Rapid Separation of Amino-Acids using UHPLC with Pre-Column Derivatization

Introduction

Here we report on a method using a hiqh-speed automated amino-acid analysis using an X-LC system with a 2.0 mm I.D. x 50 mm L column packed with 2.0-µm dia. packing material. We have successfully reduced the original analysis time from 38 min to 7 min. This is an improvement by a factor of more than 5. Pre-column derivalization with OPA (orthophthalic aldehyde) was performed automatically using the X-LC 3059AS auto-sampler.

Amino-acid analysis is becoming more important in a variety of application fields, ranging from food analysis to protein science. A number of separation and detection methods are currently used. Among them, a combination of pre-column derivatization with OPA (orthophthalic aldehyde) and separation using reversed-phase with a C18 column and fluorescence detection is generally preferred due to the simplicity and high sensitivity.

A typical conventional method uses a 4.6 mm 1.0mm I.D x 150 or 250 mm L column packed with 5-µm dia. packing material that requires up to 40 mins to complete the analysis.

Experimental

Automated Pre-column Derivatization System

The derivatization of amino acids was performed using an autosampler program. The procedure for derivatizatlon is described in the text. Figure 2 shows the flow diagram of the auto-sampler.

Figure 2. Flow diagram of the autosampler (X-LC 3059AS)

Procedure for automated pre-column derivatization

Wash needle
Draw reagent and sample with air segmentation
Eject to reaction vial
Mix sample and reagent by air bubbling
Wash needle
Pause to allow reaction to complete
Draw derivatized sample
Injection onto column

(1) Optimization of derivatization conditions

Peak areas of 6 amino-acids were plotted against reaction time as shown in Figure 3. The derivatives of glycine and lysine showed appreciable decay of fluorescence intensity when reaction times became greater than 300 seconds. In addition, the lysine derivative gave a very low response compared with the derivatives of other amino acids. This phenomenon was also observed by Jones 1). We selected the reaction time of 30 seconds, as shown by the blue line.

Figure 3. Effect of reaction time on peak area

(2) Optimization of Separation Conditions

As shown in Figure 4, pH greatly affected the separation of histidine and serine. pH 5.8 was selected. The change in the column temperature greatly influenced separations shown in orange, green and blue in Figure 5. The column temperature was set to 40°C.

Optimized conditions
Eluent A: 1.0M citrate buffer (pH5.8) 3.5mL in 1L of H20
Eluent B: 1.0M citrate buffer (pH5.8) 3.5mL in 1L of CH3CN/C2H5OH/H2O (30/30/40)
Gradient condition: 1 cycle 10min
A:B = 90:10 – 90:10 – 72:28 – 72:28 – 42:58 – 42:58 – 23:77 – 0:100 – 0:100 – 90:10
0.2min 2.2min 2.5min 4.6min 5.0min 6.1min 6.15mln 7.0min 7.05min
Flow rate: 0.6mL/min
Column: X-PressPak V-C18 (2-µm dia., 2.0mm 1.0. x 5Omm)
Column temperature: 40°C
Detection: Fluorescence; Ex 345nm, Em 455nm; Gain x100

Figure 4. Effect of eluent pH on separation of histidine and serine

Figure 5. Effect of column temperature on separation of several amino acids

Results

This analysis was performed under the optimized conditions as shown in Figure 6. The blue line shows the gradient profile (%B).

Figure 7 compares X-LC with conventional HPLC. High-speed analysis was achieved using X-LC; the total analysis time was reduced to 1/5th. In addition, it is remarkable that the amount of solvent consumption was reduced to only 1/10th.

Figure 6. Analysis of standard mixture of 18 amino acids (20pmol/~L each)

Figure 7. Comparison between UHPLC and Conventional HDLC

(1) Precision of the Analysis

Reproducibility of Retention Time and Peak Area

Reproducibilities of the retention times and peak areas were calculated from the results of ten consecutive analyses of 8 amino-acid standards (20pmol each) as shown in Table 1. The relative standard deviations were between 0.048 and 0.429% for the retention times, and between 0.546 and 1.90% for the peak areas. Excellent reproducibilities of the retention times and peak areas were observed.

Linear Dynamic Range and Detection Limit

Figure 8 shows the linear dynamic range for selected amino acid derivatives. Correlation coefficients (r) were also calculated using the results from the analyses of 7 amino-acid derivatives ranging from 0.2 to 20pmol. Excellent linearity was observed.

Highly sensitive detection was achieved using a fluorescence detector; the detection limit (S/N=3) was between 21 and 49 fmol/µL as shown in Table 2.

(2) Applications in food analysis

Figure 9. Analysis of amino-acids in red wine

Figure 10. Analysis of amino-acids in aged rice vinegar.
Aged rice vinegar contains various amino-acids and minerals. Seventeen amino-acids were detected.

Figure 11. Analysis of amino-acids in functional beverage.
This beverage contained 9 essential amino-acids, i.e., His, Thr, Met, Val, Trp, Phe, lie, Leu and Lys.

Analysis of other amino-acids

Figure 12 shows a chromatogram of a standard mixture of 2 amino-acids containing GABA (y-aminobutyric acid). GABA is an important inhibitory neurotransmitter in the central nervous system and essential for brain metabolism and function. Asparagine and glutamine are contained in this standard sample besides GABA. The analysis conditions are the same as shown in Figure 6.

Figure 13 shows the chromatogram of amino-acids in white wine containing GABA. The label of this wine said that it contained lychee juice that is known as good source of GABA. The analysis conditions are the same as shown in Figure 6.

Figure 12. Analysis of standard mixture of 21 amino-acids containing GABA (20pmol/µL each)

Figure 13. Analysis of amino-acids in white wine containing GABA

Analysis of Theanine

Theanine is an amino-acid which is a derivative of glutamine. It is found almost exclusively in tea plants and it produces a feeling of relaxation. Theanine has a sweet and umami taste and shows the highest correlation to the quality of Japanese green tea. Figure 14 shows the chromatogram of standard mixture of 18 amino-acids containing theanine. The analysis condition is different from that in Figure 6. The standard mixture contained two internal standards, MBA and norvaline. MBA was used as an internal standard for the analysis of green tea, but norvaline is suitable for the analysis of black tea. Because it is difficult to separate impurity peaks from MBA.

Figure 15 shows the chromatogram of amino-acids of green tea. It should be noted that the green tea contained a large amount of theanine, and it was the major component in the amino- acids.

Analysis Condition (Theanine)

This analysis condition is aimed at the analysis of amino-acids containing theanine in tea. This analysis operates in a gradient elution mode with a binary ethanol-citrate buffer mobile phase system.

Eluent A: 1.0M citrate buffer (pH6.0) 3.5mL in 1L of C2H5OH/H2O (12/88)
Eluent B: 1.0M citrate buffer (pH6.0) 3.5mL in 1L of C2H5OH/H2O (50/50)
Gradient condition: 1cycle 14min
A:B = 100:0 – 0:85 – 0:85 – 100:0
9.0min 0.5min 0.05min
Flow rate: O.4mL/min
Column: X-PressPak V-C18 (2-µm dia., 2.0mm I.D. x 50mm)
Column temperature: 40°C
Detection: Fluorescence; Ex 340nm,Em 450nm; Gain xl 00

Figure 14. Analysis of standard mixture of 18 amino-acids containing theanine (O.4mg/L each)

Figure 15. Analysis of amino-acids in green tea

Conclusion

High-Speed Analysis
Total analysis time: ca 1/5 (38min-7min)
Reduction of Solvent Consumption
Amount ratio: ca 1/10 (60mL~6mL)
Highly Sensitive Detection
Detection limit (S/N=3): 21 – 49 fmol/L
Excellent Reproducibility (n=10)
RSD%: 0.05 – 0.43% (retention time), 0.58 – 1.90% (peak area)
Applicable in various Food Analyses

JASCO