Download Spectroscopic method for rapid diagnosis

1st International Conference on Advancements of Medicine and Health Care through Technology, MediTech2007,
27-29th September, 2007, Cluj-Napoca, ROMANIA
Spectroscopic method for rapid diagnosis
M. Culea and O. Cozar
Abstract— Metabolic fingerprinting, metabolite profiling or metabolite target analysis are some strategies used in the metabolomics
analysis for rapid diagnosis by using different spectroscopic methods. Quantitative bioanalysis by using selected ion monitoring was
used for diagnosis of phenylketonuria and other metabolic diseases by screening small volumes of 20 µl of plasma or blood spots of
neonatal blood. The blood samples were derivatized as trifluoroacetylbutyl esters and analyzed by gas chromatography coupled with
mass spectrometry in the SIM mode. Regression curves for some standard amino acids are used for quantitative determination of
valine, leucine, proline, phenylalanine and tyrosine. Samples were separated on a Rtx-5MS capillary column, 30 m x 0.25 mm,
0.25µm film thickness, using a temperature program from 50oC(1min), then 6oC/min to 100 oC, 4oC/min to 200 oC, 20oC/min to 300
o
C in scan mode or 50oC (1min), then 20oC/min to 310 oC in the SIM mode. The instrument conditions were: transfer line
temperature: 250oC, injector temperature: 200 oC; ion source temperature 250 oC; Splitter: 10:1. Electron energy was 70eV and
emission current, 100µA. Quantitative determination of the amino acids of interest was made in the SIM mode by selecting the
highest ions of the mass spectrum of five amino acids od interest and the m/z 183 for the internal standard, 15N-labelled isoleucine.
Keywords: metabolic disease, fingerprint, GC/MS.
and other aminoacidemia by GC-MS is low cost method
[1]. The method is useful for diagnosis of metabolic
diseases as PKU by determination of phenylalanine (Phe)
and tyrosine (Tyr) in blood and maple syrup urine disease
(MSUD) by determination of valine (Val), leucine (Leu)
and proline (Pro) in blood. The analysis of the five amino
acids in blood samples by GC-MS is useful in the
diagnosis of the both diseases
1. Introduction
GC-MS demonstrates to be an indispensable method for
diagnosing inborn error of metabolism and is widely
recognized for its effectiveness in related fields [1].
Phenylketonuria (PKU) is a metabolic disease usually
caused by phenylalanine hydroxylase deficiency which
causes an increase of phenylalanine in plasma and a
decreased tyrosine, a specific pattern of blood amino
acids. The normal catabolism of phenylalanine in
mammals requires its conversion in tyrosine in liver.
Maple syrup urine disease (MSUD), other metabolic
disease, could be diagnosed by some branched amino
acids determination in blood [2]. Newborn amino acids
determination by gas chromatography–mass spectrometry
is a useful method for diagnosis of inborn errors in
metabolism [3].
2. Materials and methods
2.1 Chemical and samples
Acetyl chloride was purchased from Fluka, trifluoroacetic
anhydride was obtained from Merck (Darmstadt,
Germany), ion exchange resin Dowex 50W-X8 50–100
mesh was purchased from Fluka. Amino acids standards
were purchased from Sigma. [15N]-isoleucine (99%) was
produced by chemical synthesis. All other chemicals
were from Comchim (Bucharest).
The blood samples were obtained from patients and
voluntiers from the Pediatric Clinic III Cluj-Napoca.
Written informed consents were obtained from each
subject parent prior to this study.
PKU is general screened by a bacterial inhibition assay
(BIA) of elevate blood phenylalanine levels on newborn
filter paper samples of blood specimens. There are many
chromatographic methods for screening PKU, but also
mass spectrometry as electrospray mass spectrometry,
ESI-MS-MS.
A rapid method by profiling some amino acids and their
quantitative determination, a minim invasive method,
using 20 µl of blood was developed. Volatile derivatives
of amino acids were analyzed in very small volumes of
plasma or whole blood by using filter paper blood
specimens and GC-MS technique. PKU relies on the
amino acid profiling by mass spectrometry detection of
phenylalanine. Neonatal screening for phenylketonuria
2.2 Extraction procedure and derivatization
Two different method of extraction and derivatisation were
compared.
Extraction and derivatization procedure 1. The amino acids
were purified on a Dowex 50W-W8 exchange resin, on a
2x40mm column and eluted with 4M NH4OH. A two step
derivatization procedure was applied: esterification with
butanol-acetyl chloride (4:1 v/v) for 1 h at 110oC and
trifluoroacetylation with 200 µl trifluoroacetic anhydride at
60oC for 30 min. [2].
M. Culea is with the Babes-Bolyai University of Cluj-Napoca,
Romania, phone: +40-264-405-300; fax: +40-264-591-906; e-mail:
mculea @ phys.ubbcluj.ro.
T. C. Author is with Babes-Bolyai University of Cluj-Napoca,
Romania, phone: +40-264-405-300; fax: +40-264-591-906; e-mail:
cozar @ phys.ubbcluj.ro.
389
1st International Conference on Advancements of Medicine and Health Care through Technology, MediTech2007,
27-29th September, 2007, Cluj-Napoca, ROMANIA
4 0 AAs td s ca n # 2 2 2
R T: 7 .5 1
T: + c Fu l l m s [ 5 0 .0 0 -6 5 0 .0 0 ]
69
10 0
3
95
Extraction and derivatization procedure 2. The blood
was placed in a screw cap vial with 200 µl methanol/HCl
0.1% and extraction was obtained either 1h at 4oC or by
sonication 1 min. 100 µl were placed in a vial and
derivatized after the addition of the internal standard.
Amino acids in blood samples or standard samples were
derivatized as trifluoroacetyl butyl ester derivative.
Derivatization was followed in two steps, in screw-cap
tubes. Dry samples were esterified with 0.1 ml butanol:
acetyl chloride, 4:1, v/v) for 30 min at 100 °C. The excess
reagent was removed with a stream of nitrogen. The
amino group was acetylated with 100 µl trifluoacetic
anhydride (TFAA) at 100 °C for 30 min. After cooling,
the excess reagent was removed under nitrogen at ice
temperature and ethyl acetate was added.
AV:
1
NL:
1 .7 9 E7
182
140
90
85
80
75
R
e
l
a
t
i
v
e
A
b
u
n
d
a
n
c
e
70
65
60
55
50
45
40
35
30
25
153
20
114
15
5
183
166
1 26
70
10
1 71
139
67
58
71
78
96
93
82
113
99
141
138
115
154
142
184
180
1 94
0
60
80
100
120
140
160
180
197
200
m /z
4 0 AAs td s ca n # 2 6 5
R T: 8 .3 8
T: + c Fu l l m s [ 5 0 .0 0 -6 5 0 .0 0 ]
AV:
1
NL:
1 .6 4 E7
166
10 0
95
90
4
85
80
75
R
e
l
a
t
i
v
e
A
b
u
n
d
a
n
c
e
70
65
60
55
50
45
40
35
30
25
20
15
16 7
10
69
5
96
71
68
78
0
60
98
91
85
80
109
114
118
100
128
139
120
1 48
159
140
164
17 0
177
160
19 3
182
198
180
203
200
m /z
4 0 AAs td s ca n # 3 3 1
R T: 9 .7 3
T: + c Fu l l m s [ 5 0 .0 0 -6 5 0 .0 0 ]
AV:
1
NL:
2 .3 1 E7
91
10 0
95
90
5
85
80
75
R
e
l
a
t
i
v
e
A
b
u
n
d
a
n
c
e
70
65
60
1 48
55
50
45
40
35
30
25
147
1 03
20
131
15
77
65
10
69
5
63
60
119
92
78
104
102
89
96
76
0
80
118
113
100
204
149
132
120
120
146
150
158
140
178
1 71
160
198
189
180
200
m /z
AA_ 0 3 1 1 2 4 1 5 4 9 2 1 # 1 9 9 6
R T:
T: + c Fu l l m s [ 3 0 .0 0 -3 5 0 .0 0 ]
2 4 .2 8
AV:
1
N L:
6 .8 9 E5
260
10 0
203
95
90
6
85
80
75
R
e
l
a
t
i
v
e
A
b
u
n
d
a
n
c
e
70
65
60
55
50
45
40
35
30
25
20
69
10
Gly+15N-Gly
117
134
100
1
215
164
146
259
3 28
231
244
176
150
NL:
200
m /z
262
310
290
250
334
300
343
350
1 .5 9 E7
85
80
75
70
80
65
60
55
50
45
40
35
30
Thr
Ser
25
8.39
220
15
10.11
60
24.63
91
80
1 05
1 00
120
135
147
14 0
163
160
165
18 0
180
194
200
m /z
214
220
232
241
240
260
260
2 69
283
280
290
300
311
326
333
320
His
Figure 3. The mass spectra of Pro, Phe, Tyr and
uncompleted derivatized Tyr.
Met
Trp
Phe
24.90
40
77
79
36.30
Val
50
69
Lys
21.40
55
57
0
Glu
60
45
108
119
10
12.48
65
164
20
5
70
Asp
75
Relative Abundance
91
90
R
e
l
a
t
i
v
e
A
b
u
n
d
a
n
c
e
85
Leu
Ala
90
243
204
109
77
65
50
4 0 AAs td s ca n # 4 0 7
R T: 1 1 .2 8
AV:
T: + c Fu l l m s [ 5 0 .0 0 -6 5 0 .0 0 ]
107
10 0
7
95
Cistine
95
261
57
56
45
5
0
NL:
3.69E7
TIC MS
AA3
8.93
100
316
17 5
41
15
RT: 0.00 - 39.37
35
30
Tyr
18.75
2.71
20
27.58
10
31.18
36.99
A Trace DSQ ThermoFinnigan quadruple mass
spectrometer in the EI mode coupled with a Trace GC
was used. The capillary column Rtx-5MS was of 30 m
length x 0.25 mm, 0.25µm film thickness, by using two
different temperature programs: (1) from 50oC, 1 min,
6oC/min at 100oC, 4oC/min at 200oC, 20oC/min at 300oC,
(3min) and (2) from 50oC (1 min), then increased to
310oC, at 20 o C/min, in the selected ion monitoring mode
(SIM). Helium (99.9995%) carrier gas had a flow rate of
1 ml/min. The qualitative analysis was carried out in the
mass range 50-500 a.m.u. The following conditions were
followed: transfer line temperature 250oC, injector
temperature 200 oC; ion source temperature 250 oC;
Splitter: 10:1. Electron energy was 70eV and emission
current, 100µA.
24.30
4.36
5.51
5
2.3 Apparatus
29.64
15
Arg
25
13.57
35.93
32.86 34.83
25.80
14.18
21.73
0
0
5
10
15
20
Time (min)
25
30
35
Figure 1 Standard amino acid chromatogram: Ala, Gly, Thr,
Ser, Val, Leu, Met, Asp, Phe, Glu, Lys, Arg, His, Trp, Cys.
Table 1. The amino acids used for diagnosis of the metabolic
error diseases (PKU and MSUD)
Amino acid
Valine
Leucine
15
N-Isoleucine
(internal
standard)
Proline
Phenylalanine
Symbol
Val
Leu
Ions (SIM)
m/z 168
m/z 182
15
m/z 183
Pro
Phe
Tyrosine
Tyr
m/z 166
m/z 91,148
m/z 203, 260,316,
m/z 107,164,220
AA3 # 5 2 0
T: + c Fu l l
R T: 1 0.7 6
AV: 1
NL :
m s [ 3 0.0 0 -5 20 .0 0 ]
N-Ile (IS)
3. Results
Quantitative analysis of five amino acids in blood
samples by using two different methods of extraction,
derivatization and analysis showed similar results. The
coeficient of regression obtained by comparison the
amino acid values in gave r=0.91 (n=3). Figure 1 shows
the separation of the amino acids by using the first
method of analysis (from 0.5 ml blood) and Figure 4
shows the chromatogram of separation of five amino acid
by using the second method (minim invasive, from 20 µl
blood). In the method 2, the analysis beeing performed in
the SIM mode by using the ions selected from the
trifluoroacetyl butyl ester derivatives mass spectra:, m/z
168 for valine, m/z 182 for leucine (Figure 2), m/z 183
for 15N-isoleucine (15N-Ile), used as internal standard, m/z
166 for proline, m/z 91 and 148 for phenylalanine and
m/z 203, 260, 316 for tyrosine or m/z 107, 164, 220 for
1 .6 2E5
168
100
95
90
85
80
75
70
R
e
l
a
t
i
v
e
A
b
u
n
d
a
n
c
e
55
65
43
60
55
32
50
44
45
57
40
35
30
1 53
25
45
20
69
15
16 9
114
70
10
61
5
73
126
1 00
85
17 1
13 9
179
19 4
0
50
1 00
4 0 AAs td s ca n # 2 2 2
R T: 7 .5 1
T: + c Fu l l m s [ 5 0 .0 0 -6 5 0 .0 0 ]
69
100
3
95
AV:
1
150
NL:
22 7
2 07
200
m /z
2 40
28 2
252
250
2 87
3 15
329
30 0
341
3 50
1 .7 9 E7
182
140
90
85
80
75
R
e
l
a
t
i
v
e
A
b
u
n
d
a
n
c
e
70
65
60
55
50
45
40
35
30
25
153
20
114
15
5
1 66
126
70
10
183
171
13 9
58
67
71
78
82
93
96
99
11 3
115
138
141
142
154
180
0
60
80
100
120
140
160
180
184
194
197
200
m /z
Figure 2. The mass spectra of Val and Leu.
390
1st International Conference on Advancements of Medicine and Health Care through Technology, MediTech2007,
27-29th September, 2007, Cluj-Napoca, ROMANIA
mono trifluoroacetyl butyl ester derivative of tyrosine
(Figure 3). 25µg/ml of the internal standard was added at
each sample.
The mass spectra of the amino acids used for profiling
determination are presented in Figure 2 and Figure 3.
RT: 0.00 - 14.03
NL:
1.43E6
TIC F: MS
Ros tas 20u
l3
10.76
100
90
Relative Abundance
80
70
60
50
8.39
40
30
9.70
11.28
20
10
3.81
6.40
4.97
7.02
7.49
9.61
13.99
12.83 13.64
10.18
8.29
8.40
0
100
NL:
1.02E7
TIC F: MS
10aa2
11.28
90
80
70
7.02
9.71
60
50
40
7.50
30
10.77
20
6.40
10
3.81
0
0
1
2
3
4
4.97
5.94
5
6
7.57
7
Tim e (m in)
10.18
9.52
8
9
10
11.45
11
12.62 13.64
12
13
14
Figure 4. Amino acid screening of a blood sample and a
standard mixture of amino acids in the SIM-GC-MS mode
RT: 0.00 - 11.93
NL:
4.39E5
TIC F: MS
4514
9.68
100
90
Relative Abundance
80
70
60
50
40
30
20
8.35
10
5.08 5.45 6.68 6.98
0
100
11.25
7.45
9.38
10.00
NL:
2.33E5
8.38
TIC F: MS
40
90
80
70
60
9.69
50
The methods were validated by using 15 amino acid
standards (Figure 1), respectively 5 amino acids (Figure
4). The standards have followed the same extraction and
derivatization procedure (n=3) as samples. Precision gave
lower value than 19.81% for R.S.D., except Arg, Cys and
Met and sensitivity value lower than 10 ng of amino acid
injected. The regression curves were obtained by
injecting standard solutions containing amino acids in
concentration of 5, 10, 15, 20 and 40 µg/ml with 20
µg/ml of 15N-Gly added to each standard solution (method
1).
In the method 2, linearity was calculated by representing
the ratio of selected ion peak area for each amino acid
and the internal standard versus the amino acid standard
concentrations, in µg/ml. The regression curves were
obtained by injecting standard solutions containing amino
acids in concentration of 1, 5, 10, 20, 30 and 40 µg/ml
with 25 µg/ml of 15N-Ile added to each standard solution
(method 2), respectively per ml of blood sample. The
regression curves obtained were: Val: y = 0.0912x +
0.0555, r=0.999; Leu: y = 0.0515x - 0.0097, regression
coefficient r=0.998; Pro: y = 0.1835x - 0.1382, r=0.998;
Phe: y = 0.1119x - 0.0665, r=988; Tyr: y = 0.08x 0.1938, r=0.984. Precision and accuracy were studied by
extracting four times the standard solutions of 1, 30 and
40 µg/ml. RSD was obtained between 9- 12.9 % for
30µg/ml and 6,7 -18.6 for 40 µg/ml. For 1 µg/ml,
precision was between 45-50 %. Accuracy values were
between 4-30% for 30µg/ml and lower than 5.5 for
40µg/ml (n=4). Precision and accuracy were lower for
Tyr. The L.O.D. was lower than 0.1 µg/ml. The method
gave also very good values by using as internal standard
15
N-Gly, 25 µg, per ml.
7.01
40
30
6.72
7.48
Table 2. Precision and accuracy values (method 2)
11.28
20
7.56
10
5.06 5.48
0
0
1
2
3
4
5
6.40
6
Time (min)
8.26
7
8
8.70
9
10.19
10
Amino
acid
(n=4)
Val
Leu
Pro
Phe
Tyr
11
Figure 5. SIM-GC-MS comparison of screening of the five
amino acidc in a blood sample and a standard mixture of
amino acids (7.01 min:Val, 7.48min: Leu, 7.56:min: 15N-Ile
(IS); 8.38min: Pro; 9.69min: Phe and 10.19 and 11.28: Tyr
(complet and uncomplet derivatized).
The GC-MS was used in the scan mode in the range 50500 a.m.u. for qualitative determination of the amino
acids to obtain the mass spectra of each amino acids of
interest (Figure. 2 and 3) and to select the important ions
for the SIM mode for the more sensitive method 2.. The
peaks in the mass spectra of the five important amino
acids tested in the two metabolic disease are: m/z 168 for
Val, m/z 182 (the third important peak) for Leu, m/z 166
for Pro, m/z 91, 148 for Phe and m/z 203, 260, 316, m/z
107, 164, 220 for Tyr, as shown in Figure 2 and Figure 3.
The chromatogram of separation of the standard amino
acids is presented in Figure 1 (method 1) and Figure 4
(method 2). For Val, Leu and Pro, the important ions
selected in the SIM experiment from the trifluoroacetyl
butyl esters derivatives mass spectra correspond to the
loss of butyl ester from the molecular ion [M –
COOC4H9]+
Precision
RSD(%)
30 µg/ml
9.02
12.90
11.73
9.70
7.92
Precision
RSD(%)
40 µg/ml
12.82
6.73
8.68
18.60
8.22
Acc.
RSD(%)
30 µg/ml
3.75
0.15
13.33
24.67
30.44
Acc.
RSD(%)
30 µg/ml
0.65
1.66
2.11
1.76
5.54
Table 4 shows the comparison of the results obtained for
the amino acids average value in children blood samples
(n=53).
Table 3. The six spot amino acids values for the same blood
sample in a control child (n=6): R.S.D.(%) lower than 11.
Amino
acid
RSD(%)
Precision
(n=6)
Val
6.97
Leu
5.00
Pro
6.57
Phe
10.41
Tyr
5.06
391
1st International Conference on Advancements of Medicine and Health Care through Technology, MediTech2007,
27-29th September, 2007, Cluj-Napoca, ROMANIA
GC-MS is rapid, sensitive, selective, simple, less
expensive, measuring several amino acids at once.
Phe/Tyr values for PKU patients and controls
20
15
Phe/Tyr
Figure 6. Six blood spots used for SIM/GC/MS diagnosis
Table 4 The amino acids quantitation in control blood
samples, spot samples (n=53)
10
5
0
Amino acid
Val
Leu
Pro
Phe
Tyr
Phe/Tyr
µM
µg/ml
99.62
100.71
155.00
103.95
147.33
0.71
0
RSD (%)
11.66
13.19
17.82
17.15
26.67
0.64
29.45
45.31
52.70
68.31
85.86
21.73
24.56
23.67
134.13
38.99
21.47
16.97
19.85
89.75
19.4
23.22
25.84
25.64
142.7
35.91
22.16
33.16
65.66
131.3
57.2
21.51
19.92
31.42
119.3
33.53
1.86
3.44
4.63
3.97
2.30
3.56
8.9
10.86
10
8.95
Leu
23.16
41.3
35.59
36.57
Pro
16.34
40.09
43.52
43.68
Phe
262.33
608.71
548.85
491.65
Tyr
15.84
42.63
50.07
67.59
Phe/Tyr
16.56
14.28
10.96
7.27
40
Screening of plasma amino acids is necessary for the
diagnosis of metabolic diseases. The results showed that
GC-MS is a suitable method for PKU diagnosis in
neonatal blood samples, either by screening or
quantitation of some amino acids.
The method is a minim invasive by using very small
quantities of blood. The ratio of aliphatic amino acids to
aromatic amino acids could indicate other disorders of the
amino acid metabolism such as MSUD.
5. Acnowledgements
This paper is part of a CEEX project supported by
Romanian Scientific Research Agency.
6. References
[1] C. Deng, C. Shangb, Y. Hub , X. Zhang, Rapid
diagnosis of phenylketonuria and other aminoacedimias
by quantitative analysis od amino acids in neonatal blod
spots by gas chromatography-mass spectrometry, Journal
of Chromatography B, vol. 775,pp. 115- 120, 2002.
[2] M. Culea, A. Bucur, O. Cozar, Determination of
amino acids in blood samples by GC/MS, Analele Univ.
Vest Timisoara vol. 45, pp.89-92, 2004.
[3] C. Deng, X. Yin, L. Zhang, X. Zhang, Development of
microwave-assisted derivatization followed by gas
chromatography-mass
spectrometry
for
fast
determination of amino acids in neonatal blood samples,
Rapid Commun. Mass Spectrom; vol.19, pp. 2227-2234,
2005
[4]http://www.lifespan.org/adam/healthillustratedencyclopedi
Table 5. Amino acids values (µg/ml) in a PKU child blood
samples; spots were taken at different intervals of the day;
hour: 7, 9, 14 and 19.
Val
30
Figure 7. Phe/Tyr values from blood spots for PKU and
controls by SIM/GC/MS
Table 5 Amino acids values (µg/ml) in PKU blood samples,
spot samples
22.27
22.96
37.76
93.18
50.02
20
samples
The Comparison between intraindividual amino acid
values and interindividual variation are shown in Table 3
and 4. Good amino acid repetitivity in the same child,
R.S.D.lower than 10.4 %., was obtained Our results
obtained by using only 20 µl of blood spots showed that
PKU diagnosis could be tested by calculating the ratio
Phe /Tyr. Diagnosis of MSUD disease will be obtained
by calculating the ratio between aliphatic and aromatic
amino acids in the blood samples.
The results for some PKU patients by using method 2, are
presented in table 5.
Val
Leu
Pro
Phe
Tyr
Phe/
Tyr
10
a/1/003361.html
M. Culea, borned in Turda, 1951, is a Physicist,
PhD (1993), Associate Professor of Babes-Bolyai
University, Faculty of Physics, Department of
Biomedical Physics. The major research studies are
related of medical, clinical and environmental
applications of mass spectrometry coupled with gas
chromatography.
4. Conclusions
The BIA method is the semi quantitative method used for
PKU diagnosis but is with false positive up to 5% [1].
392