(Received for publication, October 2, 1996, and in revised form, November 7, 1996)
From the Department of Molecular Genetics, the
§ Second Department of Internal Medicine, and the
¶ Second Department of Pathology, Kumamoto University School of
Medicine, Kuhonji 4-24-1, Kumamoto 862, Japan
Nitric oxide is synthesized by nitric-oxide
synthase from arginine, a common substrate of arginase. Rat peritoneal
macrophages were cultured in the presence of bacterial
lipopolysaccharide (LPS), and expression of the inducible isoform of
nitric-oxide synthase (iNOS) and liver-type arginase (arginase I) was
analyzed. mRNAs for iNOS and arginase I were induced by LPS in a
dose-dependent manner. iNOS mRNA appeared 2 h
after LPS treatment and increased to a near maximum at 8-12 h. On the
other hand, arginase I mRNA that was undetectable prior to the
treatment began to increase after 4 h with a lag time and reached
a maximum at 12 h. Immunoblot analysis showed that iNOS and
arginase I proteins were also induced. mRNA for arginase II, an
arginase isozyme, was not detected in the LPS-activated peritoneal
cells. mRNA for CCAAT/enhancer-binding protein (C/EBP
), a
transactivator of the arginase I gene, was also induced, and the
induction was more rapid than that of arginase I mRNA. Changes in
iNOS and arginase I mRNAs were also examined in LPS-injected rats
in vivo. iNOS mRNA increased rapidly in the lung and
spleen, reached a maximum 2-6 h after the LPS treatment, and decreased
thereafter. Arginase I mRNA was induced markedly and more slowly in
both tissues, reaching a maximum in 12 h. Thus, arginase I appears
to have an important role in down-regulating nitric oxide synthesis in
murine macrophages by decreasing the availability of arginine, and the
induction of arginase I is mediated by C/EBP
.
Nitric oxide (NO)1 is a major molecule regulating blood vessel dilatation and immune response and functions as a neurotransmitter in the brain and peripheral nervous system (see Refs. 1-3 for reviews). NO is synthesized from arginine by nitric-oxide synthase (NOS), generating citrulline. Cellular NO production is absolutely dependent on the availability of arginine. This amino acid can be obtained from exogenous sources via the blood circulation, from intracellular protein degradation, or by the endogenous synthesis of arginine. Major sites of arginine synthesis in ureotelic animals are the liver, where arginine generated in the urea cycle (ornithine cycle) is rapidly converted to urea and ornithine by arginase, and the kidney, where arginine is synthesized from citrulline and released into the blood circulation (see Ref. 4 for a review). In other tissues and cell types, arginine can be generated from citrulline, which is produced as a coproduct of the NOS reaction, forming a cycle that is composed of NOS, argininosuccinate synthetase, and argininosuccinate lyase and that is termed the "citrulline-NO cycle" (5-10). The inducible isoform of NOS (iNOS) and argininosuccinate synthetase are coinduced in activated murine macrophage-like RAW 264.7 cells (8), in cultured vascular smooth muscle cells (9), and in vivo (10, 11). Argininosuccinate lyase is also induced in vivo (10, 11).
On the other hand, arginine is utilized for both the arginase and NOS
reactions. Thus, these two enzymes compete for arginine. At least two
isoforms of arginase are present. Liver-type arginase (arginase I) is
expressed almost exclusively in the liver and catalyzes the final step
of urea synthesis. Arginase I activity (12) and mRNA (13) in rat
liver increase markedly in the perinatal period, in coordination with
other urea cycle enzymes. The enzyme is regulated by dietary protein
(14) and hormones (15). Arginase I consists of three identical subunits
of about 35,000 Da. cDNA clones were isolated from rat (16, 17) and
human (18, 19) liver. The rat (20) and human (21) genes are 11.5-12 kb
long and consist of 8 exons. Promoter and enhancer regions of the rat gene were characterized (22-24). In addition to arginase I, an isozyme
(arginase II) is present in extrahepatic tissues, including kidney,
small intestine, and lactating mammary gland (25-27). The coinduction
of NOS and arginase II activities in RAW 264.7 cells activated by
lipopolysaccharide (LPS) was reported (28). We isolated cDNA for
arginase II and showed that it is 59% identical with arginase I on the
amino acid level (29). We also found that the enzyme is mitochondrial
and that its mRNA is induced by LPS, dexamethasone, and cyclic
adenosine monophosphate and is reduced by interferon- (29). To
better understand the role of arginase isoforms in NO synthesis, we
examined expression of the isoforms and iNOS in cultured rat peritoneal
cells and in LPS-treated rats using RNA blots, immunoblots, and
immunocytochemical analyses. We report here that iNOS and arginase I
mRNAs and proteins are coinduced by LPS in cultured rat peritoneal
macrophages and in the lung and spleen in vivo. The
induction of mRNA for CCAAT/enhancer-binding protein
(C/EBP
), a transactivator of the arginase I gene, is also
described.
A monoclonal antibody against mouse iNOS was obtained from Transduction Laboratories (Lexington, KY). Antiserum against human arginase I was raised in a rabbit. Purified recombinant human arginase I (30) was provided by M. Ikemoto of Kyoto University, Kyoto, Japan. A monoclonal antibody RM4 against rat macrophages will be described elsewhere (41).
Preparation and Culture of Rat Peritoneal CellsSpecific pathogen-free male Wistar rats (7 weeks of age) were given 5 ml of 10% polypeptone (Difco) intraperitoneally, and peritoneal cells were harvested 3 days after the injection (31). The cells were seeded in 10-cm culture dishes at 5 × 106 cells/dish in RPMI 1640 medium with 10% heat-inactivated fetal calf serum. The cells were cultured in the absence or presence of LPS at 37 °C under 5% CO2 in air.
Animals and LPS TreatmentMale Wistar rats (5-6 weeks of age) were given Escherichia coli LPS (serotype 0127:B8, Sigma) intraperitoneally at 1.0 mg/kg of body weight and killed at the indicated times following anesthetization with ether.
RNA Blot AnalysisTotal RNA from rat tissues and packed
peritoneal cells was prepared by the guanidium
thiocyanate-phenol-chloroform extraction procedure (32). After
electrophoresis in formaldehyde-containing agarose gels, RNAs were
transferred to nylon membranes. Hybridization was performed using as
probes digoxigenin-labeled antisense RNAs for rat iNOS (10), rat
arginase I, mouse C/EBP, or rat arginase II. The antisense RNAs were
synthesized using as templates pcDNAII-riNOS (10),
pcDNAII-rAI-1, pcDNAII-C/EBP
, and pGEM-rAII-1 (29), respectively. To obtain the plasmid pcDNAII-rAI-1, the
approximately 850-bp EcoRI-EcoRV fragment of
pARGr-2 (17) was inserted into the EcoRI and
EcoRV sites of the plasmid pcDNAII (Invitrogen, San
Diego, CA). To obtain the plasmid pcDNAII-C/EBP
, the
approximately 1.6-kb BstXI fragment of pEF-C/EBP
(33) was
inserted into the BstXI site of the pcDNAII.
Chemiluminescence signals derived from hybridized probes were detected
on x-ray films using a DIG luminescence detection kit (Boehringer
Mannheim) and quantified using the MacBas bioimage analyzer (Fuji Photo
Film Co., Tokyo, Japan).
Rat tissues were excised and homogenized in 9 volumes of 20 mM potassium HEPES buffer, pH 7.4, containing 1 mM dithiothreitol, 50 µM antipain, 50 µM leupeptin, 50 µM chymostatin, and 50 µM pepstatin. Packed rat peritoneal cells were homogenized in 9 volumes of the same buffer containing 0.5% Triton X-100. The homogenates were centrifuged at 25,000 × g for 30 min at 4 °C and the supernatants were used as tissue or cell extracts. The extracts were subjected to SDS-polyacrylamide gel electrophoresis and proteins were electrotransferred to nitrocellulose membranes. Immunodetection was performed using an ECL kit (Amersham Corp.) according to the manufacturer's protocol. Protein was determined with protein assay reagent (Bio-Rad) using bovine serum albumin as a standard.
Immunocytochemical StainingCultured peritoneal cells were prepared for immunocytochemical staining by cytocentrifuge. Briefly, 5 × 104 cells were suspended in 100 µl of RPMI 1640 medium containing 10% heat-inactivated fetal calf serum, and then the cell suspension was applied on cytospin slides by centrifugation at 600 rpm for 5 min (Cytospin 2, Shandon, Astmoor, United Kingdom). Immunocytochemical staining was done as described (10).
Rat peritoneal cells were cultured in the presence of LPS,
and expression of iNOS and arginase I mRNAs was studied. Fig.
1 shows the effects of LPS concentration on iNOS and
arginase I mRNAs. iNOS mRNA of about 4.5 kb, which was not
detectable in the absence of LPS, was evident at 1 ng/ml and gradually
increased with increasing concentrations of LPS up to 10 µg/ml.
mRNA for arginase I (liver-type arginase) of about 1.7 kb was also
induced by LPS. However, dose dependence differed from that of iNOS
mRNA. Arginase I mRNA was barely detectable with 1 ng/ml of
LPS; it was detected with 10 ng/ml and the level increased with up to 10 µg/ml. The maximal mRNA concentration seen with 10 µg/ml of LPS was comparable to that in the liver.
Fig. 2 shows the time course of induction of iNOS and
arginase I mRNAs in the peritoneal cells treated with 10 µg/ml of
LPS. iNOS mRNA was detected at 2 h, increased with time,
reached a near maximum at 8 h and a maximum at 12 h, and
remained at the near maximal level at 24 h. On the other hand,
arginase I mRNA began to increase after 4 h with a lag time
and reached a maximum at 12-24 h. Thus, iNOS and arginase I mRNAs
are coinduced by LPS in rat peritoneal cells, but arginase I mRNA
is induced more slowly than iNOS mRNA. Arginase II mRNA, which
is 57% identical with arginase I mRNA, was not detected in
LPS-treated peritoneal cells under conditions where the mRNA of 1.8 kb was clearly detected in the rat intestine (data not shown).
Induction by LPS of C/EBP
We
reported that C/EBP family members bind to both the promoter and
enhancer regions of the rat arginase I gene (22, 23) and that the
promoter is activated by C/EBP and other members (24). We then asked
whether C/EBP
mRNA would also be induced by LPS in rat
peritoneal cells. mRNA was present at a very low level prior to the
LPS treatment; it increased with time, reached a maximum at 12 h,
and then decreased (Fig. 2, A, c, and B, c). Thus, induction of C/EBP
mRNA was more rapid than that of
arginase I mRNA.
Fig. 3 shows the time course of induction of
iNOS and arginase I proteins in cultured peritoneal cells after
exposure to LPS. iNOS protein of about 150 kDa was induced 8 h
after the treatment; it increased at 24 h, then decreased. On the
other hand, two polypeptides of about 35 and 38 kDa immunoreacted with
the arginase I antibody, comigrated with the two forms of arginase I in
the liver, and were observed 8 h after the treatment, then
increased to 48 h. These two polypeptides of arginase I apparently
arose by alternative translation initiation from the two methionine
residues located 30 base pairs apart (17, 20, 34). Concentration of
arginase I protein in the LPS-treated peritoneal cells was comparable
to that in the liver; these findings were in accord with the
observation that mRNA concentrations in the activated peritoneal
cells and in the liver were comparable (see above).
Immunocytochemical Detection of iNOS and Arginase I in Peritoneal Cells
To identify cells positive for iNOS and arginase I in
peritoneal cells, immunocytochemical analysis was performed (Fig.
4). iNOS immunoreactivity was absent prior to LPS
treatment. However, most cells became strongly positive for iNOS
immunoreactivity after the treatment. Arginase I was negative prior to
the LPS treatment, but most cells became positive after the treatment. Distributions of the iNOS-positive cells, the arginase I-positive cells, and macrophages positive to the macrophage-specific antibody RM
4 were similar. However, cellular staining patterns of iNOS and
arginase I differed from that with antibody RM4; iNOS and arginase I
are cytosolic proteins, whereas the antibody RM4 is specific to
lysosomes in macrophages.2 Therefore, iNOS and arginase I
are coinduced in practically all peritoneal macrophages.
Induction of mRNAs and Proteins for iNOS and Arginase I in the Lung and Spleen of LPS-treated Rats
We then examined expression
of iNOS and arginase I in vivo. We previously reported that
iNOS mRNA was induced in various tissues of rats injected
intraperitoneally with LPS (10); the induction was the strongest in the
lung and spleen. mRNAs for arginase I as well as for iNOS in the
lung were measured at various times after LPS treatment (Fig.
5). iNOS mRNA in the lung increased to a near
maximum 2 h after the treatment, reached a maximum at 6 h,
decreased thereafter, and returned to a hardly detectable level at
24 h. Arginase I mRNA was also induced. mRNA was present at a very low level before treatment, began to increase after 2 h
with a lag time, reached a maximum at 12 h on the average, and
then decreased slowly. Concentration of the mRNA at 12 h was over 10-fold lower than that in the liver (data not shown).
iNOS mRNA in the spleen showed an increase similar to that in the lung, but the level decreased more slowly than seen in the lung (10). Arginase I mRNA was also induced in the spleen, but less markedly than in the lung, and the time course was similar to that in the lung and slower than that of iNOS mRNA in this organ (data not shown).
We reported that iNOS protein was markedly induced in the lung and spleen of LPS-treated rats (10). The induction of arginase I protein in the lung was also examined using immunoblot analysis. A polypeptide that comigrated with the larger form of arginase I was induced 24 h after LPS-treatment (data not shown). Concentration of arginase I in the lung of the LPS-treated rat was over 10-fold lower than that in the liver.
NO synthesis is regulated by depending on the availability of
arginine, substrate of the NOS reaction, as well as by NOS activity and
other factors. Arginine can be obtained via the blood
circulation or by endogenous synthesis from citrulline. Arginine
transport into cultured macrophages increases in response to LPS and to interferon- (35, 36). The cat-2 gene of the arginine
transporter in cultured vascular smooth muscle cells is stimulated by
interleukin-1
and by tumor necrosis factor-2 (37). Furthermore,
argininosuccinate synthetase and argininosuccinate lyase, which
together synthesize arginine from citrulline, are induced in stimulated
murine macrophages (8, 10) and aortic smooth muscle cells (9), and
intracellular arginine synthesis is enhanced.
Activity of arginase that degrades arginine is also induced in activated macrophages (38, 39). Arginase induced in activated RAW 264.7 cells, a mouse macrophage-like cell line, was shown to be arginase II (28, 29). On the other hand, in the present work, we noted that arginase I, not arginase II, is induced by LPS in primary cultured rat peritoneal macrophages and in the lung and spleen in vivo. This was unexpected because arginase I has been thought to be expressed almost exclusively in the liver. Differential induction of the two arginase isoforms in the mouse macrophage-like cell line and in primary cultured rat macrophages may be due to differences in animal species or to differences between the established cell line and the primary cultured cells. Arginase I mRNA was induced more slowly than iNOS mRNA. The induced arginase I probably decreases arginine availability for the NOS reaction in activated macrophages and may down-regulate the overproduction of NO. All of these results suggest that there is a complex regulation of genes encoding enzymes and transporter proteins involved in arginine metabolism that together control NO production in cells.
Arginase I mRNA was induced slowly in cultured peritoneal
macrophages and also in vivo with an apparent lag time of a
few hours. Therefore, the induction of arginase I appears to be
mediated by a transcription factor(s) that is synthesized de
novo in response to LPS stimulation. C/EBP family members bind to
the promoter and enhancer regions of the arginase I gene and activate
the promoter (Refs. 22-24; see Ref. 40 for a review), C/EBP being
the most potent. The present study also shows that C/EBP
mRNA is
induced in the LPS-stimulated peritoneal macrophages. The induction of C/EBP
mRNA is more rapid than that of arginase I mRNA. Thus, the induction of arginase I appears to be mediated, at least in part,
by the induction of C/EBP
.
We thank S. Chowdhury and K. Iwase for
providing the C/EBP RNA probe and the arginase I RNA probe,
respectively. We also thank M. Ohara for comments on the manuscript and
M. Imoto for secretarial services.