Download Regulation of Urea Synthesis by Agmatine in the Perfused Liver

AJP-Endo Articles in PresS. Published on August 13, 2002 as DOI 10.1152/ajpendo.00246.2002
MS#E00246-2002, Nissim et al 1
Regulation of Urea Synthesis by Agmatine in the Perfused Liver:
Studies With 15N
Itzhak Nissim*, Oksana Horyn, Yevgeny Daikhin, Ilana Nissim, Adam Lazarow
and Marc Yudkoff
Children Hospital of Philadelphia, Division of Child Development and Rehabilitation, Department
of Pediatrics, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104
Running Title: Regulation of Hepatic Ureagenesis
CONTENTS:
Pages: 40
Figures: 8
Tables: 2
References: 49
*
Correspondence should be addressed to:
Itzhak Nissim, Ph.D.
Division of Child Development
Abramson Pediatrics Research Ctr. Room-510C
34th Street and Civic Center Boulevard
Philadelphia, PA 19104-4318
FAX: (215)-590-5199
e-mail : [email protected]
Copyright 2002 by the American Physiological Society.
MS#E00246-2002, Nissim et al 2
ABSTRACT
Administration of arginine or a high protein diet increases the hepatic content of Nacetylglutamate (NAG) and the synthesis of urea. However, the underlying mechanism is unknown.
We have explored the hypothesis that agmatine, a metabolite of arginine, may stimulate NAG
synthesis, and, thereby, urea synthesis. We tested this hypothesis in a liver perfusion system to
determine: (i) The metabolism of L-[guanidino-15N2]arginine to either agmatine, nitric-oxide (NO)
and/or urea; (ii) Hepatic uptake of perfusate agmatine and its action on hepatic nitrogen
metabolism; and (iii) The role of arginine, agmatine or NO in regulating NAG synthesis and
ureagenesis in livers perfused with 15N-labeled glutamine and unlabeled ammonia or 15NH4Cl and
unlabeled glutamine. Our principal findings are: (i) [guanidino-15N2]agmatine is formed in the liver
from perfusate L-[guanidino-15N2]arginine. About 90% of hepatic agmatine is derived from
perfusate arginine; (ii) Perfusions with agmatine significantly stimulated the synthesis of [15N]NAG and [15N]urea from
15
N-labeled ammonia or glutamine; and (iii) The increased levels of
hepatic agmatine are strongly correlated with increased levels and synthesis of 15N-labeled NAG
and [15N]urea. These data suggest a possible therapeutic strategy encompassing the use of agmatine
for the treatment of disturbed ureagenesis, whether secondary to inborn errors of metabolism or to
liver disease.
Key Words: Arginine, N-acetylglutamate, Carbamoyl Phosphate Synthetase-I,
Hyperammonemia, Nitric oxide
MS#E00246-2002, Nissim et al 3
INTRODUCTION
The regulation of ureagenesis has posed a complex problem since the first description of the
urea cycle by Krebs and Henseleit in 1932 (20). The first and most critical step in urea synthesis is
the conversion of NH4+ and HCO3 - into carbamoyl phosphate by mitochondrial carbamoyl
phosphate synthetase I (CPS-I), ( EC 6.3.5.5) (8, 13, 16, 25), which requires N-acetylglutamate
(NAG) as an obligatory effector for its activation (8, 13, 14). Brosnan (4) has suggested that acute
increases in ureagenesis are initiated by greater availability of NAG. Increased NAG synthesis and
concomitant ureagenesis is observed after intake of arginine or a high protein diet (7, 19, 27, 44).
However, it is not clear how and why arginine or a high protein diet increases the hepatic NAG
level, and, thereby, urea synthesis. It is unknown whether arginine directly or one of its metabolites
stimulates the synthesis of NAG, and, thereby, ureagenesis.
As illustrated in Figure 1, L-arginine can be metabolized in the liver via several pathways:
(i) Hydrolysis via arginase to form urea, the end product of the urea cycle. There are two
isoenzymes of mammalian arginase (8, 17, 38, 47). Arginase I is mainly present in the cytosol of
periportal hepatocytes and is linked to urea cycle activity (2, 25, 38, 47). Arginase II is found
primarily in mitochondria (7, 17, 38, 47); (ii) Decarboxylation via mitochondrial arginine
decarboxylase (ADC) to produce agmatine and CO2 (14, 39, 47); and/or (iii) Metabolism via NOsynthase to form nitric oxide (NO) and citrulline (21, 47). The ADC and NO-synthase reactions are
quantitatively minor routes of overall arginine metabolism, but the products of these pathways (NO
and agmatine) have significant roles as signaling molecules that regulate many metabolic and
physiologic functions (21, 40, 41, 47). However, their role in the regulation of ureagenesis is
uncertain. Understanding this relationship may provide a new therapeutic strategy for treatment of
urea cycle disorders and hyperammonemia .
It has been demonstrated that the ADC pathway is present in rat kidney, brain, liver and gut
(40, 41). Recent studies have highlighted the importance of agmatine for a number of physiological
MS#E00246-2002, Nissim et al 4
processes, including: dose-dependent release of insulin from islet cells exposed to glucose,
proliferation of vascular smooth muscle, regulation of intracellular polyamine levels and cellular
proliferation, regulation of neurotransmitter receptors, modulation of opioid analgesics, and
inhibition of NO synthesis (2, 40, 41). Agmatine may stimulate the release of catecholamines from
adrenal chromaffin cells by binding to _ 2 -adrenergic receptors (14). Thus, agmatine, which is
widely distributed in mammalian tissue (37, 39), may have a role as a hormone and/or polyamine
precursor for a multiple metabolic functions. However, the significance of agmatine as a signaling
molecule in regulating the urea cycle remains to be defined.
In the current investigation we have explored the hypothesis that dietary arginine (whether
free or as a part of a high protein diet), is decarboxylated to agmatine in hepatic mitochondria, and
that agmatine, not arginine, is the regulator of NAG synthesis and the resulting activation of
mitochondrial CPS-I. As illustrated in Figure 1, both agmatine and NAG are synthesized in
mitochondria, the latter from acetyl-CoA and glutamate by N-acetylglutamate synthetase (25). The
mitochondrial location of both ADC and NAG synthetase makes agmatine a reasonable candidate
for regulation of NAG synthesis, and, thereby, activation of CPS-I.
We have tested the above hypothesis in three steps: (i) Initial studies that involve perfusions
with L-[guanidino-15N2]arginine in order to determine the hepatic uptake of arginine and its relative
metabolism to either agmatine, NO and/or urea, as indicated in Figure 1; (ii) Perfusions with
agmatine to determine its hepatic uptake and the dose dependence of agmatine action on urea
synthesis; and (iii) Studies to elucidate the role of arginine, agmatine or NO in the regulation of
NAG synthesis and ureagenesis in liver perfused with either [2-15N]- or [5-15N]-labeled glutamine
and unlabeled ammonia or 15NH4Cl and unlabeled glutamine, as we have previously described (3, 5,
31, 32). The results obtained substantiate the hypothesis that agmatine regulates NAG and urea
synthesis.
MS#E00246-2002, Nissim et al 5
Materials and Methods
I.
Liver Perfusions: Livers from fed, male Sprague Dawley rats were perfused "in situ" in the
non-recirculating mode as described by Pastor et. al (35). We employed the single-pass perfusion
with antegrade flow because this model preserves the normal lobular microcirculation of the liver
and physiologic flow direction. The model also avoids the problem of substrate recycling from
perivenous to periportal hepatocytes. Oxygen consumption was measured continuously using a
Clark electrode and the Oxygen Measuring System (Instech SYS203, Plymouth Meeting, PA). One
electrode was attached into the inflow cannula and one into the outflow. This system was linked to a
computer equipped with the WinDaq/Lite/Pro/Pro+ Data System (DATAQ Instrument, Akron,
Ohio) for online acquisition and recording of signals corresponding to inflow and outflow levels of
pO2. The basic perfusion medium was Krebs saline continuously gassed with 95% O2 / 5% CO2 and
containing lactate (2.1 mM) and pyruvate (0.3 mM) as metabolic fuels. Perfusion flow rate (3-3.5
ml/min/g), hepatic temperature, pH and pO2 (in influent and effluent media) were monitored
throughout, and oxygen consumption was calculated. After 15 min. of pre-perfusion with the basic
perfusate, we changed to a medium that contained, in addition to the lactate and pyruvate,
precursors for urea nitrogen or modulators of urea synthesis as indicated below.
II.
Experimental Design: To establish the role of arginine or its metabolites, NO or agmatine,
in the regulation of NAG and urea synthesis, we first performed experiments to determine the rate
of hepatic uptake of arginine and its relative metabolism to agmatine (ADC reaction), urea (arginase
reaction) or nitric oxide (NO-synthase reaction). To this end, after 15 minutes pre-perfusion with a
basic perfusate as indicated above, we changed to a perfusate that contained 0.3 mM NH4Cl, 1mM
unlabeled glutamine and 0.5mM L-[guanidino-15N2]arginine ((99 mole % excess, (MPE)). Separate
perfusate reservoirs, each containing different media, were used to facilitate changes in perfusate
medium. Perfusion was continued for a total of 70 min. Samples were taken from the influent and
MS#E00246-2002, Nissim et al 6
effluent media for chemical and GC-MS analyses. At the end of the perfusions, livers were freezeclamped, and used for measurements of metabolite concentrations and 15N enrichment in arginine
and agmatine.
The next series of experiments was designed to examine the effect of exogenous agmatine or
nitroprusside (SNP), a donor of NO, on total urea synthesis. We also evaluated a possible dosedependent action of either arginine, agmatine or NO on urea synthesis from ammonia and
glutamine. Perfusions were performed with: (i) agmatine at an increasing concentrations (µM) of
25, 50, 100, 250, 500 or 1000; (ii) SNP, at an increasing concentration (µM) of 25, 50 or 100 or (iii)
arginine at an increasing concentration (µM) of 250, 500 or 1000. The dose dependence was studied
in the same liver perfusion. At the indicated times, a separate perfusate reservoir containing
different media was used to facilitate changes in the concentration of each modulator.
To examine the role of arginine or its metabolites in the regulation of N-acetylglutamate
synthesis and ureagenesis under conditions and concentrations that approximate those that exist in
vivo, perfusions were performed without (control) or with addition of 100 µM agmatine, 25 µM
SNP or 500 µS arginine. The 15N-labeled precursor for urea nitrogen was either: (a) glutamine
(1mM), [5-15N]glutamine or [2-15N]glutamine (99 MPE) and NH4Cl (0.3mM), or (b) unlabeled
glutamine (1mM) and
15
NH 4Cl (0.3mM), (99 MPE). Pre-perfusions with basic medium were
performed for 15 minutes, and continued for 70 minutes with the indicated
15
N precursor or
modulator.
III.
Chemical Analyses: Samples were taken at the indicated times from influent and effluent
media for chemical determination of metabolites. At the end of the perfusion (70 min), livers were
freeze-clamped. The frozen livers were extracted into perchloric acid and the neutralized extracts
were used for chemical analysis. Adenine nucleotides (ATP, ADP and AMP) were determined
using 31P-NMR analysis as described (30). Amino acids as well as agmatine were measured by
MS#E00246-2002, Nissim et al 7
HPLC, using precolumn derivatization with o-phthalaldehyde (18). Measurements also were made
of glucose (1), urea and ammonia (29), lactate (15) and pyruvate (12).
IV.
GC-MS Methodology: GC-MS measurements of hepatic NAG levels and
15
N isotopic
enrichment in various metabolites were performed on a Hewlett Packard 5970 MSD and/or 5971
MSD coupled with a 5890 HP-GC, as described previously (3, 5, 31, 32). The following
measurements were performed:
i. )
Amino Acids, ammonia and urea: For measurement of 15N enrichment in urea and
amino acids, samples were prepared as we have previously described (3, 5, 31, 32). Briefly, a 500
µl aliquot of effluent or liver extract was purified via an AG-50 column (H+; 100-200 mesh; 0.5 x
2.5 cm), and then converted into the t-butyldimethylsilyl (t-BDMS) derivatives. The m/z 231, 232,
233, 234 of the urea t-BDMS derivative (46) were monitored for singly (Um+1) and doubly (Um+2)
labeled urea determination (3, 31, 32). Isotopic enrichment in glutamate, glutamine, aspartate and
alanine was monitored using ratios of ions at m/z of 433/432, 432/431, 419/418 and 261/260,
respectively.
15
NH3 enrichment was measured after conversion of ammonia to norvaline as
described (5). 15N enrichment in L-[guanidino-15N2]arginine was determined after derivatization
with 100 µl of tri-fluoroacetic anhydride (TFAA) and monitoring the ratio of ions at m/z 377/375 as
we have previously described (28).
ii.)
Measurement of NAG Concentration: The level of NAG in freeze-clamped livers
was determined using GC-MS and a modification of the conventional isotope dilution technique, as
described (31). First, 15N-labeled NAG was synthezised (99 MPE) by reacting [15N]glutamate (99
MPE) with acetic anhydride. The 15N-labeled NAG was used to prepare standard dilution curves by
mixing labeled and unlabeled NAG, and spiking samples for determination of NAG levels by
isotope dilution.
MS#E00246-2002, Nissim et al 8
NAG in standard solutions or samples was converted into the methyl esters as follows:
Samples were dried under N2 and azeotroped with methylene chloride. Then 100 µl of methanolic
HCl (3N) (Supelco) was added. Capped vials were heated at 60oC for 10 min., cooled and dried
under a gentle stream of N2. The residue was extracted into 1 ml of ethyl acetate and 300 µl of H2O
after vortexing for 30 sec., the organic layer was removed, dried and reconstituted in 75 µl of ethyl
acetate. Usually 1-2 µl were injected into the GC-MS, and isotopic enrichment in NAG was
determined using the ratio at m/z 159/158.
To determine the concentration of NAG in freeze-clamped liver extracts, an aliquot (500 µl)
was assayed as indicated above for 15N enrichment (I1), after 15N labeled precursor was used in the
perfusion, e.g., in experiments with 15NH4Cl or [2-15N]glutamine (in this case, the value of I1 was
between 20 - 30 MPE). A second aliquot (500 µl) was spiked with 5 nmoles of unlabeled NAG, and
the second isotopic enrichment (I2) was determined. In experiments with [5-15N]glutamine, I1
usually was below 3 MPE. Therefore, the second aliquot of 500 µl was spiked with 5 nmoles of 15N
labeled NAG and I2 was determined. NAG concentrations were calculated using the isotope dilution
technique (31, 48). With each series of measurements, a calibration curve of NAG with a known
isotopic enrichment (1-50 MPE) was prepared and analyzed by GC-MS. In nearly every
preparation, we achieved an excellent agreement between the observed and the expected
15
N
enrichment in NAG with r values around 0.9.
iii.)
Determination of
15
N-labeled Agmatine: We first synthesized [guanidino-
15
N2]agmatine as described (6). Briefly, 18mM L-[guanidino-15N2]arginine in 0.2M sodium acetate,
pH 5.2, containing 5mM pyridoxal phosphate (Sigma), 0.1% BSA and 1 I.U., bacterial ADC (Sigma
# A8134), was incubated at 37oC for 2 hour. Then, an additional 1 IU of ADC was added and
incubation was continued for another 3 hours. The reaction was stopped by addition of 5 M KOH.
Agmatine was extracted with n-butanol and dried under vacuum. Purity was determined by HPLC,
MS#E00246-2002, Nissim et al 9
and yield was almost 100%. The
15
N-labeled agmatine was used to develop the GC-MS
methodology for measurement of 15N-labeled agmatine in biological samples and for preparation of
standard isotope dilution curves by mixing labeled and unlabeled agmatine.
For GC-MS analysis of
method for determination of
15
15
N-labeled agmatine we have applied a previously described
N-labeled arginine in biological samples (28). To examine the
accuracy of this method for measurement of 15N enrichment in agmatine a standard dilution curve of
15
N-labeled agmatine, with a known isotopic enrichment (1-50 MPE) was prepared. Samples were
alkalinized with 1ml of 3 M NH4OH and then loaded into a column containing the acetate form of
Dowex 1-X8 (100-200 mesh, Bio-Rad). The column was washed with 3-4 ml H2O, and agmatine
was eluted with 3ml of 2N HCl. The eluate was dried under N2, azeotroped with CH2Cl2 and
derivatized with 100 µl of TFAA at 100oC for 10 minutes. Usually, 2-4 µl was injected into GC-MS
for analysis.
Derivatized agmatine was separated from arginine and other compounds on a capillary
column (15m x 0.25 I.D., ZB-1, Phenomenex # 023545). GC conditions were: injector temperature
250oC and temperature program 110 oC isothermal for 2 minutes and then 10oC/min. We have found
that approximately 90% of agmatine was converted into N-tri-TFA-agmatine with a major ion at
m/z 349, and about10% was converted into N-tetra-TFA with a major ion at m/z 445. We used the
N-tri-TFA-agmatine and m/z 351/349 ratio to determine isotopic enrichment in [guanidino15
N2]agmatine. When the standard isotope dilution curve was analyzed, we achieved excellent
agreement between the observed and the expected 15N enrichment in agmatine with r values of 0.9
or better. For determination of 15N-labeled agmatine following liver perfusion with L-[guanidino15
N2]arginine, a 1ml of perfusate or liver extract was alkalinized with 1ml of 3 M NH4OH and then
loaded onto a Dowex 1-X8 column (acetate). Separation, derivatization and GC-MS analysis were
completed as indicated above.
MS#E00246-2002, Nissim et al 10
V.
Determination of NO released in effluent: NO released (NO-2+NO-3) in the effluent during
liver perfusions were determined with a Nitric Oxide Analyzer (Sievers Instruments, Boulder, CO),
as indicated (36). Briefly, perfusate samples (10 µl) were injected in a reaction chamber containing
a VCl3/HCl mixture heated at 90oC, and nitric oxide was detected by chemiluminescence. With each
measurement, a standard curve analysis of known concentration of NaNO3 was injected and used
for calculation of NO-2+NO-3 in the perfusate samples.
VI.
Statistical analyses: Statistical and regression analyses were carried out using In-STAT 1.14
software for the Macintosh. We used the Student-t test or ANOVA test to compare two groups or
differences among groups as needed. A p value less than 0.05 was taken as indicating a statistically
significant difference.
VII. Materials and Animals : Male Sprague-Dawley rats (Charles-River) were fed ad lib on a
standard rat chow diet. Chemicals were of analytical grade and obtained from Sigma-Aldrich.
Enzymes and cofactors for the analysis of urea, lactate, pyruvate, glucose and ammonia were
obtained from Sigma. 15N-labeled arginine, glutamine and ammonia (99 mole % excess) were from
Isotec.
MS#E00246-2002, Nissim et al 11
Results
Hepatic uptake and metabolism of arginine - Production of agmatine: The initial series of
experiments was designed to determine the extent to which perfusate arginine is taken up and
metabolized to either urea, agmatine or nitric oxide. Figure 2 demonstrates the formation of these
metabolites and oxygen consumption during perfusion with L-[guanidino-15N2]arginine. The
constancy of oxygen consumption is an indication of the viability and stability of the perfused
livers. The rate of L-[guanidino - 15N2]arginine uptake is about 120-140 nmoles/min/g liver (Figure
2B). At the end of 70 minutes perfusion, the isotopic enrichment of hepatic arginine was 46.3 ± 9.3
MPE (Figure 2C), indicating that approximately 50% of hepatic arginine was derived from arginine
entering the liver via the portal vein .
Arginine in the effluent was about 0.45 mM and that in the perfusate 0.5 mM, a finding in
agreement with a previous estimate that approximately 15% of perfusate arginine is metabolized in
the liver (33, 34). The major metabolites of L-[guanidino-15 N 2 ]arginine are [15N2]urea and
[guanidino-15N2]agmatine (Figure 2A). The 15N enrichment in doubly labeled urea was about 10-13
MPE between 20-40 minutes and labeling decreased to 5-6 MPE by 70 minutes. The decreased
enrichment in [15N2]urea after 40 minutes of perfusion reflects increased formation of unlabeled
urea from ammonia and glutamine via the urea cycle. The total (labeled and unlabeled) urea output
between 20-70 minutes was approximately 1000 nmoles/min/g liver. Total [15N2]urea output was
about 70-90 nmoles/min/g (Figure 2A), which is about 80% of arginine uptake. Therefore, when
physiologic concentrations of glutamine and ammonia are provided, approximately 7-9% of total
urea in the effluent was formed from external arginine entering the liver via the portal vein.
Agmatine is the second major metabolite of arginine. The output of [guanidino15
N2]agmatine in the effluent was about 14-20 nmoles/min/g (Figure 2A), indicating that
approximately 10-15% of arginine uptake was recovered as agmatine in the effluent. The
MS#E00246-2002, Nissim et al 12
concentration of agmatine in the liver extract was about 130 nmoles/g, and 41.1 ± 11.2 % of this
was in the form of [guanidino-15N2]agmatine (Figure 2C). The ratio between [guanidino15
N2]agmatine and L-[guanidino-15N 2]arginine was 0.87, indicating that approximately 87% of
hepatic agmatine was derived from external arginine.
An additional pathway of arginine metabolism is via nitric oxide synthetase (21, 46). The
output of NO, measured as NO
2
+ NO -3 was 1.5-2 nmoles/min/g, which is about 1% of arginine
uptake (Figure 2A). The current value of NO output is in good agreement with a previous study
showing that, in livers perfused with 1mM arginine, the sum of NO -2 + NO
3
output was about 4
nmoles/min/g dry weight (35).
Dose dependence of agmatine action on urea synthesis: Following the demonstration that
agmatine is formed in the liver from external arginine, we next examined the hepatic uptake of
perfusate agmatine and its action on hepatic nitrogen metabolism and ureagenesis. We also
evaluated the effect of perfusate agmatine on these parameters. Figure 3 depicts the results of
perfusions with unlabeled glutamine (1mM), ammonia (0.3mM), and either 25, 50 or 100 µM
agmatine. We found that the maximum effect of agmatine on urea synthesis occurred at a
concentration of 100 µS in the perfusate. Higher levels (not shown) had little incremental effect on
urea synthesis or oxygen consumption. Similarly, no significant differences were found with an
arginine concentration above 500 µM or SNP above 25 µM (data not shown). These data indicate
that agmatine is taken up by the liver, where it stimulates oxygen consumtion and urea synthesis
from glutamine and ammonia. This action is dose dependent, with a maximum effect at perfusate
agmatine of 100 µM.
It has been shown that agmatine is taken up by system y+, the same system used for arginine
and polyamine transport (6, 42). In cultured hepatocytes, approximately 10% of agmatine was
converted to putrescine via the agmatinase pathway (6). In the current study, if we assume that 10%
MS#E00246-2002, Nissim et al 13
of agmatine uptake was metabolized to putrescine and urea, then the expected amount of urea
formed from agmatine will be about 10 nmoles/min/g, which is less than 0.1% of total urea output
(Figure 3C). Thus, the amount of urea which may be derived from agmatine is negligible. This
conclusion is further supported by the formation of [15N]urea from
15
N- labeled precursors, as
indicated below.
Effect of arginine, agmatine or nitric-oxide on hepatic metabolic activity and nitrogen
balance: Viability of the livers is documented by the constancy of oxygen consumption during the
course of perfusions (Figure 4). The addition of arginine or agmatine to the perfusate significantly
(p<0.05) increased O2 consumption between 40-70 minutes. The increased O2 consumption was
more significant with agmatine (p=0.006) than with arginine (Figure 4B).
The increased oxygen consumption in perfusions with agmatine was associated with
stimulated glutamine-N uptake (Table 1 and Figure 4A). Between 40-70 minutes of perfusion,
glutamine-N uptake (nmoles N/min/g) was increased by approximately 4-fold in perfusions with
agmatine and by 2-fold in perfusion with arginine or SNP (Table 1). There were only minor
differences in ammonia-N uptake between the various experimental groups (Table 1 and Figure
4C). The appearance of nitrogen in urea, alanine and glutamate represents the major nitrogenous
output (the release of other amino acids was minor). Therefore, we calculated the extent to which
these three compounds could account for nitrogen balance across the liver. In control experiments
the combined nitrogenous uptake of glutamine and ammonia nitrogen was 1127 nmoles of
nitrogen/min/g liver. The output of nitrogen in urea, alanine and glutamate was 1031 nmoles
N/min/g liver. In perfusions with agmatine, arginine or SNP, the uptake of nitrogen was 2652, 1610
or 1652 nmoles N/min/g liver, respectively (Table 1). The output of nitrogen was 1890, 1332 or
1225 nmoles N/min/g liver, in perfusion with agmatine, arginine or SNP, respectively (Table 1).
Thus, in control perfusions and with the addition of arginine or NSP there was almost complete
MS#E00246-2002, Nissim et al 14
recovery of nitrogen uptake in the release of urea, alanine and glutamate. In perfusions with
agmatine, the uptake of nitrogen exceeded the output by approximately 30%.
Table 2 presents the level of hepatic metabolites at the end of 70 minutes perfusion. We
present only those metabolites that are directly related to the urea cycle. The agmatine level in
control perfusions was 29.6 ± 4.2 nmoles/g, and increased by approximately 10-fold after perfusion
with agmatine and 5-fold after perfusion with arginine, but relatively little with SNP. Agmatine
significantly increased the level of hepatic ornithine, and most notably the level of NAG, from 28.9
± 9.6 nmoles/g in control to 47.6 ± 16.6 nmoles/g (p<0.05) following perfusion with agmatine. At
the end of arginine perfusion, there was a significant increase in hepatic content of citrulline,
ornithine and arginine. The NAG level increased by approximately 20%, but this difference was not
significant. In perfusions with SNP, levels of arginine, agmatine or NAG show small changes
compared with control perfusions (Table 2). However, there was a significant (p<0.05) depletion of
liver glutamine, glutamate, alanine and aspartate, associated with significantly increased ornithine.
If 1g of liver contains 53 mg mitochondrial protein (7), the mitochondrial NAG levels would be
0.55, 0.91, 0.64 and 0.51 nmoles/mg protein in control or in perfusions with agmatine, arginine and
SNP, respectively. The level of hepatic NAG in control perfusions is in good agreement with
mitochondrial NAG levels previously reported (7).
Measurements of the lactate/pyruvate ratio in liver extracts at the end of perfusions show
little differences among the study groups (data not shown), but the sum of lactate and pyruvate
uptake was approximately 1.8 ± 0.5 µmoles/min/g in control perfusions and 3.1 ± 0.2 µmoles/min/g
in perfusions with agmatine (N=4, p<0.05). Little differences were found in perfusions with
arginine or SNP compared with control. The increased uptake of lactate and pyruvate in perfusions
with agmatine was not accompanied by increased glucose output, but a 25% increase in glucose
output was found in perfusions with SNP compared with control (600 vs. 800 nmoles/min/g in
control or SNP, respectively).
MS#E00246-2002, Nissim et al 15
Effect of arginine, agmatine or nitric-oxide on 15N-labeled glutamine metabolism and
synthesis of [15N]urea : Hepatic glutamine metabolism and ammonia formation are mediated via
flux through the phosphate-dependent glutaminase (PDG) reaction (25, 29, 31, 32). In the current
study we estimated flux through PDG as the sum of
15
N-labeled ammonia, urea, alanine and
glutamate output in the effluent when [5-15N]glutamine was used as labeled precursor. We have
found that the formation of other amino acids accounted for less than 5% of [5-15N]glutamine
consumption (32). The calculated flux through the PDG reaction based upon values at 60 minutes of
perfusion (Figure 5), shows rates (nmoles/min/g) of: 426 ± 113 (control), 691 ± 190 (plus agmatine,
p=0.027), 584 ± 118 (plus arginine, p=0.062) and 433 ± 121 (plus SNP, p>0.05). Therefore, the
addition of agmatine to the perfusate significantly stimulated flux through PDG, whereas the
addition of arginine is less significant.
The PDG reaction converts [5-15N]glutamine to 15NH3, which either can be released in the
effluent as 15NH3, incorporated into glutamate by glutamate dehydrogenase (GDH) or into urea via
the CPS-I reaction (29, 31, 32). Figure 6A illustrates the production of 15N-labeled urea (sum of
Um+1 + Um+2). Between 20 and 40 minutes of perfusion, [15N]urea production from [5-15N]glutamine
amounted to 150-190 nmoles N/min/g and continued to increase to 250-280 nmoles N/min/g at 60
min in control perfusions or perfusions with SNP. In perfusions with agmatine, however, [15N]urea
production increased to about 431 nmoles N min/g (P=0.02) at 60 minutes. In perfusions with
arginine, [15N]urea production increased by approximately 25%, to about 310 nmoles N/min/g,
which is not significant (p=0.1).
To distinguish between a possible effect of agmatine on glutamine metabolism via the
mitochondrial PDG reaction or via the deamination of glutamate in the GDH reaction (29, 31, 32),
perfusions were performed with [2-15N]glutamine. As with [5-15N]glutamine, perfusions with [215
N]glutamine demonstrate a control rate of [15N]urea production of 100-150 nmoles/min/g between
20-40 minutes and 260 nmoles/min/g from 40-60 min. In perfusions with arginine, [15N]urea output
MS#E00246-2002, Nissim et al 16
was 320 nmoles N/min/g at 60 min., but these differences are not significant. However, in
perfusions with agmatine [15N]urea production increased to about 600 nmoles N/min/g from 40-60
min (p=0.001) (Figure 6B). Similarly, livers infused with SNP increased [15N] urea production from
[2-15N]glutamine to 385 nmoles N min/g from 40-60 min. (p=0.026).
Incorporation of 15NH4Cl into urea - Regulation by agmatine: To determine whether the
stimulatory action of agmatine on [15N]urea production from
15
N-labeled glutamine is directly
linked to glutamine metabolism, or is subsequent to a stimulation of carbamoyl phosphate synthesis,
perfusions were carried out with agmatine,
15
NH 4 Cl and unlabeled glutamine. Figure 6C
demonstrates that the presence of 15NH4Cl resulted in an immediate and massive production of
[15N]urea (between 20-60 minutes) of about 1000 nmoles N/min/g (control) and 1700 nmoles
N/min/g (plus agmatine, p<0.05). Agmatine, independent of an effect on glutamine metabolism,
increased [15N]urea production from 15NH4Cl by approximately 40% (p<0.05) even with 1mM
unlabeled glutamine in the perfusate. Thus, it is reasonable to propose that agmatine stimulates the
formation of
15
N-labeled carbamoyl phosphate from
15
NH 4 Cl secondary to its action on
mitochondrial NAG synthesis.
Synthesis of N-acetylglutamate - Regulation by agmatine: Figure 7 illustrates the
relationship between hepatic agmatine and NAG levels under all experimental conditions. Data
points from perfusions with [2-15N]glutamine, [5-15N]glutamine and 15NH4Cl are included in these
correlation analyses. It is evident that there was a highly significant relationship between hepatic
agmatine and NAG concentrations (p=0.0003), between agmatine and [15 N]urea synthesis
(p=0.0002), and between NAG and [15N]urea synthesis (p=0.0002). This observation indicates that
the level of NAG in the liver is strongly associated with levels of agmatine, and that agmatine may
regulate the synthesis of NAG in the mitochondria. This observation suggests that agmatine
MS#E00246-2002, Nissim et al 17
regulates NAG synthesis independent of an effect on flux through PDG or GDH, and this action of
agmatine on NAG synthesis may play a key role in the regulation of hepatic ureagenesis.
The data suggest that increased hepatic agmatine can induce NAG and urea synthesis. A
related question is: Does this change reflect increased synthesis or diminished translocation of NAG
from mitochondrion to cytosol, where it is degraded? To answer this question we have measured the
production of
15
15
N-labeled NAG in liver perfused with 15NH4Cl and unlabeled glutamine or [2-
N]glutamine and unlabeled ammonia (See Methods). With 15NH4Cl, [15N]glutamate will be formed
in the mitochondria via the reductive amination of _-ketoglutarate catalyzed by glutamate
dehydrogenase. In addition, some mitochondrial glutamate will be derived from unlabeled
glutamine via the PDG pathway. In the case of [2-15N]glutamine, [15N]glutamate will be formed via
the PDG pathway (29, 31, 32). Therefore, the 15N-labeled NAG will be derived from the overall
15
N-labeled mitochondrial glutamate pool regardless of the respective fluxes through PDG or GDH.
Figure 8 illustrates the production of 15N-labeled NAG from [2-15N]glutamine. Similar observations
were obtained with
15
NH4Cl (data not shown). These data clearly demonstrate that agmatine
significantly (p<0.001) increased the synthesis of NAG compared with perfusions without
agmatine. Arginine marginally increased (p=0.08), and SNP significantly (p=0.002) decreased the
production of 15N-labeled NAG (Figure 8).
MS#E00246-2002, Nissim et al 18
Discussion
The liver regulates whole body nitrogen metabolism and detoxifies ammonia via conversion
of excess amino-N to urea (25, 49). Although ureagenesis is a major hepatic function, the factors
that regulate this process have been the subject of conflicting views and controversies (4).
Understanding these mechanism(s) is important not only in terms of understanding ammonia
detoxification, but also because ureagenesis has provided insight into whole body nitrogen balance
and metabolism (49).
The initial step in urea synthesis is the conversion of NH4+ and HCO3- to carbamoyl
phosphate by mitochondrial CPS-I (28), which requires NAG as an obligatory effector (9, 13, 16,
25). Increased NAG synthesis and concomitant ureagenesis have been observed after the intake of
arginine or a high protein diet (8, 19, 27, 44). However, it is unknown whether arginine itself or a
metabolite of arginine stimulates the synthesis of NAG, and, thereby, ureagenesis. In the current
investigation we have explored the hypothesis that agmatine, not arginine, is responsible for
increased NAG synthesis and concomitant ureagenesis.
The principal observations supporting our hypothesis are as follows:
(i)
L-[guanidino-15N2]arginine enters the liver via the portal vein and is catabolized to
[guanidino-15N2]agmatine and [15N2]urea. About 90% of arginine uptake was
recovered in urea and agmatine output in the effluent,
(ii)
Metabolism of perfusate arginine is primarily mediated via the mitochondrial ADC
and arginase II reactions,
(iii)
Agmatine, which is naturally present in the intact liver, significantly increases in
concentration following addition of arginine to the perfusate. As much as 90% of
hepatic agmatine can be derived from arginine entering the liver via the portal vein,
MS#E00246-2002, Nissim et al 19
(iv)
The increase of hepatic agmatine strongly correlates with an increase of the newly
synthesized 15N-labeled NAG and [15N]urea production from either [2-15N] or [5-15N]
glutamine and unlabeled ammonia or 15NH4Cl and unlabeled glutamine as nitrogen
precursors,
(v)
Perfusions with agmatine significantly stimulated (p<0.0001) the synthesis of [15N]NAG and [15N]urea from 15N-labeled ammonia or glutamine. However, perfusion
with arginine marginally increased (p=0.08) NAG and urea synthesis.
The current observations suggest that the arginine entering the liver via the portal vein is
mainly metabolized through the mitochondrial ADC and arginase II reactions, and there is minimal
or no equilibrium with the arginine pool that is linked with the urea cycle. Evidence of this is the
observation that label in [15N]urea was about 5-10% of total urea output with 0.5 mM L-[[5N 2guanidino]arginine, unlabeled ammonia and glutamine in the perfusate. If the perfusate arginine is
in equilibrium with the cytosolic pool of arginine formed via the urea cycle, then the isotopic
enrichment in hepatic arginine at the end of perfusion should be similar to the isotopic enrichment
of [15N2]urea in the effluent. However, the observed enrichment in hepatic [guanidino-15N2]arginine
was about 50 MPE and enrichment in effluent urea was about 5-10 MPE at the end of perfusion.
This is expected, since the urea cycle does not allow net production of intermediates, and arginine
formed is hydrolyzed to ornithine and urea (25, 47). In addition, the percentage of [15N2]urea of the
total urea in the effluent indicates that the relative activity of arginase II is about 5-10% of the total
(sum of arginase I and II reactions) hepatic arginase. This estimate is in agreement with previous
observations indicating that in human liver arginase II is about 2% of liver arginase (17), and in rat
about 10% of total arginase activity (8).
In addition to mitochondrial formation of [15 N]urea from external L-[guanidino15
N2]arginine, the current study demonstrates that approximately 15-20% of L-[guanidino-
15
N2]arginine uptake was catabolized via the mitochondrial ADC reaction to form agmatine (Figure
MS#E00246-2002, Nissim et al 20
2A). O'Sullivan et al (33, 34), suggested that [U-14C]arginine is catabolized in the liver by ornithine
aminotransferase following its degradation via arginase I. Their conclusion is based on the release
of 14CO2 in the effluent, which was approximately 13 nmoles/min/g in liver obtained from rats fed a
normal protein diet (34), a regimen similar to that of this study. If we assume that 1 mole of
agmatine formed via the ADC reaction will be accompanied by the release of 1 mole of CO2, then
the rate of CO2 formation would be 14-20 nmoles/min/g, similar to the values obtained by
O'Sullivan et al (33, 34).
Agmatine is widely distributed in mammalian tissues (37, 40, 41), and may have a role in
multiple metabolic functions. The current observations demonstrate that agmatine entering the liver
via the portal vein stimulates the uptake of glutamine (Table 1), synthesis of NAG and urea (Figures
4-8). This possibility is consistent with an increased rate of oxygen consumption in perfusions with
agmatine, and to a less extent with arginine (Figure 4B). The increased oxygen consumption in
perfusions with agmatine may indicate (i) higher energy requirements; (ii) stimulation of substrate
supply to the electron transport chain and/or (iii) increased respiratory chain activity (45). Agmatine
may increase the activity of respiratory enzymes secondary to stimulated mitochondrial energyconsuming functions such as the synthesis of NAG, carbamoyl phosphate and urea itself.
In line with the above suggestion are data (Figure 8) demonstrating that agmatine
significantly (p<0.0001) increased formation of newly synthesized
15
15
N-labeled NAG from [2-
N]glutamine. However, arginine marginally increased (p=0.08), and SNP significantly (p=0.002)
decreased the production of NAG compared with control. The linear correlation between agmatine
and NAG levels (Figure 7), demonstrates that synthesis of NAG and, thereby, activity of CPS-I
would depend upon the concentration of agmatine. Therefore, the current data strongly support the
hypothesis that agmatine is a positive effector for mitochondrial synthesis of NAG. In addition, the
data in Figure 7 suggest that the concentration of intra-hepatic agmatine is an important feature in
the regulation of urea synthesis and ammonia detoxification. Nevertheless, it is uncertain whether
MS#E00246-2002, Nissim et al 21
increased NAG levels are secondary to increased synthesis or diminished translocation of NAG
from mitochondrion into cytosol, where it is degraded. It has been shown that efflux of NAG out of
mitochondria and its subsequent degradation are diminished in rats fed a high protein diet compared
with rats on a protein-free diet (22, 23, 26). The current data clearly demonstrate that agmatine
stimulates the synthesis of NAG (Figure 8). The diminished rate of NAG translocation in response
to a high protein diet (26) may reflect the presence in this diet of arginine, which is converted to
agmatine.
It has been suggested that hepatic glutaminase, and concomitant glutamate production is a
key factor in facilitating NAG synthesis (4). Consistent with this interpretation are the data
demonstrating that agmatine stimulates glutamine uptake (Figure 4A and Table 1) and flux through
glutaminase (Figure 5). However, the positive association between NAG synthesis, [15N]urea
synthesis and agmatine concentration is valid whether [2-15N]glutamine, [5-15N] glutamine or
15
NH4Cl was used to monitor these metabolic processes (Figure 7). Thus, agmatine may regulate
NAG synthesis independent of an effect on flux through the PDG reaction. The observed increased
flux through PDG in perfusions with agmatine may be subsequent to elevated NAG levels in the
mitochondrial matrix, as previously indicated (24). In addition, agmatine-stimulated glutamine
uptake (Table 1), may provide an additional fuel that support NAG and/or carbamoyl phosphate
synthesis, both of which require ATP. This possibility would be in line with the increased oxygen
consumption associated with agmatine induced glutamine uptake (Figure 4A, 4B and Table 1).
A significant decrease in hepatic glutamate level was observed following perfusions with
SNP (Table 2). This finding may explain the decreased NAG synthesis in perfusions with SNP
(Figure 8). The uptake of glutamine is increased with SNP (Table 1 and Figure 4), but this is not
reflected in augmented output of [15N]urea and 15NH3 from [5-15N]glutamine (i.e., flux through the
PDG pathway), which show minor differences compared with control (Figures 5 and 6).
Measurements of 15NH3 enrichment in perfusions with [2-15N]glutamine show that, in control, 15N
MS#E00246-2002, Nissim et al 22
ammonia enrichment in the effluent was about 3 atom % excess, with little differences in perfusions
with agmatine or arginine. In perfusions with SNP, 15NH3 enrichment was about 11 atom % excess,
indicating a stimulation of glutamate catabolism mediated via the GDH reaction. Therefore,
stimulated mitochondrial glutamate catabolism in perfusions with SNP may account for the
reduction of the newly synthesized NAG as shown in Figure 8.
An alternative mechanism for increased NAG synthesis by agmatine is stimulation of `oxidation and/or the pyruvate dehydrogenase and pyruvate carboxylase reactions. This possibility is
supported by the significant (p<0.05) increase in oxygen consumption in perfusions with agmatine
(Figures 3B and 4B), and increased pyruvate and lactate uptake (µmoles/min/g) from 1.7 (control)
to about 3 (plus agmatine). Stimulation of the pyruvate dehydrogenase reaction would increase
availability of acetyl-CoA for NAG synthesis, and an increase in the pyruvate carboxylase activity
would provide more oxaloacetate, and thereby, aspartate for synthesis of argininosuccinate, as we
have previously indicated in studies with [3-13C]pyruvate (31). These possibilities are currently
under investigation using 13C-labeled precursors and NMR as an analytical tool.
In conclusion, the current observations provide evidence to support the hypothesis that
availability of agmatine rather than arginine may have a major regulatory role in hepatic ammonia
detoxification and urea synthesis. The current findings may have clinical implications for the
treatment of disturbed urea synthesis and toxic hyperammonemia. Effective drugs have been
introduced for removal of waste toxic nitrogen from the body (10), but agmatine might prove a
valuable therapeutic adjunct. This agent has been proposed as a treatment for other disorders (40,
41). Agmatine can be useful especially in cases of hyperactivity of the hepatic GDH reaction, which
is associated with the hyperinsulinism/hyperammonemia syndrome (43). In this case,
hyperammonemia is secondary to diminished synthesis of NAG (43). The stimulation of NAG
synthesis by agmatine (Figure 8), lends support to the importance of agmatine as a potential
candidate for the treatment of hyperinsulinism/hyperammonemia syndrome in infants (43)
MS#E00246-2002, Nissim et al 23
ACNOWLEDEGEMENT
We thank Dr. J. T. Brosnan for helpful discussion. This work was supported by United States
National Institutes of Health Grants DK-53761 and CA-79495 (to I.N).
MS#E00246-2002, Nissim et al 24
REFERENCES:
1.
Bergemeyer, H.U., Bernt. E, Schidt F., and Stork H. D-Glucose In: Methods of
Enzymatic Analysis. H.U.Bergemeyer, Ed., Academic press, 3: PP, 1198-1201, 1974
2.
Blantz R.C., Satriano J., Gabbai F., Kelly C. Biological effects of arginine metabolites,
Acta Physiol. Scand 168: 21-25, 2000.
3.
Brosnan J.T., Brosnan M.E., Charron R. and Nissim I. A mass isotopomers study of urea
and glutamine synthesis from 15N-labeled ammonia in the perfused rat liver. J. Biol.
Chem. 271: 16199-16207, 1996.
4.
Brosnan J. T. Glutamate, at the interface between amino acid and Carbohydrate
Metabolism J. Nut. 130: 988S-990S, 2000.
5.
Brosnan J.T., Brosnan M.E., Yudkoff M., Nissim-Ilana, Daikhin Y., Lazarow A., Horyn
O., and Nissim I: Alanine metabolism in the perfused rat liver J. Biol. Chem. 276:
31876-31882, 2001.
6.
Cabella C., Gardini G., Corpillo D., Testore G., Bedino S., Solinas S.P., Cravazola C.,
Vargiu C., Grillo M.A., and Colombatto S. Transport and metabolism of agmatine in rat
hepatocyte cultures, Eur. J. Biochem. 268: 940-947, 2001.
7.
Carey G.B., Cheung Chia-Wei, Cohen N.S., Brusilow S., and Raijman L. Regulation of
urea and citrulline synthesis under physiological conditions 292: 241-247, 1993.
8.
Cheung C. W, and Raijman L. Arginine, mitochondrial arginase, and the control of
carbamyl phosphate synthesis. Arch. Biochem. Biophys. 209: 643-649, 1981.
9.
Cheung C. W., Cohen N. S. , and Raijman L. Channeling of urea cycle intermediates in
situ in permeabilized hepatocytes. J. Biol. Chem. 264:4038-4044, 1989.
10.
Come B., Kasumov T., Bradley P.A., Puchowicz M.S., Dahms W., Kerr D., Nissim, I.,
and Brunengraber H. Identification of phenylbutyrylglutamine, a new metabolite of
phenylbutyrate metabolism in humans J. Mass Spectrometry. In. Press, 2002.
11.
Curran R.D., Billiar T.R., Stuehr D.J., Hofmann K., Simmons R.L. Hepatocytes produce
nitric oxides from L-arginine in response to inflammatory products of Kupffer cells. J.
Exp. Med. 170P 1769-1774, 1989.
12.
Czok, R., Lamprecht, W. Pyruvate, phosphoenolpyruvate and D-glycerate-2-phosphate
In: Methods of Enzymatic Analysis. H.U.Bergemeyer, Ed., Academic press, 3: PP,
1446-1451, 1974.
13.
Fahien L.,Schoder J.M., Gehred G.A. and Cohen P.P. Studies on the mechanism of
action of N-acetylglutamate as an activator of carbamoyl phosphate synthetase. J. Biol.
Chem. 239: 1935-1942, 1964.
MS#E00246-2002, Nissim et al 25
14.
Gen L., Regunathan D., Barrow C.J., Esraghi J., Cooper R., and Reis D.J. Agmatine: an
endogenous clonidine-displacing substance in the brain Sicence 263: 12231-12234,
1994.
15.
Gutmann, I., Wahlefeld A W. L-(+)-Lactate In: Methods of Enzymatic Analysis.
H.U.Bergemeyer, Ed., Academic press, 3: PP, 1464-1472, 1974.
16.
Hall L.M., Metzenberg R. L. and Cohen P.P. Isolation and characterization of a naturally
ocurring cofactor of carbamyl phosphate biosynthesis. J. Biol. Chem. 230: 1013-1021,
1958.
17.
Iyer R., Jenkinson C.P., Vockley J., G., Kern R., M., Grody W., W., and Cederbaum S.
The human arginases and arginase deficiency, J. Inher. Metab. Dis. 21: 86-100, 1998.
18.
Jones, B.N and Gilligan, J.P. Ortho-phthaldialdehyde precolumn derivatization and
reversed-phase high-performance liquid-chromatography of polypeptide hydrolysates
and physiological fluids J. Chromatogr. 266, 471-482, 1983.
19.
Kamemoto E.S., Atkinson D.E. Modulation of the activity of rat liver acetylglutamate
synthase by pH and arginine concentration, Arch. Biochem. Biophy. 243: 100-107,
1985.
20.
Krebs, H. A., and Henseleit K. Investigation of urea metabolism in the animal body.
Hoppe-Syeler's Z. Physiol. Chem. 210: 33-36, 1932.
21.
Marletta M.A. Nitric oxide synthase structure and mechanism J Biol Chem 268: 966969, 1993.
22.
Meijer A. J., and Van Woerkom G. M. Turnover of N-acetylglutamate in isolated rat
hepatocytes Biochimica et Biophysica Acta 721: 240-246, 1982.
23.
Meijer A. J., Van Woerkom G. M., Wanders R. J. A., and Lof C. Transport of Nacetylglutamate in rat-liver mitochondria, Eur. J. Biochem. 124: 325-330, 1982.
24.
Meijer A. J., and Verhoeven A.J. Regulation of hepatic glutamine metabolism. Biochem.
Soc. Trans. 14: 1001-1004, 1986.
25.
Meijer A.J., Lamers W.H. and Chamuleau R.A.F.M Nitrogen metabolism and ornithine
cycle function . Physiological Reviews 70: 701-748, 1990.
26.
Morita T., Miura S., Mori M., Tatibana M. Transport of the precursor for rat-liver
ornithine carbamoyltransferase into mitochondria in vitro E. J. Biochem. 91: 563-569,
1982.
27.
Morimoto H., Brady J.F., and Atkinson D.E. Effect of Level of Dietary Protein on
Arginine-Stimulated Citrulline Synthesis, Biochem. J. 272: 671-675, 1990.
MS#E00246-2002, Nissim et al 26
28.
Nissim I.,Yudkoff M., Terwilliger T., and Segal S. Rapid determination of [guanidine15
N]Arginine in plasma with Gas Chromatography-Mass Spectrometry: Application to
human metabolic studies, Anal. Biochem.131: 75-82, 1983.
29.
Nissim I., Cattano C., Nissim Ilana and Yudkoff M. Relative role of the glutaminase,
glutamate dehydrogenase, and AMP-deaminase pathway in hepatic ureagenesis: studies
with 15N and GC-MS. Arch. Biochem. Biophys. 292: 393-401, 1992.
30.
Nissim, I, Weinberg J.M. Glycine attenuates Fanconi syndrome induced by maleate or
ifosfamide in rats (1996) Kidney Int. 49: 684-695, 1996.
31.
Nissim I., Yudkoff M., and Brosnan J.T. Regulation of [15N] urea synthesis from [515N] glutamine: Role of pH, hormones and pyruvate. J. Biol. Chem. 271: 31234-31242,
1996.
32.
Nissim I., M.E. Brosnan,Yudkoff M., Ilana Nissim and Brosnan J.T: Studies of hepatic
glutamine metabolism in the perfused rat liver with 15N-labeled glutamine. J. Biol.
Chem. 274: 28958-28965, 1999.
33.
O'Sullivan D., Brosnan J.B., and Brosnan M.E. Hepatic zonation of the catabolism of
arginine and ornithine in the perfused rat liver, Biochem. J. 330: 627-632, 1998.
34.
O'Sullivan D., Brosnan J.B., and Brosnan M.E. Catabolism of arginine and ornithine in
the perfused rat liver: Effect of dietary protein and of glucagons, Am. J. Physiol. 278:
E516-E521, 2000.
35.
Pastor C.M., and Billiar T.R., Nitric oxide causes hyporeactivity to phenylephrine in
isolated perfused livers from endotoxin-Treated Rats, Am. J. Physiol. 268: G177-G182,
1995.
36.
Paxinou E., Weisse M., Chen Q., Souza J.M., Hertkon C., Selak M., Daikhin E.,
Yudkoff M., Sowa G., Sessa W.C., and Ischiropoulos H. Dynamic regulation of
metabolism and respiration by endogenously produced nitric oxide protects against
oxidative stress, Proc. Natl. Acad. Sci. 98: 11575-11580, 2001.
37.
Rasch W., Regunathan S., Li G., and Reis D.J. Agmatine, the bacterial amine, is widely
distributed in mammalian tissues, Life Sci. 56: 2319-2330, 1995.
38.
Reczkowski R. S., and Ash D.E. Rat liver arginase: kinetic mechanism, alternate
substrates, and inhibitors. Arch.. Biochem. Biophy. 312: 31-37, 1994.
39.
Regunathan S., and Reis D.J. Characterization of arginine decarboxylase in rat brain and
liver: Distinction from ornithine decarboxylase, J. Neurochemistry 74: 2201-2208, 2000.
40.
Reis D.J., and Regunathan S. Is agmatine a novel neurotransmitter in brain? Trends
Pharmacl. Sci. 5:187-193, 2000.
41.
Satriano J., Kelly C.J., and Blantz R.C. An emerging role for agmatine, Kidney Inter. 56:
1252-1253, 1999.
MS#E00246-2002, Nissim et al 27
42.
Satriano J. Isome M., Casero R.A., Thomson S.C., and Blantz R.C. Polyamine transport
system mediates agmatine transport in mammalian cells, Am. J. Physiol. Cell Physiol.
281: C329-C334, 2001.
43.
Stanely C. A., Lieu BS, Hsu BYL, Alberto BB, Greenberg CR, Hopwood NJ, Perlman
K, Rich BH, Zammarchi E, and Poncz M. Hyperinsulinism and hyperammonemia in
infants with regulatory mutations of the glutamate dehydrogenase gene. New England J.
Medicine, 338:1352-1357,1998.
44.
Stewart P.M., Walser M. Short term regulation of ureagenesis. J. Biol. Chem. 255:
5270-5280, 1980.
45.
Weinberg JM, Venkatachalam MA, Roeser NF, and Nissim I. Mitochondrial dysfunction
during hypoxia/reoxygenation and its correction by anaerobic metabolism of citric acid
cycle intermediates. Proc. Nat. Acad. Sci., USA, 97:2826-2831, 2000.
46.
Wolfe R.R. Measurment of urea kinetics in vivo by means of a constant tracer infusion
of di-15N-urea.. Am. J. Physiol. 240: E428-E434, 1981.
47.
Wu G., and Morris S.M. Arginine metabolism: Nitric oxide and beyond, Biochem. J.
336: 1-17, 1998.
48.
Yang D., Puchowicz A., David F., Powers L., Halperin M.L., and Brunengraber H. 15N
enrichment of ammonium, glutamine-amide and urea, measured via Mass Isotopomer
analysis of hexamethylenetetramine, J. Mass Spectrom. 34: 1130-1136, 1999.
49.
Young V.R., and Ajami A.M. Glutamine: The emperor or his clothes? J. Nutr. 131:
2449S-2459S, 2001.
MS#E00246-2002, Nissim et al 28
TABLE 1
Effect of Agmatine, Arginine or SNP on Nitrogen Balance.
Experimental Group
Control
(+) Agmatine
(+) Arginine
(+) SNP
( nmoles / g /min )
Nitrogen Uptake
From Glutamine-N
466 ± 64
1877 ± 150
812 ± 50
1020 ± 218
From Ammonia-N
661 ± 24
775 ± 29
798 ± 20
632 ± 19
As Urea-N
961 ± 87
1745 ± 99
1228 ± 120
1146 ± 24
*As Amino-N
70 ± 13
145 ± 24
104 ± 15
79 ± 18
Nitrogen Output
`
Values are means ± S.D of data points presented in Figure 4 between 40-70 minutes of perfusions.
Values of total glutamine and urea N are nmoles/min/g X 2.
*Amino-N represents the sum of glutamate and alanine nitrogen.
MS#E00246-2002, Nissim et al 29
Table 2
Hepatic Content of Agmatine, N-Acetylglutamate and Amino Acids Following 70 Minutes of
Perfusion With Either Agmatine, Arginine or SNP
Experimental Group
Metabolite
Control (n=15)
(+)Agmatine (n=18)
(+)Arginine (n=9)
(+)SNP (n=6)
Agmatine1
29 ± 4
311 ± 61.8**
129 ± 40**
27 ± 3
N-Acetylglutamate1
28.9 ± 8.6
Ornithine1
46.2 ± 15.7
71.9 ± 33.7**
134 ± 37**
74 ± 8**
Citrulline1
66 ± 25
57 ± 26
96 ± 34**
52 ± 15
Arginine1
51 ± 21
69 ± 17
148 ±37**
75 ± 22
Glutamine2
2.4 ± 0.7
2.7 ± 0.8
1.8 ± 0.3
1.6 ± 0.4**
Glutamate2
1.31 ± 0.4
1.6 ± 0.7
1.5 ± 0.5
0.8 ± 0.2**
Alanine2
0.9 ± 0.2
0.9 ± 0.3
0.8 ± 0.1
0.5 ± 0.1**
Aspartate2
0.29 ± 0.1
0.36 ± 0.1
0.28 ± 0.1
0.19 ± 0.1**
47.6 ± 16.6**
33.6 ± 8.5
26.8 ± 6.7
Perfusions performed as described in Materials and Methods. Livers were perfused with either
agmatine (100 µM), arginine (500 µM) or SNP (25 µM). Perfusions were terminated as described
and processed for quantitation of the indicated amino acid and agmatine. Perfusate also included
glutamine (1 mM), NH4Cl (0.3 mM), lactate (2.1 mM) and pyruvate (0.3 mM). Each result is
expressed as the mean value ± SD. The number of experimental determinations is given in
parentheses for each experimental group.
**
P < 0.05, that is, significantly different from control perfusions.
1 – nmol/g wet weight
2 - µmol/g wet weight
MS#E00246-2002, Nissim et al 30
FIGURE LEGENDS
FIGURE 1:
Schematic presentation of external L-[guanidino-15N2]arginine metabolism
and formation of [guanidino-15N2]agmatine and [15N 2]urea in the liver. This schema reflects
perfusions with L-[guanidino-15N2]arginine and unlabeled glutamine (1mM) and ammonia (0.3mM).
Enzymatic steps presented are: 1, nitric oxide synthetase; 2, arginine decarboxylase; 3, arginase II;
4, N-acetylglutamate synthetase; 5, glutamate dehydrogenase (GDH); 6, phosphate-dependentglutaminase (PDG); 7, carbamoyl phosphate synthetase I (CPS-I); and 8, arginase I. ? indicates a
questionable pathway ; + , indicates stimulatory action or co-factor.
FIGURE 2: Uptake of perfusate L-[guanidino-15N2]arginine (------), and B, the output in
the effluent, of [15N2]urea (---L--); [guanidino - 15N2]agmatine (---6--) ; and nitric oxide as
NO-2+NO-3 (------). Also is shown oxygen consumption during the course of perfusions (----).
Bottom panel C, represents the 15N enrichment in arginine and agmatine in liver extracts at the end
of 70 minutes perfusions. Bars are means ± S.D for three livers.
FIGURE 3: Dose dependence of agmatine uptake (A), oxygen consumption (B) and total
urea N output (C), during the course of perfusions with increasing concentration of agmatine.
Perfusions were carried out with a basic perfusate for 15 minutes and then perfusate was replaced
by perfusate contained 1mM glutamine, 0.3 mM ammonia and the indicated concentration of
agmatine. The dose dependence was studied in the same liver, and at the indicated time a separate
perfusate reservoir containing the indicated agmatine concentration was used. Bars are means ± S.D
for three livers.
MS#E00246-2002, Nissim et al 31
FIGURE 4:
Total nitrogen balance and O2 consumption in perfused liver. Livers were
perfused without (control) or with either agmatine (100 µM), SNP (25 µM) or arginine (500 µM)
and 1mM [2-15N]- or [5-15N]-labeled glutamine and 0.3 mM unlabeled ammonia or 15NH4Cl and
unlabeled glutamine as nitrogen precursors. Since all perfusions were performed with the same
concentrations of substrates (i.e., 1 mM glutamine, 0.3 mM NH4Cl, 2.1 mM lactate and 0.3 mM
pyruvate), it was possible to combine those with 15N-labeled glutamine or ammonia. The data are
the means ± S.D, for control perfusions (--O--, N=15), plus agmatine (--G--, N=18), plus arginine (-6,--, N=9) and plus SNP (--▲--, N=6). Values of total glutamine and urea N are nmoles/min/g of
these metabolites times 2. Panel A, represents glutamine N uptake; B, O2 consumption; C, ammonia
uptake; and D, nitrogen output.
FIGURE 5: Flux through the PDG reaction during the course of liver perfusions. Values
for PDG were estimated from perfusions with [5-15N]glutamine as detailed in (31, 32). The data are
the means ± S.D, for control perfusions (--O--, N=5), plus agmatine (--G--, N=5), plus arginine (-6,--, N=5) and plus SNP (--▲--, N=4).
FIGURE 6:
Total [15N]urea production (sum of Um+1 and Um+2), during the course of liver
perfusion with [5-15N]glutamine (panel A), [2-15N]glutamine (panel B) and 15NH4Cl (panel C).
Perfusion conditions are as indicated in Figure 4. The output of 15N-labeled urea is the product of
15
N enrichment (mole % excess/100) times one half of total urea N (nmoles N/min/g) for Um+1, and
times total urea N for Um+2. The data are the means ± S.D, for control perfusions (--O--, N=3-5),
plus agmatine (--G--, N=4-6), plus arginine (--6,--, N=3-5) and plus SNP (--▲--, N=3-4).
MS#E00246-2002, Nissim et al 32
FIGURE 7: The dependence of NAG levels and urea synthesis on agmatine concentrations
in freeze-clamped liver at the end of perfusions. Data points included in these correlations are from
perfusions with [2-15N]glutamine, [5-15N]glutamine and 15NH4Cl of control, plus agmatine, arginine
and SNP perfusions. The bottom panels represent the relationship between hepatic agmatine or
NAG in freeze-clamped liver extracts and [15N]urea in the effluent at the end of perfusions.
FIGURE 8:
Appearance of 15N-labeled N-acetylglutamate in extracts of livers that were
freeze-clamped at the end of perfusions with [2-15N]glutamine. Amount of 15N-labeled NAG is the
product of isotopic enrichment (MPE/100) times total concentration (nmoles/g). The data are the
means ± S.D, for control perfusions (N=4), plus agmatine (N=4), plus arginine (N=3) and plus SNP
( N=3). p values indicate the degree of significance compared with control level.
MS#E00246-2002, Nissim et al 33
Figure 1
Nissim et. al.
Cytosol
15NO
Mitochondrion
+ [15N] Citrulline
3
[15N2] Urea + Ornithine
1
L-[guanidino-15N2] Arginine
15N
2
2- Arginine
[guanidino-15N2] Agmatine
CO2
(?)
}
(?)[15N2] Urea
[14N2] Urea
8
Urea
ArginineCycle
Ornithine
NAG
HCO–3
CP
7
+
NH
NH4Cl
Glutamine
+
Acetyl-CoA
4
+
5
4
+
Glutamine
6
Glutamate
MS#E00246-2002, Nissim et al 34
MS#E00246-2002, Nissim et al 35
MS#E00246-2002, Nissim et al 36
MS#E00246-2002, Nissim et al 37
MS#E00246-2002, Nissim et al 38
Nissim et. al.
Total [15N] Urea Nitrogen Output
(nmoles/min/g)
Figure 6
600
550
500
450
400
350
300
250
200
150
100
50
0
A
15 20
25 30 35 40 45
50 55 60
Total [15N] Urea Nitrogen Output
(nmoles/min/g)
Time (min)
600
550
500
450
400
350
300
250
200
150
100
50
0
B
15
20
25 30 35 40 45
50 55
60
Total [15N] Urea Nitrogen Output
(nmoles/min/g)
Time (min)
2500
2250
C
2000
1750
1500
1250
1000
750
500
250
0
15 20 25 30 35 40 45 50 55 60
Time (min)
65
70
MS#E00246-2002, Nissim et al 39
MS#E00246-2002, Nissim et al 40