrosine (3-HMT), meta-carboxyphenylalanine, acetylornithine, and α-aminoadipic acid (Table 4, Figure 5a). Several compounds, including baikiain, 3-HMP, 3-HMT, 5-HPA, 4-HPA, and PA, have been previously reported in C. spinosa [44]. Finally, M. pudica seeds contained six NPAAs, including GABA, α-aminoadipic acid, mimosine, γ-glutamyl-leucine, γ-glutamyl-3-phenyl-L-alanine, and pipecolic acid as shown in Table 4.

Figure 5.

Extracted ion chromatograms (EIC) for each analyte with mass spectrometry information in MS and MS-MS modes of (a) C. spinosa (tara) seed powder samples; (b) P. vulgaris (common bean) soaked seed samples.

3.6. Non-Proteinogenic Sulfur Amino Acids

The analysis of seeds of 21 species across different genera within the Fabaceae family revealed a broad distribution and accumulation of S-methyl-L-cysteine and its γ-glutamyl derivative. S-Methyl-L-cysteine was detected in all the examined Phaseolus and Vigna species, as well as in C. cajan, L. culinaris, C. arietinum, and G. max. Within the genus Vigna, S-methyl-L-cysteine distribution aligns with phylogenetic groups [40].

V. radiata and P. vulgaris represent closely related members of the Phaseoleae tribe within the Faboideae family, followed by G. max and C. cajan [45]. S-Methyl-L-cysteine and γ-glutamyl-S-methyl-cysteine (GMC) were detected in V. radiata, V. unguiculata, V. umbellata, and V. aconitifolia, but only trace amounts were observed in V. mungo, V. angularis, G. max, C. cajan, and L. culinaris. Conversely, γ-glutamyl methionine was detected in major amounts in V. mungo and trace amounts in C. cajan, but not detected in V. angularis and V. radiata.

S-Methyl-homoglutathione was detected in V. radiata., V. umbellata, and V. aconitifolia, but in low amounts in V. mungo, and not detected in V. angularis and V. unguiculata. Similarly, L-homoglutathione was detected in V. radiata., V. umbellata, V. aconitifolia, and V. angularis, but in low amounts in V. mungo and not detected in V. unguiculata. These compounds were not detected in M. pruriens, M. sativa, M. pudica, C. spinosa, and C. bonducella (Table 4).

All the samples contained γ-aminobutyric acid, but C. cajan exhibited lower concentrations of α/β/γ-aminobutyric acid than γ-aminobutyric acid alone. Low levels of α-aminobutyric acid have also been identified in Phaseolus species (Figure 5b). Further, the NPAAs identified in P. vulgaris are represented in Figure 5b with notable amounts of pipecolic acid, GABA, and S-methyl-L-cysteine. Bell [26] reported that Phaseolus species accumulate significantly higher concentrations of GMC than Vigna spp. The common bean (P. vulgaris) is known to store sulfur-containing NPAAs, such as GMC [46,47], S-methyl-L-cysteine [48,49,50], and S-methyl-homoglutathione [51]. Several grain legumes, including Phaseolus and Vigna species, store sulfur in the form of NPAAs within their seeds, demonstrating that in P. vulgaris, free S-methyl cysteine levels are elevated during early seed development, subsequently declining as GMC accumulates [52].

Four species of Phaseolus, L. culinaris and two Vigna species were identified to contain both S-methyl cysteine and GMC in their seeds. The Phaseolus, Lens, and Vigna genera encompass species that are important for human consumption, including the common bean (P. vulgaris), mung bean (V. radiata), and cowpea (V. unguiculata). S-Methyl-homoglutathione was detected in mature seeds of V. radiata and P. vulgaris [46,51]. S-Methyl-homoglutathione was observed in Phaseolus, L. culinaris, and Vigna species.

Forty-seven free NPAAs and twenty free PAAs were identified across various legume samples in this study. Most NPAAs (26/45) were definitively identified based on accurate mass spectrum, retention time, and MS-MS fragment ions. While reference standards were unavailable for the remaining 21 NPAAs, they were tentatively identified based on accurate mass spectra and fragment ions. Most samples contained all 20 PAAs. All 22 species showed the presence of GABA and α-aminoadipic acid. Nineteen of twenty-two species contained γ-glutamyl-phenyl-L-alanine, while eighteen contained pipecolic acid, and sixteen contained γ-glutamyl-leucine and S-methyl-L-cysteine. Fifteen samples showed the presence of γ-acetyl-L-ornithine. Fourteen contained γ-glutamyl-S-methyl cysteine, and nine contained homoserine. Other compounds were detected in fewer samples (Table 4).

With amino acids playing a crucial role in protein synthesis and metabolic homeostasis, amino acid profiling has become a valuable tool for identifying and monitoring primary metabolic disorders. These disorders may manifest biochemically as metabolic acidosis, hyperammonemia, hypoglycemia (with appropriate or increased ketosis), multi-system dysfunction, developmental delay, encephalopathy, coma, or death [53].

3.7. Statistical Analysis

The statistical analyses used to determine the distribution of proteinogenic and non-proteinogenic amino acids in the soaked bean samples include two-way ANOVA along with a Tukey multiple comparison testing analysis. It is observed that the “p” value is <0.05 for different free proteinogenic and non-proteinogenic amino acids in legume seeds, which were listed under Table 1. This defines the significance of the results on proteinogenic and non-proteinogenic amino acids in the edible soaked seeds of the Leguminosae family.

4. Conclusions

Legumes are a major protein source in many cultures and offer a rich supply of non-animal protein for the future. Twenty free proteinogenic amino acids (PAAs) form the building blocks of proteins, while free non-proteinogenic amino acids (NPAAs) have diverse roles in physiology, ranging from beneficial to harmful. The distribution of NPAAs varies among legume species, often serving species-specific or unique marker compounds. A liquid chromatography quadrupole time-of-flight high-resolution mass spectrometry (LC-QToF-MS) method was developed to simultaneously analyze free proteinogenic and non-proteinogenic amino acids as their respective 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC) derivatives in 22 legume (Fabaceae) seeds. The distribution of PAAs and NPAAs was determined across 13 genera (Phaseolus, Vigna, Lens, Pisum, Lathyrus, Mimosa, Medicago, Vicia, Cajanus, Caesalpinia, Cicer, Mucuna, and Glycine) based on retention time indices, accurate mass, and fragmentation ion matching in comparison with respective reference standards. Seeds of legumes showed a diverse range of non-proteinogenic amino acids with different chemical and biochemical properties. At least 7–14 NPAAs were identified in each seed sample, with 47 NPAAs and 20 PAAs detected. Most were newly identified, especially the sulfur-containing non-proteinogenic amino acids. Despite the benefits or nutritional properties, toxicities with non-proteinogenic amino acids may sometimes pose a risk to humans or livestock. All the seeds analyzed in this study showed different absorption rates and quantities of water absorbed, which are important quality parameters useful for many industrial applications. The Bradford assay provided the soluble protein concentration, whereas the standard SDS-PAGE method was employed to resolve the protein polypeptides within proteomic mixtures. The resulting method is expected to aid in the qualitative and quantitative analysis of commercial plant-based protein products and could serve as a valuable tool for the rapid dereplication of NPAA-based causative agents if specific protein-based products are implicated in adverse effects. Furthermore, this analytical technique allows for the frequent monitoring and analysis of toxic compounds to help in the prevention or reduction of exposure. Ultimately, this work will contribute to a more comprehensive understanding of legume amino acids (PAAs and NPAAs) composition and its biological effects.

Considering the minimal regulatory standards for plant-based food products, it is imperative to address the safety profile of food products, especially those which contain edible beans or seeds. The findings of this study highlight the need for analyzing NPAAs in food products since they plausibly cause metabolic changes, and may elicit toxicity in humans.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/foods14040611/s1: Figure S1. Classification of legume samples based on family, sub-family, genus and species used for LC-MS analysis; Figure S2. Extracted ion chromatograms and mass spectrum for protein, and non-protein amino acids.

Author Contributions

B.A.: methodology, investigation, data curation, formal analysis, writing—original draft, and writing—review and editing. K.K.: formal analysis, investigation, and writing—review and editing. I.P.: data curation, formal analysis, and writing—review and editing. K.K.T.: data curation. A.G.C. and Y.-H.W.: writing—review and editing, and data curation. I.A.K.: conceptualization, resources, funding acquisition, investigation, and project administration. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

Data available upon request.

Conflicts of Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Funding Statement

This research is supported in part by “Science Based Authentication of Botanical Ingredients” funded by the Center for Food Safety and Applied Nutrition, US Food and Drug Administration grant number 5U01FD004246, and “Discovery & Development of Natural Products for Pharmaceutical & Agricultural Applications” funded by the United States Department of Agriculture, Agricultural Research Service, Specific Cooperative Agreement No. 58-6060-6-015.