Clinical Science (2009) 117, 2330 (Printed in Great Britain) doi:10.

1042/CS20080444 23

Arginine, citrulline and nitric oxide

metabolism in sepsis

Christina C. KAO , Venkata BANDI, Kalpalatha K. GUNTUPALLI,

Manhong WU , Leticia CASTILLO and Farook JAHOOR

USDA/Agricultural Research Service, Childrens Nutrition Research Center, Department of Pediatrics, Baylor College of
Medicine, One Baylor Plaza, Houston, TX 77030, U.S.A., and Section of Pulmonary, Critical Care, and Sleep Medicine,
Department of Medicine, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, U.S.A.

A B S T R A C T

Arginine has vasodilatory effects, via its conversion by NO synthase into NO, and
immunomodulatory actions which play important roles in sepsis. Protein breakdown affects
arginine availability and the release of asymmetric dimethylarginine, an inhibitor of NO synthase,
may therefore affect NO synthesis in patients with sepsis. The objective of the present study
was to investigate whole-body in vivo arginine and citrulline metabolism and NO synthesis rates,
and their relationship to protein breakdown in patients with sepsis or septic shock and in healthy
volunteers. Endogenous leucine flux, an index of whole-body protein breakdown rate, was
measured in 13 critically ill patients with sepsis or septic shock and seven healthy controls using
an intravenous infusion of [1-13 C]leucine. Arginine flux, citrulline flux and the rate of conversion
of arginine into citrulline (an index of NO synthesis) were measured with intravenous infusions of
[15 N2 ]guanidino-arginine and [5,5-2 H2 ]citrulline. Plasma concentrations of nitrite plus nitrate,
arginine, citrulline and asymmetric dimethylarginine were measured. Compared with controls,
patients had a higher leucine flux and higher NO metabolites, but arginine flux, plasma asymmetric
dimethylarginine concentration and the rate of NO synthesis were not different. Citrulline flux
and plasma arginine and citrulline were lower in patients than in controls. Arginine production
was positively correlated with the protein breakdown rate. Whole-body arginine production and
NO synthesis were similar in patients with sepsis and septic shock and healthy controls. Despite
increased proteolysis in sepsis, there is a decreased arginine plasma concentration, suggesting
inadequate de novo synthesis secondary to decreased citrulline production.

INTRODUCTION by NOS (NO synthase), and the arginineNO system

Arginine is a dispensable amino acid in normal health, but
may be conditionally essential in stressed states such as
sepsis [1]. In addition to acting as a substrate for protein
synthesis, arginine serves important physiological roles,
including ammonia detoxification to urea, cell growth
and differentiation, wound healing and immune function.
Arginine is also the only substrate for synthesis of NO
is important in the regulation of vascular tone and blood
pressure [2]. Low plasma arginine has been correlated
with a worse prognosis in patients with sepsis [3],
suggesting that there may be an overall increase in the
requirement for arginine that is not met by endogenous
production. At physiological concentrations of arginine,
eNOS (endothelial NOS) should be saturated; however,
supplementation of arginine in humans increases eNOS

Key words: arginine, citrulline, isotope, nitric oxide, septic shock.

Abbreviations: ADMA, asymmetric dimethylarginine; BMI, body mass index; DANS, 5-(dimethylamino)-1-napthalene
sulfonamide; GCRC, General Clinical Research Center; ICU, intensive care unit; IS, internal standard; -KICA, -oxoisocaproic
acid; MICU, medical ICU; NOS, NO synthase; eNOS, endothelial NOS; NOx, NO metabolites (nitrite and nitrate).
Correspondence: Dr Christina K. Kao (email ck692121@bcm.tmc.edu).


C The Authors Journal compilation 
C 2009 Biochemical Society
24 C. C. Kao and others
activity, a discrepancy known as the l-arginine paradox
[4]. Thus plasma arginine levels and arginine uptake can
limit NO synthesis [5]. Decreased NO synthesis may
underlie the poorer outcome seen in patients with sepsis
with low plasma arginine.
Although increased NO synthesis has been implicated
in the vasodilation and resistance to vasopressor drugs
seen in septic shock [6], a multicentre trial of the NOS
inhibitor NG -methyl-l-arginine in patients with septic
shock was stopped early because of increased 28-day mor-
tality in the drug-treated group [7]. On the other hand, in
patients with acute lung injury, of whom 25 % had sepsis,
higher urine NOx (NO metabolites; nitrites and nitrates),
indicating increased NO synthesis, were associated with
improved outcomes [8]. Taken together, these findings
suggest that adequate NO synthesis, and therefore argin-
ine availability, is critical for survival in sepsis. However,
the actual picture is not clear because the limited studies
of in vivo arginine and NO production in humans with
sepsis have produced conflicting results. Although one
study reported a faster NO synthesis but no difference
in arginine flux between children with sepsis and healthy
adults [9], another study reported a lower arginine flux
but no difference in NO synthesis in hypotensive adults
with sepsis compared with healthy controls [10].
Arginine is derived from the diet, de novo synthesis and
protein breakdown. Citrulline, the precursor for arginine
synthesis, is a non-protein amino acid produced from
intestinal glutamine metabolism [11]. In humans, the ma-
jor source of arginine is whole-body protein breakdown,
since de novo arginine synthesis constitutes only 515 %
of endogenous arginine flux [12]. Along with arginine
availability, endogenous inhibitors of NOS, including
ADMA (asymmetric dimethylarginine), may affect NO
synthesis [13]. Dimethylarginines are synthesized by the
methylation of arginine residues in proteins and are sub-
sequently released by proteolysis [14]. We hypothesized
that the rate of protein breakdown in patients with
sepsis would affect both arginine availability and ADMA
release and therefore the rate of NO synthesis. In the
present study we tested this hypothesis by determining
whole-body arginine and citrulline fluxes, the rate of NO
synthesis and plasma ADMA concentration and their
relationship to protein breakdown in patients with sepsis
or septic shock and in healthy controls.
MATERIALS AND METHODS
Subjects
The study was reviewed and approved by the Institutional
Review Board at the Baylor College of Medicine in
Houston, TX, U.S.A. Thirteen adult patients admitted
with sepsis to the MICU [medical ICU (intensive
care unit)] of Ben Taub General Hospital in Houston,
TX, U.S.A were enrolled in the study. Sepsis was

C The Authors Journal compilation 
C 2009 Biochemical Society
defined according to the International Sepsis Definitions
Conference [15]. All patients had clinical evidence of
infection and met the definition of sepsis. Five of the 13
patients also satisfied the criteria for septic shock. All
MICU patients had been diagnosed with sepsis in the
48-h period prior to the study, and nine of the patients
were studied within 24 h of admission. None of the
patients were receiving systemic corticosteroids or had a
diagnosis of malignancy or end-stage renal disease.
Seven healthy adult volunteers participated in the
study as control subjects. All control subjects were in
good health as established by medical history, physical
examination and blood chemistry measurements. They
were selected to be similar in age, gender and BMI (body
mass index) to the patients. All patients and controls were
enrolled after written, informed consent was obtained.
The present study was part of a larger investigation of
metabolic changes in patients with sepsis.
Isotope tracer infusion
Tracer infusions were performed in healthy subjects in
the adult GCRC (General Clinical Research Center) of
Baylor College of Medicine, and the patients with sepsis
were studied in the MICU of Ben Taub General Hospital.
Endogenous leucine flux (an index of whole-body
protein breakdown), arginine flux and citrulline
flux were measured with the tracers [1-13 C]leucine,
[15 N2 ]guanidino-arginine and [5,5-2 H2 ]citrulline respect-
ively. The conversion of [15 N2 ]guanidino-arginine into
[15 N]citrulline, an index of NO synthesis, was determ-
ined. Sterile solutions of [13 C]leucine, [15 N2 ]arginine,
[2 H2 ]citrulline and [15 N]citrulline (Cambridge Isotope
Laboratories) were prepared in 9 g/l of saline.
After an overnight fast, the healthy subjects were
admitted to the GCRC and an intravenous catheter was
placed in an antecubital vein for isotope infusions and in
a hand vein of the contralateral arm for blood sampling.
The hand was heated to arterialize blood samples. After a
baseline blood sample was obtained and primed constant
infusions of [15 N2 ]arginine (prime = 8 mol/kg of body
weight, infusion = 8 mol kg1 of body weight h1 )
and [2 H2 ]citrulline (prime = 1 mol/kg of body weight,
infusion=1 mol kg1 of body weight h1 ) were
administered for 6 h. The citrulline pool was also
primed with [15 N]citrulline (prime = 0.16 mol/kg of
body weight). At 2 h, a primed constant infusion of
[1-13 C]leucine (prime = 6 mol/kg of body weight,
infusion = 6 mol kg1 of body weight h1 ) was
started and maintained for 4 h. Additional blood samples
were collected at 30 min intervals during the last 90 min
of the infusions.
The same tracer infusions were performed in patients
with sepsis who had been fasting for at least 8 h. The
tracers were given through a pre-existing central venous
catheter or through an intravenous catheter placed in
the antecubital fossa. Blood samples were obtained

from pre-existing arterial or central venous catheters.
Although blood samples were drawn from arterial and
central venous sites in patients with sepsis and heated
dorsal hand veins in healthy controls, the use of
heated dorsal hand vein sampling as a surrogate for direct
arterial sampling has previously been validated [16,17].
Sample analysis
The blood samples were drawn into prechilled tubes
containing sodium fluoride and potassium oxalate.
The tubes were centrifuged immediately at 4 C, and the
plasma was removed and stored immediately at 70 C
for later analysis.
The plasma isotopic enrichment of -KICA (-
oxoisocaproic acid; -ketoisocaproic acid), a surrogate
of intracellular leucine, was measured by negative
chemical ionization GC/MS of its pentafluorobenzyl
derivative and monitoring of ions at m/z 129 and 130.
The plasma arginine and citrulline isotopic enrichments
were measured by tandem LC/MS. Plasma arginine
and citrulline were converted into their DANS [5-
(dimethylamino)-1-napthalene sulfonamide] derivatives
and analysed on a triple quadrupole mass spectrometer
(TSQ Quantum Ultra; Thermo Fisher Scientific),
equipped with an ESI (electrospray ionization) source,
a Survey pump (Thermo Fisher Scientific) and a HTC
PAL autosampler (Leap Technologies). The ions were
then analysed by SRM (selected reaction monitoring)
mode. The transitions observed were precursor ion m/z
408 to product ion m/z 170 at 34 eV for arginine, and
precursor ion m/z 409 to product ion m/z 392 at 14 eV
for citrulline. Instrumental control, data acquisition and
analysis were performed by the XCalibur (version 2.0)
software package (Thermo Fisher Scientific).
Plasma arginine, citrulline and glutamine concen-
trations were measured by standard ion-exchange
chromatography. Plasma concentrations of NOx were
measured by in vitro isotope dilution as previously
described [10]. Plasma concentrations of ADMA were
also measured by in vitro isotope dilution. Briefly,
200 l of the baseline plasma sample was spiked with
a known quantity of [2 H7 ]ADMA (Cambridge Isotope
Laboratories) as an IS (internal standard). ADMA was
then converted into its DANS derivative. The samples
were analysed by LC/MS on an Atlantis T3 3 m
2.1 mm 150 mm column (Waters). ADMA and SDMA
(symmetric dimethylarginine) were well separated
by HPLC. Plasma ADMA and its IS were analysed by
selected reaction monitoring at precursor ion m/z 436 to
product ion m/z 170 for ADMA, and precursor ion m/z
443 to product ion m/z 170 for its [2 H7 ]IS.
Calculations
The rate of appearance or total flux (Q) of leucine,
arginine and citrulline were calculated from the steady-
Arginine and NO in sepsis 25
state equation (eqn 1):
1
Q (mol kg of body weight h1 )
= (IEinf /IEplateau ) i (1)
where IEinf is the isotopic enrichment (mole percentage
excess) of leucine, arginine or citrulline in the infusate,
IEplateau is the isotopic enrichment of -KICA, arginine
or citrulline in plasma at the isotopic steady-state, and i
is the infusion rate of the tracer in mol kg1 of body
weight h1 . Endogenous leucine, arginine and citrulline
fluxes (the rates of production of the given amino acids by
the body) were determined by subtracting their infusion
rate, i, from Q.
Under steady-state conditions, the rate of appearance
of arginine equals the rate of disappearance. Therefore
(eqn 2):
1
Arginine clearance (ml kg of body weight min1 )
= QArg /plasma arginine concentration (2)
The NO synthesis rate was calculated from the rate
of conversion of arginine into citrulline via the NOS
reaction as previously described (eqn 3) [18]:
1
NO synthesis (mol kg of body weight h1 )
= QArgCit = QCit IECit /IEArg (QArg /IArg + QArg )
(3)
Where QArg and QCit are the fluxes of arginine and
citrulline, IECit is the plasma enrichment of the M + 1
isotopomer of citrulline (that is, ureido-[15 N]citrulline
derived from [15 N2 ]arginine), IEArg is the plasma
enrichment of the M + 2 isotopomer of arginine and IArg
is the rate of infusion of [15 N2 ]arginine.
Statistical analysis
Continuous variables were summarized by group as
means + S.E.M. unless otherwise indicated. Differences
between groups of subjects were assessed by the unpaired
Students t test. Tests were considered statistically
significant if P < 0.05. Correlations were performed using
Pearsons correlation (r). Data analysis was performed
with STATA software (version 9).
RESULTS
Subject characteristics
The characteristics of the patients and the controls are
presented in Table 1. There were no significant differences
in age, gender, ethnicity, weight or BMI between the two
groups. The diagnoses of the 13 patients with sepsis were
as follows: pneumonia in five patients, urosepsis in four
patients, cellulitis in three patients and septic arthritis
in one patient. Blood cultures drawn at admission from
11 of the 13 patients were positive. In eight patients,

C The Authors Journal compilation 
C 2009 Biochemical Society
26 C. C. Kao and others

Table 1 Demographic characteristics, vital signs and blood had a previous diagnosis of Type 2 diabetes mellitus. The
chemistry indices in controls (n = 7) compared with patients mean APACHE II score of the patients was 20 + 2. There
with sepsis (n = 13) was no significant difference in APACHE II scores in
Values are means +
S.D. P < 0.01, patients with sepsis compared with controls patients with shock compared with those without shock.
(measured by an unpaired Students t test); P < 0.05, patients with sepsis Four patients died during hospitalization.
compared with controls (measured using an unpaired Students t test). ALT,
alanine aminotransferase; AST, aspartate aminotransferase. Leucine, arginine, citrulline and NO
Characteristic/measurement Controls Patients with sepsis kinetics
Because leucine is an essential amino acid, in the fasted
Gender (male/female) (n) 4/3 8/5 state its flux is derived only from whole-body protein
Ethnicity (hispanic/non-hispanic) (n) 3/4 5/8 breakdown. Therefore the endogenous flux of leucine is
Age (years) 49 +
6 54 +
10 a reflection of the rate of whole-body proteolysis. Endo-
Weight (kg) 72.7 +
14.1 75.9 +
23.9 genous leucine flux was significantly higher in patients
BMI (kg/m2 ) 25.4 +
3.7 26.0 +
7.2 with sepsis than in the controls (P = 0.02; Table 2). There
Temperature ( C) 35.8 +
0.7 37.6 +
1.2

was no significant difference in arginine flux between
Mean arterial pressure (mmHg) 88 +
10 78 +
16 patients and controls (P = 0.60), but the plasma concen-
Heart rate (beats/min) 75 +
14 112 + 15

tration of arginine was significantly lower in patients
Respiration (breaths/min) 18 +
2 23 +
9 with sepsis compared with controls (P < 0.001; Table 2).
Leucocyte count (K/l) 6.3 +
1.6 15.3 +
8.4 Arginine clearance was significantly higher in patients
Creatinine (mg/dl) 0.9 +
0.2 1.7 +
0.8 compared with controls (P = 0.001; Table 2). Citrulline
AST (units/l) 18 +
5 88 +
99 flux was significantly lower in patients than in controls
ALT (units/l) 16 +
6 47 +
20

(P < 0.001), and plasma concentrations of citrulline and
its precursor, glutamine, were significantly lower in the
patients compared with controls (P < 0.001; Table 2). In
blood cultures grew Gram-positive organisms. Of these, patients with sepsis, arginine flux was positively correl-
five were Streptococci, two were Staphylococci and one ated with endogenous leucine flux (Table 3). Arginine flux
was Enterococcus. The remaining three patients had was also significantly correlated with endogenous leucine
blood cultures that grew Gram-negative organisms. Five flux in healthy controls (r = 0.92, P = 0.003).
patients met the criteria for septic shock. Six patients The rate of NO synthesis was not different in patients
were intubated and received mechanical ventilation with sepsis than in controls (Table 2), or in patients with
(three patients with septic shock and three patients with septic shock compared with patients without shock
sepsis but no shock). Four of the patients with sepsis (0.19 +
0.06 compared with 0.20 +
1
0.06 mol kg of

Table 2 Whole-body endogenous leucine, arginine and citrulline kinetics, the rate of NO synthesis, and plasma
concentrations of arginine, citrulline and glutamine in controls compared with patients with sepsis
(A) Whole-body endogenous leucine, arginine and citrulline kinetics and the rate of NO synthesis from 15 N-labelled citrulline in controls (n = 7) compared with
patients with sepsis (n = 13). (B) Plasma concentrations of arginine, citrulline and glutamine in controls (n = 7) compared with patients with sepsis (n = 13). Values
are means +
S.E.M. P < 0.05, patients with sepsis compared with controls (measured using an unpaired Students t test).
(A)

Parameter Controls Patients with sepsis P value

Endogenous leucine flux (mol kg1 of body weight h1 ) 89.7 +

3.0 130.8 +
11.8 0.02
Arginine flux (mol kg1 of body weight h1 ) 48.7 +
2.8 53.0 +
5.5 0.60
Arginine clearance (ml kg1 of body weight min1 ) 9.7 +
0.8 23.6 +
2.6 0.001
Citrulline flux (mol kg1 of body weight h1 ) 10.6 +
0.8 4.4 +
0.5 < 0.001
NO synthesis (mol kg1 of body weight h1 ) 0.15 +
0.04 0.20 +
0.04 0.47
(B)

Plasma concentration (mol/l) Controls Patients with sepsis P value

Arginine 85.5 +
3.3 40.2 +
3.8 < 0.001
Citrulline 21.4 +
2.5 10.2 +
0.8 < 0.001
Glutamine 354.1 +
21.8 151.8 +
19.0 < 0.001
NOx 27.1 +
5.2 55.0 +
7.7 0.02


C The Authors Journal compilation 
C 2009 Biochemical Society

Table 3 Correlations between different parameters in
patients with sepsis (n = 13)
P < 0.05. ALT, alanine aminotransferase; AST, aspartate aminotransferase.
Pearsons
Parameter correlation (r )
Endogenous leucine flux compared with 0.79
endogenous arginine flux
Arginine flux compared with NO synthesis 0.52
Endogenous leucine flux compared with NO 0.17
synthesis
NO synthesis compared with mean arterial 0.0016
pressure
Plasma ADMA compared with NO synthesis 0.27
Plasma arginine compared with plasma citrulline 0.77
Plasma citrulline compared with plasma 0.69
glutamine
Plasma ADMA compared with endogenous leucine 0.09
flux
Plasma ADMA compared with arginine flux 0.14
Plasma ADMA compared with creatinine 0.35
Plasma ADMA compared with ALT 0.27
Plasma ADMA compared with AST 0.28
body weight h1 ; P = 0.92). There was no significant
correlation between the rate of NO synthesis and
mean arterial pressure. However, plasma NOx levels
were significantly higher in patients with sepsis than
in controls (P = 0.02; Table 2). There was a positive
correlation between endogenous arginine flux and
NO synthesis rate which just failed to reach statistical
significance (P = 0.07; Table 3).
Although there was no difference in plasma concen-
trations of ADMA between patients with sepsis and
controls, patients with sepsis had a wider range of
ADMA concentrations (Figure 1). In the patients
with sepsis, the plasma ADMA concentration was not
significantly correlated with endogenous leucine flux,
the rate of NO synthesis, arginine flux, renal function
(creatinine) or liver function [ALT (alanine transaminase)
and AST (aspartate transaminase)] (Table 3). Three of
the four patients with the highest ADMA concentrations
died during their hospitalization.
DISCUSSION
The present study aimed to compare arginine and
citrulline flux, plasma ADMA concentration and the rate
of NO synthesis in patients with sepsis and septic shock
and healthy volunteers, and to determine the relationship
of these measurements to each other and to protein
breakdown. The results show that, although leucine flux,
an index of protein breakdown rate, was significantly
higher in patients with sepsis than in healthy controls,
Arginine and NO in sepsis 27
P value
0.001
0.07
0.57
1.0
Figure 1 Plasma concentrations of ADMA in patients with
0.38
sepsis (n = 13) compared with controls (n = 7)
0.002
In the patient cohort, black-filled circles represent patients who survived and
0.009
grey-filled circles represent patients who died.
0.78
0.65 there were no differences in arginine flux, plasma
0.24 ADMA concentration and the rate of NO synthesis
0.38 between patients and controls. However, higher arginine
0.35 clearance was associated with a lower plasma arginine
concentration, and both citrulline flux and its plasma
concentration were significantly lower in patients
than in controls. Furthermore, arginine production
was positively correlated with the protein breakdown
rate in patients with sepsis. These findings suggest
that, although more arginine is being released from
protein breakdown in patients with sepsis, its plasma
concentration is decreased because of increased clearance.
Decreased conversion of citrulline into arginine, leading
to inadequate de novo arginine synthesis, may also be
a contributing factor. Finally, synthesis of NO at the
whole-body level does not seem to be altered by sepsis.
In the fasted state, all of leucine and the majority
of arginine flux will be derived from the breakdown of
body proteins [12]. The positive correlation between
endogenous arginine and leucine flux was therefore
expected, and parallel increases in both arginine and
leucine fluxes have been shown in hypercatabolic
states, such as burn injury [19,20]. In the present study,
however, patients with sepsis had higher leucine flux,
indicating faster protein breakdown, but similar arginine
flux compared with controls. These results agree with
those of Argaman et al. [9] who reported that children
with sepsis had higher leucine flux, but similar arginine
flux compared with healthy adults. One explanation may
be that there is some compartmentalization of arginine
produced; for instance, arginine present in the urea cycle
may not be in equilibrium with the rest of the body pool
[21]. Alternatively, impaired de novo arginine synthesis
may explain the lack of increase in arginine flux despite
its accelerated release from proteolysis. De novo arginine
synthesis was not measured directly, because of the use
of the [5,5-2 H2 ]citrulline tracer. Castillo et al. [18] found

C The Authors Journal compilation 
C 2009 Biochemical Society
28 C. C. Kao and others
only a low level of arginine labelling with this tracer,
which was attributed to recycling of the 2 H2 label [18].
De novo arginine synthesis in patients with sepsis may
have been impaired because of decreased citrulline syn-
thesis. In animal studies, it has been shown that citrulline
is produced from intestinal glutamine and proline meta-
bolism, and the majority is taken up by the kidney and
converted into arginine [22,23]. A recent study in fasting
non-cirrhotic patients confirmed in vivo the presence
of the glutaminecitrullinearginine pathway in humans
and demonstrated a significant correlation between renal
citrulline uptake and arginine release [24]. In a study
in healthy humans, citrulline produced from glutamine
contributed to 64 % of de novo arginine synthesis [25]. In
the present study, our finding that patients with sepsis had
significantly lower citrulline flux and lower plasma con-
centration compared with healthy controls strongly sug-
gests that decreased citrulline production led to decreased
de novo arginine synthesis. This argument is further sup-
ported by the strong correlation between citrulline and
arginine concentrations, a finding also seen in critically ill
children [26]. Since gut citrulline synthesis is dependent
on glutamine and proline availability, the decreased
citrulline production seen in patients with sepsis probably
reflects a decrease in glutamine and/or proline availability
or metabolism. In support of this, we found that plasma
glutamine concentrations were markedly lower in the
patients with sepsis compared with the healthy controls,
and glutamine concentrations were strongly correlated
with citrulline concentrations. Furthermore, others have
reported lower plasma proline concentrations in patients
with sepsis [27]. Since glutamine is the primary precursor
of de novo arginine synthesis, it has been suggested
that glutamine supplementation may serve as a way of
stimulating arginine synthesis in sepsis [28]. However,
further information is needed on the metabolic interre-
lationships between glutamine, citrulline and arginine in
sepsis.
Argaman et al. [9] previously found that, despite
unchanged arginine flux, children with sepsis had a negat-
ive arginine balance owing to limited de novo synthesis
and increased oxidative losses. The authors concluded
that arginine is conditionally indispensable in sepsis. In
the present study, there was no difference in endogenous
arginine flux between the patients with sepsis and healthy
controls, plasma arginine concentration was markedly
lower in patients with sepsis, a finding which agrees with
previous reports [10,27]. The lower plasma arginine con-
centration is secondary to increased clearance of arginine,
which may be due to greater hepatic extraction or uptake
by peripheral tissues [27]. Arginine supplementation may
increase plasma arginine concentrations in sepsis and
restore positive arginine balance.
In earlier studies, the plasma concentration of NOx
was found to be markedly increased in patients with septic
shock [29,30], which led to the generally accepted hypo-

C The Authors Journal compilation 
C 2009 Biochemical Society
thesis that increased NO synthesis was responsible for
the hypotension associated with sepsis [6]. However, in a
large clinical trial, inhibition of NOS in sepsis improved
blood pressure and vascular resistance, but increased
mortality [7], suggesting that NO may be beneficial in
sepsis. In the present study, we found increased plasma
NOx concentrations in patients with sepsis compared
with controls, but no difference in the rate of NO
synthesis. Our findings are in accordance with a previous
study by our group [10], which found increased NOx
levels in hypotensive adults with sepsis compared with
controls, but no difference in NO synthesis. The in-
crease in NOx concentrations, without a concomitant
increase in the rate of NO synthesis, may be due to altered
renal function and extracellular volume changes in sepsis,
which may affect the estimation of NO production using
NOx concentrations [31]. Alternatively, measurement
of NO production by stable isotope methods may
not account for possible compartmentalization of
arginine intracellularly and within organs or differential
production of NO by different isoforms of NOS [32].
Although at the whole-body level NO synthesis is neither
increased nor decreased in patients with sepsis, further
understanding of the regulation of NO synthesis in sepsis,
particularly at an organ-specific level, is needed. For
example, maintaining adequate NO synthesis in sepsis
may be necessary for protection of the kidney by causing
local vasodilation and inhibiting platelet aggregation and
leucocyte adhesion [33].
Another regulator of NO synthesis in sepsis
is ADMA. In critically ill patients, increased plasma
ADMA has been shown to be a risk factor for ICU mor-
tality [34], death during ICU stay, duration of ICU
stay, duration of inotropic and vasopressor treatment,
mean APACHE II score and duration of ventilatory
support [35]. The major mechanism for elimination of
ADMA from the body is degradation by the enzyme
DDAH (dimethylarginine dimethylaminohydrolase)
[14]. Nijveldt et al. [36] hypothesized that ADMA
concentration increases in critically ill patients because
of increased release from proteolysis and decreased
elimination secondary to renal and hepatic failure. In the
present study, plasma concentrations of ADMA were not
different between patients and controls, although patients
had a much wider range of ADMA levels. There was no
difference in ADMA concentrations in patients who died
compared with those who survived; however, this was a
small group of subjects and was probably underpowered
to detect such a difference. Of clinical importance
is the fact that, of the four patients with the highest
ADMA levels, three of them died, which may support
previous evidence that ADMA is a marker of mortality;
however, there was no correlation between ADMA
concentrations and the rate of NO synthesis or renal or
hepatic function. It is possible that a static measurement
of plasma ADMA concentration is insufficient to reveal

the metabolic relationship between ADMA and NO
synthesis.
In summary, whole-body arginine production and
NO synthesis are similar in patients with sepsis
and septic shock compared with healthy controls.
Decreased de novo synthesis of arginine secondary to
decreased citrulline production may be an important
factor in arginine homoeostasis in sepsis. However,
the extent of contribution of the organ-specific or
intracellular pool of arginine to NO formation has not yet
been established. Further studies in sepsis are needed to
determine the interrelationships between arginine, citrul-
line and glutamine metabolism and the potential utility of
supplementation with arginine and/or glutamine. Finally,
our limited data on ADMA concentrations suggest that
it may be a marker of mortality in sepsis, and this should
be investigated in a larger cohort of patients.
ACKNOWLEDGEMENTS
We are grateful to the nursing staff of the General Clinical
Research Center at Baylor College of Medicine for their
care of the subjects and to Margaret Frazer, Melanie Del
Rosario and Megan Frelich for their assistance analysing
the samples in the laboratory.
FUNDING
This work was supported by the National Institutes
of Health [grant numbers T32HL007747 (to C. C. K.),
M01-RR00188 (to the GCRC)]; and with federal funds
from the U.S. Department of Agriculture, Agricultural
Research Service [Cooperative Agreement Number
58-6250-6001].
REFERENCES
1 Luiking, Y. C., Poeze, M., Dejong, C. H., Ramsay, G. and
Deutz, N. E. (2004) Sepsis: an arginine deficiency state?
Crit. Care Med. 32, 21352145
2 Bruckdorfer, R. (2005) The basics about nitric oxide. Mol.
Aspects Med. 26, 331
3 Freund, H., Atamian, S., Holroyde, J. and Fischer, J. E.
(1979) Plasma amino acids as predictors of the severity and
outcome of sepsis. Ann. Surg. 190, 571576
4 Tsikas, D., Boger, R. H., Sandmann, J., Bode-Boger, S. M.
and Frolich, J. C. (2000) Endogenous nitric oxide synthase
inhibitors are responsible for the l-arginine paradox. FEBS
Lett. 478, 13
5 Cynober, L. A. (2002) Plasma amino acid levels with a note
on membrane transport: characteristics, regulation, and
metabolic significance. Nutrition 18, 761766
6 Landry, D. W. and Oliver, J. A. (2001) The pathogenesis of
vasodilatory shock. N. Engl. J. Med. 345, 588595
Arginine and NO in sepsis 29
7 Lopez, A., Lorente, J. A., Steingrub, J., Bakker, J.,
McLuckie, A., Willatts, S., Brockway, M., Anzueto, A.,
Holzapfel, L., Breen, D. et al. (2004) Multiple-center,
randomized, placebo-controlled, double-blind study of the
nitric oxide synthase inhibitor 546C88: effect on survival in
patients with septic shock. Crit. Care Med. 32, 2130
8 McClintock, D. E., Ware, L. B., Eisner, M. D.,
Wickersham, N., Thompson, B. T. and Matthay, M. A.
(2007) Higher urine nitric oxide is associated with
improved outcomes in patients with acute lung injury.
Am. J. Respir. Crit. Care Med. 175, 256262
9 Argaman, Z., Young, V. R., Noviski, N., Castillo-Rosas, L.,
Lu, X. M., Zurakowski, D., Cooper, M., Davison, C.,
Tharakan, J. F., Ajami, A. and Castillo, L. (2003) Arginine
and nitric oxide metabolism in critically ill septic pediatric
patients. Crit. Care Med. 31, 591597
10 Villalpando, S., Gopal, J., Balasubramanyam, A., Bandi,
V. P., Guntupalli, K. and Jahoor, F. (2006) In vivo arginine
production and intravascular nitric oxide synthesis in
hypotensive sepsis. Am. J. Clin. Nutr. 84, 197203
11 Moinard, C. and Cynober, L. (2007) Citrulline: a new
player in the control of nitrogen homeostasis. J. Nutr. 137,
1621S1625S
12 Wu, G. and Morris, Jr, S. M. (1998) Arginine metabolism:
nitric oxide and beyond. Biochem. J. 336, 117
13 Boger, R. H. (2004) Asymmetric dimethylarginine, an
endogenous inhibitor of nitric oxide synthase, explains the
l-arginine paradox and acts as a novel cardiovascular risk
factor. J. Nutr. 134, 2842S2847S
14 Tran, C. T., Leiper, J. M. and Vallance, P. (2003) The
DDAH/ADMA/NOS pathway. Atherosclerosis Suppl. 4,
3340
15 Levy, M. M., Fink, M. P., Marshall, J. C., Abraham, E.,
Angus, D., Cook, D., Cohen, J., Opal, S. M., Vincent, J. L.
and Ramsay, G. (2003) 2001 SCCM/ACCP/
ATS/SIS International Sepsis Definitions Conference.
Crit. Care Med. 31, 12501256
16 Abumrad, N. N., Rabin, D., Diamond, M. P. and Lacy,
W. W. (1981) Use of a heated superficial hand vein as an
alternative site for the measurement of amino acid
concentrations and for the study of glucose and alanine
kinetics in man. Metab. Clin. Exp. 30, 936940
17 Copeland, K. C., Kenney, F. A. and Nair, K. S. (1992)
Heated dorsal hand vein sampling for metabolic studies: a
reappraisal. Am. J. Physiol. 263, E1010E1014
18 Castillo, L., Chapman, T. E., Sanchez, M., Yu, Y. M.,
Burke, J. F., Ajami, A. M., Vogt, J. and Young, V. R. (1993)
Plasma arginine and citrulline kinetics in adults given
adequate and arginine-free diets. Proc. Natl. Acad. Sci.
U.S.A. 90, 77497753
19 Yu, Y. M., Ryan, C. M., Castillo, L., Lu, X. M., Beaumier,
L., Tompkins, R. G. and Young, V. R. (2001) Arginine and
ornithine kinetics in severely burned patients: increased
rate of arginine disposal. Am. J. Physiol. Endocrinol.
Metab. 280, E509E517
20 Yu, Y. M., Young, V. R., Castillo, L., Chapman, T. E.,
Tompkins, R. G., Ryan, C. M. and Burke, J. F. (1995)
Plasma arginine and leucine kinetics and urea production
rates in burn patients. Metab. Clin. Exp. 44, 659666
21 Castillo, L., Beaumier, L., Ajami, A. M. and Young, V. R.
(1996) Whole body nitric oxide synthesis in healthy men
determined from [15 N] arginine-to-[15 N]citrulline
labeling. Proc. Natl. Acad. Sci. U.S.A. 93, 1146011465
22 Windmueller, H. G. and Spaeth, A. E. (1981) Source and
fate of circulating citrulline. Am. J. Physiol. 241,
E473E480
23 Wu, G. (1997) Synthesis of citrulline and arginine from
proline in enterocytes of postnatal pigs. Am. J. Physiol.
272, G1382G1390
24 van de Poll, M. C., Siroen, M. P., van Leeuwen, P. A.,
Soeters, P. B., Melis, G. C., Boelens, P. G., Deutz, N. E. and
Dejong, C. H. (2007) Interorgan amino acid exchange in
humans: consequences for arginine and citrulline
metabolism. Am. J. Clin. Nutr. 85, 167172
25 Ligthart-Melis, G. C., van de Poll, M. C., Boelens, P. G.,
Dejong, C. H., Deutz, N. E. and van Leeuwen, P. A. (2008)
Glutamine is an important precursor for de novo
synthesis of arginine in humans. Am. J. Clin. Nutr. 87,
12821289

C The Authors Journal compilation 
C 2009 Biochemical Society
30 C. C. Kao and others
26 van Waardenburg, D. A., de Betue, C. T., Luiking, Y. C.,
Engel, M. and Deutz, N. E. (2007) Plasma arginine and
citrulline concentrations in critically ill children: strong
relation with inflammation. Am. J. Clin. Nutr. 86,
14381444
27 Druml, W., Heinzel, G. and Kleinberger, G. (2001) Amino
acid kinetics in patients with sepsis. Am. J. Clin. Nutr. 73,
908913
28 Vermeulen, M. A., van de Poll, M. C., Ligthart-Melis,
G. C., Dejong, C. H., van den Tol, M. P., Boelens, P. G. and
van Leeuwen, P. A. (2007) Specific amino acids in the
critically ill patientexogenous glutamine/arginine: a
common denominator? Crit. Care Med. 35, S568S576
29 Evans, T., Carpenter, A., Kinderman, H. and Cohen, J.
(1993) Evidence of increased nitric oxide production in
patients with the sepsis syndrome. Circ. Shock 41, 7781
30 Ochoa, J. B., Udekwu, A. O., Billiar, T. R., Curran, R. D.,
Cerra, F. B., Simmons, R. L. and Peitzman, A. B. (1991)
Nitrogen oxide levels in patients after trauma and during
sepsis. Ann. Surg. 214, 621626
31 Luiking, Y. C. and Deutz, N. E. (2003) Isotopic
investigation of nitric oxide metabolism in disease. Curr.
Opin. Clin. Nutr. Metab. Care 6, 103108
Received 29 August 2008/4 November 2008; accepted 23 December 2008
Published as Immediate Publication 23 December 2008, doi:10.1042/CS20080444

C The Authors Journal compilation 
C 2009 Biochemical Society
32 van Eijk, H. M., Luiking, Y. C. and Deutz, N. E. (2007)
Methods using stable isotopes to measure nitric oxide
(NO) synthesis in the l-arginine/NO pathway in health
and disease. J. Chromatogr. B Analyt. Technol. Biomed.
Life Sci. 851, 172185
33 Cauwels, A. (2007) Nitric oxide in shock. Kidney Int. 72,
557565
34 Nijveldt, R. J., Teerlink, T., van der Hoven, B., Siroen,
M. P., Kuik, D. J., Rauwerda, J. A. and van Leeuwen, P. A.
(2003) Asymmetrical dimethylarginine (ADMA) in
critically ill patients: high plasma ADMA concentration is
an independent risk factor of ICU mortality. Clin. Nutr.
22, 2330
35 Siroen, M. P., van Leeuwen, P. A., Nijveldt, R. J., Teerlink,
T., Wouters, P. J. and van den Berghe, G. (2005)
Modulation of asymmetric dimethylarginine in critically ill
patients receiving intensive insulin treatment: a possible
explanation of reduced morbidity and mortality? Crit.
Care Med. 33, 504510
36 Nijveldt, R. J., Siroen, M. P., Teerlink, T. and van Leeuwen,
P. A. (2004) Elimination of asymmetric dimethylarginine
by the kidney and the liver: a link to the development of
multiple organ failure? J. Nutr. 134, 2848S2852S