41 Amino acids analysis in serum by LC-MS/MS

Introduction

Amino acids are found throughout the body as they are the basic structural units that build proteins. Amino acids in serum have recently emerged as biomarkers for diagnostic purposes. Indeed, some disorders result in the accumulation or the deficiency of 1 or more amino acids in biological fluids such as serum. In addition, recent studies introduce amino acid profiling data as a potential new “omics” input into the metabolomics subset, to further investigate interaction networks.
Here we describe the analysis of 41 amino acids in serum by LC-MS/MS, with the aim of validating this method’s accuracy and robustness in our laboratory.

LC-SRM Analytical Strategy

We used a combination of solvents to extract amino acids from small samples of 25µl of serum. Then, we performed targeted LC-MS/MS analysis by LC-SRM (Selected Reaction Monitoring) on a triple quadrupole mass spectrometer. The precursor ion corresponding to an amino acid is first selected in the Q1 quadrupole, then fragmented in the Q2 collision cell. Specific fragment ions are selected in the Q3 quadrupole (Figure 1). The precursor ion/fragment ion pair is called a SRM “transition” and provides specificity and sensitivity to the quantitative analysis.

Figure 1 : Schematic diagram of Selected Reaction Monitoring

The multiplex capabilities of SRM make it the method of choice for the simultaneous analysis of 41 amino acids (Table 1) and their corresponding internal standards. Quantitative and qualitative detection was achieved for all compounds.

1-Methylhistidine	Glycine	4-Hydroxyproline	Proline	Arginine
alpha-Aminobutyric acid	Isoleucine	Carnosine	Tryptophane
Alanine	Lysine	gamma-Aminobutyric acid	Argininosuccinic acid
Allo-Isoleucine	Sarcosine	Homocitrulline	Glutamic acid
Asparagine	Valine	Methionine	Threonine
Glutamine	beta-Alanine	Phenylalanine	Tyrosine
Histidine	Citrulline	Phosphoethanolamine	Cystathionine
Phosphoserine	Cystine	3-Methylhistidine	Ethanolamine
Taurine	Hydroxylysine	alpha-Aminoadipic acid	Ornithine
beta-Aminoisobutyric acid	Leucine	Anserine	Serine

Table 1 : 41 amino acids screened by LC-SRM

Results

The following chromatogram shows data obtained for the screening of 41 amino acids in serum by rapid solvent extraction followed by quantitative LC-SRM.

Figure 2: Chromatogram of all 41 screened components

Mass spectrometry is an analytical technique that measures the mass-to-charge ratio (m/z) of ions. The specificity of the MS targeted analysis allows qualitative and quantitative analysis. Regarding amino acid profiling, spectrophotometric detectors were previously used but were unable to discriminate components with very similar retention time. LC-SRM mass spectrometry, on the other hand, is able to discriminate between leucine and isoleucine molecule fragment despite their equivalent mass and very similar retention time (Figure 3).

Figure 3 : example of isoleucine (top) and leucine (bottom) extracted chromatograms in serum

Internal standards corresponding to labeled heavy amino acids (in red) are used to validate peak retention times.
To evaluate the accuracy and the robustness of our method for amino acids analysis in serum sample, we determined inter-assay precision and accuracy for the 41 components. Two quality controls (QCs) were performed for each serum component (from 2 distinct serum samples), twice a day for 5 days (10 points for each component). The obtained accuracy parameters are listed in the tables below and summarized on figure 4:

QC 1
Component Name	Theoretical Concentration (µg/ml)	Num. Values	Mean	Standard deviation	CV %	Accuracy
1-Methylhistidine	0.65	10 of 10	0.69	0.03	4.1	105.6
alpha-Aminobutyric acid	0.98	10 of 10	1.03	0.04	3.5	105.0
Alanine	25.30	10 of 10	26.24	0.86	3.3	103.7
Allo-Isoleucine	1.68	10 of 10	1.76	0.07	4.2	104.6
Asparagine	7.79	10 of 10	8.23	0.38	4.6	105.7
Glutamine	80.10	10 of 10	86.14	4.11	4.8	107.5
Histidine	10.20	10 of 10	10.55	0.60	5.7	103.4
Phosphoserine	1.69	10 of 10	1.78	0.08	4.7	105.3
Taurine	9.90	10 of 10	10.40	0.43	4.2	105.0
beta-Aminoisobutyric acid	0.71	10 of 10	0.73	0.03	4.8	103.3
Arginine	7.56	10 of 10	7.70	0.51	6.6	101.9
Cystathionine	1.93	10 of 10	2.08	0.12	5.6	107.8
Ethanolamine	1.27	10 of 10	1.34	0.08	5.9	105.2
Glycine	25.90	10 of 10	26.34	1.27	4.8	101.7
Isoleucine	6.15	10 of 10	6.34	0.21	3.4	103.1
Lysine	8.37	10 of 10	8.51	0.51	5.9	101.7
Sarcosine	0.66	10 of 10	0.69	0.02	3.2	103.9
Valine	14.20	10 of 10	14.55	0.46	3.1	102.5
beta-Alanine	1.59	10 of 10	1.69	0.06	3.6	106.4
Citrulline	4.01	10 of 10	4.09	0.22	5.4	101.9
Cystine	3.31	10 of 10	3.44	0.32	9.4	104.0
Hydroxylysine	1.21	10 of 10	1.24	0.04	3.3	102.8
Leucine	15.30	10 of 10	16.23	0.68	4.2	106.1
Serine	18.60	10 of 10	19.62	0.64	3.2	105.5
Threonine	10.10	10 of 10	10.72	0.47	4.4	106.2
Tyrosine	10.20	10 of 10	10.91	0.52	4.8	107.0
4-Hydroxyproline	1.95	10 of 10	2.04	0.09	4.7	104.5
Carnosine	1.70	10 of 10	1.77	0.06	3.6	104.1
gamma-Aminobutyric acid	0.49	10 of 10	0.51	0.04	8.3	103.1
Homocitrulline	2.89	10 of 10	3.01	0.10	3.3	104.2
Methionine	2.67	10 of 10	2.79	0.07	2.7	104.4
Phenylalanine	21.60	10 of 10	22.33	1.05	4.7	103.4
Phosphoethanolamine	1.54	10 of 10	1.62	0.10	6.2	105.1
3-Methylhistidine	2.01	10 of 10	2.12	0.08	3.7	105.6
alpha-Aminoadipic acid	0.87	10 of 10	0.91	0.04	4.3	104.9
Anserine	1.09	10 of 10	1.12	0.05	4.2	102.7
Argininosuccinic acid	3.10	10 of 10	3.71	0.65	17.5	119.8
Glutamic acid	14.70	10 of 10	16.03	0.96	6.0	109.0
Ornithine	7.24	10 of 10	7.48	0.33	4.4	103.3
Proline	16.80	10 of 10	17.49	0.87	5.0	104.1
Tryptophane	5.77	10 of 10	6.28	0.52	8.3	108.9

QC 2
Component Name	Theoretical Concentration (µg/ml)	Num. Values	Mean	Standard deviation	CV %	Accuracy
1-Methylhistidine	1.73	10 of 10	1.84	0.07	3.7	106.1
alpha-Aminobutyric acid	5.19	2 of 2	5.37	0.19	3.6	103.5
Alanine	69.70	2 of 2	71.87	2.22	3.1	103.1
Allo-Isoleucine	11.40	2 of 2	11.94	0.45	3.7	104.7
Asparagine	21.50	2 of 2	22.67	1.20	5.3	105.5
Glutamine	183.00	2 of 2	192.94	6.79	3.5	105.4
Histidine	28.70	2 of 2	30.05	1.77	5.9	104.7
Phosphoserine	8.12	2 of 2	8.50	0.19	2.2	104.7
Taurine	34.50	2 of 2	35.67	1.03	2.9	103.4
beta-Aminoisobutyric acid	2.38	2 of 2	2.46	0.11	4.4	103.5
Arginine	27.50	2 of 2	27.84	1.44	5.2	101.2
Cystathionine	6.54	2 of 2	6.61	0.43	6.6	101.1
Ethanolamine	7.43	2 of 2	7.82	0.28	3.6	105.2
Glycine	63.60	2 of 2	66.19	2.19	3.3	104.1
Isoleucine	26.00	2 of 2	26.71	0.84	3.2	102.7
Lysine	46.50	2 of 2	49.09	2.11	4.3	105.6
Sarcosine	1.89	2 of 2	1.98	0.07	3.8	104.8
Valine	43.20	2 of 2	44.73	1.41	3.2	103.5
beta-Alanine	5.33	2 of 2	5.63	0.21	3.7	105.5
Citrulline	19.10	2 of 2	19.45	0.86	4.4	101.8
Cystine	20.20	2 of 2	20.86	2.79	13.4	103.3
Hydroxylysine	3.43	2 of 2	3.60	0.12	3.3	105.0
Leucine	48.80	2 of 2	50.67	1.14	2.2	103.8
Serine	47.30	2 of 2	48.90	1.69	3.5	103.4
Threonine	35.30	2 of 2	37.23	1.20	3.2	105.5
Tyrosine	33.60	2 of 2	35.82	1.48	4.1	106.6
4-Hydroxyproline	11.90	2 of 2	12.40	0.42	3.4	104.2
Carnosine	5.00	2 of 2	5.23	0.27	5.2	104.7
gamma-Aminobutyric acid	1.18	2 of 2	1.28	0.09	7.3	108.4
Homocitrulline	7.01	2 of 2	7.31	0.39	5.4	104.2
Methionine	12.80	2 of 2	13.35	0.83	6.2	104.3
Phenylalanine	67.10	2 of 2	69.71	2.24	3.2	103.9
Phosphoethanolamine	9.29	2 of 2	9.59	0.89	9.3	103.3
3-Methylhistidine	8.89	2 of 2	9.21	0.33	3.6	103.6
alpha-Aminoadipic acid	2.11	2 of 2	2.19	0.07	3.3	103.8
Anserine	2.63	2 of 2	2.73	0.14	5.1	103.8
Argininosuccinic acid	19.90	1 of 2	22.20	2.88	13.0	111.6
Glutamic acid	63.00	2 of 2	66.10	2.68	4.1	104.9
Ornithine	30.20	2 of 2	31.34	1.38	4.4	103.8
Proline	48.80	2 of 2	50.97	2.45	4.8	104.4
Tryptophane	24.90	2 of 2	27.82	2.53	9.1	111.7

Figure 4: Median accuracy (Right) of QC1 (n = 10) and QC2 (n = 2) as to the concentration of 41 amino acids in serum (100* experimental conc. / theoretical conc.) , and the related Coefficients of Variation (CV in %)

Coefficients of Variation (CVs) remain well below 20% for both QCs and all 41 compounds (5 days and two sessions per day) (the maximal value is 13.4%). The mean CV is 4.6% and 3.8% for both QCs, respectively. Our method tends to slightly overestimate amino acid concentrations; the mean accuracy is 104.5 and 104.2 for both QCs, respectively. Overall, accuracy and precision results validate LC-MS for the profiling of these 41 amino acids in serum.

Concluding remark

Amino acid profiling of biological samples generates specific and relevant data that could be advantageously used for clinical diagnostic purposes. Here we show that amino acid profiling in serum can be achieved in our laboratory with a robust LC-MS based method.