Translate this page into:
Evaluation of the fluorescent chiral derivatization reagent DBMA for targeting amino functional groups: An application in the analysis of DL-amino acids in the saliva of patients with diabetes
* Corresponding author: E-mail address: junzhemin23@163.com (J.Z. Min); zhengmingshan@ybu.edu.cn (M. Zheng)
-
Received: ,
Accepted: ,
Abstract
The study developed a novel chiral fluorescence (FL) derivatization reagent, (R)-1-(7-(N,N-dimethylsulfamoyl)-benzo[c][1,2,5]oxadiazol-4-yl)-2-methylpyrrolidine-2-carboxylic acid (DBMA), targeting the amino functional group. Eighteen DL-amino acids were tested for chiral separation efficiency using ultra-performance liquid chromatography-FL (UPLC-FL). DBMA and DL-amino acids formed diastereomers following reaction conducted at 40°C for 30 min, employing 1-(3-dimethylaminopropyl)-3-ethylcarbohydrazide (EDC) and hydroxybenzotriazole (HOBt) condensation agents. The analysis was performed using a BEH C18 column (2.1 × 100 mm, 1.7 μm) with detection wavelengths set at 460 nm (excitation) and 550 nm (emission). The mobile phase consisted of two components: an aqueous phase containing 10 mM ammonium acetate with 0.05% formic acid (FA), and an organic phase comprising either 0.1% FA in acetonitrile or 0.1% FA in MeOH. The results demonstrated that the Rs was 1.51-4.40, indicating a favorable chiral separation effect. Furthermore, the method was extended to the simultaneous detection of four DL-amino acids (DL-Asp, DL-Thr, DL-Ile, and DL-Lys) from 31 healthy volunteers and 19 diabetic patients. The R2 was ≥ 0.9991 in the 5-2000 μM concentration range, with a limit of detection ranging from 2.5 to 25 pmol. The intraday and interday precisions ranged from 1.13% to 9.25%. Diabetic patients exhibited notably elevated average concentrations of D-Asp, DL-Thr, L-Ile, and DL-Lys in their saliva, in contrast to healthy volunteers. The disparity was determined to have statistical significance (p < 0.01). Furthermore, the ratios of D/L-Asp, D/L-Thr, and D/L-Lys were statistically significant (p < 0.01). DBMA, a novel fluorescent chiral derivatizing agent, enables enantioselective detection of D/L-amino acid biomarkers in biological matrices. This introduces an innovative chiral derivatization agent for studying chiral metabolomics.
Keywords
DBMA
Diabetic
DL-amino acid
Fluorescent chiral probe
UPLC-FL
1. Introduction
Diabetes type 2 is a long-term metabolic illness. Currently, the primary clinical indicators utilized in assessing patients with diabetes include levels of glycated hemoglobin (HbA1c), blood glucose, and urinary glucose [1-2]. Blood tests pose certain infection risks and often suffer from poor patient compliance, whereas urine tests, while convenient, are susceptible to missed diagnoses and exhibit lower accuracy. Compared with routine blood and urine tests that require professional containers such as blood anticoagulant tubes, saliva collection only needs to use inexpensive 5 mL ordinary test tubes. In addition, blood tests require professional medical staff to operate, while the collection of saliva samples can be completed by patients themselves, which greatly reduces labor costs. It is worth emphasizing that compared with blood and urine samples, saliva sample collection does not require acupuncture or invasive procedures, reducing the pain and infection risk for patients. It is more suitable for children, the elderly, or people who need frequent testing. Recent metabolomics studies have elucidated a close association between amino acid levels in human saliva and diabetes [3-4], with specific amino acids potentially serving as novel biomarkers for diabetes diagnosis. For example, threonine (Thr), alanine (Ala), tyrosine (Tyr), and glutamate (Glu) have been found to be significantly elevated in the bloodstream of individuals diagnosed with diabetes [5]. However, in saliva, the role of chiral amino acids has not yet been thoroughly investigated in academic literature on diabetes. Saliva collection combines procedural simplicity with strong participant adherence while maintaining biochemical correlations with bloodborne metabolite concentrations. Therefore, saliva is anticipated to serve as a substitute or complementary biological sample with potential clinical applications, thus emerging as one of the non-invasive biological samples with promising clinical prospects.
Common methods for detecting enantiomers include high-performance liquid chromatography (HPLC) [6-7], gas chromatography (GC) [8], supercritical fluid chromatography (SFC) [9], and capillary electrophoresis (CE) [10]. Enantiomer separation methods based on HPLC technology can be divided into direct and indirect separation methods [11]. Chiral stationary phase technology and chiral mobile phase additives represent direct separation approaches that employ chiral columns or additive agents, respectively, yet face clinical implementation challenges due to prohibitive operational expenditure. Besides, extensive experimentation is required to optimize the concentration of additives and gradient conditions when using the manual mobile phase method; otherwise, the peak broadening and resolution of the analyte may be affected [12]. Compared to direct methods, indirect methods (pre-column derivatization methods) may have more cumbersome operations, but diastereomers labeled by derivatization can be separated on inexpensive C18 chromatography columns [13]. Derivatization labeling enhances analytical performance by improving analyte detectability while mitigating matrix interference, establishing its utility in modern bioanalytical methodologies.
Currently, the types of derivatization reagents used mainly include ultraviolet (UV), mass spectrometry (MS), and fluorescence (FL) chiral derivatization reagents. UV derivatization reagents mainly include OTPTHE [14], FDAA [15], and (S)-Nap Btz [16]. The above derivatization reagents can target the labeling of primary or secondary amine functional groups, facilitating the resolution of amino acid diastereomers in both plasma samples and food matrices. The detection limit (LOD) can reach the pmol level. Despite HPLC-UV being able to detect amino acid enantiomers, their selectivity is insufficient, and the result of the matrix effect fails to satisfy method validation for biological analysis. In recent years, the high selectivity and sensitivity offered by MS detectors have propelled the development of MS derivatizing agents. The reported chiral MS derivatization reagents mainly include DMT-(S)-Pro-Osu [17], L-PGA-Osu [18], and (S)-COXA-Osu [19]. MS detection enhances the sensitivity of labeled amino acid diastereomers to attomolar-level LOD, positioning MS-based derivatization methods as the gold standard for ultratrace amino acid quantification in biofluids like plasma and saliva. However, MS is inevitably expensive, and their universal access is difficult to achieve. The FL derivatization reagent, after derivatization of the amino acid, can make the diastereoisomer of the amino acid carry a fluorophore and can be detected by an FL detector. The commonly used fluorescent chiral derivatization reagents include DBD-PyNCS [20], FLEC [21], DATAN [22], and DBD-M-Pro [23]. Employing these derivatization compounds facilitates high-sensitivity detection of amino acids in plasma and food specimens, attaining LOD at the pmol level. However, some derivatization reagents have the problem of long reaction times and high temperatures [21]. Therefore, rapid and high-throughput labeling methods are of great significance for early screening. The above derivatization reagents are summarized in Table S1.
Given these considerations, a novel chiral fluorescent labeling reagent named (R)-1-(7-(N,N-dimethylaminosulfonyl)-benzo[c][1,2,5]oxadiazol-4-yl)-2-methylpyrrolidin-2-carboxylic acid (DBMA) was designed and synthesized for specific targeting of amino functional groups. Furthermore, this chiral derivatization agent demonstrated successful application in the chromatographic separation of 18 pairs of DL-amino acids. We developed an innovative DBMA-based ultra-performance liquid chromatography coupled with FL detection (UPLC-FL) method for the simultaneous determination of four D/L-amino acids in human saliva. This methodology was successfully applied to quantify chiral amino acids in biological specimens from 19 diabetes patients (PD) and 31 healthy volunteers (HVs), demonstrating its potential as a non-invasive screening tool for early-stage diabetes detection.
2. Materials and Methods
2.1. Materials and reagents
6-Aminohexanoic acid (Internal stand, IS), L-Ala, and D-lysine (D-Lys) for internal standardization were acquired from Sigma-Aldrich (St. Louis, MO, USA). D-Thr and D-Ser were obtained from Alfa-Aesar (Shanghai, China). D-isoleucine (D-Ile) and D-Alanine (D-Ala) was provided by Fluka (Everett, WA, USA). Dibei Biotechnology (Shanghai, China) provided L-Gln. 4-(N,N-Dimethylaminosulfonyl)-7-fluoro-2,1,3-benzoxadiazole (DBD-F), L-methionine (L-Met), DL-phenylalanine (DL-Phe), DL-tryptophan (DL-Trp), L-Ile, DL-Glu, D-Gln, DL-valine (DL-Val), D-Asp, DL-threonine (DL-Thr), DL-cysteine (DL-Cys), DL-leucine (DL-Leu), L-Lys, DL-Tyr, D/L-proline (DL-Pro), DL-arginine (DL-Arg), L-Ser, DL- histidine (DL-His), and formic acid (FA) were were obtained from Dibei Biotechnology (Shanghai, China). Anhydrous Na2CO3, anhydrous Na2SO4, borax, ammonium acetate and ammonium formate were procured from Aladdin Reagent. Co. (Shanghai, China). (R)-2-Methylproline (R-MPC) was provided by Haohong Biomedical Technology (Shanghai, China). Tokyo Kasei Industrial (Tokyo, Japan) provided D-Met, L-Asp, and 1-(3-Dimethylaminopropyl)-3-ethylcarbohydrazide (EDC). 1-Hydroxybenzotriazole (HOBt) was obtained from Macklin (Shanghai, China). Ethyl acetate (EA) was purchased from Haoran Chemical Technology (Yanji, China.). Concentrated hydrochloric acid was provided by Nanjing Chemical Reagent Co. (Nanjing, China). Thermo Fisher Scientific (New Jersey, USA) supplied the HPLC-grade acetonitrile (CH3CN) and MeOH (CH3OH) used in this study. The ultrapure and distilled water samples were recently produced through a Milli-Q purification apparatus (Millipore, Xiamen, China).
2.2. UPLC-FL conditions
A FL detector was coupled with an ultra-performance liquid chromatography (UPLC) system, and amino acid separation and detection were carried out (UPLC H-class, Waters, USA). The separation of DL-amino acids was performed on an ACQUITY UPLCTM BEH C18 column (100 × 2.1 mm, 1.7 μm) under controlled conditions: column temperature 40°C, flow rate 0.35 mL/min. The mobile phase system comprised (A) aqueous 0.1% FA, (B1) 0.1% FA/CH3CN, and (B2) 0.1% FA/CH3OH. FL detection wavelengths set at 460 nm (excitation) and 550 nm (emission).
2.3. Synthesis of fluorescent derivatization reagent DBMA
Solutions of DBD-F (0.41 mM) and R-MPC were separately prepared in CH3CN and 0.25 M Na2CO3, respectively. The two solutions were combined in a 1:1 (v/v) ratio and allowed to react for 60 min at 30°C. After the reaction, a vacuum concentration system was employed to dry the solvent at 35°C. Being dissolved in 200 μL of ultrapure water, the residue was moved to a separating funnel. An equal volume of EA was added to remove any unreacted DBD-F. Subsequently, the water-based layer underwent hydrochloric acid to bring the pH to 1-2. The extraction procedure was carried out three times with an equal volume of EA added. Purified orange-yellow DBMA powder was obtained by drying the EA layer over anhydrous sodium sulfate. The synthetic pathway and DBMA has been illustrated in Figure 1.
- Synthetic routes of DBMA and UHPLC-HRMS mass chromatogram and spectrum of the novel chiral fluorescent derivatization reagent DBMA. *Indicates an asymmetric carbon.
2.4. Derivatization of DBMA with DL-amino acid standard
Acetonitrile-water (1:1, v/v) was added to the 1.5 mL tube containing 40 μL of 5.0 mM DBMA, 20 μL of 100 mM EDC, 20 μL of 100 mM HOBt, 30 μL of 100 mM borax buffer solution (pH=9.5), and 10 μL of 2.0 mM amino acid to reach a final volume of 200 μL. After thorough mixing, amino acid derivatives were obtained by reacting at 40°C for 2 h. A 0.22-μm nylon membrane was used for the filtering process, and it was stored at 4°C until needed. The formula for the derivatization reaction and the chemical structure formula for DL-amino acids have been displayed in Figure 2.
- Derivatization reaction of DL-amino acid with DBMA. * Indicates an asymmetric carbon.
2.5. Saliva collection and pretreatment
Saliva samples were acquired from 23 HVs were male and eight females (aged 20 to 30 years), and eight PDs were male and 11 were female (aged 27 to 71 years), with fasting blood glucose levels of 5.6-27.4 mmol/L. The Ethics Committee of Yanbian University kindly supplied all saliva specimens, and all participants provided written informed consent. Yanbian University’s Medical Ethics Committee sanctioned this research (No.2024271).
Saliva collection necessitated that volunteers fast for 1 h before the experiment. Following a 10-min session of tooth brushing, the test tube was methodically filled with saliva through a gradual dripping technique. Before being used, the participant’s saliva was collected and kept at -20°C. Each 500 μL saliva sample was blended in 1000 μL of acetonitrile and vortexed for 30 seconds. The protein precipitation process involved centrifuging the solutions at 10,000 rpm, 10 min. After being dehydrated, the supernatant was kept at 35°C in an Eppendorf tube. Furthermore, the dried residue was dissolved in acetonitrile and then supplemented with 10 μL of 1.0 mM internal standard (IS), 40 μL of 5.0 mM DBMA, 20 μL of 100 mM EDC and HOBt, and 30 μL of 100 mM borax. The solution was reacted at 40°C, 30 min. A nylon membrane with a pore size of 0.22 μm was used for filtering, and subsequent analysis was performed through UPLC-FL.
2.6. Validation of the method
2.6.1. Determination of calibration curve and LOD
Stock solutions (2000 μM) of DL-Asp, DL-Thr, DL-Ile, and DL-Lys were produced separately in acetonitrile-water (1:1, v/v). Subsequently, the stock solutions of DL-Asp and DL-Lys were further diluted to working standard 25-1000 μM, while DL-Ile and DL-Thr were diluted to form working standard solutions with concentrations ranging from 5-2000 μM and 25-2000 μM, respectively. The process of amino acid derivatization was carried out in accordance with Section 2.5. The internal standard approach was used to generate the calibration curve in this investigation. The mass chromatogram of LOD (signal-to-noise (S/N) ratio = 3) was calculated by contrasting the peak intensity of low concentrations with the background noise level.
2.6.2. Measurement of interday and intraday precision
The lower limit of quantitation (LLOQ), low-quality control (LQC), middle-quality control (MQC), and high-quality control (HQC) samples containing DL-Asp, DL-Thr, DL-Ile, and DL-Lys at concentrations of 25 (5), 50, 125, and 250 μM were independently prepared as described in Section 2.6.1. After derivatization, the samples’ relative error (RE) and root mean square deviation (RSD) were calculated to evaluate the precision and accuracy of intraday and interday measurements. There were six duplicates of each concentration (n = 6).
2.6.3. Recovery of DL-amino acids in human saliva
Saliva can be mixed with known amounts of standard amino acid solutions to calculate recovery rates. Evaluation of extraction recovery involved comparing the peak areas observed post-extraction with those recorded before extraction for QC samples. Following the method outlined in Section 2.6.1, three distinct concentrations of QC samples were prepared: LQC (DL-Asp, DL-Thr, DL-Ile, and DL-Lys at 125, 125, 39.51, and 11.71 μM, respectively), MQC (DL-Asp, DL-Thr, DL-Ile, and DL-Lys at 250, 250, 88.89, and 39.51 μM, respectively); and HQC (DL-Asp, DL-Thr, DL-Ile, L-Lys, and D-Lys at 50, 500, 133.34, 59.26, and 82.83 μM, respectively). The recovery percentage of salivary amino acids was calculated, considering the significant improvements observed following the derivatization process.
2.7. DL-amino acids were determined in the saliva of HVs and PDs
Herein, saliva served as a biological sample. Altogether, 50 saliva samples were collected, comprising samples from 31 HVs and 19 PDs. In this study, DL-Asp, DL-Thr, DL-Ile, and DL-Lys were analyzed. Following the natural thawing of saliva, the saliva was pretreated as described in Section 2.5. Following derivatization and filtering via a 0.22-μm organic filter membrane, saliva samples were analyzed using UPLC-FL.
2.8. Statistical analysis
Welch’s t-test and Mann-Whitney U test were employed for statistical comparisons, with significance thresholds established at p < 0.05. Linear regression is done using GraphPad Prism version 7.0, and the Pearson correlation coefficient (r) is used to determine the correlations between the two variables.
3. Results and Discussion
3.1. Synthesis of the fluorescent chiral probe DBMA
Our team created a new fluorescent probe, DBMA, which features a lipid-soluble fluorescent group (DBD). By implementing structural alterations, both signal responsiveness and chromatographic residence time of the substance can be effectively enhanced. The structural formulas in this reaction have been depicted in Figure 1. In addition, the presence of carboxyl groups and chiral centers in DBMA enables it to identify amino groups and efficiently distinguish DL-amino acids. Following derivatization with amino acids, DBMA yielded a pair of non-enantiomeric compounds, facilitating the resolution of chiral amino compounds on an affordable reverse-phase chromatography column (Figure 1). It presents the extracted ion chromatogram (XIC) and MS results. The retention time (RT) and mass to charge ratio (m/z) for DBMA were 12.91 min and 355.10614 Da [M+H]+, respectively. The 1H-NMR (500 MHz, DMSO-d6) spectrum revealed peaks at δ13.04 (s, 1H), 7.85 (d, J = 8.3 Hz, 1H), 6.14 (d, J = 8.3 Hz, 1H), 4.06 (d, J = 94.8 Hz, 2H), 2.72 (s, 6H), 2.35-2.30 (m, 1H), 2.23-2.19 (m, 1H), 2.15-2.11 (m, 2H), and 1.67 (s, 3H). Similarly, the 13C-NMR (125 MHz, DMSO-d6) spectrum exhibited peaks at δ173.90, 146.19, 138.74, 138.61, 107.60, 105.01, 102.41, 67.94, 53.16, 40.52, 36.83(2C), 22.05(2C).
3.2. Optimization of derivative conditions for DBMA
In this study, the derivatization efficiency of DBMA was evaluated using DL-Thr, DL-Asp, DL-Ile, and DL-Lys. The chemical structural formula of DL-amino acids and the general formula of the derivatization reaction have been shown in Figure 2. To determine the optimal derivatization conditions, we investigated various reaction times (10-240 min) and temperatures (20-80°C) separately; the results have been shown in Figure 3. Setting the reaction duration to 60 min led to a steady rise in the peak area of DL-amino acids as the reaction temperature rose, stabilizing at 40°C, suggesting this temperature was ideal for derivatization. Subsequently, we investigated the peak areas of eight amino acids under 13 distinct reaction times while maintaining the reaction temperature at 40°C. The results indicated that the peak area reached its maximum value when the reaction time was set to 30 min. Therefore, it was established that 30 min at 40°C represented the conditions for the derivatization process.
- Effect of temperature and time on the derivatization reaction of DL-amino acids and DBMA. (a) Temperature of derivatization reaction; (b) Duration of derivatization reaction.
3.3. Evaluation of chiral separation efficiency in 18 DL-amino acids
To give a more comprehensive evaluation of DBMA chiral separation effectiveness, we concentrated on the 18 DL-amino acid separation efficiency across a range of gradients in the mobile phase. UPLC-FL analysis of amino acid derivative solutions and blank controls was performed to identify chromatographic peaks corresponding to novel products formed through derivatizing reagent-amino acid interactions. The gradient and content of the mobile phase were then changed to effectively separate the desired DL-amino acid derivatives. All 18 amino acids were eluted at their respective gradients in less than 30 min, as indicated in Table 1. The separation selectivity (α) ranged from 1.03 to 1.23, and the resolution value (Rs) ranged from 1.51 to 4.40, indicating complete separation. Figure 4 illustrates that during amino acid separation by chromatography, L-amino acids consistently eluted before D-amino acids among the 18 amino acids.
| No. | Amino compounds | Mobile phase | k | a | Rs | |
|---|---|---|---|---|---|---|
| L-Isomer | D-Isomer | |||||
| 1 | Met | I (20-45) | 11.16 | 12.84 | 1.15 | 2.76 |
| 2 | Phe | I (20-45) | 14.03 | 16.92 | 1.21 | 4.40 |
| 3 | Tip | I (20-45) | 14.95 | 16.93 | 1.13 | 2.81 |
| 4 | Asp | I (15-24) | 9.68 | 10.47 | 1.08 | 2.26 |
| 5 | Thr | I (15-23) | 13.25 | 13.89 | 1.05 | 1.83 |
| 6 | lie | I (20-45) | 12.43 | 15.27 | 1.23 | 3.75 |
| 7 | Val | I (15-40) | 14.37 | 16.70 | 1.16 | 3.85 |
| 8 | Pro | I (27-32) | 6.11 | 7.40 | 1.21 | 2.13 |
| 9 | Leu | I (15-65) | 10.80 | 12.45 | 1.15 | 2.04 |
| 10 | Ala | I (25-56) | 4.30 | 4.87 | 1.13 | 2.28 |
| 11 | Glu | I (15-40) | 6.69 | 7.16 | 1.07 | 2.80 |
| 12 | Gin | I (15-40) | 5.49 | 5.77 | 1.05 | 2.42 |
| 13 | Tyr | I (50-57) | 4.49 | 5.13 | 1.14 | 1.68 |
| 14 | Lys | I (40-40) | 5.16 | 5.61 | 1.09 | 1.56 |
| 15 | His | I (30-41) | 18.06 | 18.69 | 1.03 | 1.68 |
| 16 | Ser | I (14-14) | 22.10 | 23.19 | 1.05 | 1.66 |
| 17 | Cys | I (20-60) | 14.63 | 15.77 | 1.08 | 2.82 |
| 18 | Arg | n (20-50) | 8.32 | 8.67 | 1.04 | 1.51 |
Capacity factor k=(tR-t0)/t0; Separation factor α=(tR2-t0)/(tR1-t0); Resolution value Rs= 2(tR2-tR1)/(W1 + W2). Mobile phase Ⅰ: 10 mM CH3COONH4-0.05% FA in H2O; Mobile phase Ⅱ: (A) 0.1% FA in CH3CN; (B) 0.1% FA in CH3OH.
- UPLC-FL chromatograms of separation of derivative DL-amino acid enantiomers.
3.4. Method validation
As shown in Table 2, the calibration curve exhibited a robust linear relationship for DL-Asp, DL-Thr, DL-Ile, and DL-Lys in the concentration range of 5-2000 μM. The standard curve was drawn with the amino acid concentration as the abscissa and the ratio of amino acid to IS peak area as the ordinate. The R2 values ranged between 0.9991 and 0.9998. The LOD (S/N = 3) for DL-Asp, DL-Thr, DL-Ile, and DL-Lys were 25, 25, 5, and 2.5 pmol, respectively. Table 3 compiles the results for precision and accuracy for the intraday and interday periods. The table makes it clear that the intraday precision for the four chiral amino acids ranged from 0.17% to 11.74%, while the interday precision spanned from 1.13% to 9.78%. The recovery of the four chiral amino acids has been summarized in Table 4, indicating an average recovery rate ranging from 82.73% to 109.88%. The findings confirm the effectiveness of the technique used in this research for concurrently analyzing DL-Asp, DL-Thr, DL-Ile, and DL-Lys. The results demonstrate notable linearity, repeatability, and accuracy, thus highlighting the robustness and reliability of the analytical approach utilized. Moreover, the method shows potential for identifying and isolating chiral amino acids from human saliva.
| Amino acids | Calibration range (µM) | Linear equation | Linearity (R2) | CV (%) | LOD (pmol) | LOQ (pmol) |
|---|---|---|---|---|---|---|
| L-Asp | 25-1000 | y=0.000045x+0.000504 | 0.9995 | 2.282-8.139 | 25 | 70 |
| D-Asp | 25-1000 | y=0.000084x+0.000849 | 0.9991 | 0.9619-6.524 | 25 | 70 |
| L-Thr | 25-2000 | y=0.000061x-0.000019 | 0.9995 | 0.3868-7.241 | 25 | 70 |
| D-TTir | 25-2000 | y=0.000045x+0.000511 | 0.9993 | 1.052-9.548 | 25 | 70 |
| L-Ile | 5.0-2000 | y=0.000447x-0.007783 | 0.9992 | 0.5320-6.506 | 5 | 12 |
| D-Ile | 5.0-2000 | y=0.000215x+0.001916 | 0.9998 | 0.8123-11.16 | 5 | 12 |
| L-Lys | 25-1000 | y=0.001010x+0.001464 | 0.9991 | 0.05362-2.601 | 2.5 | 10 |
| D-Lys | 25-1000 | y=0.000567x-0.001929 | 0.9995 | 0.3229-3.000 | 2.5 | 10 |
| Amino acids | Amount (µm) | Intra-day assay | Inter-day assay | ||
|---|---|---|---|---|---|
| Mean ± SD (µM) | CV% (n=6) | Mean ± SD (µM) | CV% (n=6) | ||
| L-Asp | 25 | 23.459 ± 2.753 | 11.735 | 23.196 ±2.268 | 9.777 |
| 50 | 47.741 ±3.195 | 6.693 | 47.046 ± 4.352 | 9.250 | |
| 125 | 123.289 ±2.015 | 1.635 | 127.747 ± 7.407 | 5.798 | |
| 250 | 308.852 ±4.348 | 1.408 | 305.600 ± 23.243 | 7.606 | |
| D-Asp | 25 | 24.013 ± 1.564 | 6.513 | 23.982 ±2.106 | 8.781 |
| 50 | 44.348 ± 1.691 | 3.812 | 44.718 ± 2.189 | 4.896 | |
| 125 | 118.823 ±2.058 | 1.732 | 116.865 ± 6.466 | 5.533 | |
| 250 | 265.125 ±4.626 | 1.745 | 274.055 ± 13.639 | 4.977 | |
| L-Thr | 25 | 23.861 ± 2.053 | 8.603 | 23.983 ±2.163 | 9.018 |
| 50 | 54.768 ± 0.900 | 1.643 | 54.420 ± 2.892 | 5.315 | |
| 125 | 122.438 ±2.401 | 1.961 | 121.128 ±2.621 | 2.164 | |
| 250 | 251.704± 4.301 | 1.709 | 251.834 ± 3.424 | 1.360 | |
| D-Thr | 25 | 26.586 ± 1.486 | 5.589 | 26.013 ± 1.965 | 7.553 |
| 50 | 71.315 ±2.441 | 3.423 | 68.246 ±4.782 | 7.001 | |
| 125 | 129.205 ±4.868 | 3.768 | 128.742 ± 6.427 | 4.992 | |
| 250 | 242.115 ±5.028 | 2.077 | 240.710 ± 10.211 | 4.242 | |
| L-Ile | 5 | 5.965 ±0.342 | 5.733 | 6.135 ± 0.463 | 7.546 |
| 50 | 57.554 ± 0.362 | 0.456 | 57.753 ±0.801 | 1.386 | |
| 125 | 127.497 ±2.851 | 2.236 | 126.946 ± 3.104 | 2.445 | |
| 250 | 244.565 ±0.655 | 0.268 | 243.006 ± 3.709 | 1.526 | |
| D-Ile | 5 | 6.156 ±0.312 | 5.068 | 6.356 ± 0.361 | 5.679 |
| 50 | 47.708 ± 0.636 | 1.332 | 47.041 ± 1.393 | 2.962 | |
| 125 | 122.918 ± 1.208 | 0.983 | 123.038 ± 3.626 | 2.947 | |
| 250 | 269.822 ±0.472 | 0.175 | 264.093 ± 7.103 | 2.690 | |
| L-Lys | 25 | 24.689 ± 1.972 | 7.987 | 24.988 ± 1.064 | 4.258 |
| 50 | 41.806 ± 0.652 | 1.558 | 41.971 ±0.474 | 1.129 | |
| 125 | 127.541 ±0.216 | 0.169 | 127.460 ± 1.689 | 1.325 | |
| 250 | 260.0894 ± 1.050 | 0.403 | 257.105 ±4.714 | 1.833 | |
| D-Lys | 25 | 25.019 ± 1.268 | 5.068 | 25.139 ± 1.762 | 7.009 |
| 50 | 48.527 ± 0.886 | 1.826 | 48.766 ± 0.766 | 1.571 | |
| 125 | 130.571 ±0.608 | 0.465 | 130.846 ± 1.501 | 1.147 | |
| 250 | 263.837 ± 1.640 | 0.622 | 261.049 ±4.127 | 1.581 | |
| Endogenous amino acids | Spiked amount (µM) | Measured Mean ± SD (µM) | Recovery (%) | Mean recovery (%) |
|---|---|---|---|---|
| L-Asp | 0 | 0 | - | |
| 125.0 | 121.58 ±9.11 | 97.26 | 104.26 ± 6.40 | |
| 250.0 | 262.56 ± 14.01 | 105.75 | ||
| 500.0 | 548.93 ± 12.07 | 109.79 | ||
| D-Asp | 0 | 84.60 ± 10.25 | - | |
| 125.0 | 214.12 ± 8.19 | 103.62 | ||
| 250.0 | 360.73 ±9.89 | 110.45 | 109.88 ±6.00 | |
| 500.0 | 662.40 ± 7.44 | 115.56 | ||
| L-Thr | 0 | 108.378 ±0.778 | - | |
| 125.0 | 230.87 ±6.80 | 98.00 | ||
| 250.0 | 382.75 ±2.49 | 109.75 | 106.92 ± 7.90 | |
| 500.0 | 673.45 ± 14.72 | 113.01 | ||
| 0 | 1963.36 ± 18.34 | - | ||
| D-Tlr |
125.0 250.0 |
2078.06 ±7.51 2228.66 ±21.87 |
94.88 106.12 |
99.28 ±6.00 |
| 500.0 | 2447.57 ±9.87 | 96.84 | ||
| L-Ile | 0 | 2.95 ± 0.027 | - | |
| 39.5 | 36.40 ± 1.04 | 84.65 | ||
| 88.9 | 82.44 ±0.50 | 92.74 | 87.65 ± 4.44 | |
| 133.3 | 117.01 ±3.36 | 85.54 | ||
| D-Ile | 0 | 1.50 ±0.06 | - | |
| 39.5 | 34.93 ±0.54 | 84.60 | ||
| 88.9 | 73.44 ±0.13 | 80.93 | 83.57 ±2.30 | |
| 133.3 | 115.07 ±0.46 | 85.17 | ||
| L-Lys | 0 | 4.619 ± 0.13 | - | |
|
11.7 39.5 |
17.44 ±0.40 43.67 ±0.53 |
109.51 98.80 |
103.77 ±5.4 | |
| 59.3 | 65.66 ±0.04 | 103.01 | ||
| D-Lys | 0 | 7.25 ±0.23 | - | |
| 11.7 | 17.16 ± 0.09 | 84.66 | ||
| 39.5 | 55.05 ±0.52 | 80.68 | 82.73 ± 1.99 | |
| 88.9 | 80.88 ±0.70 | 82.83 |
3.5. Analysis of the D/L ratio and comparison of D- and L-amino acid levels in saliva between HVs and PDs
In this study, we quantified the saliva samples from 31 HVs and 19 PDs based on the fluorescent probe DBMA and UPLC-FL. After comparing the retention duration to those of standard samples, the presence of DL-Asp, DL-Tyr, DL-Ile, and DL-Lys in the saliva sample was validated. Figure 5(a) displays the separation chromatogram that shows how the derivative amino acid standards were separated. Additionally, the chromatograms of saliva samples from HVs and PDs have been shown in Figures 5(b) and 5(c).
- UPLC-FL chromatograms of four DL-amino acids present in human saliva. (a) Amino acid standards; (b) Saliva of healthy volunteers (HVs); (c): Saliva of patients with diabetes (PDS).
To further compare the content of amino acids in saliva between HVs and PDs, we conducted quantitative and statistical analyses of the levels of Asp, Thr, Ile, and Lys, respectively. Some studies have shown that compared with healthy people, the contents of amino acids such as Lys, Ile, and Asp in the serum and plasma of patients with type 2 diabetes are significantly increased, which may have a certain correlation with the onset of type 2 diabetes and can be used as potential biomarkers for the onset and progression of diabetes [5,24,25]. Furthermore, Yu et al. found that Thr metabolism is associated with abnormal glycosylation of certain immune-related proteins in the saliva of diabetic patients [26]. However, at present, there are few reports on the correlation between the optical isomers of the above four amino acids in saliva and diabetes. Therefore, in this study, the four chiral amino acids Lys, Thr, Asp, and Ile in saliva were taken as the research objects to explore their relationship with type 2 diabetes. Figure 6 illustrates that saliva from diabetes patients had significantly more of six amino acids (p < 0.01) than that of healthy volunteers (D-Ile and L-Asp excluded). The levels of D-Thr and L-Thr showed a significant increase (p < 0.0001), suggesting an association with diabetes. Moreover, there was a notable correlation between the content of L-Ile in saliva and diabetes. In the research reports on chiral amino acids and diabetes, Wang et al. found that compared with healthy volunteers, the content of L-Thr in the hair of diabetic patients showed a significant upward trend [27]. Furthermore, Zhao et al. found that L-Lys is correlated with DCD [28]. However, since diabetes is a metabolic disease, patients with diabetes may suffer from various complications, such as chronic kidney disease. Kimura et al. analyzed plasma samples from patients with diabetes and kidney disease and found that the level of D-Asp in the serum of diabetic patients showed certain differences from that of healthy people [29]. The content variation trends of D-Asp, L-Thr, and L-Lys screened in the saliva of diabetic patients in this study are consistent with the above studies. In addition, this study also found that L-Ile, D-Thr, and D-Lys in the saliva of diabetic patients showed a significant upward trend, which provided certain references for the diagnosis and treatment of diabetes. Furthermore, a statistical study of the saliva of diabetes patients showed statistically significant variations in the amounts of D-Asp, L-Ile, and D-Lys. Combining the D-Thr, L-Thr, and L-Ile readings can help with diabetes early detection and warning.
- Comparison of DL-Asp, DL-Ile, DL-Thr, and DL-Lys in the saliva of HVs. (n = 31) and PDs (n = 19) (**p < 0.01, ***p < 0.001, ****p < 0.0001; ns: not significant).
We performed a statistical study to ascertain the amino acid composition in the saliva of patients with HVs and diabetes based on the D/L ratios of chiral amino acids. Findings indicated notable disparities in the D/L ratios for Asp, Thr, and Lys, with statistical significance (p < 0.01), indicating their potential to be used as novel biomarkers for disease screening (Figure 7). Although D/L-Ile also exhibited robust statistical variance, the exceedingly low D-Ile content in human saliva-below the LOQ of this method-rendered the detection of D-Ile inconclusive for some samples. Therefore, the analysis of the D/L-Ile ratio lacks meaningful interpretation. In future research, we will focus on improving detection sensitivity and expanding the sample size to conduct an in-depth evaluation of the clinical efficacy of DL-Ile.
- Ratios of D/L-Asp, D/L-Thr, D/L-Ile, and D/L-Lys concentrations in the saliva of 31 HVs and 19 PDs. (**p < 0.01, ***p < 0.001, ****p < 0.0001, ns: not significant).
4. Conclusions
Successful creation of a new fluorescent chiral derivatization reagent, DBMA, aimed at amino functional groups. Enantiomers of 18 DL-amino acids, modified using DBMA, were entirely isolated using a C18 reversed-phase column in UPLC-FL (Rs ≥ 1.51), demonstrating a high efficiency of chiral separation. Based on pre-column derivatization, four DL-amino acids in human saliva were simultaneously chirally separated using UPLC-FL, which is a unique approach. In diabetic patients’ saliva, the levels of D-Asp, DL-Thr, L-Ile, and DL-Lys were markedly greater than in HVs (p < 0.01), and D/L-Lys (p < 0.01), and the ratios of D/L-Asp, D/L-Thr, and D/L-Lys were significantly different (p < 0.01). These findings suggest that they may serve as new markers for the early warning of diabetes. Assessing the survival of four D/L-amino acid enantiomers in human saliva can serve as a potential biomarker for diabetes. This work offers a novel analytical tool and strategy for the comprehensive investigation of the condition.
Acknowledgment
The present research was supported by the National Natural Science Foundation of China (32160234, 32360243), and the Science and Technology Development Project of Jilin Province of China (YDZJ202201ZYTS596). We thank Bullet Edits Limited for the linguistic editing and proofreading of the manuscript.
CRediT authorship contribution statement
Xuanran Cong: Methodology, Data curation, Writing – original draft. Xiaoxi Man: Data curation, Writing – original draft. Shuyun Xiao: Data curation, Writing – original draft. Yuxuan Li: Investigation, Validation, Xi-Ling Li: Validation. Mingshan Zheng: Writing – review & editing. Jun Zhe Min: Conceptualization, Writing – review & editing, Funding acquisition.
Declaration of competing 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.
Declaration of generative AI and AI-assisted technologies in the writing process
The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.
Supplementary data
Supplementary material to this article can be found online at https://dx.doi.org/10.25259/AJC_127_2025
References
- Determination of salivary glucose in healthy adults. Medicina Oral, Patologia Oral Y Cirugia Bucal. 2009;14:e510-e513. https://doi.org/10.4317/medoral.14.e510
- [Google Scholar]
- Caries and saliva in 12-18-year-old diabetics and controls. Scandinavian Journal of Dental Research. 1992;100:310-313. https://doi.org/10.1111/j.1600-0722.1992.tb01077.x
- [Google Scholar]
- Diabetes screening: Detection and application of saliva 1,5-anhydroglucitol by liquid chromatography-mass spectrometry. The Journal of Clinical Endocrinology and Metabolism. 2020;105:dgaa114. https://doi.org/10.1210/clinem/dgaa114
- [Google Scholar]
- On-site test for cannabinoids in oral fluid. Clinical Chemistry. 2012;58:1418-1425. https://doi.org/10.1373/clinchem.2012.189001
- [Google Scholar]
- Metabolite biomarkers of type 2 diabetes mellitus and pre-diabetes: A systematic review and meta-analysis. BMC Endocrine Disorders. 2020;20:174. https://doi.org/10.1186/s12902-020-00653-x
- [Google Scholar]
- Chiral HPLC for efficient resolution of enantiomers. Chemical Society Reviews. 2008;37:2593-2608. https://doi.org/10.1039/b808881k
- [Google Scholar]
- Recent progresses in protein-based chiral stationary phases for enantioseparations in liquid chromatography. Journal of Chromatography B. 2008;875:12-19. https://doi.org/10.1016/j.jchromb.2008.05.022
- [Google Scholar]
- Gas chromatographic enantioseparation of unfunctionalized chiral alkanes: A challenge in separation science (overview, state of the art, and perspectives) Chirality. 2009;21:183-198. https://doi.org/10.1002/chir.20638
- [Google Scholar]
- Synthesis of (S)-naproxen-benzotriazole and its application as chiral derivatizing reagent for microwave-assisted synthesis and indirect high performance liquid chromatographic separation of diastereomers of penicillamine, cysteine and homocysteine. Journal of Chromatography. A. 2011;1218:3648-3653. https://doi.org/10.1016/j.chroma.2011.04.016
- [Google Scholar]
- Capillary electrophoresis and column chromatography in biomedical chiral amino acid analysis. Analytical and Bioanalytical Chemistry. 2009;394:695-706. https://doi.org/10.1007/s00216-009-2792-y
- [Google Scholar]
- Approaches for multi-gram scale isolation of enantiomers for drug discovery. Expert Opinion on Drug Discovery. 2017;12:1253-1269. https://doi.org/10.1080/17460441.2017.1383981
- [Google Scholar]
- Recent advances in chiral selectors immobilization and chiral mobile phase additives in liquid chromatographic enantio-separations: A review. Journal of Chromatography. A. 2023;1706:464214. https://doi.org/10.1016/j.chroma.2023.464214
- [Google Scholar]
- Indirect enantioseparations: Recent advances in chiral metabolomics for biomedical research. International Journal of Molecular Sciences. 2022;23:7428. https://doi.org/10.3390/ijms23137428
- [Google Scholar]
- Evaluation of chiral separation efficiency of a novel OTPTHE derivatization reagent: Applications to liquid-chromatographic determination of DL-serine in human plasma. Chirality. 2019;31:1043-1052. https://doi.org/10.1002/chir.23133
- [Google Scholar]
- A sensitive and economical method for simultaneous determination of D/L- amino acids profile in foods by HPLC-UV: Application in fermented and unfermented foods discrimination. Food Chemistry. 2023;410:135382. https://doi.org/10.1016/j.foodchem.2022.135382
- [Google Scholar]
- Synthesis of (S)-naproxen-benzotriazole and its application as chiral derivatizing reagent for microwave-assisted synthesis and indirect high performance liquid chromatographic separation of diastereomers of penicillamine, cysteine and homocysteine. Journal of Chromatography. A. 2011;1218:3648-3653. https://doi.org/10.1016/j.chroma.2011.04.016
- [Google Scholar]
- Towards the chiral metabolomics: Liquid chromatography-mass spectrometry based DL-amino acid analysis after labeling with a new chiral reagent, (S)-2,5-dioxopyrrolidin-1-yl-1-(4,6-dimethoxy-1,3,5-triazin-2-yl)pyrrolidine-2-carboxylate, and the application to saliva of healthy volunteers. Analytica Chimica Acta. 2015;875:73-82. https://doi.org/10.1016/j.aca.2015.02.054
- [Google Scholar]
- Relative quantification of enantiomers of chiral amines by high-throughput LC-ESI-MS/MS using isotopic variants of light and heavy L-pyroglutamic acids as the derivatization reagents. Analytica Chimica Acta. 2013;773:76-82. https://doi.org/10.1016/j.aca.2013.02.026
- [Google Scholar]
- Succinimidyl (3-[(benzyloxy)carbonyl]-5-oxo-1,3-oxazolidin-4-yl)acetate on a triazole-bonded phase for the separation of dl-amino-acid enantiomers and the mass-spectrometric determination of chiral amino acids in rat plasma. Journal of Chromatography A. 2019;1585:131-137. https://doi.org/10.1016/j.chroma.2018.11.061
- [Google Scholar]
- Determination of l-tryptophan and l-kynurenine derivatized with (R)-4-(3-isothiocyanatopyrrolidin-1-yl)-7-(N,N-dimethylaminosulfonyl)-2,1,3-benzoxadiazole by LC-MS/MS on a triazole-bonded column and their quantification in human serum. Biomedical Chromatography : BMC. 2016;30:1481-1486. https://doi.org/10.1002/bmc.3709
- [Google Scholar]
- Selectivity evaluation of phenyl based stationary phases for the analysis of amino acid diastereomers by liquid chromatography coupled with mass spectrometry. Journal of Chromatography. A. 2019;1590:80-87. https://doi.org/10.1016/j.chroma.2018.12.068
- [Google Scholar]
- Novel strategy for untargeted chiral metabolomics using liquid chromatography-high resolution tandem mass spectrometry. Analytical Chemistry. 2021;93:5805-5814. https://doi.org/10.1021/acs.analchem.0c05325
- [Google Scholar]
- Simultaneous determination of free DL-amino acids in natto with novel fluorescent derivatization by UPLC-FL. Food Analytical Methods. 2021;14:1099-1109. https://doi.org/10.1007/s12161-020-01959-1
- [Google Scholar]
- Nine amino acids are associated with decreased insulin secretion and elevated glucose levels in a 7.4-year follow-up study of 5,181 finnish men. Diabetes. 2019;68:1353-1358. https://doi.org/10.2337/db18-1076
- [Google Scholar]
- Impaired amino acid metabolism and its correlation with diabetic kidney disease progression in type 2 diabetes mellitus. Nutrients. 2022;14:3345. https://doi.org/10.3390/nu14163345
- [Google Scholar]
- Integrated glycomics strategy for the evaluation of glycosylation alterations in salivary proteins associated with type 2 diabetes mellitus. RSC Advances. 2020;10:39739-39752. https://doi.org/10.1039/d0ra05466f
- [Google Scholar]
- Simultaneous determination of free DL-amino acids in human hair with a novel DBD-M-Pro derivatization by UHPLC-HRMS: An application in diabetes patients. Journal of Pharmaceutical and Biomedical Analysis. 2024;251:116425. https://doi.org/10.1016/j.jpba.2024.116425
- [Google Scholar]
- Cerebrospinal fluid amino acid metabolite signatures of diabetic cognitive dysfunction based on targeted mass spectrometry. Journal of Alzheimer’s Disease : JAD. 2022;86:1655-1665. https://doi.org/10.3233/JAD-215725
- [Google Scholar]
- Chiral amino acid metabolomics for novel biomarker screening in the prognosis of chronic kidney disease. Scientific Reports. 2016;6:26137. https://doi.org/10.1038/srep26137
- [Google Scholar]