America n Journal of Analy tic al Chemistry, 2011, 2, 303-313
doi:10.4236/ajac.2011.23038 Published Online July 2011 (http://www.scirp.org/journal/ajac)
303
Determinati on of Asymm et ric Di m et hy larg inine and
Symmetric D imethyl argi ni ne in Biological Samples of Mice
Using LC/MS/MS
Daisuke Saigusa1, Mai Takahashi1, Yoshitom i Kanemitsu1, Ayako Ishida1, Takaaki Abe2,
Tohru Yamakuni3, Naoto Suzuki1, Yoshihisa Tomioka1*
1Laboratory of Oncology, Pharmacy Practice and Sciences, Graduate School of Pharmaceutical Sciences,
Tohoku University, Sendai, Japan
2Division of Medical Sciences, Graduate School of Biomedical Engineering, Tohoku University, Sendai, Japan
3Department of Pharmacotherapy, Graduate School of Pharmaceutical Sciences, Tohok u Uni ve r s i ty , Sendai, Japan
Received March 7, 2011; revised Apri l 2, 2011; accepted April 30, 2011
Abstract
Here in, we present a novel method of asymmetric dimethylarginine (ADMA) and symmetric dimethylargi-
nine (SDMA) determination within biological samples using protein precipitation and LC/MS/MS. Chroma-
tographic separation of ADMA and SDMA was succe ssfully performed using a silica column with optimized
elution, or mobile phase, of 10 mM ammonium acetate buffer H2O/methanol/acetonitrile (20/35/45, v/v) at
pH 4. The calibration ranges were 0.50 - 50.0 µg·mL–1, and good linearities were obtained for all compounds
(
r
> 0.99). The i nt ra- and inter-assay accuracies with recoveries and precisions at three concentration levels
(i.e. 1.00, 5.00 and 25.0 µg·mL–1) were better than 86.9% and 7.36%, respectively. The analytical perfor-
mance of the method was evaluated by determination of compounds in plasma, urine and tissues from male
BALBc/J mice. For the first time, we were able to characterize the distribution of ADMA, SDMA and AD-
MA/SDMA in plasma, urine, brain, heart, kidneys, liver, lungs, pancreas and spleen. Additionally, we dem-
onstrated that the ADMA/SDMA r atio in the brain was approximately 10-fold lower than all the other bio-
logical samples. Only 10 µL of plasma, 1 µL of urine and about 25 mg of tissues were required. These re-
sults suggest that the developed methodology was useful in ADMA and SDMA determination within bio-
logical samples.
Keywords: Asymmetric Dimethylarginine, Symmetric Dimethylarginine, Creatin in e , Arginine, Tissue,
Liquid Chromatography/Tandem Mass Spectrometry.
1. Introduction
Asymmetric dimethylarginine (ADMA) and symmetric
dimethylarginine (SDMA) are methylated by protein ar-
ginine (Arg) methyltransferases (PRMTs) from Arg and
metabolized by dimethylarginine dimethylaminohydro-
lase (DDAH) (Figure 1) [1,2]. ADMA and SDMA are
endogenous uremic toxins that are associated with chro-
nic kidney disease (CKD) and renal inflammation [3-6].
In fact, Toyoh ara, et al. reported that ADMA was a bio-
marker for CKD patients [7]. In previous studies, the
biological reactions associated with these conditions
have been primarily attributed to ADMA, whereas the
role of SDMA has been overlooked. However, it has
recently been reported that a ratio of Arg, ADMA and
SDMA is impor tant in the pathophysiological analysis of
cardiovascular diseases, reduced renal functions and ot h-
er diseases [6,8 ,9]. Therefore, a methodology for detect-
ing both ADMA and SDMA i s warranted. The analytical
methods of ADMA and SDMA have been developed
using several techniques. The enzyme-linked immuno-
sorb ent assay (ELISA) method has been shown to detect
cross-reacti vity with SDMA [10], albeit with low sensi-
tivity. Given that ADMA and SDMA are structural iso-
mers and the molecular weight is identical at 202.1,
chromatographic separation using high performance liq-
uid chromatography (HPLC) with ultra violet (UV), ra-
dioimmunoassay and fluorescence (FL) detection was

D. SAIGUSA ET AL. 304
shown to be necessary [11-15]. The first HPLC method
for dimethylarginine detection was reported in 1999 [12].
In 2000, an HPLC method with FL detection was devel-
oped, which had high sensitivity, reliability, and good
separation of ADMA and SDMA [13 ,15]. Unfortunately,
these HPLC methods are very time consuming. Since
2000, ADMA and SDMA have also been detected by
liquid chromatography/mass spectrometry (LC/MS) and
liquid chromatography/tandem mass spectrometry (LC/
MS/MS) [16-21]. Although detection of a selective pro-
duct ion with MS/MS fragmenta tio n can be difficult,
Bishop, et a l. and Zotti, et a l. have been reported that the
selective product ions for ADMA and SDMA are at m/z
203→46 and 203→172, respectively [17-19]. Addition-
ally, though these methods are more sensitive, ADMA
and SDMA have the different pattern of dissociation
using various MS system. Di Gangi, et al. have devel-
oped a reliable method using ultra per-formance LC/MS
/MS that is simple and has a short analytical time [21].
Thi s met ho d is sensitive, however preparation for deriva-
tization is ti me consuming, and ADMA and SDMA ha ve
not been separated successfully. The use of capillary
electrophoresis (CE) for ADMA and SDMA analysis has
also been reported. In fact, recent reports describe using
CE/MS/MS and CE/U V for det erminat io n o f AD M A a nd
SDMA [22,23]. Due to the high number of theoretical
plates, these methods have good separation of ADMA
and SDMA, high sensitivity and reliability. Ho wever, CE
is only appropriate for determining ADMA and SDMA
in plasma (i.e . it is not applicable in tissues or any bio-
logical samples). Additionally, homoarginine is not a
suitable internal standard (IS) because it is detected in
biological samples. Thus, a simple, un-deriva- tive,
highly sensitive and reliable method using LC/MS /MS
for ADMA and SDMA determination, which can also
separate isomers and determine ADMA and SDMA in
biological samples, is warranted. The purpose of the
present study is to: 1) develop a simple, sensitive and
reliable method for ADMA, SDMA, Arg and creatinine
(Cr) determination in plas ma, urine a nd tis sues, and 2) to
determine the distribution of ADMA and SDMA in bio-
logical samples of mice. Herein, we have described the
chromatographic separation of ADMA and SDMA fol-
lowing HP LC optimization.
2. Experimental
2.1. Chemicals
An ADMA standard was obtained from Sigma-Aldrich
(St. Louis, MO). SDM A and Cr standards were obtained
from Wako Pure Chemical Industries (Tokyo, Japan).
Arg and L-arginine-13C6 hydrochloride (Arg-13C6) were
obtained from Tokyo Chemical Industry (Tokyo, Japan)
and Cambridge Isotope Laboratories (Andover, MA),
respectively. Creatinine-d3 (methyl-d3) (Cr-d3) was ob-
tained from Toronto Research Chemicals (North York,
Ontario, Canada). Methanol (MeOH) and acetonitrile
(CH3CN) of LC/MS grade were obtained from Kanto
Chemical (Tokyo, Japan). Ammonium acetate
(CH3COONH4), acetic acid (CH3COOH) and for mic acid
(HCOOH) of LC/MS grade were obtained from Wako
Pure Chemical Industries. Ultrapure-grade water was
prepared with Purelab Ultra from Organo (Tokyo, Ja-
pan).
2.2. Mass spectrometry system and conditi ons
The MS system was a Thermo Fisher Scientific TSQ
Quantum Ultra triple quadrupole mass spectrometer
equipped with a heated electrospray ionization (HESI)
source. The operating conditions were optimized for
each compound by continuously infusing standard solu-
tions dissolved in water (10.0 µg·mL–1) at a rate of 5
µL·min–1. Our final analytical conditions for MS are
summarized in Table 1. HESI was performed in a posi-
tive ion mode (pos) for ADMA, SDMA, Arg, Cr, IS1
and IS2. Samples were analyzed using the selected reac-
tion monitoring (SRM) mode, and employing the transi-
tion of the (M + H)+ precursor ions to their product ions.
The MS/MS transitions were determined in the full scan
mode (m/z 30 - 250). For the MS/MS analysis, the opti-
mized tube lens offsets and collision energies for colli-
sion-induced dissociation (CID) of ADM A, SDM A, Ar g,
Cr, IS1 and IS2 are summarized in Table 1. The pos
HESI spray voltages were 1,500 V, the heated capillary
temperature was 380˚C, the sheath gas pressure was 65
psi, the auxiliary gas setting was 20 psi and the heated
vaporizer temperature was 380˚C. Both the sheath gas
and auxiliary gas were nitrogen gas. The collision gas
was argon at a pressure of 1.2 mTorr. The LC/MS/MS
Figure 1. Chemical structure of asymmetric di me thyl argi-
nine (ADMA) and symmetric dime thyl arginine (SDMA) .

D. SAIGUSA ET AL. 305
Table 1. Analy t i cal co nditions of HPLC and MS systems for determining ADMA, SDMA, Arg and Cr.
HPLC system
Analytical column
Guard column
Mobile phase
Flow rate
Oven temperat ure
Divert va lve
MS system
Ionization
Spray voltage
Vaporizer temperature
She at h gas pr essure
Auxiliary gas pressure
Cap illary temp er ature
Collision gas pressure
Tub e lens of f set
Collision energy
NANOSPACE SI-2 (Shiseido)
Mightysil Si 60 (250 × 3 mm I.D., 5 µm par ticle size)
CA P CEL LPA K C18 MGII (10 × 2 mm I.D., 3 µm particl e size)
10 mM CH3COONH4-H2O/MeOH/CH3CN=20/ 35/45 (pH = 4)
400 µL·min–1
40˚C
0 - 4 min: waste, 4 - 14.5 min: detector, 14.5 - 15 min: waste
TSQ quantum ultra (Thermo Fisher Scientific)
HESI (+)
1500 V
380˚C
65 psi
20 psi
380˚C
1.2 mTorr
ADMA, SDMA: 64, Arg: 54, Cr: 55, IS1: 54 and IS2: 55
ADMA, SDM A: 2 9 e V (m/ z 203.1 > 70.1), Arg: 24 eV ( m/z 175.1 > 70.1),
Cr: 18 eV (m/ z 114.1 > 44. 3) , IS1: 24 eV (m/z 181.1 > 74.1),
IS2: 18 eV (m/z 117.1 > 47. 3) , IS 1 : Arg -13C6, IS2: Cr -d3
syste m was c ontro lled by the Xcalibur software (Thermo
Fisher Scientific, San Jose, CA) and data were collected
with the same software.
2.3. Liquid chromatography system and condi-
tions
A NANOSPACE SI-2 LC system comprising an LC
pump, auto-sampl er, column oven and on-line degasser
(Shiseido, T okyo, Japan) wa s used. T he separations were
performed on a Mightysil Si (250 mm × 2 mm I.D., 5 µm
particle size) analytical column coupled with a CAP-
CELL PAK C18 MG II (10 mm × 2 mm I.D., 3 µ m pa r-
ticle size) (Shiseido, Tokyo, Japan) guard column main-
tained at 40˚C (Ta b le 1). The effect of the ratio of or-
ganic solutio n in mobile phase o n retention was tested b y
varying the percentage of MeOH and CH3CN. The per-
centages of MeOH and CH3CN were 10, 20, 30, 40, 50,
60, 70 and 80% while keeping the CH3COONH4 concen-
tration constant at 10 mM in the mobile phases. The
mixture of MeOH and CH3CN was as follows: 10 mM
CH3COONH4-H2O/MeOH/CH3CN. MeOH and CH3CN
ratios were varied between 10 and 80% while keeping
H2O at 20%. The pH of the mobile phase was between
3.5 and 6.7, and the flow rate was between 300 and 1000
μL·min–1. Retention time (R.T.) and resolution (Rs) val-
ues were used to evaluate the retention and separation of
ADMA and SDMA, and the R.T. and Rs values of
ADMA and SDMA were plotted against the content of
MeOH and CH3CN in the mobile phase. The Rs value
was calculated from equation mentioned below:
(R. T. of ADMA)(R. T. of SDMA)
Rs0.5(half-width of ADMA)(half-width of SDMA)
−
=×+
2.4. Calibration
All peaks were integrated automatically by the Xcalibur
software. The ADMA, SDMA and Arg amounts were
calculated fro m the calibration curves using the ratios of
their peak areas to that of IS1, and the Cr amounts were
calculated from the calibration curve using the ratios to
IS2. T he range used for the calibratio n c urves o f ADMA,
SDMA, Arg and Cr was between 0.500 - 50.0 µg·mL–1
(i.e. 2.47 - 247, 2.47 - 247, 2.87 - 287 and 4.42 - 442
µmol·L-1, resp ectively).
2.5. Validation of the analytical method
To determine the accuracy and precision, the newly de-
veloped method was validated at three concentrations
(1.00, 5.00 and 25.0 µg·mL–1) using fiv e samples on
three different days. The accuracies and precisions of the
method were determined through intra- and inter-day
analyses. Accuracy was calculated from the percentage
deviation from the mean of the true value, and precision
was expressed as the relative error and coefficient of
variation (CV, %). The data were validated based on
FDA’s Guidance for Industry: Bioanalytical Method
Validation guidelines.
2.6. Animals and collection of biological samples
Male BALBc/J mice (
3n=
) were kept in a r oom wit h a
12-h/12-h light-dark cycle (light cycle from 9:00 to
21:00) at 23˚C - 25˚C and provided water and food ad
libitum. All procedures used for LC/MS/MS were ap-
proved by the committee on the Care and Use of Expe-
rimental Animals, Tohoku University, in accordance
with the Guide for the Care and Use of Laboratory Ani-
mals publis hed by the U.S. N ational Institutes of Healt h.
To demonstrate the bio logical distr ibutio n of ADM A and
SDMA, we collected 9 biological sa mples (plasma, urine
and 7 tissues). Mice were first anesthetized with so-

D. SAIGUSA ET AL. 306
Figure 2. Precursor ion mass spectra of ADMA (a) and SDMA (b), and product ion mass spectra of ADMA (c) and SDMA
(d).
dium pentobarbital. Blood was then collected via retro-
orbital bleeding, transferred into a 1.5-mL plastic tube
with 30 µL of 10 mM ethylenediaminetetraacetic acid,
immediately centrifuged at 16,400 × g for 10 min, and
the supernatant was transferred into a new plastic tube.
Mice were then rapidly perfused transcardially with cold
saline through the left ventricle and thereby sacrificed.
Subsequently, urine and tissues (brain, liver, kidneys,
lungs, pancreas, spleen and he art ) were q uic kl y r e mo ve d.
All plasma, urine and pulverized tissues were stored at
–80˚C.
2.7. Sample preparation of plasma and urine
Sample preparation of plasma and urine for LC/MS/MS
was performed as follows. Plasma (10 µL) and urine (1
µL) were transferred into a 1.5-mL plasti c t ub e . Then, 5 0
µL of internal standard 1 (IS1; Arg-13C6 at 10 µg·mL–1)
and IS2 (Cr-d3 at 1 µg·mL–1), and 250 µL of 0.1%
HCOOH/CH3CN were added. The resulting mixture was
homogenized for 30 s in an ultrasonic bath. After centri-
fugation at 16,400 × g for 10 min, the supernatant was
transferred into a new p lastic t ube a nd evaporated at 60˚C
until dry under nitrogen gas stream. The residue was re-
constituted in 50 µL of mobile phase, vortexed for 30 s,
and passed through a filter (pore size: 0.2 µm, YMC).
Subsequently, 1 µL of the filtered solution was injected
into the LC/MS/MS system for analysis. The concentra-
tions of the individual compounds were calculated from
a regression of the calibration curves. The values were
calculated as the mean ± stand ar d d eviation (SD). Cr was
used as a biomarker to correct for different volumes or
urine produced per day [24]. The corrected values of
ADM A, SDM A and Arg in urine were calculated by the
concentration ratios of the individual compounds
(µg·mL –1) to the Cr concentration in ur ine (µg·mL–1).
2.8. Sample preparation of tissues
For LC/MS/MS analysis, tissue preparation based on
previously described methodology [25]. Approximately
25 mg of each tissue was transferred into a 2-mL plastic
tube, and 1000 µL of 0.1% HCOOH/CH3CN, 50 µL of
internal standard 1 (IS1; Arg-13C6 at 10 µg·mL–1) and IS2
(Cr-d3 at 1 µg·mL–1) were added. The resulting mixture
was homogenized for 30 s by a soni cator. After centrifu-

D. SAIGUSA ET AL. 307
Figure 3. Effect of MeOH percentage in mobile phase (H2O/MeOH) on retention time (R.T.) and resolution (Rs) (a, b) for
ADMA and SDMA. Effect of CH3CN percentag e in mobile ph ase ( H2O/MeOH/CH3CN) on R.T. and Rs (c, d). Effect of pH on
R.T. and Rs (e, f). Effect of flow rate on R.T. and Rs (g, h).
gation at 16,400 × g for 10 min, the supernatant was
transferred into a new plastic tube and the preparations
were following as described above.
2.9. Statistical analyses
For statistical analysis, we use Micro soft Office Excel
2007 software. The values were calculated as the me an ±
SD.
3. Resul t s and discussions
3.1. MS/MS and LC optimization
The ionization of ADMA, SDMA, Arg and Cr was per-
formed in positive ion mode. The optimized HESI of
ADMA, SDMA, Arg, Cr, IS1 and IS2 produced abun-
dant [M + H]+ ions at m/z 203.1, 203.1, 175.1, 114.1,
181.1 and 117.1, respectively. The conditions for MS/
MS detection were optimized for maximum product ion
formation through infusion analyses. The precursor and
product ion mass spectra of ADMA and SDMA in a
product ion scan mode were shown in Figure 2. The
selective product ion of ADMA and SDMA that has the
highest intensity from the precursor ion was the same at
m/z 203.1→70.1. Martens-Lobenhoffer et al. reported
the collision induced dissociation (CID) process of
ADMA and SDMA [26], and some pervious methods
selected different product ion [18,19]. In o ur MS system,
the product ions were as follows: m/z 40.1 for ADMA
and m/z 129.1 for SDMA. However, these ions were not
selective and not the highest i ons for ADMA and SDMA
determination. Consequently, a chromatographic separa-
tion of ADMA and SDMA was necessary in our study.
Quantification analyses were performed in the SRM
mode owing to the high selectivity and sensitivity of
SRM data acquisition, in which transitions from the pre-
cursor ion into the product ion were monitored at: m/z
203.1→70.1 for ADMA and SDMA; m/z 175.1→70.1
for Arg; m/z 114.1→44.3 for Cr; m /z 181.1→74.1 for
IS1; and m/z 117.1→47.3 for IS2.
The results of retention and chromatographic separa-
tion are shown in Fig ure 3. Previously, Paglia, et al.
described chromatographic separation tech n iques for

D. SAIGUSA ET AL. 308
Table 2. Linearity and correlation c oefficients of A DMA, SDMA, Arg and Cr.
Calibration range
Equaitiona
Correlat ion coefficient
ADMA
SDMA
Arg
Cr
0.50 - 50.0 µg·mL–1 (2.86 – 286 μm ol ·L–1)
0.50 - 50.0µg·mL–1 (4.27 – 427 μ mol·L–1)
0.50 - 50.0 µg mL–1 (2.87 – 287 μmol·L–1)
0.50 - 50.0 µg·mL–1 (4.42 – 442 μm o l·L–1)
y = 0.30x + 0.040
y = 0.67x + 0.014
y = 0.12x – 0.013
y = 0.82x + 0.018
r = 0.998
r = 0.996
r = 0.999
r = 0.999
a. x, analyte concentration (μmol·L–1); y, peak area ratio
Table 3. Accuracy of determination method f o r ADMA, SDMA, Arg and Cr.
Added per sample
(µg·mL–1)
Intr a-da y (n = 5) Inter-day (n = 3)
Day 1 Day 2 Day 3
ADMA
SDMA
Arg
Cr
1.00
5.00
25.0
1.00
5.00
25.0
1.00
5.00
25.0
1.00
5.00
25.0
1.05 ± 0.01
5.45 ± 0.05
27.0 ± 0.20
1.10 ± 0.01
5.65 ± 0.05
27.1 ± 0.20
0.90 ± 0.49
4.75 ± 0.06
25.2 ± 0.26
1.00 ± 0.04
5.05 ± 0.15
22.1 ± 0.15
0.95 ± 0.01
5.45 ± 0.04
25.3 ± 0.20
0.95 ± 0.02
5.60 ± 0.05
25.5 ± 0.20
0.90 ± 0.32
4.90 ± 0.09
25.1 ± 0.35
1.00 ± 0.06
5.10 ± 0.14
23.3 ± 0.43
0.95 ± 0.01
5.15 ± 0.05
24.9 ± 0.25
0.95 ± 0.01
5.85 ± 0.05
25.9 ± 0.25
0.90 ± 0.30
4.55 ± 0.06
24.3 ± 0.48
1.05 ± 0.04
4.90 ± 0.15
23.1 ± 0.60
0.98 ± 0.06
5.35 ± 0.17
25.7 ± 1.12
1.00 ± 0.09
5.70 ± 0.13
26.1 ± 0.83
0.90 ± 0.01
4.73 ± 0.18
24.9 ± 0.49
1.02 ± 0.03
5.02 ± 0.10
22.8 ± 0.64
Table 4. Precision of determination method for ADMA, SDM A, Arg and Cr.
Added per sample
(µg·mL–1)
Intr a-da y (n = 5) Inter-day (n = 3)
Day 1 Day 2 Day 3
ADMA
SDMA
Arg
Cr
1.00
5.00
25.0
1.00
5.00
25.0
1.00
5.00
25.0
1.00
5.00
25.0
0.19%
0.88%
0.69%
0.28%
1.00%
0.79%
5.84%
1.22%
0.97%
3.53%
3.40%
1.10%
1.02%
0.66%
0.80%
1.36%
0.82%
0.86%
3.08%
1.85%
1.35%
6.16%
2.58%
1.86%
0.87%
0.86%
0.93%
1.11%
0.99%
1.09%
3.30%
1.02%
1.99%
3.52%
3.02%
2.52%
6.32%
2.98%
4.20%
7.36%
2.86%
4.02%
1.66%
0.43%
0.49%
1.43%
0.43%
0.74%
ADMA and SDMA using a silica column [20]. Thus,
additional conditions of the mobile phase were tested to
obtain better retention and separation of ADMA and
SDMA.
Using either MeOH or CH3CN as the mobile phase did
not result in a good separation of ADMA and SDMA.

D. SAIGUSA ET AL. 309
Table 5. Value of ADMA, SDMA, Arg, Cr and ADMA/SDMA in biological samples (n=3).
Biolo gica l samples ADMA SDMA Arg Cr ADMA/SDMA
plasma (µg·mL–1)
urine (µg·mL–1)
(mg/mgCr)
brain (µg·mL–1)
hea r t (µg·mL–1)
kidney (µg·mL–1)
liver (µg·mL–1)
lung (µg·mL–1)
pancreas (µg·mL–1)
spleen (µg·mL–1)
0.289 ± 0.031
43.0 ± 8.4
0.085 ± 0.012
0.207 ± 0.026
1.17 ± 0.83
10.4 ± 0.3
7.50 ± 1.3
0.977 ± 0.256
3.14 ± 0.91
33.3 ± 4.71
0.047 ± 0.004
12.4 ± 1.8
0.023 ± 0.007
0.113 ± 0.011
0.069 ± 0.047
0.485 ± 0.122
0.596 ± 0.224
0.047 ± 0.016
0.237 ± 0.088
1.30 ± 0.36
15.4 ± 0.3
27.2 ± 20.7
0.046 ± 0.03
39.9 ± 0.4
34.6 ± 0.4
142 ± 9
3.91 ± 1.27
22.6 ± 6.8
129 ±116
168 ± 14
0.795 ± 0.060
553 ± 109
96.7 ± 2.4
110 ± 30
9.27 ± 1.96
4.04 ± 0.27
4.92 ± 1.56
37.8 ± 6.9
7.48 ± 1.26
6.13 ± 0.11
3.48 ± 0.63
1.86 ± 0.40
17.6 ± 0.16
22.3 ± 5.5
13.4 ± 3.8
21.4 ± 2.9
13.6 ± 1.5
26.1 ± 3.4
Retention in MeOH was much shorter than in CH3CN
because of interactions of ADMA and SDMA with the
silica column was stronger in MeOH than in CH3CN.
Although the Rs value increased depending on the per-
centage of MeOH, all of the peaks had poor shapes due
to tailing (data not shown). On the other hand, using a
mixture of MeOH and CH3CN rather than MeOH or
CH3CN alone resulted in better separation, retention and
peak shapes. In fact, 45% of CH3CN produced the best
retention and Rs value (Rs = 1.62). Because of hydro-
philic interac tions, the d istr ibutio n o f ADMA and SDMA
between the silica column and MeOH was too weak and
CH3CN was too strong. Thus, a mixture of MeOH and
CH3CN results in the best retention and separation of
ADMA and SDMA. Furthermore, retention time is im-
portant for high-throughput analysis. In 45% of CH3CN,
retention time was significantly increased for both AD-
MA and SDMA due to the strong hydrophobicity of
CH3CN, which retained the targeted compounds on the
silica column.
Additionally, pH dramatically affected the retention,
separation and peak shapes. ADMA and SDMA were
strongly retained at pH 5 and over. The Rs value at pH
3.5 was very low (Rs = 1.03) and the peak shapes and
intensity were worse at pH 4.5 and over. The flow rate
was determined from the Rs value of ADMA and
SDMA. The optimal Rs value was 1.5 at a flow rate of
400 µL·min –1. The best retention, separation and peak
shapes were achieved at pH 4 and a flow rate of 400
µL·min–1. These results indicate that the optimal mobile
phase consists of 10 mM CH3COONH4
H2O/MeOH/CH3CN (20/35/45, v/v) adjusted to pH 4
using CH3COOH. Separations were performed on a
Mightysil Si (250 mm × 2 mm I.D., 5 µm particle size)
analytical column coupled with a CAPCELL PAK C18
MG II (10 mm × 2 mm I.D., 3 µm particle size) (Shisei-
do, Tokyo, Japan) guard column maintained at 40˚C. A
valve was used to divert the LC effluent to waste during
the first 4 min and last 0.5 min of the chromatographic
run. Typical chromatograms of aqueous standard solu-
tions are shown in Figure 4. The retention times of
ADMA, SDMA, Arg, Cr, IS1 and IS2 were 13.1, 12.0,
9.2, 5.1, 9.2 and 5.1 min, respectively.
3.2. Limit of quantification (LLOQ) and lower
limit of detection (LLOD)
The lower limit of quantification (LLOQ) was consi-
dered as the lowest concentration that was measurable
with a CV of >20 and a signal-to-noise (S/N) ratio of
>10. The lower limit of detec tion (LLOD) was defined as
the concentration with an S/N of 3. The analytical pro-
cedure was sensitive with LLOQ and LLOD values for
ADM A, SD M A, Ar g and Cr of 2.47, 2.47, 2.87 and 4.42
µmol·L–1 and 0.742, 0.742, 0.862 and 1.33 µmol·L-1,
respectively. The LLOD values obtained using our me-
thodology were similar to those previously reported for
ADMA and SDMA determination in biological samples
[23]. However, the previously reported methodology
required 100 µL of plasma, whereas our methodology
required only 10 µL of plasma and 1 µL of urine. These
results suggest that the sensitivity of our novel met ho-
dology was 10-fold greater than the previously reported
met hodology.
3.3. Linearity
The linearity of the calibration curves was evaluated us-
ing seven concentrations (0.250, 0.500, 1.00, 2.50, 5.00,
25.0 and 50.0 µg·mL–1). The calibration curves of AD-
MA, SDMA, Arg and Cr were all linear over a range of
0.500 - 50.0 µg·mL–1 (2.47 - 247, 2.47 - 247, 2.87 - 287
and 4.42 - 442 µmol·L–1, respectively) (Table 2). A li-
near regression analysis was performed on these portions

D. SAIGUSA ET AL. 310
of the curves, and it was found that the correlation coef-
ficient was grea te r than 0.99 9 for all analytes.
3.4. Accuracy and precision
The results for the accuracy and precision are shown in
Tables 3 and 4, respectively. The intra- and inter-day
accuracies ranged from 86.9% to 112% for the three
concentrations used with all of the compounds. The in-
tra- and inter-day precisions ranged from 0.19% to
7.36% for the three concentrations used with all of the
compounds. These results indicate that our methodology
has good re liability and repeatability. Validation was de -
termined by spiking the standard compounds in plasma.
Generally, validation of analytical met hods is necessary,
since the effects of ion suppression are variable in dif-
ferent types of tissues, and this is achieved by spiking
each tissue sample. In our methodology, matrix effects
could be avoided by using a silica column and the opti-
mal mobile phase, which allows for chromatographic se-
paration of ADMA and SDMA in several biological
Figure 4. SRM chromatograms of m/z 114.1→44.3 for Cr,
m/z 117.1→47.3 for Cr-d3, m/z 175.1→70.1 for Arg, m/z
181.1→74.1 for Arg-13C6 and m/z 203.1→70.1 for ADMA
and SDMA from s tandard compounds.
samples.
3.5. Tissue distribution of ADMA and SDMA in
BALBc/J mice
ADMA, SDMA, Arg and Cr levels determined from
various tissue samples of mice are shown in Table 5.
Although, all compounds were detected, some samples
contained analytes at concentration levels outside the
linear working range. These samples were adjusted into
the working range of the calib ration and were reana-
lyzed. The SRM c hromatograms o f biolo gical sa mple are
sho wn in Figure 5. ADMA and S DMA levels were high
in urine, kidneys, liver, pancreas and spleen. With the
exception of low liver levels, Arg had similar urine and
tissue levels as ADMA and SDMA. Cr levels were high
in plasma, urine, brain, heart and pancreas. In previous
studies, Arg/ADMA or ADMA + SDMA/ monome t hy-
larginine (MMA) ratios were described to have an im-
portant role evaluating disease progression [27]. There-
for e, we calculated ADMA/SDMA ratios for tiss ue com-
parisons (Table 5). ADMA/SDMA ratios in the brain
Figure 5. SRM chromatograms of m/z 114.1→44.3 for Cr,
m/z 117.1→47.3 for Cr-d3, m/z 175.1→70.1 for Arg, m/z
181.1→74.1 for Arg-13C6 and m/z 203.1→70.1 for ADMA
and SDMA from a biological sample (plasma).

D. SAIGUSA ET AL. 311
and plasma were lower than in the other tissues. The
present study is the first to demonstrate lower level of
ADMA/SDMA in t he brain. DDAH is an enzyme that is
known to metabolize endogenous nitric oxide synthase
(NOS) inhibitors such as MMA and ADMA to citrulline
[28]. The gene expression of DDAH I and II has been
reported by Tra et a l., and D DAH I have been more dis-
tributed in the brain than the other organs [29]. In addi-
tion, DDAH I and NOS were up-regulated in neurons
following nerve injury [30]. These reports indicated that
ADMA might be expeditiously metabolized by DDAH I
in the b rain, a nd the central nerves system might be p ro-
tected by the low concentration of ADMA from the af-
fection of NOx via the inhibition with NOS. Although
the reason and mechanism for these findings cannot be
elucidated in the present study, the data suggests that the
brain a nd pl as ma ma y ha ve high levels of SDMA in rela-
tion to ADMA. It has been suggested that the relative
value is just as important as the absolute value for re-
vealing the mechanisms of ADMA and SDMA. Thus,
our novel methodology proved to be useful in detecting
ADMA and SDMA in various biological samples with
high sensiti vity and selectivity .
4. Concluding remarks
We have developed a methodology for ADMA and
SDMA determination in biological samples using LC/
MS/MS. Furthermore, it was validated that this metho-
dology can determine ADMA and SDMA levels with
high sensitivity and reliability. This metho dology is the
first of its kind that can determine tissue distribution of
ADMA and SDMA with good separation. Additionally,
this methodology requires small sample sizes, specifi-
cally only 10 µL of plasma, 1 µL of urine and about 25
mg of tissues were required in the present study. Fur-
thermore, the ADMA/SDMA ratio was found to be lower
in the brain than in any other tissues. Although the me-
chanism for this finding is unclear, the developed me-
thodology was use ful in determining ADMA a nd SDM A
levels in biolo gical samples.
5. References
[1] A. J. Pope, K. Karuppiah and A. J. Cardounel, “Role of
the PRMT-DDAH-ADMA axis in the Regulation of En-
dothelial Nitric Oxide Production,” Pharmacological Re-
serch, Vol. 60, No. 6, 2009, pp. 461-465.
doi:10.1016/j.phrs.2009.07.016
[2] C. Y. Ivashchenko, B. T. Bradley, Z. H. Ao, J . Leiper, P.
Vallance and D. G. Johns, “Regulation of the ADMA-
DDAH System in Endothelial Cells: A Novel Mechanism
for the Sterol Response Element Binding Proteins,
SREBP 1c and-2,” American Journal of Physiology-Heart
and Circulatory Physiology, Vol. 298, 2010, pp. 251-258.
doi:10.1152/ajpheart.00195.2009
[3] F. Kronenberg, “Emerging Risk Factors and Markers of
Chronic Kidney Disease Progression,” Nature Reviews
Nephrology, Vol. 5, 2009, pp. 677-689.
doi:10.1038/nrneph.2009.173
[4] J. T. Kielstein, D. Fliser and H. Veldink, “Asymmetric
Dimethylarginine and Symmetric Dimethylarginine: Axis
of Evil or Useful Alliance?” Seminars in Dialysis, Vol.
22, No. 5, 2009, pp . 346-350.
doi:10.1038/nrneph.2009.173
[5] L. Tarnow, P. Hovind, T. Teerlink, C. D. A. Stehouwer
and H. H. Parving, “Elevated Plasma Asymmetric Dime-
thylarginine as a Marker of Cardiovascular Morbidity in
Early Diabetic Nephropathy in Type 1 Diabetes,” Di-
abetes Care, Vol. 27, No. 3, 20 04, pp. 765-769.
doi:10.2337/diacare.27.3.765
[6] S. Abhary, N. Kasmeridis, K. P. Burdon, A. Kuot, M. J.
Whiting, W. P. Yew, N. Petrovsky and J. E. Craig, “Di-
abetic Retinopathy is Associated With Elevated Serum
Asymmetric and Symmetric Dimethylarginines,” Di-
abetes Care, Vol. 32, No. 11, 2009, pp. 2084-2086.
doi:10.2337/dc09-0816
[7] T. Toyohara, T. Suzuki, R. Morimoto, Y. Akiyama, T.
Souma, H. O. Shiwaku, Y. Takeuchi, E. Mishima, M.
Abe, M. Tanemoto, S. Masuda, H. Kawano, K . Maernura,
M. Nakayama, H. Sato, T. Mikkaichi, H. Yamaguchi, S.
Fukui, Y. Fukumoto, H. Shimokawa, K. Inui, T. Terasaki,
J. Goto, S. Ito, T. Hishinuma, I. Rubera, M. Tauc, Y. Fu-
jii-Kuriyama, H. Yabuuchi, Y. Moriyama, T. Soga and T.
Abe, “SLCO4C1 Transporter Eliminates Uremic Toxins
and Attenuates Hypertension and Renal Inflammation,”
Journal of the American Society of Nephrology, Vol. 20,
2009, pp. 2546-2555. doi:10.1681/ASN.2009070696
[8] J. T. Kielstein, R. H. Boger, S. M. Bode-Boger, J. Scha-
ffer, M. Barbey, K. M. Koch and J. C. Frolich, “Asym-
metric Dimethylarginine Plasma Concentrations Differ in
Patients with End-Stage Renal Disease: Relationship to
Treatment Method and Atherosclerotic Disease,” Journal
of the American Society of Nephrology, Vol. 10, 1999,
pp. 594-600.
[9] E. Schwedhelm, R. Maas, J. Tan-Andresen, F. Schulze,
U. Riederer and R. H. Boger, “High-Throughput Liquid
Chromatographic-Tandem Mass Spectrometric Det ermi-
nation of Arginine and Dimethylated Arginine Deriva-
tives in Human and Mouse Plasma,” Journal of Chroma-
tography B-Analytical Technologies in the Biomedical
and Life Sciences, Vol. 851, No. 1-2, 2007, pp. 211-219.
doi:10.1016/j.jchromb.2006.11.052
[10] F. Schulze, R. Wesemann, E. Schwedhelm, K. Sydow, J.
Albsmeier, J. P. Cooke and R. H. Boger, “Determination
of Asymmetric Dimethylarginine (ADMA) Using a Nov-
el ELISA Assay,” Clinical Chemistry and Laboratory
Medicine, Vol. 42, No. 12, 2004, pp. 1377-1383.
doi:10.1515/CCLM.2004.257
[11] A. Zinellu, S. Sotgia, E. Zinellu, A. Pinna, F. Carta, L.
Gaspa, L. Deiana and C. Carru, “High-Throughput CZE-
UV Determination of Airginine and Dimethylated Argi-

D. SAIGUSA ET AL. 312
nines in Human Plasma,” Electropho r es is, Vol. 28, No.
12, 2007, pp. 1942-1948. doi:10.1002/elps.200600534
[12] N. Rawal, Y. J. Lee, J. N. Whitaker, J. O. Park, W. K.
Paik and S. Kim, “Urinary-Excretion of
N-G-Dimethylarginines in Multiple-Sclerosis Pa-
tients-Preliminary-Ob- servations,” Journal of the Neu-
rological Sciences, Vol. 129, No. 2, 1995, pp. 186-191.
doi:10.1016/0022-510X(94)00277-U
[13] S. Sotgia, A. Zinellu, G. A. P inna, L. Deiana and C. Car-
ru, “A New Selective Pre-Column Ninhydrin-Based De-
rivatization for a RP-HPLC Determination of Plasma
Asymmetric Dimet hyl-L-Arginine (ADMA) by Fluores-
cence Detection,” Amino Acids, Vol. 34, No. 4, 2008, pp.
677-682. doi:10.1007/s00726-007-0001-x
[14] Y. Ohike, K. Kozaki, K. Iijima, M. Eto, T. Kojima, E.
Ohga, T. Santa, K. Imai, M. Hashimoto, M. Yoshizumi
and Y. Ouchi, “Amelioration of Vascular Endothelial
Dysfunction in Obstructive Sleep Apnea Syndrome by
Nasal Continuous Positive Airway Pressure-Possible In-
volvement of Nitric Oxide and Asymmetric NG, NG-
Dimethylarginine,” Ci rculation Journal, Vol. 69, No. 2,
2005, pp. 221-226. doi:10.1253/circj.69.221
[15] Y. Dobashi, T. Santa, K. Nakagomi and K. Imai, “An
Automated Analyzer for Methylated Arginines in Rat
Plasma by High-Performance Liquid Chromatography
with Po st column Fluorescence Reaction,” Analyst, Vol.
127, No. 1, 2002, pp. 54-59. doi:10.1039/b106828h
[16] K. Vishwanathan, R. L. Tackett, J. T. Stewart and M. G.
Bartlett, “Determination of Arginine and Methylated Ar-
ginines in Human Plasma by Liquid ChromAtogra-
phy-Tandem Mass Spectrometry,” Journal of Chromato-
graphy B, Vol. 748, No. 1, 2000, pp. 157-166.
doi:10.1016/S0378-4347(00)00399-6
[17] R. Maas, J. Tan-Andreesen, E. Schwedlhelm, F. Schulze
and R. H. Boger, “A Stable-Isotope Based Technique for
the Determination of Dimethylarginine Dimethylamino-
Hydrolase (DDAH) Activity in Mouse Tissue,” Journal
of Chromatography B, Vol. 851, No. 1-2, 2007, pp. 693-
701. doi:10.1016/j.jchromb.2007.01.020
[18] M. J. Bishop, B. Crow, D. Nor ton, E . Pali akov, J . George
and J. A. Bralley, “Direct Analysis of Un-Derivatized A-
symmetric Dimethylarginine (ADMA) and L-Arginine
from Plasma Using Mixed-Mode Ion-Exchange Liquid
Chromatography-Tandem Mass Spectrometr y,” Journal
of Chromatography B-Analytical Technologies in the
Biomed ical and Life Sci ences, Vol. 859, No. 2, 2007, pp.
164-169. doi:10.1016/j.jchromb.2007.09.024
[19] M. Zotti, S. Schiavone, F. Tricarico, M. Colaianna, O.
D'Apolito, G. Paglia, G. Corso and L. Trabace, “Deter-
mination of Dimethylarginine Levels in Rats Using
HILIC- MS/MS: An in Vivo Microdialysis Study,” Jou r-
nal of Sep ar at i on Science, Vol. 31, 2008, pp. 2511-2515.
doi:10.1002/jssc.200800147
[20] G. Paglia, O. D'Apolito, F. Tricarico, D. Garofalo and G.
Corso, “Evaluation of Mobile Phase, Ion Pairing, and
Temperature Influence on an HILIC-MS/MS Method for
L-Arginine and its Dimethylated Derivatives Detect ion,”
Journal of Separation Science, Vol. 31, No. 13, 2008, pp.
2424-2429. doi:10.1002/jssc.200800142
[21] I. M. Di Gangi, L. Chiandetti, A. Gucciardi, V. Moret, M.
Natural e and G. Giordano, “Simultaneous Quantitative
Determination of N-G,N-G-Dimethyl-L-Arginine or Asy-
mmetric Dimethylarginine and Related Pathway's Meta-
bolites in Biological Fluids by Ultrahigh-Performance
Liquid Chromatography/Electrospray Ionization-Tandem
Mass Spectrometry,” Analytica Chimica Acta, Vol. 677,
No. 2, 2010, pp. 140-148. doi:10.1016/j.aca.2010.08.011
[22] C. Desiderio , D. V. Rosset ti, I. Messana, B. Giardina and
M. Castagnola, “Analysis of Arginine and Methylated
Metabolites in Human Plasma by Field Amplified Sample
Injection Capillary Electrophoresis Tandem Mass Spec-
trometry,” Electrophoresis, Vol. 31, No. 11, 2010, pp.
1894-1902. doi:10.1002/elps.200900690
[23] A. Zinellu, S. Sotgia, M. F. Usai, G. Pintus, L. Deiana
and C. Carru, “Improved Method for Plasma ADMA,
SDMA, and Arginine Quantification by Field-Amplified
Sample Injection Capillary Electrophoresis UV Detec-
tion,” Analytical and Bioanalytical Chemistry, Vol. 23,
2010.
[24] D. Saigusa, N. Suzuki, M. Takahashi, K. Shiba, S. Tana-
ka, T. Abe, T. Hishinuma and Y. Tomioka, “Simultane-
ous Determination of Guanidinosuccinic Acid and Gua-
nidinoacetic Acid in Urine Using High Performance Liq-
uid Chromatography/Tandem Mass Spect rometry,” Ana-
lytica Chimica Acta, Vol. 677, No. 2, 2010, pp. 169- 175.
doi:10.1016/j.aca.2010.08.005
[25] A. Nakajima, D. Saigusa, N. Tetsu, T. Yamakuni, Y.
Tomioka and T. Hishinuma, “Neurobehavioral effects of
tetrabromobisphenol A, a brominated flame retardant, in
mice,” Toxicology Letters, Vol. 189, No. 1, 2009, pp. 78-
83. doi:10.1016/j.toxlet.2009.05.003
[26] J. Martens-Lobenhoffer, E. Schwedhelm and D. Tsikas,
“Quantification of Arginine and its Mono- and Dimethy-
lated Analogs NMMA, ADMA and SDMA in Biological
Fluids by LC-MS/MS: Is LC superfluous?” Journal of
chromatography B, Analytical technologies in the bio-
medica l and life sciences, Vol. 877, No. 27, 2009, pp.
3261-3266. doi:10.1016/j.jchromb.2009.07.003
[27] Z. Ajtay, A. Nemeth, E. Sulyok, A. Cziraki, S. Szabados,
J. Martens-Lobenhoffer, F. Awiszus, C. Szabo an d S. M.
Bode-Boger, “E ffect s of Stent Implementation on Plasma
Levels of Asymmetric Dimethylarginine in Patients with
or without ST-Segment Elevation Acute Myocardial In-
farction,” International Journal of Molecular Medicine,
Vol. 25, No. 4, 2010, pp. 617-624.
[28] M. Kimoto, G. S. Whitley, H. Tsuji and T. Ogawa,
“Detection of NG, NG-Dimethylarginine Dimethylami-
nohidrolase in Human Tissues Us ing a Monoclonal Anti-
body,” Journal of Biochamistry, Vol. 117, No. 2, 1995,
pp. 237-238. doi:10.1093/jb/117.2.237
[29] C.T.L. Tran, M.F. Fox, P. Vallance, and J. M. Leiper,
“Chromosomal Localozation, Gene Structure, and Ex-
pression Pattern of DDAH1: Comparison with DDAH2
and Implications for Evolutionary Origins,” Genomics,
Vol. 68, No. 1, 2000, pp. 101-105.
doi:10.1006/geno.2000.6262

D. SAIGUSA ET AL. 313
[30] S. Nakagomi, S. Kiryu-Seo, M. Kimoto, P. C. Emson and
H. Kiyama, “Dimethylarginine Dimethylaminohidrolase
(DDAH) as a Nerve-Injury-Associated Mol ecule: mRNA
Localozation with Neuronal NO Synthase (nNOS) in
Axotomizwd Motoneurons,” European Journal of Neuro-
science, Vol. 11, No. 6, 1999, pp. 2160-2166.
doi:10.1046/j.1460-9568.1999.00634.x