Abstract
Quantification of amino acids in biological samples is a critical tool for studying metabolism. Although many methods for amino acid analysis exist, important considerations include ease of sample preparation, dynamic range, reproducibility, instrument availability, and throughput. Here we present a simple, rapid, and robust method for analysis of amino acids by chemical derivatization and liquid chromatography – mass spectrometry (LC-MS). We provide a detailed protocol for analysis of 20 proteinogenic amino acids in biological samples which will enable straightforward implementation on modern LC-MS instruments.
Introduction
The most widely used method for amino acid analysis involves cation exchange chromatography followed by post-column derivatization with ninhydrin and detection by spectroscopy. Although robust, these methods require specialized instrumentation, have a narrow dynamic range, and have limited throughput. With the widespread adoption of liquid chromatography – mass spectrometry (LC-MS), many LC-MS based methods have been developed for quantification of amino acids. One of the major challenges to development of amino acid analysis methods by LC-MS is that amino acids have diverse chemical properties and do not retain well on traditional reverse phase columns. To circumvent this limitation, three general strategies have been developed: i) chemical derivatization where amino acids are covalently linked to a chemical that improves retention on reverse phase columns, ii) ion-pairing agents which are included in LC solvents to promote amino acid retention on reverse phase columns, and iii) hybrid columns that retain polar and non-polar amino acids or the use of multiple columns chosen to retain polar or non-polar amino acids.
Our laboratory, like many others, uses our LC-MS system for a wide range of targeted and untargeted applications. Therefore, it is critical that amino acid analysis methods do not require specialized column setups, multi-position valves, dual pumps, or solvent additives that may interfere with other assays run on the instrument. Ion-pairing agents are notoriously difficult to remove from tubing, columns, valves, etc. and can cause ion suppression in other analytical modes, so their use is advised for dedicated LC instruments. Hybrid columns or methods that use multiple columns requires a specialized instrument setup with multiple pumps and multi-port column selection valves which are not widely available on standard LC-MS instruments. Due to these limitations, our lab has chosen to employ pre-column chemical derivatization with 5-dimethylaminonaphthalene-1-sulfonyl chloride (dansyl chloride). Dansyl chloride (DNS) derivatization has been extensively used to analyze metabolites bearing primary and secondary amines, including amino acids (1–3), biogenic amines (4–6), polyamines (7), adenine nucleosides (8) among others. The derivatization method (Figure 1) is simple, robust, and enables retention of all amino acids on a single reverse phase column and also boosts signal in positive mode electrospray ionization. Rather than using internal standards for each amino acid which can be time-consuming and costly, we obtain robust stable isotope dilution mass spectrometry results using a limited set of internal standards. This method can be implemented on a wide variety of different LC-MS instruments using standard reverse phase columns and requiring limited assay validation. Although this method may not be suitable for all applications, we find that it exhibits excellent performance for routine amino acid analysis in a wide range of biological samples.
Figure 1.
Dansylation reaction. Dansyl chloride is incubated with amino acid standards or samples at room temperature in a sodium carbonate buffer for 60 minutes. Free amines react with dansyl chloride, yielding dansylated reaction products, which are well-retained on reverse phase columns. The tertiary amine also helps boost signal for dansylated products in positive mode electrospray ionization.
Here we provide a detailed method for the analysis of all 20 proteinogenic amino acids within biological samples using dansyl chloride derivatization. We provide parameters and chromatography conditions for analysis using a quadrupole time-of-flight mass spectrometer which can easily be adapted for use on other mass spectrometers such as triple quadrupoles and orbitraps.
2. Materials
2.1. Equipment
Eppendorf Thermomixer with adapter for 96-well plates (optional).
Vortex mixer.
Centrifuge with rotor capable of holding 96-well plates.
Fume hood.
Liquid chromatography – mass spectrometry system (triple quadrupole, quadrupole time-of-flight (Q-TOF) or orbitrap).
Reverse-phase C18 column.
Computer equipped with data analysis software.
2.2. Supplies
Polypropylene V-bottom 96-well plates (USA Scientific, 1833–9600).
Multichannel pipettes.
15 mL and 50 mL Falcon tubes.
Corning microplate lids (Corning, 3931).
Pierceable 96-well sealing mat (Thermo Scientific, AB0566) or Zone-free plate sealing film (Excel Scientific, ZAF-PE-50).
2.3. Reagents
All the solvents and water used in calibration curve, sample preparation, and LC-MS run should be HPLC grade or higher (see Note 1). Use of chemicals with 98% + purity is preferred. Methanol, acetonitrile, and ammonium hydroxide should be handled in a fume hood.
Amino Acid standards: make amino acid mixtures or use commercially available amino acid stocks, for example, Amino Acid Standard Solution from Sigma (Sigma AAS18). (see Note 2).
Internal standards (ISTD): stable isotope labeled amino acids are used as internal standards (see Note 3). Internal standards are dissolved in water as final 20 mM stocks and kept at −20 °C. Dilute ISTD to 0.1 mM as working solution before adding to samples or AA-STD.
Acetonitrile/Methanol (ACN/MeOH, 3:1) Extraction solution: 30 mL of acetonitrile (ACN) and 10 mL of methanol (MeOH) are freshly mixed and stored in 50 mL Falcon tubes to be used within 24 hours.
Dansyl chloride (DNS) derivatization reagents:
100 mM sodium carbonate/bicarbonate buffer, pH 9.8. Dissolve 240.9 mg sodium bicarbonate (mw: 84.01 g/mol) and 226.0 mg sodium carbonate anhydrous (mw: 105.99 g/mol) in 45 mL water, vortex to dissolve completely. Add water to final volume of 50 mL. Filter through a 0.22 μm filter and store at 4 °C. Warm up to room temperature before use.
50 mM DNS in 100% ACN. Dissolve 134.9 mg dansyl chloride (mw: 269.75 g/mol) in 10.0 mL ACN in 15 mL Falcon tube by vortexing. The stock will be slightly opaque. Store in dark and use within 24 hours (see Note 4). Immediately before derivatization, mix DNS and 100 mM sodium carbonate/bicarbonate buffer, pH 9.8 at 1:1 ratio. DNS is not stable at high pH. The mixed derivatization reagents should be used as soon as possible.
10% (v/v) ammonium hydroxide in water. Ammonium hydroxide is used to quench DNS reaction. Add 10 μL ammonium hydroxide (Honeywell Fluka, 4427310X1ML) in 90 μL water, mix well and use within 24 hours.
40% (v/v) acetonitrile with 0.01% formic acid. Mix 4 mL of acetonitrile with 6 mL of water in a 15 mL Falcon tube, then add 1 μL of formic acid. Mix well. Keep the cap tightly closed and use within 24 hours.
3. Methods
Reagent volumes are provided in Table 1. The provided example is in 96-well format. The volume can be increased proportionally for preparing samples in larger scales.
Table 1.
Reagent volumes for extraction and dansylation
| Components | Vol. (μL) | Ratio | Note |
|---|---|---|---|
| Samples or AA-STD | 25 | Total 1 vol. | The ratio of sample and ISTD can be changed as long as the final concentration of ISTD is the same in all samples and AA-STD. Matrix can be added here to account for matrix effect. |
| ISTD | 25 | ||
| 3:1 ACN/MeOH | 150 | 3 vol. | |
| Total | 200 | ||
| ACN/MeOH extracted samples or AA-STD | 25 | 1 vol. | Sample can be diluted in 50% ACN before added to derivatization reaction. |
| 50 mM DNS-Cl | 25 | 1 vol. | |
| 100 mM sodium carbonate/bicarbonate, pH 9.8 | 25 | 1 vol. | |
| Total | 75 | ||
| 10% ammonium hydroxide | 7.5 | 0.3 vol. | Quench the reaction. |
| Quenched DNS derivatization mixture | 8 | Dilution factor can be adjusted as needed. | |
| 40% acetonitrile (ACN) with 0.01% formic acid | 112 | ||
| Total | 120 |
1. Make amino acid standard curve (AA-STD)
Dilute amino acid stocks to final 1 mM of each amino acid in water and use as amino acid standard 1 (AA-St1). In a 96-well V bottom polypropylene plate, add 150 μL of AA-St1 to well A01, and aliquot 75 μL water to wells A02-A12. Make a 2-fold serial dilution of AA-St1 through wells A02-A11 in water by mixing 75 μL of amino acid with 75 μL of LC-MS water by pipetting up and down eight times, total 11 levels. Level 12 is water only and serves as a reagent blank (see Note 5).
2. Add ISTD to samples and AA-STD.
Mix 25 μL of sample or AA-STD (1 volume) with 25 μL of ISTD (1 volume of working solution) in 96-well V-bottom polypropene plate (see Note 6).
3. Acetonitrile/Methanol (ACN/MeOH) extraction.
Add 150 μL of ACN/MeOH to 50 μL of sample/ISTD and AA-STD/ISTD (3:1 volume ratio), mix by pipetting up and down 5 times. Cover the plate with a microplate lid to reduce evaporation. Spin down at 5,000 × g for 10 min, room temperature. Transfer 120 μL of supernatant to a new plate. Use 25 μL of extracted supernatant (sample can be further diluted in 50% ACN if the expected concentration of amino acids is too high) for DNS derivatization (see Note 7), save the remaining supernatant at −20 °C for short term storage or −80 °C for long term storage.
4. DNS derivatization in dark.
Immediately before use, 1:1 mix 100 mM sodium carbonate-bicarbonate mixture, pH 9.8 and 50 mM dansyl chloride (see Note 8) and aliquot 50 μL DNS/carbonate-bicarbonate buffer per well in a 96-well V-bottom plate. Then add 25 μL of ACN/MeOH extract from the step above (see Note 9), mix well by pipetting up and down 5 times. Seal the plate with sealing mat or cover the plate with a microplate lid to prevent evaporation. Incubate in Thermo mixer at 25 °C with lid and shake at 300 rpm for 60 min. If thermo mixer is not accessible, incubate the reaction in dark (i.e., in a drawer) for 30 min. Then shake it for 1–2 min on a shaker or mix the reaction by pipetting up and down. Continue the incubation in dark for total 1 h.
Briefly spin down the plates at 1,000 × g for 1 min at room temperature. Add 7.5 μL of 10% (v/v) ammonium hydroxide to each well (the vol. of 10% ammonium hydroxide = 1/10 vol. of total derivatization volume). Incubate in Thermo Mixer at 300 rpm, room temperature, for another 5 min to consume excess DNS (see Note 10). Briefly spin down the plates at 1,000 × g for 1 min at room temperature. Mix 8 μL of quenched reaction with 112 μL of 0.01% formic acid/40% ACN (see Note 11). Seal with pierceable 96-well sealing mats or Zone-free plate sealing films (see Note 12).
5. Separation by liquid chromatography and compound detection by electrospray ionization mass spectrometry.
Tune instrument according to manufacturer’s guidelines.
Inject samples via refrigerated autosampler into solvent and elute compounds using the gradient shown in Table 2.
Acquire data in positive ionization full scan mode using parameters listed in Table 2.
Table 2.
Liquid chromatography - mass spectrometry (LC-MS) parameters
| Dansylation derivatized positive mode | |||
|---|---|---|---|
|
| |||
| Parameters | Conditions | ||
|
| |||
| LC-MS | Agilent infinity II UPLC with 6545XT Q-TOF | ||
| Column | Waters BEH C18 1.7-mm particle size C18 column (2.1 × 100 mm) | ||
|
| |||
| Gradient | Time (min) | A (%) 0.1% formic acid in water | B (%) 0.1% formic acid in methanol |
| Begin | 0 | 55 | 45 |
| 7 | 50 | 50 | |
| 11 | 10 | 90 | |
| 12 | 10 | 90 | |
| 12.1 | 1 | 99 | |
| 13.7 | 1 | 99 | |
| 13.8 | 55 | 45 | |
| End | 16 | 55 | 45 |
|
| |||
| Flow rate | 0.4 mL/min | ||
| Injection volume | 2 μL | ||
| Column temperature | 60 °C | ||
| Ionization mode | ESI, positive mode | ||
| Gas temperature | 300 °C | ||
| Drying gas flow | 6 L/min | ||
| Nebulizer pressure | 30 psi | ||
| Sheath gas temperature | 275 °C | ||
| Sheath gas flow | 11 L/min | ||
| Capillary voltage | 4000 V | ||
| Fragmentor voltage | 180 V | ||
6. Data analysis
Use software from manufacturer for quantification of compounds by assigning retention times, accurate masses, and internal standards as listed in Table 3.
Manually inspect peaks to ensure integration is appropriate and adjust peak integration as necessary. Representative chromatograms for 20 amino acids are provided in Figure 2.
Table 3.
Retention times, quantifier ions, and measurement ranges for dansylated amino acid analysis by LC-MS.
| Amino Acid | Type | ISTDa | Dansylationb | Quantifier Ion (m/z) | RT (min) | Ion Polarity | Regression | Weight | Routine measurement range (nM)c |
|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||
| Arg-D7 | ISTD | n/a | 1 Dansyl | 415.2139 | 1.14 | Positive | n/a | n/a | |
| Gln-D5 | ISTD | n/a | 1 Dansyl | 385.1594 | 1.45 | Positive | n/a | n/a | |
| Pro-D7 | ISTD | n/a | 1 Dansyl | 356.1662 | 3.98 | Positive | n/a | n/a | |
| Phe-D5 | ISTD | n/a | 1 Dansyl | 404.1693 | 7.73 | Positive | n/a | n/a | |
| Leu-D10 | ISTD | n/a | 1 Dansyl | 375.2163 | 7.94 | Positive | n/a | n/a | |
| Lys-D8 | ISTD | n/a | 2 Dansyl | 621.2651 | 9.91 | Positive | n/a | n/a | |
| His-13C6-15N3 | ISTD | n/a | 2 Dansyl | 631.1907 | 10.5 | Positive | n/a | n/a | |
| Tyr-D4 | ISTD | n/a | 2 Dansyl | 652.2084 | 10.8 | Positive | n/a | n/a | |
|
| |||||||||
| Arg | Target | Arg-D7 | 1 Dansyl | 408.1706 | 1.14 | Positive | Linear | 1/x | 10–2500 |
| Asn | Target | Gln-D5, Phe-D5 | 1 Dansyl | 366.1124 | 1.38 | Positive | Linear | 1/x | 20–2500 (can be higher) |
| Gln | Target | Gln-D5 | 1 Dansyl | 380.1280 | 1.46 | Positive | Linear | 1/x | 20–2500 (can be higher) |
| Ser | Target | Arg-D7, Phe-D5 | 1 Dansyl | 339.1015 | 1.74 | Positive | Linear | 1/x | 40–2500 (can be higher) |
| Glu | Target | Gln-D5, Phe D5 | 1 Dansyl | 381.1121 | 1.75 | Positive | Linear | 1/x | 20–2500 (can be higher) |
| Asp | Target | Gln-D5, Phe-D5 | 1 Dansyl | 367.0964 | 1.76 | Positive | Linear | 1/x | 20–2500 (can be higher) |
| Gly | Target | Arg-D7, Phe-D5 | 1 Dansyl | 309.0909 | 2.08 | Positive | Linear | 1/x | 10–2500 |
| Thr | Target | Phe-D5 | 1 Dansyl | 353.1171 | 2.23 | Positive | Linear | 1/x | 20–2500 |
| Ala | Target | Arg-D7, Phe-D5 | 1 Dansyl | 323.1066 | 2.68 | Positive | Linear | 1/x | 20–2500 |
| Pro | Target | Pro-D7, Phe-D5 | 1 Dansyl | 349.1222 | 4.06 | Positive | Linear | 1/x | 10–2500 |
| Met | Target | Phe-D5 | 1 Dansyl | 383.1099 | 4.72 | Positive | Linear | 1/x | 5–2500 |
| Val | Target | Phe-D5 | 1 Dansyl | 351.1379 | 5.10 | Positive | Linear | 1/x | 20–2500 |
| Trp | Target | Trp-D5, Phe-D5 | 1 Dansyl | 438.1488 | 5.62 | Positive | Linear | 1/x | 20–2500 |
| Ile | Target | Leu-D10 | 1 Dansyl | 365.1535 | 7.69 | Positive | Linear | 1/x | 20–2500 |
| Phe | Target | Phe-D5, Leu-D10 | 1 Dansyl | 399.1379 | 7.87 | Positive | Linear | 1/x | 5–2500 |
| Leu | Target | Leu-D10 | 1 Dansyl | 365.1535 | 8.10 | Positive | Linear | 1/x | 20–2500 |
| Cystine | Target | Leu-D10 | 2 Dansyl | 707.1338 | 9.41 | Positive | Linear | 1/x | 40–2500 (can be higher) |
| Lys | Target | Lys-D8 | 2 Dansyl | 613.2155 | 9.92 | Positive | Linear | 1/x | 10–2500 |
| His | Target | His-13C6-15N3 | 2 Dansyl | 622.1795 | 10.5 | Positive | Linear | 1/x | 10–2500 |
| Tyr | Target | Tyr-D4 | 2 Dansyl | 648.1839 | 10.8 | Positive | Linear | 1/x | 10–2500 |
Internal standards are listed that in our hands work well for the indicated analytes. Other isotopes are likely to work as internal standards for these analytes and the reader is encouraged to explore more affordable or widely available isotopes during method development/implementation.
Amino acids that have two free amines or other reactive groups (such as the hydroxyl of tyrosine) become doubly dansylated.
Concentration ranges that we use in our laboratory are listed. Concentrations listed are for the diluted dansylated amino acids prior to injection on the LC-MS. The range could be broader based on the instrument used and the range of standard concentrations applied.
Figure 2.
Representative extracted ion chromatograms for dansylated amino acids. A standard solution of amino acids (Sigma AA-S18 plus Trp, Asn, and Gln) was dansylated as described in the method. Extracted ion chromatograms for each analyte is shown from the level 3 standard which corresponds to ~630 nM each at a final concentration prior to injection (Cystine is ~315 nM).
Acknowledgements.
We thank Manhong Wu (Stanford) for sharing initial dansylation protocols with us which we subsequently modified. This work was supported by NIH grants K08-DK110335, R35-GM142873 and R01-AT011396 and a career development award by the American Society of Nephrology.
Notes
Pour stock solutions out into a clean glassware or disposable falcon tubes for short term storage. Plastic serological pipettes are not compatible in acetonitrile and should be avoid for liquid transfer. Polypropylene pipettes are safe to use for organic solvent.
Except tyrosine, all other amino acids are dissolved in LC-MS grade water and kept at −20 °C. Up to 50 mM of tyrosine is dissolved in 0.2 N HCl with heating. Tryptophan, asparagine, and glutamine are not stable, so individual freshly prepared stocks are recommended. Cysteine is quickly oxidized to cystine in solution, and only the detection of cystine is provided in this method.
The best ISTD is the stable isotope version of the same amino acid, but amino acid isotopes with similar chemical properties can be used as ISTD for other amino acids (see Table 3).
Because we want to have excess DNS in the derivatization reaction, we make 50 mM of DNS stock, which is not completely soluble and will stay slightly turbid in ACN. The DNS solution will clear up after it mixes with carbonate/bicarbonate buffer. Both DNS and dansyl derivatives are light sensitive and light exposure should be minimized whenever possible.
The concentration of AA-St1 is adjusted based on the highest estimated amino acid concentration in samples. It should be at least 2-fold greater than the highest sample concentration. In our hands, derivatization reduces matrix effects, especially for in vitro culture samples. However, other sample matrices (i.e. plasma) may affect peak shape and data quality. In these cases, we recommend adding the same matrix (i.e. charcoal treated human serum for plasma samples) in the standard curve so that the proportion of the matrix in standards matches that in the samples. If matrix is added to the standard curve, one should make sure that no or only very small amounts of amino acids of interest are present in the matrix. If amino acids are present in the matrix for the STD and cannot be easily removed, one can generate standard curves using isotope labeled amino acids, carefully choosing an appropriate ISTD, and apply this standard curve for quantitation of the unlabeled amino acid in samples.
The ratio of sample and ISTD can be adjusted as long as the final concentration of ISTD is the same in all samples and AA-STD. The suggested concentration of internal standard is ~1/10–1/20 of the highest concentration of analyte in AA-St1.
Both ACN and MeOH are drippy, handle them quickly to ensure accurate pipetting. If electronic pipettes are available, add in an air gap step after taking up the liquid to minimize dripping.
Dansyl chloride hydrolyzes easily at high pH (dark yellow turns to light yellow or even clear), so the mixture needs to be used immediately, and make mixture for one plate each time.
The total amount of free amines in the sample or standard should be less than the amount of DNS, preferably only 1/3 of DNS to ensure full derivatization of all amino acids. Sample can be diluted in 50% ACN prior to derivatization.
After quenching the reaction with ammonium hydroxide, the deep yellow solution turns light yellow in a few minutes.
The dilution of the derivatization reaction can be adjusted. Make sure the final ACN percentage is around 40% to prevent precipitation of dansyl-derivatives which may occur in the refrigerated autosampler.
Because of the high proportion of ACN in final sample diluent, pierceable 96-well sealing mat is recommended to minimize evaporation in autosampler when many samples are being analyzed (e.g., more than one plate).
References
- 1.Furst P, Pollack L, Graser TA, Godel H, Stehle P. Appraisal of four pre-column derivatization methods for the high-performance liquid chromatographic determination of free amino acids in biological materials. J Chromatogr. 1990;499:557–69. [DOI] [PubMed] [Google Scholar]
- 2.Negro A, Garbisa S, Gotte L, Spina M. The use of reverse-phase high-performance liquid chromatography and precolumn derivatization with dansyl chloride for quantitation of specific amino acids in collagen and elastin. Anal Biochem. 1987;160(1):39–46. [DOI] [PubMed] [Google Scholar]
- 3.Tapuhi Y, Schmidt DE, Lindner W, Karger BL. Dansylation of amino acids for high-performance liquid chromatography analysis. Anal Biochem. 1981;115(1):123–9. [DOI] [PubMed] [Google Scholar]
- 4.Yamada H, Yamahara A, Yasuda S, Abe M, Oguri K, Fukushima S, et al. Dansyl chloride derivatization of methamphetamine: a method with advantages for screening and analysis of methamphetamine in urine. J Anal Toxicol. 2002;26(1):17–22. [DOI] [PubMed] [Google Scholar]
- 5.Nirogi R, Komarneni P, Kandikere V, Boggavarapu R, Bhyrapuneni G, Benade V, et al. A sensitive and selective quantification of catecholamine neurotransmitters in rat microdialysates by pre-column dansyl chloride derivatization using liquid chromatography-tandem mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci. 2013;913–914:41–7. [DOI] [PubMed] [Google Scholar]
- 6.Learey JJ, Crawford-Clark S, Bowen BJ, Barrow CJ, Adcock JL. Detection of biogenic amines in pet food ingredients by RP-HPLC with automated dansyl chloride derivatization. J Sep Sci. 2018;41(24):4430–6. [DOI] [PubMed] [Google Scholar]
- 7.Molins-Legua C, Campins-Falco P, Sevillano-Cabeza A, Pedron-Pons M. Urine polyamines determination using dansyl chloride derivatization in solid-phase extraction cartridges and HPLC. Analyst. 1999;124(4):477–82. [DOI] [PubMed] [Google Scholar]
- 8.Goodwin KJ, Gangl E, Sarkar U, Pop-Damkov P, Jones N, Borodovsky A, et al. Development of a quantification method for adenosine in tumors by LC-MS/MS with dansyl chloride derivatization. Anal Biochem. 2019;568:78–88. [DOI] [PubMed] [Google Scholar]