- Email: [email protected]
Enhanced utilization and altered metabolism of arginine in inflammatory macrophages caused by raised nitric oxide synthesis
The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
Enhanced utilization and altered metabolism of arginine in inflammatory macrophages caused by raised nitric oxide synthesis Nándor Müllner, Ágnes Lázár, András Hrabák∗ Department of Medical Chemistry, Molecular Biology and Pathobiochemistry, Semmelweis University, VIII. Puskin u. 9., P.O. Box 260, Budapest H-1444, Hungary Received 3 July 2001; received in revised form 15 February 2002; accepted 20 February 2002
Abstract Nitric oxide (NO) production was increased in macrophages during inflammation. Casein-elicitation of rodents causing a peritoneal inflammation offered a good model to study alterations in the metabolism of l-arginine, the precursor of NO synthesis. The utilization of l-arginine for NO production, arginase pathway and protein synthesis were studied by radioactive labeling and chromatographic separation. The expression of NO synthase and arginase was studied by Western blotting. Rat macrophages utilized more arginine than mouse macrophages (228 ± 27 versus 71 ± 12.8 pmol per 106 macrophages). Arginine incorporation into proteins was low in both species (<15% of labeling). When NO synthesis was blocked, arginine was utilized at a lower general rate, but l-ornithine formation did not increase. The expression of enzymes utilizing arginine increased. NO production was raised mainly in rats (1162 ± 84 pmol citrulline per 106 cells) while in mice both arginase and NO synthase were active in elicited macrophages (677 ± 85 pmol ornithine and 456 ± 48 pmol citrulline per 106 cells). We concluded, that inflammation induced enhanced l-arginine utilization in rodent macrophages. The expressions and the activities of arginase and NO synthase as well as NO formation were increased in elicited macrophages. Specific blocking of NO synthesis did not result in the enhanced effectivity of the arginase pathway, rather was manifested in a general lower rate of arginine utilization. Different rodent species reacted differently to inflammation: in rats, high NO increase was found exclusively, while in mice the activation of the arginase pathway was also important. © 2002 Elsevier Science Ltd. All rights reserved. Keywords: Resident/inflammatory peritoneal macrophages; Arginine metabolism; Arginine uptake/incorporation; Arginase; Nitric oxide synthase
1. Introduction
Abbreviations: FBS, foetal bovine serum; HBSS, Hank’s balanced salt solution; LPS, lipopolysaccharide; NOS II, inducible nitric oxide synthase; PAGE, polyacrylamide gel electrophoresis; PBS, phosphate buffered saline; PDTC, pyrrolidine dithiocarbamate; SDS, sodium dodecyl sulfate; SMT, S-methyl-isothiourea; TLC, thin layer chromatography ∗ Corresponding author. Tel.: +36-1-266-2755/4089; fax: +36-1-266-2615. E-mail address: [email protected] (A. Hrab´ak).
The metabolism of l-arginine is very important in cells producing nitric oxide (NO). Macrophages represent one of these cell types. They metabolize l-arginine via multiple pathways: inducible nitric oxide synthase (NOS II) produces NO and l-citrulline (Cit) [1], arginase hydrolyses l-arginine to urea and l-ornithine (Orn) [2] and l-arginine is also incorporated into cellular proteins. Orn and Cit are precursors of other amino acids and their derivatives. The activities of these pathways in macrophages depend on
1357-2725/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S 1 3 5 7 - 2 7 2 5 ( 0 2 ) 0 0 0 2 8 - 6
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
the activation state of the cell influenced by various cytokines and bacterial endotoxin (lipopolysaccharide, LPS) in vitro [3]. Arginine metabolism shows significant changes during inflammation, influencing NO production in vivo [4]. The reciprocal regulation of NOS II and arginase has been demonstrated both in the induction processes [5] and also in substrate utilization [6]. Previous studies have demonstrated the changes in NOS II and arginase activities after various cytokine treatments in vitro [7] and the mutual inhibition of NOS II and arginase by the products of their catalyzed reactions [8,9]. However, after in vivo stimulations, the distribution of products derived from l-arginine in macrophages of different species during inflammatory reactions has not been studied directly. Since macrophages are involved in inflammatory reactions, the comparison of the arginine utilization in resident and elicited macrophages may give interesting informations about the biochemical mechanism of inflammation. Here we report a comparative study concerning the distribution of the products derived from l-arginine under physiological conditions (pH 7.4, 0.1 mM Arg concentration), in macrophages isolated from control (resident) and casein-elicited (inflammatory) mice and rats, in which different activities of arginase and NOS have been observed previously [10]. We have chosen in vivo i.p. casein-elicitation as a model because it provoked inflammation and informations about the altered activities of arginine utilizing enzymes have been available [10,11]. The changes in the expression of arginase and NOS, as the background of these metabolic alterations were also studied. Experiments to decrease NO production in two different ways (inhibition of NOS induction by pyrrolidine dithiocarbamate and direct inhibition of its activity by S-methyl-isothiourea) were performed to obtain informations showing whether the block of a certain (NOS) pathway of l-arginine metabolism favored alternative metabolic routes or not.
1081
mice and Wistar male rats (Charles River, Hungary) as described earlier [10]. Casein-injected animals were sacrificed on the 4th day after treatment. Isolated peritoneal cells (3×106 per well) were adhered to 24-well Corning plastic dishes for 2 h and then cultured for 24 h at 37 ◦ C in a 5% CO2 atmosphere in 0.5 ml RPMI 1640 medium containing 5% foetal bovine serum (FBS), penicillin (100 IU/ml) and streptomycin (100 g/ml). This period of culturing is sufficient to induce NO synthesis and to elevate arginase activity in elicited cells. After culturing media were removed, cells were washed three times with phosphate buffered saline (PBS, pH 7.4), and kept in 100 l Hank’s balanced salt solution (HBSS) for 2 h to deplete the cells of l-arginine derived from the culture medium. HBSS was renewed before further experiments. The composition of the adherent cell population was tested by light microscopic detection of a NaF-sensitive esterase found only in macrophages. The method of Ornstein et al. [12] was used. Briefly, cells were adhered in 24-well Nunclon plates to Thermanox cover slips (Nunc Inc.) and stained with a buffered solution of naphthol-DS-acetate and Fast Blue, finally studied under a light microscope. In the absence of NaF all cells are positive (blue) while in the presence of NaF only cells differing from macrophages (i.e. the neutrophils) are positive. Two-hundred cells were counted for esterase activity in both cases. 2.2. Measurement of the total protein content in adhered cells After adherence non-adherent cells were removed by multiple washing (PBS) and adherent cells were solubilized in 200 l 2N KOH. Protein was precipitated by excess HClO4 and dissolved in 0.5N NaOH and measured with coomassie blue by the Bradford method [13] using bovine serum albumin as standard. 2.3. Labeling and separation of labeled fractions from macrophages
2. Materials and methods 2.1. Cells, microscopy Macrophages were removed from the peritoneal cavity of untreated and casein-elicited CD-CR-1 male
The 14 C-labeled l-arginine (final concentration 0.1 mM; Amersham, specific activity 100 MBq/mmol) was added to the cells for 60 min at 37 ◦ C. The uptake of labeled arginine was stopped with 20 l 1 M unlabeled l-arginine (in HBSS, pH 7.4, final concen-
1082
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
tration 0.1 M) and cells were washed four times with PBS to remove excess labeling. Ten microliters of the removed first supernatant was separated by thin layer chromatography (TLC) and measured for Arg, Orn and Cit content (extracellular samples, see the following paragraphs). Cells were solubilized in 200 l cold 2 M KOH, 10 l samples were removed to measure the entered radioactivity and the macromolecular fraction was precipitated by adding 0.5 ml 1 M HClO4 . After centrifugation, an aliquot of the supernatant was measured for radioactivity to calculate the amount of labeled arginine taken up by the cells without being incorporated into proteins. In order to characterize the latter, the precipitate was washed once with cold 0.5 M HClO4 , then dissolved in 300 l 0.5 M NaOH and an aliquot was measured for radioactivity. Another sample series was dissolved in 100 l 1N KOH,
precipitated with 100 l 1 M HClO4 and 10 l mixture of unlabeled l-Arg, l-Orn and l-Cit (5–5 mM) was added. After centrifugation 100 l supernatant was tested for radioactivity and 30 l aliquots were submitted to thin layer chromatography to separate Arg, Orn, Cit [14]. Spots were visualized with 1% ninhydrine solution and then cut out. We placed them into 3 ml mixtures of a toluene-based scintillation cocktail and Triton-X-100 (2:1) and then measured for radioactivity by a Packard Trikarb LH 2100 liquid scintillation spectrometer. For each sample four tests were performed: (1) a background sample (“0 min”, i.e. labeled arginine was added after the addition of 1 M Arg), to subtract from the counts measured in other samples; (2) 60 min sample to test the distribution of the labeling in various subfractions; (3) the same as above but derived from macrophages
Fig. 1. Scheme of the separation of various sample fractions for the study of the distribution of macrophages.
14 C-l-arginine
label in murine peritoneal
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
cultured for 24 h in the presence of 50 M pyrrolidine dithiocarbamate (PDTC), an inhibitor of NOS II transcription via NF-B transcription factor; (4) the same as above but in the presence of 0.5 mM S-methyl-isothiourea (SMT), an inhibitor of NOS II activity, during the labeled arginine load without previous PDTC treatment. The scheme of the whole experiment is shown in Fig. 1. 2.4. Measurement of NO synthase and arginase activities in macrophages Macrophages were cultured for 24 h in RPMI–5% FBS and then cells were washed twice with PBS. Cells were lysed by 5 mM HEPES (pH 7.4)–1 mM NADPH–10 M tetrahydrobiopterin and 20 M 14 C-l-arginine (final concentration) was added for 60 min. Reaction was stopped with 5–5 mM Arg–Orn– Cit, an aliquot was chromatographed on TLC plates as described. Another aliquot was used to measure protein by Bradford’s method [13]. Arginase and NOS specific activities were expressed as pmol Orn and Cit per min per mg protein, respectively. 2.5. NO synthase and arginase content in macrophages In order to check the presence and functional state of NOS II, nitrite production was also tested in the supernatants of cultured macrophages by the Griess reaction [15]. Expression of arginine utilizing enzymes was tested by Western blotting. After 24 h culturing in RPMI–5% FBS, 3 × 106 cells were solubilized in a sample buffer containing 3% sodium dodecyl sulfate (SDS) and after heating samples were separated by SDS-polyacrylamide gel electrophoresis (PAGE). Separated samples were transferred to nitrocellulose membrane and then treated with a primary mouse anti-iNOS and anti-arginase I antibody (Transduction Laboratories, England). Secondary antibody was an anti-mouse antibody conjugated with peroxidase (Sigma). Finally samples were developed with ECL solution (Amersham) and fluorescence was detected on a Cronex-6 X-ray film (Agfa/Forte, Hungary). Bands were quantitatively evaluated by an Ultroscan LKB gel densitometer and bands were scanned by Scan Maker II scanner to form computer files for photography.
1083
The presence of arginase isoenzymes was also studied by gel electrophoresis according to Spector et al. [16]. Briefly, 7.5% SDS-free polyacrylamide gels were used and after electrophoresis gels were cut into 5 mM slices, extracted by a Tris–Glycine buffer and 180 l extracts were mixed with 20 l 2 mM [14 C]-l-arginine for 60 min at 37 ◦ C at pH 8.3 to measure arginase activities. Reaction was stopped by adding 200 l 1N HCl and released urea was separated on small Dowex 50 × 8 resin columns. The non-bound [14 C]-urea was measured in a liquid scintillation spectrometer. 2.6. Statistical evaluation of the results Results are presented as mean plus S.E.M. values. Significances were calculated by ANOVA test for multiple comparisons. Since conclusions can be based mainly on the comparison of corresponding sample pairs, the significance of the differences in corresponding pairs was calculated by adjusted t-tests with P-values corrected by the Bonferroni method and by the Mann–Whitney unpaired two-tailed t-test. Differences were considered as significant when P < 0.05. Corresponding pairs for comparison are: mouse and rat samples both from resident and elicited animals; resident and elicited samples both from mouse and rats; mouse- and rat-elicited samples without or with PDTC and SMT, respectively. The number of experiments are reported in figure and table legends.
3. Results 3.1. Identification of adhered cells and measurement of their protein content The cell composition of the peritoneal exudate depends on the time elapsed after the induction of inflammation. After 3–4 days, macrophages are preponderant. This was checked by light microscopy using a non-specific esterase staining (see Section 2). Both in mouse and rat adherent cell populations were almost exclusively macrophages (less than 5% was stained in the presence of NaF while 85–90% of the total population was positive in the absence of NaF). Total protein content of adhered cells was also measured. Macrophages prepared from injected
1084
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
Fig. 2. Distribution of l-arginine derived labeling in macrophages. Abbreviations: MR, mouse resident; MC, mouse casein-elicited; RR, rat resident; RC, rat casein-elicited; P indicates PDTC-treated samples; S indicates samples measured in the presence of 0.5 mM SMT; 3 × 106 macrophages were adhered and cultured and then studied for 14 C-labeling distribution. ANOVA test gave P < 0.0001 for each comparison except for acid-insoluble, precipitated fraction (P > 0.05). Significant differences of corresponding pairs calculated by Bonferroni method: P < 0.001 ∗∗∗ for MC–RC and RC–RCP, P < 0.01 ∗∗ for RR–RC, P < 0.05 ∗ for RCP–RCS both in total and acid-soluble samples (n = 9).
animals contained more protein than resident cells, however, the increase was not significant. The increase was similar in mouse and rat cells. Resident mouse and rat macrophages contained 2.94 ± 0.71 and 3.75 ± 0.65 g protein per 106 cells, respectively, while the protein contents of elicited mouse and rat cells were 3.60 ± 0.38 and 4.08 ± 0.88 g per 106 macrophages. 3.2. Intracellular labeling of the subfractions The amount of intracellular 14 C-arginine labeling in mouse resident peritoneal macrophages was significantly lower than in rat cells (Fig. 2). Labeling was markedly decreased in each PDTC-treated sample and less markedly in the presence of SMT. Casein-elicitation augmented the amount of intracel-
lular labeling both in mouse and rat cells, a strong significance was found. The bulk of entered amino acid was found in both species in the acid-soluble fraction; incorporation of l-arginine into proteins did not exceed 15% and in elicited cells this proportion was even lower. These results show that in macrophages arginine was rather transformed to other amino acids and other small molecules instead of having been incorporated into proteins, although de novo protein synthesis was also detected: PDTC and actinomycin D treatment significantly blocked NOS II transcription indicated by the marked decrease of nitrite production (Table 1) and PDTC decreased the amount of both NOS II and arginase (Fig. 4). Nevertheless, neither PDTC nor SMT treatment decreased further the low incorporation of l-arginine significantly.
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
1085
Table 1 Nitrite production in the supernatants of macrophage cultures Samples
Resident Mouse
Control PDTC Actinomycin Da
4.2 ± 1.9 2.1 ± 0.6 N.D.
Elicited Rat
Mouse 6.7∗
11.9 ± 3.3 ± 1.1 N.D.
Rat 2.0∗∗
13.4 ± 3.4 ± 3.3 0.6
38.0 ± 7.0∗∗ 8.3 ± 2.7 4.7
Cells were cultured for 24 h in RPMI 1640 medium containing 5% foetal bovine serum, then an aliquot of the supernatants was determined directly using the Griess method [15]. Values are given in nmol nitrite per 106 cells ± S.E.M., n = 9. Elicited mouse and rat samples were significantly different (P < 0.01 ∗∗ ). Resident and elicited samples were also significantly different (for mouse P < 0.01 ∗∗ , for rat P < 0.05 ∗ ). Elicited, PDTC-treated samples are significantly different compared to controls: for elicited mouse samples, P < 0.05 ∗ , for elicited rat samples, P < 0.01 ∗∗ . a For actinomycin D (20 M), the average of two measurements are shown (n = 2).
3.3. Transformation of l-arginine into other small molecules The utilization of l-arginine in the synthesis of other amino acids has been measured both in the intracellular fraction and in the extracellular medium because a considerable part of the products may be transported out of the cells during the 60 min incubation. Results indicate that most of the ornithine and citrulline labeling and other amino acids can be found in the extracellular space (>90%). For this reason only the total (extracellular plus intracellular) and intracellular samples are presented separately (Fig. 3). The 14 C-label derived from l-arginine was measured in l-ornithine (produced by the arginase pathway) and l-citrulline (produced by NOS pathway) [14] (Fig. 3). The 14 C-labeling not found in these three amino acids was considered as other small acid-soluble molecules (mainly amino acids) derived from the added l-arginine. In mouse resident macrophages, the arginase pathway represented a minor, but measurable metabolic route; Orn production increased from 319 ± 36 to 677 ± 85 pmol per 106 cells (Fig. 3) in casein-elicited mouse cells. NO synthase pathway, representing a similar proportion in total was enhanced by elicitation from 303 ± 31 to 456 ± 48 pmol per 106 cells. These increases were simultaneous with the decrease of the transformation of Arg into other soluble products and also with the decrease of Arg amount. The absolute amounts of l-ornithine and other amino acids were markedly decreased in the PDTC-treated samples and in the presence of SMT, too.
In rat macrophages the originally negligible Orn content showed a small increase, while Cit production was markedly enhanced followed by a marked decrease of l-arginine and other amino acids. The changes in the intracellular compartment were different: neither Arg nor Orn were found, but a more marked Cit formation was measured together with a net increase in the fraction of other amino acids (Fig. 3B). Finally, in PDTC-treated samples, all amino acids were present at very low intracellular concentrations. SMT produced similar effects (data are shown only for casein-elicited rat cells, where the absolute amounts were evaluable). 3.4. The cellular contents and activities of arginine utilizing macrophage enzymes Both mouse and rat resident and casein-elicited macrophages were studied for their arginase and NO synthase expression by Western blotting. NO synthase expression increased markedly in both species, particularly in casein-elicited rat macrophages (Fig. 4A). This result was supported by the enzyme activities and nitrite productions (Tables 1 and 2). The expression of arginase isoenzymes was also detected in control and elicited cells. A 40 kDa arginase positive band, in which the expression was more marked in mice, remained unchanged in elicited cells. A band with a lower molecular mass (approximately 35 kDa) increased markedly in elicited mouse macrophages and less markedly in elicited rat cells in a good agreement with our previous observations according to which arginase had a low activity in rat macrophages [10].
1086
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
Arginase bands were also weaker in PDTC-treated samples. The activities of arginase and NOS II were also measured. They were in good accordance with the protein amounts (Table 2) except rat arginase, where the activity was much lower than it was expected on the basis of Western blotting. PDTC decreased arginase activity in mice but in smaller extent than its protein amount. Our electrophoretic experiments also suggest the presence of two arginase positive bands in mouse-elicited macrophages (Fig. 5): arginase II has a higher, arginase I a lower mobility [16] while the enzyme activity of arginase I is higher. 4. Discussion
Fig. 3. Distribution of amino acids derived from l-arginine in various peritoneal macrophages. (A) Distribution (extracellular plus intracellular), measured in TLC-separated macrophage supernatants after 60 min incubation with 0.1 mM 14 C-Arg. (B) Intracellular distribution measured in TLC-separated supernatants as mentioned, calculated for 106 cells in both cases. Explanation of symbols is shown over the bars (from left to right: Arg, Orn, Cit, others). MR, murine resident; MC, murine casein; RR, rat resident; RC, rat casein; RCP, rat casein, treated with 50 M PDTC; RCS rat casein, treated with 0.5 mM SMT. Differences are significant evaluated by ANOVA (P < 0.0001), except for Arg in intracellular pools. Significant differences found in corresponding pairs calculated by Bonferroni method. In total samples: for Arg RR–RC, RC–RCP and RC–RCS P < 0.01 ∗∗ , MC–RC P < 0.05 ∗ ; for Orn, MR–MC and MC–RC P < 0.001 ∗∗∗ , MR–RR P < 0.01 ∗∗ ; for Cit, MC–RC, RR–RC, RC–RCP and RC–RCS P < 0.001 ∗∗∗ ; for other compounds, RC–RCP and RC–RCS P < 0.01 ∗∗ . In intracellular samples: for Orn, MR–MC and MC–RC P < 0.001 ∗∗∗ ; for Cit, MC–RC, RR–RC, RC–RCP and RC–RCS P < 0.001 ∗∗∗ , MR–RR P < 0.01 ∗∗ ; for other compounds, MC–RC, RR–RC, RC–RCP and RC–RCS P < 0.001 ∗∗∗ (n = 6).
The uptake of l-arginine into macrophages is mediated by cationic amino acid transporters (CAT) [17,18]. Arginine is utilized via various pathways, and serves partly for the synthesis of other small molecules (mainly amino acids) partly for de novo protein synthesis. The proportions of arginine utilized by these different pathways depends probably on the species and also on the activation state of the cells. The comparison of arginine metabolism in resident and elicited macrophages gives interesting informations about the biochemical mechanism of inflammation because macrophages are involved in it. The higher intracellular labeling of rat cells by l-arginine compared to mouse cells (Fig. 2) can be a consequence of their higher NOS content and activity (Fig. 4A, Table 2) requiring more substrate than is recycled from l-citrulline [19,20]. It has been proved that the rate of arginine transport is driven by the substrate requirement in macrophages activated by LPS or/and cytokines in vitro [21,22]. The mechanism of the stimulating effect of casein-elicitation is not clear yet; probably various cytokines are produced in the cell cultures derived from the peritoneal cavity during adherence and the following 24 h culturing period. These effectors can induce NO synthase: PDTC or actinomycin D added to these cultures abolishes or significantly decreases nitrite production (Table 1) and NOS II expression (Fig. 4), demonstrating the role of de novo NOS II synthesis. The observed metabolic changes are the combined results of the following effects: (1) activation of substrate transport, (2) actual level of substrates and
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
1087
Fig. 4. Western blot studies of enzyme expression in peritoneal macrophages. (A) Expression of NOS II, 5 l SDS-treated sample was tested (corresponding to 105 cells); lane A: mouse resident, lane B: mouse casein-elicited, lane C: rat resident, lane D: rat casein-elicited, lane E: same as D, treated with PDTC. (B) Expression of arginase, 10 l SDS-treated sample was tested (corresponding to 2 × 105 cells for mouse and to 4 × 105 cells for rat); lane A: mouse resident, lane B: mouse casein-elicited, lane C: same as B, treated with PDTC, lane D: rat resident, lane E: rat casein-elicited. Five experiments were performed and the most representative patterns are presented.
cofactors in the cells, (3) effects of the products on enzyme activities, and (4) induction of NOS II or/and arginase. The arginine uptake of elicited rat macrophages during 5 min incubation was 23.8 pmol per 106 cells versus 2.8 pmol per 106 in resident cells (our unpublished observations). Substrates and cofactors are present at the necessary level in cultured macrophages
using culture media and FBS. Finally, the effect of products on enzyme activities have not been tested in these studies, but it is known that high amounts of NO and l-ornithine inhibits NOS II and arginase, respectively, and a cross-inhibition of the products can also be observed [23]. The inhibition of NOS II induction by PDTC and the decrease of its activity by SMT causes the de-
Table 2 NOS II and arginase amounts and activities in peritoneal macrophages Samples
Mouse
Rat
Resident NOS amount Arginase amount NOS activity Arginase activity
178 455 32 227
± ± ± ±
53 94 12 38
Elicited 803 864 234 736
68∗
± ± 136∗ ± 25∗∗∗ ± 86∗∗∗
PDTC N.D. 288 ± 106 40 ± 9 510 ± 91
Resident 212 305 47 10
± ± ± ±
55 100 18 8
Elicited 2477 510 394 66
PDTC 184∗∗∗
± ± 68 ± 16∗∗∗ ± 16∗∗∗
379 ± 84 N.D. 72 ± 18 38 ± 23
NOS and arginase amounts (n = 5) are given in arbitrary units (produced by LKB gel densitometer). For arginase the sum of the two bands is shown, because the instrument could not integrate the close bands separately. NOS and arginase activities (n = 9) are given in pmol Cit or Orn per mg protein per min, respectively. For PDTC treatment, rat NOS II and mouse arginase amounts were tested only, because these activities are higher in the corresponding species. MR, RR, mouse and rat resident; MC and RC, mouse- and rat-elicited samples. ANOVA values were P < 0.0001 ∗∗∗∗ except arginase amounts (P = 0.0054 ∗∗ ). Significant differences of corresponding pairs by Bonferroni corrections were the following: for NOS amounts, MC–RC, RR–RC and RC–RCP P < 0.001 ∗∗∗ , MR–MC P < 0.05 ∗ ; for arginase amounts, MC–MCP P < 0.05 ∗ . Significant differences in NOS activity for corresponding pairs: MR–MC, MC–MCP, MC–RC, RR–RC and RC–RCP P < 0.001 ∗∗∗ , in arginase activity for corresponding pairs: MR–MC, MC–RC, MCP–RCP P < 0.001 ∗∗∗ . N.D. = not determined.
1088
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
Fig. 5. Gel electrophoretic study of arginases in elicited mouse macrophages. Gel slices were measured for arginase as described in methods and radioactivities of 14 C-urea are shown. Total radioactivity is the sum of label found in the gel slices. Liver isoform (I) has a lower mobility compared to the extrahepatic (II) isoform [16]. Three experiments were performed, the most representative one is shown.
crease of citrulline synthesis in each sample (Fig. 3, data are shown only for elicited rat macrophages). However, this does not coincide with an increased ornithine formation (via arginase), disproving a direct competition between arginase and NO synthase pathways. This may be due to the limited arginine concentration (the final concentration of Arg was 0.1 mM, corresponding approximately to the serum level of this amino acid, mimicking physiological conditions) which is not favorable for arginase having a KM of 2–8 mM order of magnitude. In addition, the optimal pH value for arginase is 9.7 different from the physiological value. Therefore, arginase cannot utilize the substrate not used by NOS II and a general marked decrease in amino acid metabolism is found (Fig. 2). Interestingly, the significant difference (Fig. 2, P < 0.001 ∗∗∗ ) between the general intracellular l-arginine utilization of mouse- and rat-elicited cells is not observed when PDTC-treated elicited mouse–rat pairs were compared. These results support that l-arginine utilization is dependent almost exclusively on NO synthesis, at least at physiological arginine concentrations (50–100 M). Results are similar when SMT, a NOS inhibitor is added directly to macrophages during the uptake of l-arginine. The catabolism of arginine may lead to the formation of proline, glutamate, glutamine as the precur-
sor of these amino acids is l-ornithine produced by arginase [24]. In mouse macrophages l-arginine is catabolized into l-ornithine in a higher amount compared to rat cells (Fig. 3), because arginase activity is much higher in mice than in rat macrophages [10,25]. Nevertheless, this finding cannot explain the much greater differences in enzyme activities (Fig. 4B, Table 2). It is possible that the catalytic efficiency of the rat enzyme is also lower or a tightly bound inhibitor is acting. Two mechanisms are known which can be responsible for the inhibition of NO synthesis: PDTC prevents the synthesis of the enzyme inhibiting its induction mediated by transcription factor NF-B, while SMT is a partially specific direct NOS II inhibitor. The incorporation of arginine into proteins is low in each sample and it is not markedly changed in PDTC-treated cells (Fig. 2, P > 0.05) indicating that the expression of proteins mediated by NF-B does not represent a high proportion of the total de novo protein synthesis. SMT does not affect arginine incorporation, either. The pre-treatment with PDTC has the advantage that the synthesis of NOS II is inhibited, however, acting on NF-B it may influence the expression of numerous other proteins. This possibly indirect effect is demonstrated on arginase (Fig. 4B, Table 2). The advantage to use SMT is its specific direct inhibitory effect on NOS II, without affecting arginase activity [26], but it can also decrease l-arginine uptake as it has been found in case of arginine analogues [27]. For this reason, experiments performed with both SMT and PDTC are also demonstrated showing that results are similar (Fig. 3). In macrophages there are two arginase isoenzymes. Both have been detected in mice, while only arginase I has been found in rat cells [28]. Arginase II, which is localized in mitochondria and is inducible by LPS and cytokines [5,29] seems to play an important role in the metabolic processes starting from l-arginine [30]. In addition, different cytokines are involved in the induction of NOS II and arginase suggesting an inverse relationship of the two enzymes leading to a NOS/arginase balance [7]. However, in our studies, the inhibition of NOS II induction by PDTC did not increase the synthesis of ornithine. This is probably the consequence of using different experimental systems: other authors have investigated the role of cytokines on enzyme expression, while our studies were performed to follow
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
the utilization of the common substrate. In our experimental system, the possible metabolic role of arginase has been studied under physiological but not optimal conditions (pH 7.4, 0.1 mM final concentration). In our Western blot samples two arginase bands are observed: a band of 40 kDa is detected in both species, its amount in mice is higher. The other band (approximately 30–35 kDa) is detected in lower amounts and its intensity is higher in elicited macrophages, particularly in mice. Therefore, this isoform seems to be inducible in casein-elicited animals and it is possibly the subunit of the arginase II isoform. The primary antibody used in Western blotting was produced against arginase I, however, based on Swissprot protein data bank studies the antigenic sequences of arginase I and II are identical to 80%. Therefore, a cross-reaction, or less likely, a microheterogeneity of arginase I isoenzyme [31,32] may be responsible for this Western blot pattern (Fig. 4). As a summary, the total arginase activity is almost doubled in elicited cells, this is reflected in the increase of ornithine formation from labeled arginine in mice (Fig. 3). Although rat macrophages also contain detectable arginase, its activity is too low at this physiological substrate concentration and ornithine formation is observed only in elicited rat macrophages (Fig. 3). It is to be noted that PDTC decreased the amount of arginases, too. Since the participation of NF-B is not known in the induction of macrophage arginase, this effect can rather be due to an indirect effect of NF-B on the transcription of arginase. The decreased amount of arginase can also explain that the block of NOS II induction does not result in an increase of Orn formation in the presence of PDTC. The similar effect of SMT may be due to its possible effect on the entry of Arg into macrophages. As a conclusion, our results demonstrate that the in vivo i.p. casein-elicitation (a model of inflammation) leads to a higher l-arginine utilization, serving mainly for NO synthesis and that inflammatory rat macrophages utilize more l-arginine than mouse cells. This must be the consequence of their larger NOS II expression, their arginase content being lower than that of mouse cells, therefore higher amounts of arginase cannot increase the utilization of the substrate so markedly as NOS II. These results can serve as further evidences for the involvement of NOS II in inflammatory reactions and in the etiology of various diseases [4,33–35].
1089
Acknowledgements The authors thank Dr. T. Garzó for the helpful discussion in the preparation of the manuscript, Miss Judit Szabó and Mr. Antal Holly for their skillful technical assistance. This work was supported by the grants of ETT 246/2000 (Hungarian Ministry of Health) and OTKA T 029095 (National Foundation of Scientific Research). References [1] D.S. Bredt, S.H. Snyder, Nitric oxide: a physiologic messenger molecule, Annu. Rev. Biochem. 63 (1994) 175– 195. [2] E. Schneider, M. Dy, The role of arginase in the immune response, Immunol. Today 6 (1985) 136–140. [3] M.A. Marletta, Nitric oxide: biosynthesis and biological significance, Trends Biochem. Sci. 14 (1989) 488–492. [4] J. MacMicking, Q.-W. Xie, C. Nathan, Nitric oxide and macrophage function, Annu. Rev. Immunol. 15 (1997) 323– 350. [5] I.M. Corraliza, G. Soler, K. Eichmann, M. Modolell, Arginase induction by suppressors of nitric oxide synthesis (IL-1, IL-10 and PGE2 ) in murine bone-marrow-derived macrophages, Biochem. Biophys. Res. Commun. 206 (1995) 667–673. [6] C. Hey, J.-L. Boucher, S. Vadon-Le Goff, G. Ketterer, I. Wessler, K. Racké, Inhibition of arginase in rat and rabbit alveolar macrophages by N-hydroxy-d, l-indospicine, effects on l-arginine utilization by nitric oxide synthase, Br. J. Pharmacol. 121 (1997) 395–400. [7] M. Munder, K. Eichmann, M. Modolell, Alternative metabolic states in murine macrophages reflected by nitric oxide synthase/arginase balance: competitive regulation by CD4+ T cells correlates with Th1/Th2 phenotypes, J. Immunol. 160 (1998) 5347–5354. [8] F. Daghigh, J.K. Fukuto, D.E. Ash, Inhibition of rat liver arginase by an intermediate in NO biosynthesis, NG -hydroxy-l-arginine: implications for the regulation of nitric oxide biosynthesis by arginase, Biochem. Biophys. Res. Commun. 202 (1994) 174–180. [9] C. Szabó, G.J. Southan, M. Thiemermann, J.R. Vane, The mechanism of the inhibitory effect of polyamines on the induction of nitric oxide synthase: role of aldehyde metabolites, Br. J. Pharmacol. 113 (1994) 757–766. [10] A. Hrabák, Á. Temesi, I. Csuka, F. Antoni, Inverse relation in the de novo arginase synthesis and nitric oxide production in murine and rat peritoneal macrophages in long-term cultures in vitro, Comp. Biochem. Biophys. 103B (1992) 839–845. [11] A. Hrabák, F. Antoni, I. Csuka, Differences in the arginase activity produced by resident and stimulated murine and rat peritoneal macrophages, Int. J. Biochem. 23 (1991) 997–1003. [12] L. Ornstein, H. Ansley, A. Saunders, Improving manual differential white cell counts with cytochemistry, Blood Cells 2 (1976) 557–568.
1090
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
[13] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding, Anal. Biochem. 72 (1976) 248–254. [14] W.C. Sessa, M. Hecker, J.A. Mitchell, J.R. Vane, The metabolism of l-arginine and its significance for the biosynthesis of endothelium-derived relaxing factor. l-Glutamine inhibits the generation of l-arginine by cultured endothelial cells, Proc. Natl. Acad. Sci. U.S.A. 87 (1990) 8607–8611. [15] L.C. Green, D.A. Wagner, J. Glogowski, P.L. Skipper, J.S. Wishnok, S.R. Tannenbaum, Analysis of nitrate, nitrite and (15 N)nitrate in biological fluids, Anal. Biochem. 126 (1982) 131–138. [16] E.B. Spector, S.C.H. Rice, S. Moedjono, B. Bernard, S.D. Cederbaum, Biochemical properties of arginase in human adult and fetal tissues, Biochem. Med. 28 (1982) 165–175. [17] E.I. Closs, CATs, a family of three distinct mammalian cationic amino acid transporters, Amino Acids 11 (1996) 193–208. [18] D.K. Kakuda, M.J. Sweet, C.L. MacLeod, D.A. Hume, D. Markovich, CAT2-mediated l-arginine transport and nitric oxide production in activated macrophages, Biochem. J. 340 (1999) 549–553. [19] A. Hrabák, M. Idei, Á. Temesi, Arginine supply for nitric oxide synthesis and arginase is mainly exogeneous in elicited murine and rat macrophages, Life Sci. 55 (1994) 797–805. [20] C.A. Schott, G.A. Gray, J.C. Stoclet, Dependence of endotoxin-induced vascular hyporeactivity on extracellular l-arginine, Br. J. Pharmacol. 108 (1993) 38–43. [21] R.G. Bogle, A.R. Baydoun, J.D. Pearson, S. Moncada, G.E. Mann, l-arginine transport is increased in macrophages generating nitric oxide, Biochem. J. 284 (1992) 15–18. [22] W.W. Simmons, E.I. Closs, J.M. Cunningham, T.W. Smith, R.A. Kelly, Cytokines and insulin induce cationic amino acid transporter (CAT) expression in cardiac myocytes. Regulation of l-arginine transport and NO production by CAT-1, CAT-2A and CAT-2B, J. Biol. Chem. 271 (1996) 11694–11702. [23] A. Hrabák, T. Bajor, Á. Temesi, G. Mészáros, The inhibitory effect of nitrite, a stable product of nitric oxide (NO) formation, on arginase, FEBS Lett. 390 (1996) 203–206. [24] G. Wu, S.M. Morris, Arginine metabolism: nitric oxide and beyond, Biochem. J. 336 (1998) 1–17.
[25] M. Fishman, Functional heterogeneity among peritoneal macrophages. III. No evidence for the role of arginase in the inhibition of tumor cell growth by supernatants from macrophages or macrophage subpopulation cultures, Cell. Immunol. 55 (1980) 174–184. [26] A. Hrabák, T. Bajor, G.J. Southan, A.L. Salzman, C. Szabó, Comparison of the inhibitory effect of isothiourea and mercaptoalkylguanidine derivatives on the alternative pathways of arginine metabolism in macrophages, Life Sci. 60 (1997) PL395–PL401. [27] E.I. Closs, F.Z. Basha, A. Habermeier, U. Förstermann, Interference of l-arginine analogues with l-arginine transport mediated by the y+ carrier hCAT-2B, Nitric Oxide 1 (1997) 65–73. [28] C.A. Louis, J.S. Reichner, W.L. Henry Jr., B. Mastrofrancesco, T. Gotoh, M. Mori, J.E. Albina, Distinct arginase isoforms expressed in primary and transformed macrophages: regulation by oxygen tension, Am. J. Physiol. 274 (1998) R775–R782. [29] W.W. Wang, C.P. Jenkinson, J.M. Griscavage, R.M. Kern, N.S. Arabolos, R.E. Byrns, S.D. Cederbaum, Co-induction of arginase and nitric oxide synthase in murine macrophages activated by lipopolysaccharide, Biochem. Biophys. Res. Commun. 210 (1995) 1009–1016. [30] C.P. Jenkinson, W.W. Grody, S.D. Cederbaum, Comparative properties of arginases, Comp. Biochem. Physiol. 114B (1996) 107–132. [31] A. Diez, J.M. Fuentes, F. Prada, M.L. Campo, G. Soler, Immunological identity of the two different molecular mass constitutive subunits of liver arginase, Biol. Chem. Hoppe-Seyler 375 (1994) 537–541. [32] Z. Spolarics, J.S. Bond, Multiple molecular forms of mouse liver arginase, Arch. Biochem. Biophys. 260 (1988) 469– 479. [33] E. Moilanen, H. Vaapatalo, Nitric oxide in inflammation and immune response, Ann. Med. 27 (1995) 359–367. [34] G. Cirino, Multiple controls in inflammation. Extracellular and intracellular phospholipase A2, inducible and constitutive cyclooxygenase, and inducible nitric oxide synthase, Biochem. Pharmacol. 55 (1998) 105–111. [35] K.-D. Kröncke, K. Fehsel, V. Kolb-Bachofen, Inducible nitric oxide synthase in human diseases, Clin. Exp. Immunol. 113 (1998) 147–156.
Enhanced utilization and altered metabolism of arginine in inflammatory macrophages caused by raised nitric oxide synthesis Nándor Müllner, Ágnes Lázár, András Hrabák∗ Department of Medical Chemistry, Molecular Biology and Pathobiochemistry, Semmelweis University, VIII. Puskin u. 9., P.O. Box 260, Budapest H-1444, Hungary Received 3 July 2001; received in revised form 15 February 2002; accepted 20 February 2002
Abstract Nitric oxide (NO) production was increased in macrophages during inflammation. Casein-elicitation of rodents causing a peritoneal inflammation offered a good model to study alterations in the metabolism of l-arginine, the precursor of NO synthesis. The utilization of l-arginine for NO production, arginase pathway and protein synthesis were studied by radioactive labeling and chromatographic separation. The expression of NO synthase and arginase was studied by Western blotting. Rat macrophages utilized more arginine than mouse macrophages (228 ± 27 versus 71 ± 12.8 pmol per 106 macrophages). Arginine incorporation into proteins was low in both species (<15% of labeling). When NO synthesis was blocked, arginine was utilized at a lower general rate, but l-ornithine formation did not increase. The expression of enzymes utilizing arginine increased. NO production was raised mainly in rats (1162 ± 84 pmol citrulline per 106 cells) while in mice both arginase and NO synthase were active in elicited macrophages (677 ± 85 pmol ornithine and 456 ± 48 pmol citrulline per 106 cells). We concluded, that inflammation induced enhanced l-arginine utilization in rodent macrophages. The expressions and the activities of arginase and NO synthase as well as NO formation were increased in elicited macrophages. Specific blocking of NO synthesis did not result in the enhanced effectivity of the arginase pathway, rather was manifested in a general lower rate of arginine utilization. Different rodent species reacted differently to inflammation: in rats, high NO increase was found exclusively, while in mice the activation of the arginase pathway was also important. © 2002 Elsevier Science Ltd. All rights reserved. Keywords: Resident/inflammatory peritoneal macrophages; Arginine metabolism; Arginine uptake/incorporation; Arginase; Nitric oxide synthase
1. Introduction
Abbreviations: FBS, foetal bovine serum; HBSS, Hank’s balanced salt solution; LPS, lipopolysaccharide; NOS II, inducible nitric oxide synthase; PAGE, polyacrylamide gel electrophoresis; PBS, phosphate buffered saline; PDTC, pyrrolidine dithiocarbamate; SDS, sodium dodecyl sulfate; SMT, S-methyl-isothiourea; TLC, thin layer chromatography ∗ Corresponding author. Tel.: +36-1-266-2755/4089; fax: +36-1-266-2615. E-mail address: [email protected] (A. Hrab´ak).
The metabolism of l-arginine is very important in cells producing nitric oxide (NO). Macrophages represent one of these cell types. They metabolize l-arginine via multiple pathways: inducible nitric oxide synthase (NOS II) produces NO and l-citrulline (Cit) [1], arginase hydrolyses l-arginine to urea and l-ornithine (Orn) [2] and l-arginine is also incorporated into cellular proteins. Orn and Cit are precursors of other amino acids and their derivatives. The activities of these pathways in macrophages depend on
1357-2725/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S 1 3 5 7 - 2 7 2 5 ( 0 2 ) 0 0 0 2 8 - 6
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
the activation state of the cell influenced by various cytokines and bacterial endotoxin (lipopolysaccharide, LPS) in vitro [3]. Arginine metabolism shows significant changes during inflammation, influencing NO production in vivo [4]. The reciprocal regulation of NOS II and arginase has been demonstrated both in the induction processes [5] and also in substrate utilization [6]. Previous studies have demonstrated the changes in NOS II and arginase activities after various cytokine treatments in vitro [7] and the mutual inhibition of NOS II and arginase by the products of their catalyzed reactions [8,9]. However, after in vivo stimulations, the distribution of products derived from l-arginine in macrophages of different species during inflammatory reactions has not been studied directly. Since macrophages are involved in inflammatory reactions, the comparison of the arginine utilization in resident and elicited macrophages may give interesting informations about the biochemical mechanism of inflammation. Here we report a comparative study concerning the distribution of the products derived from l-arginine under physiological conditions (pH 7.4, 0.1 mM Arg concentration), in macrophages isolated from control (resident) and casein-elicited (inflammatory) mice and rats, in which different activities of arginase and NOS have been observed previously [10]. We have chosen in vivo i.p. casein-elicitation as a model because it provoked inflammation and informations about the altered activities of arginine utilizing enzymes have been available [10,11]. The changes in the expression of arginase and NOS, as the background of these metabolic alterations were also studied. Experiments to decrease NO production in two different ways (inhibition of NOS induction by pyrrolidine dithiocarbamate and direct inhibition of its activity by S-methyl-isothiourea) were performed to obtain informations showing whether the block of a certain (NOS) pathway of l-arginine metabolism favored alternative metabolic routes or not.
1081
mice and Wistar male rats (Charles River, Hungary) as described earlier [10]. Casein-injected animals were sacrificed on the 4th day after treatment. Isolated peritoneal cells (3×106 per well) were adhered to 24-well Corning plastic dishes for 2 h and then cultured for 24 h at 37 ◦ C in a 5% CO2 atmosphere in 0.5 ml RPMI 1640 medium containing 5% foetal bovine serum (FBS), penicillin (100 IU/ml) and streptomycin (100 g/ml). This period of culturing is sufficient to induce NO synthesis and to elevate arginase activity in elicited cells. After culturing media were removed, cells were washed three times with phosphate buffered saline (PBS, pH 7.4), and kept in 100 l Hank’s balanced salt solution (HBSS) for 2 h to deplete the cells of l-arginine derived from the culture medium. HBSS was renewed before further experiments. The composition of the adherent cell population was tested by light microscopic detection of a NaF-sensitive esterase found only in macrophages. The method of Ornstein et al. [12] was used. Briefly, cells were adhered in 24-well Nunclon plates to Thermanox cover slips (Nunc Inc.) and stained with a buffered solution of naphthol-DS-acetate and Fast Blue, finally studied under a light microscope. In the absence of NaF all cells are positive (blue) while in the presence of NaF only cells differing from macrophages (i.e. the neutrophils) are positive. Two-hundred cells were counted for esterase activity in both cases. 2.2. Measurement of the total protein content in adhered cells After adherence non-adherent cells were removed by multiple washing (PBS) and adherent cells were solubilized in 200 l 2N KOH. Protein was precipitated by excess HClO4 and dissolved in 0.5N NaOH and measured with coomassie blue by the Bradford method [13] using bovine serum albumin as standard. 2.3. Labeling and separation of labeled fractions from macrophages
2. Materials and methods 2.1. Cells, microscopy Macrophages were removed from the peritoneal cavity of untreated and casein-elicited CD-CR-1 male
The 14 C-labeled l-arginine (final concentration 0.1 mM; Amersham, specific activity 100 MBq/mmol) was added to the cells for 60 min at 37 ◦ C. The uptake of labeled arginine was stopped with 20 l 1 M unlabeled l-arginine (in HBSS, pH 7.4, final concen-
1082
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
tration 0.1 M) and cells were washed four times with PBS to remove excess labeling. Ten microliters of the removed first supernatant was separated by thin layer chromatography (TLC) and measured for Arg, Orn and Cit content (extracellular samples, see the following paragraphs). Cells were solubilized in 200 l cold 2 M KOH, 10 l samples were removed to measure the entered radioactivity and the macromolecular fraction was precipitated by adding 0.5 ml 1 M HClO4 . After centrifugation, an aliquot of the supernatant was measured for radioactivity to calculate the amount of labeled arginine taken up by the cells without being incorporated into proteins. In order to characterize the latter, the precipitate was washed once with cold 0.5 M HClO4 , then dissolved in 300 l 0.5 M NaOH and an aliquot was measured for radioactivity. Another sample series was dissolved in 100 l 1N KOH,
precipitated with 100 l 1 M HClO4 and 10 l mixture of unlabeled l-Arg, l-Orn and l-Cit (5–5 mM) was added. After centrifugation 100 l supernatant was tested for radioactivity and 30 l aliquots were submitted to thin layer chromatography to separate Arg, Orn, Cit [14]. Spots were visualized with 1% ninhydrine solution and then cut out. We placed them into 3 ml mixtures of a toluene-based scintillation cocktail and Triton-X-100 (2:1) and then measured for radioactivity by a Packard Trikarb LH 2100 liquid scintillation spectrometer. For each sample four tests were performed: (1) a background sample (“0 min”, i.e. labeled arginine was added after the addition of 1 M Arg), to subtract from the counts measured in other samples; (2) 60 min sample to test the distribution of the labeling in various subfractions; (3) the same as above but derived from macrophages
Fig. 1. Scheme of the separation of various sample fractions for the study of the distribution of macrophages.
14 C-l-arginine
label in murine peritoneal
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
cultured for 24 h in the presence of 50 M pyrrolidine dithiocarbamate (PDTC), an inhibitor of NOS II transcription via NF-B transcription factor; (4) the same as above but in the presence of 0.5 mM S-methyl-isothiourea (SMT), an inhibitor of NOS II activity, during the labeled arginine load without previous PDTC treatment. The scheme of the whole experiment is shown in Fig. 1. 2.4. Measurement of NO synthase and arginase activities in macrophages Macrophages were cultured for 24 h in RPMI–5% FBS and then cells were washed twice with PBS. Cells were lysed by 5 mM HEPES (pH 7.4)–1 mM NADPH–10 M tetrahydrobiopterin and 20 M 14 C-l-arginine (final concentration) was added for 60 min. Reaction was stopped with 5–5 mM Arg–Orn– Cit, an aliquot was chromatographed on TLC plates as described. Another aliquot was used to measure protein by Bradford’s method [13]. Arginase and NOS specific activities were expressed as pmol Orn and Cit per min per mg protein, respectively. 2.5. NO synthase and arginase content in macrophages In order to check the presence and functional state of NOS II, nitrite production was also tested in the supernatants of cultured macrophages by the Griess reaction [15]. Expression of arginine utilizing enzymes was tested by Western blotting. After 24 h culturing in RPMI–5% FBS, 3 × 106 cells were solubilized in a sample buffer containing 3% sodium dodecyl sulfate (SDS) and after heating samples were separated by SDS-polyacrylamide gel electrophoresis (PAGE). Separated samples were transferred to nitrocellulose membrane and then treated with a primary mouse anti-iNOS and anti-arginase I antibody (Transduction Laboratories, England). Secondary antibody was an anti-mouse antibody conjugated with peroxidase (Sigma). Finally samples were developed with ECL solution (Amersham) and fluorescence was detected on a Cronex-6 X-ray film (Agfa/Forte, Hungary). Bands were quantitatively evaluated by an Ultroscan LKB gel densitometer and bands were scanned by Scan Maker II scanner to form computer files for photography.
1083
The presence of arginase isoenzymes was also studied by gel electrophoresis according to Spector et al. [16]. Briefly, 7.5% SDS-free polyacrylamide gels were used and after electrophoresis gels were cut into 5 mM slices, extracted by a Tris–Glycine buffer and 180 l extracts were mixed with 20 l 2 mM [14 C]-l-arginine for 60 min at 37 ◦ C at pH 8.3 to measure arginase activities. Reaction was stopped by adding 200 l 1N HCl and released urea was separated on small Dowex 50 × 8 resin columns. The non-bound [14 C]-urea was measured in a liquid scintillation spectrometer. 2.6. Statistical evaluation of the results Results are presented as mean plus S.E.M. values. Significances were calculated by ANOVA test for multiple comparisons. Since conclusions can be based mainly on the comparison of corresponding sample pairs, the significance of the differences in corresponding pairs was calculated by adjusted t-tests with P-values corrected by the Bonferroni method and by the Mann–Whitney unpaired two-tailed t-test. Differences were considered as significant when P < 0.05. Corresponding pairs for comparison are: mouse and rat samples both from resident and elicited animals; resident and elicited samples both from mouse and rats; mouse- and rat-elicited samples without or with PDTC and SMT, respectively. The number of experiments are reported in figure and table legends.
3. Results 3.1. Identification of adhered cells and measurement of their protein content The cell composition of the peritoneal exudate depends on the time elapsed after the induction of inflammation. After 3–4 days, macrophages are preponderant. This was checked by light microscopy using a non-specific esterase staining (see Section 2). Both in mouse and rat adherent cell populations were almost exclusively macrophages (less than 5% was stained in the presence of NaF while 85–90% of the total population was positive in the absence of NaF). Total protein content of adhered cells was also measured. Macrophages prepared from injected
1084
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
Fig. 2. Distribution of l-arginine derived labeling in macrophages. Abbreviations: MR, mouse resident; MC, mouse casein-elicited; RR, rat resident; RC, rat casein-elicited; P indicates PDTC-treated samples; S indicates samples measured in the presence of 0.5 mM SMT; 3 × 106 macrophages were adhered and cultured and then studied for 14 C-labeling distribution. ANOVA test gave P < 0.0001 for each comparison except for acid-insoluble, precipitated fraction (P > 0.05). Significant differences of corresponding pairs calculated by Bonferroni method: P < 0.001 ∗∗∗ for MC–RC and RC–RCP, P < 0.01 ∗∗ for RR–RC, P < 0.05 ∗ for RCP–RCS both in total and acid-soluble samples (n = 9).
animals contained more protein than resident cells, however, the increase was not significant. The increase was similar in mouse and rat cells. Resident mouse and rat macrophages contained 2.94 ± 0.71 and 3.75 ± 0.65 g protein per 106 cells, respectively, while the protein contents of elicited mouse and rat cells were 3.60 ± 0.38 and 4.08 ± 0.88 g per 106 macrophages. 3.2. Intracellular labeling of the subfractions The amount of intracellular 14 C-arginine labeling in mouse resident peritoneal macrophages was significantly lower than in rat cells (Fig. 2). Labeling was markedly decreased in each PDTC-treated sample and less markedly in the presence of SMT. Casein-elicitation augmented the amount of intracel-
lular labeling both in mouse and rat cells, a strong significance was found. The bulk of entered amino acid was found in both species in the acid-soluble fraction; incorporation of l-arginine into proteins did not exceed 15% and in elicited cells this proportion was even lower. These results show that in macrophages arginine was rather transformed to other amino acids and other small molecules instead of having been incorporated into proteins, although de novo protein synthesis was also detected: PDTC and actinomycin D treatment significantly blocked NOS II transcription indicated by the marked decrease of nitrite production (Table 1) and PDTC decreased the amount of both NOS II and arginase (Fig. 4). Nevertheless, neither PDTC nor SMT treatment decreased further the low incorporation of l-arginine significantly.
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
1085
Table 1 Nitrite production in the supernatants of macrophage cultures Samples
Resident Mouse
Control PDTC Actinomycin Da
4.2 ± 1.9 2.1 ± 0.6 N.D.
Elicited Rat
Mouse 6.7∗
11.9 ± 3.3 ± 1.1 N.D.
Rat 2.0∗∗
13.4 ± 3.4 ± 3.3 0.6
38.0 ± 7.0∗∗ 8.3 ± 2.7 4.7
Cells were cultured for 24 h in RPMI 1640 medium containing 5% foetal bovine serum, then an aliquot of the supernatants was determined directly using the Griess method [15]. Values are given in nmol nitrite per 106 cells ± S.E.M., n = 9. Elicited mouse and rat samples were significantly different (P < 0.01 ∗∗ ). Resident and elicited samples were also significantly different (for mouse P < 0.01 ∗∗ , for rat P < 0.05 ∗ ). Elicited, PDTC-treated samples are significantly different compared to controls: for elicited mouse samples, P < 0.05 ∗ , for elicited rat samples, P < 0.01 ∗∗ . a For actinomycin D (20 M), the average of two measurements are shown (n = 2).
3.3. Transformation of l-arginine into other small molecules The utilization of l-arginine in the synthesis of other amino acids has been measured both in the intracellular fraction and in the extracellular medium because a considerable part of the products may be transported out of the cells during the 60 min incubation. Results indicate that most of the ornithine and citrulline labeling and other amino acids can be found in the extracellular space (>90%). For this reason only the total (extracellular plus intracellular) and intracellular samples are presented separately (Fig. 3). The 14 C-label derived from l-arginine was measured in l-ornithine (produced by the arginase pathway) and l-citrulline (produced by NOS pathway) [14] (Fig. 3). The 14 C-labeling not found in these three amino acids was considered as other small acid-soluble molecules (mainly amino acids) derived from the added l-arginine. In mouse resident macrophages, the arginase pathway represented a minor, but measurable metabolic route; Orn production increased from 319 ± 36 to 677 ± 85 pmol per 106 cells (Fig. 3) in casein-elicited mouse cells. NO synthase pathway, representing a similar proportion in total was enhanced by elicitation from 303 ± 31 to 456 ± 48 pmol per 106 cells. These increases were simultaneous with the decrease of the transformation of Arg into other soluble products and also with the decrease of Arg amount. The absolute amounts of l-ornithine and other amino acids were markedly decreased in the PDTC-treated samples and in the presence of SMT, too.
In rat macrophages the originally negligible Orn content showed a small increase, while Cit production was markedly enhanced followed by a marked decrease of l-arginine and other amino acids. The changes in the intracellular compartment were different: neither Arg nor Orn were found, but a more marked Cit formation was measured together with a net increase in the fraction of other amino acids (Fig. 3B). Finally, in PDTC-treated samples, all amino acids were present at very low intracellular concentrations. SMT produced similar effects (data are shown only for casein-elicited rat cells, where the absolute amounts were evaluable). 3.4. The cellular contents and activities of arginine utilizing macrophage enzymes Both mouse and rat resident and casein-elicited macrophages were studied for their arginase and NO synthase expression by Western blotting. NO synthase expression increased markedly in both species, particularly in casein-elicited rat macrophages (Fig. 4A). This result was supported by the enzyme activities and nitrite productions (Tables 1 and 2). The expression of arginase isoenzymes was also detected in control and elicited cells. A 40 kDa arginase positive band, in which the expression was more marked in mice, remained unchanged in elicited cells. A band with a lower molecular mass (approximately 35 kDa) increased markedly in elicited mouse macrophages and less markedly in elicited rat cells in a good agreement with our previous observations according to which arginase had a low activity in rat macrophages [10].
1086
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
Arginase bands were also weaker in PDTC-treated samples. The activities of arginase and NOS II were also measured. They were in good accordance with the protein amounts (Table 2) except rat arginase, where the activity was much lower than it was expected on the basis of Western blotting. PDTC decreased arginase activity in mice but in smaller extent than its protein amount. Our electrophoretic experiments also suggest the presence of two arginase positive bands in mouse-elicited macrophages (Fig. 5): arginase II has a higher, arginase I a lower mobility [16] while the enzyme activity of arginase I is higher. 4. Discussion
Fig. 3. Distribution of amino acids derived from l-arginine in various peritoneal macrophages. (A) Distribution (extracellular plus intracellular), measured in TLC-separated macrophage supernatants after 60 min incubation with 0.1 mM 14 C-Arg. (B) Intracellular distribution measured in TLC-separated supernatants as mentioned, calculated for 106 cells in both cases. Explanation of symbols is shown over the bars (from left to right: Arg, Orn, Cit, others). MR, murine resident; MC, murine casein; RR, rat resident; RC, rat casein; RCP, rat casein, treated with 50 M PDTC; RCS rat casein, treated with 0.5 mM SMT. Differences are significant evaluated by ANOVA (P < 0.0001), except for Arg in intracellular pools. Significant differences found in corresponding pairs calculated by Bonferroni method. In total samples: for Arg RR–RC, RC–RCP and RC–RCS P < 0.01 ∗∗ , MC–RC P < 0.05 ∗ ; for Orn, MR–MC and MC–RC P < 0.001 ∗∗∗ , MR–RR P < 0.01 ∗∗ ; for Cit, MC–RC, RR–RC, RC–RCP and RC–RCS P < 0.001 ∗∗∗ ; for other compounds, RC–RCP and RC–RCS P < 0.01 ∗∗ . In intracellular samples: for Orn, MR–MC and MC–RC P < 0.001 ∗∗∗ ; for Cit, MC–RC, RR–RC, RC–RCP and RC–RCS P < 0.001 ∗∗∗ , MR–RR P < 0.01 ∗∗ ; for other compounds, MC–RC, RR–RC, RC–RCP and RC–RCS P < 0.001 ∗∗∗ (n = 6).
The uptake of l-arginine into macrophages is mediated by cationic amino acid transporters (CAT) [17,18]. Arginine is utilized via various pathways, and serves partly for the synthesis of other small molecules (mainly amino acids) partly for de novo protein synthesis. The proportions of arginine utilized by these different pathways depends probably on the species and also on the activation state of the cells. The comparison of arginine metabolism in resident and elicited macrophages gives interesting informations about the biochemical mechanism of inflammation because macrophages are involved in it. The higher intracellular labeling of rat cells by l-arginine compared to mouse cells (Fig. 2) can be a consequence of their higher NOS content and activity (Fig. 4A, Table 2) requiring more substrate than is recycled from l-citrulline [19,20]. It has been proved that the rate of arginine transport is driven by the substrate requirement in macrophages activated by LPS or/and cytokines in vitro [21,22]. The mechanism of the stimulating effect of casein-elicitation is not clear yet; probably various cytokines are produced in the cell cultures derived from the peritoneal cavity during adherence and the following 24 h culturing period. These effectors can induce NO synthase: PDTC or actinomycin D added to these cultures abolishes or significantly decreases nitrite production (Table 1) and NOS II expression (Fig. 4), demonstrating the role of de novo NOS II synthesis. The observed metabolic changes are the combined results of the following effects: (1) activation of substrate transport, (2) actual level of substrates and
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
1087
Fig. 4. Western blot studies of enzyme expression in peritoneal macrophages. (A) Expression of NOS II, 5 l SDS-treated sample was tested (corresponding to 105 cells); lane A: mouse resident, lane B: mouse casein-elicited, lane C: rat resident, lane D: rat casein-elicited, lane E: same as D, treated with PDTC. (B) Expression of arginase, 10 l SDS-treated sample was tested (corresponding to 2 × 105 cells for mouse and to 4 × 105 cells for rat); lane A: mouse resident, lane B: mouse casein-elicited, lane C: same as B, treated with PDTC, lane D: rat resident, lane E: rat casein-elicited. Five experiments were performed and the most representative patterns are presented.
cofactors in the cells, (3) effects of the products on enzyme activities, and (4) induction of NOS II or/and arginase. The arginine uptake of elicited rat macrophages during 5 min incubation was 23.8 pmol per 106 cells versus 2.8 pmol per 106 in resident cells (our unpublished observations). Substrates and cofactors are present at the necessary level in cultured macrophages
using culture media and FBS. Finally, the effect of products on enzyme activities have not been tested in these studies, but it is known that high amounts of NO and l-ornithine inhibits NOS II and arginase, respectively, and a cross-inhibition of the products can also be observed [23]. The inhibition of NOS II induction by PDTC and the decrease of its activity by SMT causes the de-
Table 2 NOS II and arginase amounts and activities in peritoneal macrophages Samples
Mouse
Rat
Resident NOS amount Arginase amount NOS activity Arginase activity
178 455 32 227
± ± ± ±
53 94 12 38
Elicited 803 864 234 736
68∗
± ± 136∗ ± 25∗∗∗ ± 86∗∗∗
PDTC N.D. 288 ± 106 40 ± 9 510 ± 91
Resident 212 305 47 10
± ± ± ±
55 100 18 8
Elicited 2477 510 394 66
PDTC 184∗∗∗
± ± 68 ± 16∗∗∗ ± 16∗∗∗
379 ± 84 N.D. 72 ± 18 38 ± 23
NOS and arginase amounts (n = 5) are given in arbitrary units (produced by LKB gel densitometer). For arginase the sum of the two bands is shown, because the instrument could not integrate the close bands separately. NOS and arginase activities (n = 9) are given in pmol Cit or Orn per mg protein per min, respectively. For PDTC treatment, rat NOS II and mouse arginase amounts were tested only, because these activities are higher in the corresponding species. MR, RR, mouse and rat resident; MC and RC, mouse- and rat-elicited samples. ANOVA values were P < 0.0001 ∗∗∗∗ except arginase amounts (P = 0.0054 ∗∗ ). Significant differences of corresponding pairs by Bonferroni corrections were the following: for NOS amounts, MC–RC, RR–RC and RC–RCP P < 0.001 ∗∗∗ , MR–MC P < 0.05 ∗ ; for arginase amounts, MC–MCP P < 0.05 ∗ . Significant differences in NOS activity for corresponding pairs: MR–MC, MC–MCP, MC–RC, RR–RC and RC–RCP P < 0.001 ∗∗∗ , in arginase activity for corresponding pairs: MR–MC, MC–RC, MCP–RCP P < 0.001 ∗∗∗ . N.D. = not determined.
1088
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
Fig. 5. Gel electrophoretic study of arginases in elicited mouse macrophages. Gel slices were measured for arginase as described in methods and radioactivities of 14 C-urea are shown. Total radioactivity is the sum of label found in the gel slices. Liver isoform (I) has a lower mobility compared to the extrahepatic (II) isoform [16]. Three experiments were performed, the most representative one is shown.
crease of citrulline synthesis in each sample (Fig. 3, data are shown only for elicited rat macrophages). However, this does not coincide with an increased ornithine formation (via arginase), disproving a direct competition between arginase and NO synthase pathways. This may be due to the limited arginine concentration (the final concentration of Arg was 0.1 mM, corresponding approximately to the serum level of this amino acid, mimicking physiological conditions) which is not favorable for arginase having a KM of 2–8 mM order of magnitude. In addition, the optimal pH value for arginase is 9.7 different from the physiological value. Therefore, arginase cannot utilize the substrate not used by NOS II and a general marked decrease in amino acid metabolism is found (Fig. 2). Interestingly, the significant difference (Fig. 2, P < 0.001 ∗∗∗ ) between the general intracellular l-arginine utilization of mouse- and rat-elicited cells is not observed when PDTC-treated elicited mouse–rat pairs were compared. These results support that l-arginine utilization is dependent almost exclusively on NO synthesis, at least at physiological arginine concentrations (50–100 M). Results are similar when SMT, a NOS inhibitor is added directly to macrophages during the uptake of l-arginine. The catabolism of arginine may lead to the formation of proline, glutamate, glutamine as the precur-
sor of these amino acids is l-ornithine produced by arginase [24]. In mouse macrophages l-arginine is catabolized into l-ornithine in a higher amount compared to rat cells (Fig. 3), because arginase activity is much higher in mice than in rat macrophages [10,25]. Nevertheless, this finding cannot explain the much greater differences in enzyme activities (Fig. 4B, Table 2). It is possible that the catalytic efficiency of the rat enzyme is also lower or a tightly bound inhibitor is acting. Two mechanisms are known which can be responsible for the inhibition of NO synthesis: PDTC prevents the synthesis of the enzyme inhibiting its induction mediated by transcription factor NF-B, while SMT is a partially specific direct NOS II inhibitor. The incorporation of arginine into proteins is low in each sample and it is not markedly changed in PDTC-treated cells (Fig. 2, P > 0.05) indicating that the expression of proteins mediated by NF-B does not represent a high proportion of the total de novo protein synthesis. SMT does not affect arginine incorporation, either. The pre-treatment with PDTC has the advantage that the synthesis of NOS II is inhibited, however, acting on NF-B it may influence the expression of numerous other proteins. This possibly indirect effect is demonstrated on arginase (Fig. 4B, Table 2). The advantage to use SMT is its specific direct inhibitory effect on NOS II, without affecting arginase activity [26], but it can also decrease l-arginine uptake as it has been found in case of arginine analogues [27]. For this reason, experiments performed with both SMT and PDTC are also demonstrated showing that results are similar (Fig. 3). In macrophages there are two arginase isoenzymes. Both have been detected in mice, while only arginase I has been found in rat cells [28]. Arginase II, which is localized in mitochondria and is inducible by LPS and cytokines [5,29] seems to play an important role in the metabolic processes starting from l-arginine [30]. In addition, different cytokines are involved in the induction of NOS II and arginase suggesting an inverse relationship of the two enzymes leading to a NOS/arginase balance [7]. However, in our studies, the inhibition of NOS II induction by PDTC did not increase the synthesis of ornithine. This is probably the consequence of using different experimental systems: other authors have investigated the role of cytokines on enzyme expression, while our studies were performed to follow
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
the utilization of the common substrate. In our experimental system, the possible metabolic role of arginase has been studied under physiological but not optimal conditions (pH 7.4, 0.1 mM final concentration). In our Western blot samples two arginase bands are observed: a band of 40 kDa is detected in both species, its amount in mice is higher. The other band (approximately 30–35 kDa) is detected in lower amounts and its intensity is higher in elicited macrophages, particularly in mice. Therefore, this isoform seems to be inducible in casein-elicited animals and it is possibly the subunit of the arginase II isoform. The primary antibody used in Western blotting was produced against arginase I, however, based on Swissprot protein data bank studies the antigenic sequences of arginase I and II are identical to 80%. Therefore, a cross-reaction, or less likely, a microheterogeneity of arginase I isoenzyme [31,32] may be responsible for this Western blot pattern (Fig. 4). As a summary, the total arginase activity is almost doubled in elicited cells, this is reflected in the increase of ornithine formation from labeled arginine in mice (Fig. 3). Although rat macrophages also contain detectable arginase, its activity is too low at this physiological substrate concentration and ornithine formation is observed only in elicited rat macrophages (Fig. 3). It is to be noted that PDTC decreased the amount of arginases, too. Since the participation of NF-B is not known in the induction of macrophage arginase, this effect can rather be due to an indirect effect of NF-B on the transcription of arginase. The decreased amount of arginase can also explain that the block of NOS II induction does not result in an increase of Orn formation in the presence of PDTC. The similar effect of SMT may be due to its possible effect on the entry of Arg into macrophages. As a conclusion, our results demonstrate that the in vivo i.p. casein-elicitation (a model of inflammation) leads to a higher l-arginine utilization, serving mainly for NO synthesis and that inflammatory rat macrophages utilize more l-arginine than mouse cells. This must be the consequence of their larger NOS II expression, their arginase content being lower than that of mouse cells, therefore higher amounts of arginase cannot increase the utilization of the substrate so markedly as NOS II. These results can serve as further evidences for the involvement of NOS II in inflammatory reactions and in the etiology of various diseases [4,33–35].
1089
Acknowledgements The authors thank Dr. T. Garzó for the helpful discussion in the preparation of the manuscript, Miss Judit Szabó and Mr. Antal Holly for their skillful technical assistance. This work was supported by the grants of ETT 246/2000 (Hungarian Ministry of Health) and OTKA T 029095 (National Foundation of Scientific Research). References [1] D.S. Bredt, S.H. Snyder, Nitric oxide: a physiologic messenger molecule, Annu. Rev. Biochem. 63 (1994) 175– 195. [2] E. Schneider, M. Dy, The role of arginase in the immune response, Immunol. Today 6 (1985) 136–140. [3] M.A. Marletta, Nitric oxide: biosynthesis and biological significance, Trends Biochem. Sci. 14 (1989) 488–492. [4] J. MacMicking, Q.-W. Xie, C. Nathan, Nitric oxide and macrophage function, Annu. Rev. Immunol. 15 (1997) 323– 350. [5] I.M. Corraliza, G. Soler, K. Eichmann, M. Modolell, Arginase induction by suppressors of nitric oxide synthesis (IL-1, IL-10 and PGE2 ) in murine bone-marrow-derived macrophages, Biochem. Biophys. Res. Commun. 206 (1995) 667–673. [6] C. Hey, J.-L. Boucher, S. Vadon-Le Goff, G. Ketterer, I. Wessler, K. Racké, Inhibition of arginase in rat and rabbit alveolar macrophages by N-hydroxy-d, l-indospicine, effects on l-arginine utilization by nitric oxide synthase, Br. J. Pharmacol. 121 (1997) 395–400. [7] M. Munder, K. Eichmann, M. Modolell, Alternative metabolic states in murine macrophages reflected by nitric oxide synthase/arginase balance: competitive regulation by CD4+ T cells correlates with Th1/Th2 phenotypes, J. Immunol. 160 (1998) 5347–5354. [8] F. Daghigh, J.K. Fukuto, D.E. Ash, Inhibition of rat liver arginase by an intermediate in NO biosynthesis, NG -hydroxy-l-arginine: implications for the regulation of nitric oxide biosynthesis by arginase, Biochem. Biophys. Res. Commun. 202 (1994) 174–180. [9] C. Szabó, G.J. Southan, M. Thiemermann, J.R. Vane, The mechanism of the inhibitory effect of polyamines on the induction of nitric oxide synthase: role of aldehyde metabolites, Br. J. Pharmacol. 113 (1994) 757–766. [10] A. Hrabák, Á. Temesi, I. Csuka, F. Antoni, Inverse relation in the de novo arginase synthesis and nitric oxide production in murine and rat peritoneal macrophages in long-term cultures in vitro, Comp. Biochem. Biophys. 103B (1992) 839–845. [11] A. Hrabák, F. Antoni, I. Csuka, Differences in the arginase activity produced by resident and stimulated murine and rat peritoneal macrophages, Int. J. Biochem. 23 (1991) 997–1003. [12] L. Ornstein, H. Ansley, A. Saunders, Improving manual differential white cell counts with cytochemistry, Blood Cells 2 (1976) 557–568.
1090
N. Müllner et al. / The International Journal of Biochemistry & Cell Biology 34 (2002) 1080–1090
[13] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding, Anal. Biochem. 72 (1976) 248–254. [14] W.C. Sessa, M. Hecker, J.A. Mitchell, J.R. Vane, The metabolism of l-arginine and its significance for the biosynthesis of endothelium-derived relaxing factor. l-Glutamine inhibits the generation of l-arginine by cultured endothelial cells, Proc. Natl. Acad. Sci. U.S.A. 87 (1990) 8607–8611. [15] L.C. Green, D.A. Wagner, J. Glogowski, P.L. Skipper, J.S. Wishnok, S.R. Tannenbaum, Analysis of nitrate, nitrite and (15 N)nitrate in biological fluids, Anal. Biochem. 126 (1982) 131–138. [16] E.B. Spector, S.C.H. Rice, S. Moedjono, B. Bernard, S.D. Cederbaum, Biochemical properties of arginase in human adult and fetal tissues, Biochem. Med. 28 (1982) 165–175. [17] E.I. Closs, CATs, a family of three distinct mammalian cationic amino acid transporters, Amino Acids 11 (1996) 193–208. [18] D.K. Kakuda, M.J. Sweet, C.L. MacLeod, D.A. Hume, D. Markovich, CAT2-mediated l-arginine transport and nitric oxide production in activated macrophages, Biochem. J. 340 (1999) 549–553. [19] A. Hrabák, M. Idei, Á. Temesi, Arginine supply for nitric oxide synthesis and arginase is mainly exogeneous in elicited murine and rat macrophages, Life Sci. 55 (1994) 797–805. [20] C.A. Schott, G.A. Gray, J.C. Stoclet, Dependence of endotoxin-induced vascular hyporeactivity on extracellular l-arginine, Br. J. Pharmacol. 108 (1993) 38–43. [21] R.G. Bogle, A.R. Baydoun, J.D. Pearson, S. Moncada, G.E. Mann, l-arginine transport is increased in macrophages generating nitric oxide, Biochem. J. 284 (1992) 15–18. [22] W.W. Simmons, E.I. Closs, J.M. Cunningham, T.W. Smith, R.A. Kelly, Cytokines and insulin induce cationic amino acid transporter (CAT) expression in cardiac myocytes. Regulation of l-arginine transport and NO production by CAT-1, CAT-2A and CAT-2B, J. Biol. Chem. 271 (1996) 11694–11702. [23] A. Hrabák, T. Bajor, Á. Temesi, G. Mészáros, The inhibitory effect of nitrite, a stable product of nitric oxide (NO) formation, on arginase, FEBS Lett. 390 (1996) 203–206. [24] G. Wu, S.M. Morris, Arginine metabolism: nitric oxide and beyond, Biochem. J. 336 (1998) 1–17.
[25] M. Fishman, Functional heterogeneity among peritoneal macrophages. III. No evidence for the role of arginase in the inhibition of tumor cell growth by supernatants from macrophages or macrophage subpopulation cultures, Cell. Immunol. 55 (1980) 174–184. [26] A. Hrabák, T. Bajor, G.J. Southan, A.L. Salzman, C. Szabó, Comparison of the inhibitory effect of isothiourea and mercaptoalkylguanidine derivatives on the alternative pathways of arginine metabolism in macrophages, Life Sci. 60 (1997) PL395–PL401. [27] E.I. Closs, F.Z. Basha, A. Habermeier, U. Förstermann, Interference of l-arginine analogues with l-arginine transport mediated by the y+ carrier hCAT-2B, Nitric Oxide 1 (1997) 65–73. [28] C.A. Louis, J.S. Reichner, W.L. Henry Jr., B. Mastrofrancesco, T. Gotoh, M. Mori, J.E. Albina, Distinct arginase isoforms expressed in primary and transformed macrophages: regulation by oxygen tension, Am. J. Physiol. 274 (1998) R775–R782. [29] W.W. Wang, C.P. Jenkinson, J.M. Griscavage, R.M. Kern, N.S. Arabolos, R.E. Byrns, S.D. Cederbaum, Co-induction of arginase and nitric oxide synthase in murine macrophages activated by lipopolysaccharide, Biochem. Biophys. Res. Commun. 210 (1995) 1009–1016. [30] C.P. Jenkinson, W.W. Grody, S.D. Cederbaum, Comparative properties of arginases, Comp. Biochem. Physiol. 114B (1996) 107–132. [31] A. Diez, J.M. Fuentes, F. Prada, M.L. Campo, G. Soler, Immunological identity of the two different molecular mass constitutive subunits of liver arginase, Biol. Chem. Hoppe-Seyler 375 (1994) 537–541. [32] Z. Spolarics, J.S. Bond, Multiple molecular forms of mouse liver arginase, Arch. Biochem. Biophys. 260 (1988) 469– 479. [33] E. Moilanen, H. Vaapatalo, Nitric oxide in inflammation and immune response, Ann. Med. 27 (1995) 359–367. [34] G. Cirino, Multiple controls in inflammation. Extracellular and intracellular phospholipase A2, inducible and constitutive cyclooxygenase, and inducible nitric oxide synthase, Biochem. Pharmacol. 55 (1998) 105–111. [35] K.-D. Kröncke, K. Fehsel, V. Kolb-Bachofen, Inducible nitric oxide synthase in human diseases, Clin. Exp. Immunol. 113 (1998) 147–156.