Departments of
* Preclinical and Clinical Pharmacology and
Experimental Pathology and Oncology, University of Florence, Italy
Correspondence: Prof. Flavio Moroni, Dipartimento di Farmacologia Preclinica e Clinica, Università di Firenze, Viale Pieraccini 6, 50139 Firenze, Italy. E-mail: moronif{at}ds.unifi.it
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(TNF-) in interferon-
(IFN-)-activated murine macrophages of the BAC1.2F5 cell line. IFN- (100 U/ml) not only increased the synthesis and release of NO and TNF- from these cells but also induced indoleamine-2,3-dioxygenase, the rate-limiting enzyme of TRP catabolism. This led to an increased metabolic flow through the kynurenine pathway and significantly decreased TRP levels in macrophage incubation media. Low TRP concentrations in the media, however, modified IFN- effects. In TRP-"starved" cultures, in fact, the IFN--mediated NO synthase induction was significantly reduced, and the increased TNF- synthesis and release were not affected. Our results suggest that a reduced local TRP availability may modify macrophage function and possibly the outcome of immune responses.
Key Words: nitric oxide • indoleamine-dioxygenase • arginine • nicotinamide
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Activated macrophages may also release a plethora of biologically active molecules, such as tumor necrosis factor (TNF-), nitric oxide (NO), proteases, and components of the complement system. These molecules may modify the function of neighboring cells and start cellular processes leading to apoptosis or necrosis [7
, 8
]. TNF- and NO, in particular, are two key macrophage-derived mediators of inflammatory responses in mammals. Increasing evidence suggests that both mediators are double-edged swords, with proinflammatory and anti-inflammatory propensities [9
, 10
].
We previously reported that activated macrophages release NO and a number of neuro-active KYN metabolites and that this release may have a key role in the development of inflammatory neurological disorders [11 , 12 ]. It is interesting that NO synthesis from arginine (ARG) and KYN synthesis from TRP are strictly interconnected and may play important roles in cellular homeostasis: Neoformed NO inhibits IDO by interacting with superoxide [13 ]. Conversely, IDO, by promoting the synthesis of NAD(P), provides one of the cofactors for inducible NO synthanse (iNOS), the enzyme primarly responsible for NO synthesis [10 ].
In the course of the above-mentioned studies aimed at evaluating the importance of different substrates [nicotinamide (NA), NAD, TRP, or ARG] in the production of toxic compounds by activated macrophages, we observed that TRP availability may selectively inhibit NO but not TNF- synthesis and release. It seems, therefore, that IDO induction, by controlling the local availability of TRP, may regulate macrophage function and possibly, the outcome of inflammatory and immune responses.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell stimulation
Cells were grown to confluence, culture medium was removed, and cell monolayers were washed with M-CSF-free, serum-supplemented DMEM and incubated in this medium for 16 h. Cells were then treated for different time lengths with interferon- (IFN-), 100 U/ml. In some experiments, after the 16-h incubation in M-CSF-free medium, cell monolayers were extensively washed with phosphate-buffered saline (PBS) and then incubated in a "defective" DMEM (see below) in the absence of serum as well as M-CSF; after 1 h equilibration, cells were finally stimulated for 16 h with IFN-, in the presence of various combinations of TRP, NA, and ARG, separately added to the medium.
Determination of IDO activity
IFN--activated cells were scraped and centrifuged at 3000 rpm for 5 min. The cell pellet was resuspended and sonicated in 0.1 M phosphate buffer, pH 7.4, containing 1 mM phenylmethylsulfonyl fluoride (PMSF) and 2 µg/ml aprotinin and leupeptin. The cell lysate was then centrifuged at 12,000 rpm for 10 min at 4°C. The supernatant was used as a cytosol preparation. IDO activity was evaluated according to Takikawa et al. [15
]. The reaction mixture consisted of 50 µl cytosol preparation and 50 µl 0.1 M phosphate buffer, pH 6.5, containing 20 mM ascorbate, 50 µM methylene blue, 200 µg catalase, and 2 mM TRP. After 30 min of incubation at 37°C, the reaction was terminated by adding 100 µl 20% (w/v) trichloroacetic acid, and the mixture was centrifuged at 14,000 rpm for 5 min.
Measurement of TRP and KYN concentrations
TRP concentration in culture medium of IFN--activated cells was determined by high-pressure liquid chromatography (HPLC) and fluorimetric detection. Medium was deproteinized by mixing with an equal volume of 10% (w/v) trichloroacetic acid. HPLC separation was obtained with a reverse-phase column (Spherisorb S5 ODS2, 25 cm) and a mobile phase (1 ml/min flow rate) composed of 5% acetonitrile, 100 mM phosphate buffer, pH 3.6, and 1 mM EDTA. Detection was performed with a Perkin-Elmer (Foster City, CA) model LC 240 fluorimeter; excitation and emission wavelengths were 313 and 420 nm, respectively.
KYN was measured using HPLC and UV detection as described by Holmes [16 ]. Briefly, HPLC separation was obtained with a reverse-phase column (Spherisorb S5 ODS2, 10 cm) and a mobile phase (1 ml/min flow rate) composed of 2% acetonitrile, 0.1 mM ammonium acetate, and 100 mM acetic acid. KYN was detected at 365 nm with a UV detector (Perkin-Elmer model LC 90).
Measurement of NO synthesis and release
The levels of NO accumulation in the culture medium of IFN--activated cells were measured using a micro-plate version of the Greiss reaction [17
]. Briefly, 100 µl medium was added to an equal volume of Greiss reagent (0.1% naphthylenediamine-dihydrochloride in water/1% sulfanilamide in 5% phosphoric acid, w/v). After 10 min, the absorbance at 550 nm was measured in a Bio-Rad (Hercules, CA) micro-plate reader (mod. 350 UV). The amount of nitrite formed was calculated from a calibration curve obtained using sodium nitrite as a standard.
Measurement of TNF- synthesis and release
The amount of TNF- produced by IFN--activated cells was measured in cell lysates and culture media by enzyme-linked immunosorbent assay (ELISA; Bender MedSystems, Vienna, Austria, cat no. BMS607), using the protocol indicated by the manufacturer.
Protein separation and immunoblotting
IFN--activated cells were washed twice with PBS, scraped off the plate, and centrifuged at 2000 rpm for 5 min. Pellets were resuspended in 50 µl 20 mM Tris buffer, pH 8, containing 1 mM PMSF, 2 µg/ml aprotinin and leupeptin, and 1% Triton X-100, and the suspension was sonicated three times on ice. Protein content was determined by the bicinchoninic acid assay using the Pierce kit (Rockford, IL) according to the manufacturer’s instructions, and the samples were normalized on the basis of protein content (30 µg protein/sample). Laemmli buffer containing 100 mM 2-mercaptoethanol was then added, and the samples were boiled for 10 min and subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis in 8% acrylamide mini-gels (200 V, 45 min). The separated proteins were transferred onto a nitrocellulose membrane (Hybond-ECL, Amersham, Milan, Italy) by electroblotting (100 V, 90 min). Equal lane loading was evaluated by staining the membranes with Ponceau-S nonpermanent red stain. After recording the images, the membranes were washed in PBS containing 0.1% Tween-20 (T-PBS) to remove the stain and were processed for immunoblotting. To estimate the level of iNOS expression, membranes were blocked [3 h, room temperature (RT)] with T-PBS containing 5% bovine serum albumin (BSA; T-PBS/5% BSA) and were incubated overnight at 4°C in a 1:1000 dilution of rabbit anti-iNOS antibodies (Sigma Chemical Co., St. Louis, MO). Membranes were then incubated for 1 h in a 1:5000 dilution of horseradish peroxidase-conjugated anti-rabbit immunoglobulin antibodies in T-PBS/2% BSA. After a final wash in T-PBS, membranes were incubated (1 min, RT) in a chemiluminescent reagent [the enhanced chemiluminescence (ECL) protein detection system, Amersham], and the peroxidase-coated protein bands were visualized on Hyperfilm-ECL (Amersham) after 1 min exposure.
Reagents
Murine recombinant (mr)IFN- was from PeproTech EC (London, UK). MrM-CSF was prepared in our laboratory by expression in bacteria, his tag conjugation, and affinity purification on Ni2+-nitrilotriacetic acid agarose columns (Quiagen, Hilden, Germany). DMEM, TRP, ARG, NA, and KYN were from Sigma Chemical Co. Fetal bovine serum and the special, defective DMEM lacking TRP, ARG, NA, and phenol red were from Gibco-BRL (San Giuliano Milanese, MI, Italy). All the remaining compounds were from Merck (Darmstadt, Germany).
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
on IDO activity and TRP catabolism in BAC-1.2F5 cells, IDO activity underwent a sixfold increase in 6 h (17±3 pmol/mg prot/min) and a 15-fold increase in 24 h (42±9 pmol/mg prot/min; Fig. 1A
). This increase was associated with a parallel decrease of TRP concentration in the culture medium. Figure 1B
shows that IFN--treated cells caused a linear, time-dependent decrease in TRP concentration in their incubation medium (from 75±6 to 36±4 µM in 48 h). In IFN--treated cells, a number of KYN-metabolizing enzymes were also induced, and KYN levels reached 0.26 ± 0.035 µM in 24 h [11
].
 View larger version (14K): [in a new window] |
Figure 1. Induction of IDO and TRP catabolism in IFN--activated macrophages. Confluent BAC-1.2F5 cell cultures were stimulated with 100 U/ml IFN-, and the effects of IFN- on IDO activity and TRP concentration were measured at different times after stimulation. (A) IDO activity significantly increased as early as 6 h after exposure to IFN- and linearly increased to 24 h. Values are means ± SEM of three experiments, each performed in duplicate. (B) IFN- induced a time-dependent decrease in TRP concentration in the culture medium, which in 48 h, was reduced by  50%. Values are means ± SEMof three experiments, each performed in duplicate.
|
-induced NO synthesis while incubated, in the absence of serum, in defective DMEM supplemented with various combinations of TRP (75 µM), ARG (400 µM), and NA (35 µM). The effects of these compounds on NO synthesis were estimated after 16 h by measuring nitrite concentrations in the culture medium as an index of NO formation (Fig. 2A
). No evidence of cell distress or death could be observed by phase-contrast microscopy and the trypan blue exclusion test when these cells were incubated for 2 days in plain defective medium. The addition of TRP, but not ARG or NA, caused a 12-fold increase in NO synthesis. The addition of NA and ARG also increased NO formation. When ARG and/or NA were added together with TRP, NO formation increased to levels that did not differ from those obtained with TRP alone. To investigate if reduced nicotinamide adenine dinucleotide phosphate (NADPH) formation through the KYN pathway could promote iNOS activity, we activated macrophages in the presence of ARG and KYN (100 µM). The latter compound is a cell-permeable TRP metabolite and a possible NA precursor. The addition of KYN did not increase the synthesis of NO to levels higher than those induced by ARG alone. Taken together, these results suggested that in activated macrophages, TRP availability was the main determinant of NO synthesis and that this effect was not a result of NAD formation through the KYN pathway.
 View larger version (25K): [in a new window] |
Figure 2. Effects of ARG, NA, TRP, and KYN on the IFN--induced iNOS expression and nitrite production. Confluent BAC-1.2F5 cell cultures were stimulated with 100 U/ml IFN- for 16 h in a medium lacking NA, TRP, ARG, and serum (see Materials and Methods). (A) Production of nitrites. Values are means ± SEM of three experiments, each performed in duplicate. The addition of TRP (75 µM) was sufficient to enable IFN--activated macrophages to produce nitrites; on the contrary, ARG (400 µM) or NA (35 µM) had no effects on nitrite formation, and their combination led to significant increases. Notably, the addition of ARG and/or NA to TRP had no effect on the formation of nitrites in the presence of TRP alone. Also, the combination of ARG plus KYN (100 µM) was unable to increase the formation of nitrites induced by IFN-. (B) Expression of iNOS. The intensities of the electrophoretic bands (above) were measured by densitometry (below). One experiment representative of four is shown. In accordance with the formation of nitrites, the IFN--induced expression of iNOS strictly depended on the presence of TRP in the incubation medium. Note that in the absence of TRP, the combination of ARG plus NA had no effect.
|
-induced iNOS expression in defective DMEM, and iNOS expression was evaluated by immunoblotting (Fig. 2B)
. IFN--activated cells were able to express iNOS in the absence of ARG, TRP, and NA. The addition of ARG (400 µM) or NA (35 µM) to the medium was unable to further induce iNOS expression. However, in the presence of 75 µM TRP, iNOS expression significantly increased, and the addition of ARG and/or NA to TRP did not further increase iNOS expression. These results suggested that iNOS expression may be impaired when TRP concentrations in the medium are quite low and that ARG or NA cannot compensate for the TRP shortage.
Dependence of iNOS expression and NO synthesis on TRP concentration
To further investigate the relationship between TRP availability and the induction of iNOS, we evaluated the effects of different TRP concentrations on the IFN--induced iNOS expression and NO synthesis (Fig. 3A
). We found that TRP concentrations ranging from 0.1 to 3 µM were not sufficient to allow for iNOS induction, and 10 µM TRP determined a significant increase of this induction, which did not increase further with higher TRP concentrations. Figure 3B
shows that nitrite production and iNOS expression depend on TRP availability. It seems that a threshold concentration of TRP may strictly regulate iNOS expression.
 View larger version (16K): [in a new window] |
Figure 3. Dose-effect of TRP on the IFN--induced iNOS expression and nitrite production. Confluent BAC-1.2F5 cell cultures were stimulated with 100 U/ml IFN- for 16 h in a medium lacking ARG, NA, and serum (see Materials and Methods) and containing different concentrations of TRP. (A) Expression of iNOS. The intensities of the electrophoretic bands (above) were measured by densitometry (below). One experiment representative of two is shown. The lane labeled ST refers to IFN--treated cultures in DMEM, containing 75 µM TRP. Under stimulation with IFN-, iNOS was poorly expressed with TRP concentrations in the range of 0–3 µM and markedly induced with 10–100 µM TRP. (B) Production of nitrites. Values are means ± SEM of three experiments, each performed in duplicate. Values were correlated with the levels of iNOS expression, corresponding to the densitometric values shown in A, expressed as a percentage of those obtained stimulating the cells in DMEM.
|
-induced production of TNF--inducible proteins, we determined the effects of different TRP concentrations on the synthesis and release of TNF-, one of the most important targets for activation stimuli in macrophages. As shown in Table 1
, cell treatment with IFN- in the absence of TRP induced the appearance of TNF- in cell lysates and culture media. The addition of TRP did not significantly influence TNF- production. |
View this table: [in a new window] |
Table 1. Dose-Effect of TRP on the IFN--Induced TNF- Production
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
synthesis and release increase during macrophage activation, a process that plays a key role in the successful eradication of pathogens. In the course of studies aimed at evaluating the importance of neurotoxic compounds released from activated macrophages [11
, 18
], we noticed that culture availability of TRP could significantly affect NO release from these cells. We report here that the concentrations of TRP in the culture medium may significantly modify NO but not TNF- synthesis and release from activated cells.
A number of interconnections between the pathways leading to NO from ARG and to NAD from TRP (the KYN pathway) have been described previously. Indeed, picolinic acid, a TRP metabolite, is able to facilitate iNOS expression by interacting with a hypoxia-responsive element present in the promoter region of the gene [19
]. Conversely, 3OH-anthranilic acid, another TRP metabolite, inhibits iNOS expression and activity [20
]. It has also been shown that NO inhibits IDO activity, and this could be an additional link between the two pathways [13
]. It seems, therefore, that regulation of the expression and activity of iNOS in macrophages is a finely tuned event [21
, 22
]. Our results, by demonstrating that TRP availability in the culture medium has an important role in the induction of iNOS but does not affect TNF- biosynthesis or release, provide further evidence of the importance of the interconnections between ARG and TRP metabolic pathways in the regulation of macrophage function.
IFN- rapidly induced IDO expression in BAC-1.2F5 cells, an observation in keeping with previous experiments performed in different macrophage cell lines [23
24
25
26
27
28
]. IDO induction resulted in a significant decrease in TRP concentration in the medium, supporting the concept that macrophages, once activated, may significantly reduce TRP availability in the extracellular environment. This local "TRP starvation" has been previously reported as being able to affect neighboring T cell function and to also cause immune tolerance [5
]. After a 48-h stimulation with IFN-, TRP concentration in culture medium decreased by 50%. We estimated that BAC-1.2F5 cells metabolized TRP at an approximate rate of 6 nmol/106 cells/h. This massive TRP degradation corroborates the hypothesis that macrophages may reduce TRP availability in the extracellular medium to such an extent that neighboring T cells are TRP-starved and anergized [5
].
In activated macrophages, contrary to other cells, ARG did not increase NO synthesis, suggesting that macrophages are able to synthesize this amino acid. It has indeed been reported that arginino-succinate synthetase and arginino-succinate lyase, two of the enzymes involved in urea metabolism, may operate with NOS in the so-called citrulline-NO cycle to synthesize ARG from L-citrulline [29 ]. It is interesting that in activated macrophages, these enzymes are coinduced with iNOS [30 , 31 ]. Considering that ARG availability is one of the rate-limiting factors for NO synthesis, our data demonstrate that the citrulline-NO cycle is fully efficient and confers to iNOS an absolute independence from the ARG supply.
It is interesting that when macrophages cultured in the presence of ARG were supplemented with NA, NO synthesis increased. As these effects were not a result of iNOS induction, it is possible that NA promoted iNOS activity by increasing NADPH availability. In keeping with this hypothesis, it was recently reported that in activated macrophages, NAD contents are strongly enhanced by inhibiting NAD use with poly(ADP-ribose)synthetase inhibitors [32
], thus suggesting that the availability of pyridine cofactors may limit NAD(P)-dependent enzyme activity. As IFN--activated BAC-1.2F5 cells release measurable amounts of QUIN, it is possible that the enzyme QUIN-phosphorybosil transferase, deputed to convert QUIN to NAD, is not abundantly expressed in these cells [11
]. It has indeed been demonstrated that macrophages able to rapidly synthesize NAD from TRP do not release QUIN after IFN- stimulation [32
].
Our data showed that NO production was strictly dependent on TRP supply and that TRP mainly acted by regulating the expression of iNOS. The correlation between TRP availability and iNOS expression was not linear: TRP concentrations in the medium ranging from 0.3 to 3 µM were not sufficient to allow iNOS induction, and the maximal induction levels were obtained at 10 µM, suggesting that a critical threshold of TRP concentration is necessary to allow this process. It is interesting that this concentration is in the same range as IDO Km (13 µM) [15
]. On this basis, we speculate that only when IDO is saturated is TRP available for iNOS synthesis. An exclusive role of TRP with respect to the other amino acids in macrophage homeostasis is suggested by the fact that tryptophanyl-tRNA synthetase is the only inducible tRNA-synthetase gene. It is intriguing that this enzyme is typically induced in IFN--activated macrophages, and its expression pattern is strictly regulated by the efficiency of the ongoing protein synthesis [33
]. This probably represents a strategy developed by macrophages throughout evolution to allow for protein synthesis to continue, despite low, free TRP concentrations. This may obviously occur when IDO, an enzyme appearing quite early during evolution [34
], remains activated for long periods.
Macrophages are frequently activated in peripheral and scarcely vascularized tissues where the availability of substrates and enzymatic cofactors is crucial for the successful eradication of pathogens. However, macrophages are the first to face the toxicity of their own antimicrobial products, such as NO and superoxide ion, and therefore developed various strategies to avoid a self-inflicted, oxidative damage and survive in offensive conditions. In agreement with this, regulation of iNOS expression and activity is one of the most finely tuned events during macrophage activation.
In conclusion, the catabolism of TRP has a pivotal role in processes related to macrophage activation. It promotes NO formation, allowing for the synthesis of pyridine-nucleotide cofactors but also limits iNOS expression when TRP levels fall below a critical threshold. The powerful immunomodulatory role of KYN pathway induction is in line with previous studies [5 ] and suggests that TRP starvation is a plausible scenario in the control of immune responses.
Received September 13, 2002; accepted September 23, 2002.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and IL-12: a balancing act in macrophage functioning Microbes Infect 3,121-129[CrossRef][Medline]-activated macrophages J. Leukoc. Biol. 68,260-266-activated murine macrophages and microglial cells J. Immunol. 159,419-424[Abstract] in murine cloned macrophages and microglial cells J. Neurochem. 66,996-1004[Medline]-activated murine macrophages and microglial cells J. Immunol. 159,419-426This article has been cited by other articles:
 |
|
 |
|
  S. A. Sarkar, R. Wong, S. I. Hackl, O. Moua, R. G. Gill, A. Wiseman, H. W. Davidson, and J. C. Hutton Induction of Indoleamine 2,3-Dioxygenase by Interferon-{gamma} in Human Islets Diabetes, January 1, 2007; 56(1): 72 - 79. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Nittler, D. Hocking-Murray, C. K. Foo, and A. Sil Identification of Histoplasma capsulatum Transcripts Induced in Response to Reactive Nitrogen Species Mol. Biol. Cell, October 1, 2005; 16(10): 4792 - 4813. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Martins Jr., A. C. F. Ferreira, A. L. Skorupa, S. C. Afeche, J. Cipolla-Neto, and L. F. B. P. Costa Rosa Tryptophan consumption and indoleamines production by peritoneal cavity macrophages J. Leukoc. Biol., June 1, 2004; 75(6): 1116 - 1121. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. El-Gayar, H. Thuring-Nahler, J. Pfeilschifter, M. Rollinghoff, and C. Bogdan Translational Control of Inducible Nitric Oxide Synthase by IL-13 and Arginine Availability in Inflammatory Macrophages J. Immunol., November 1, 2003; 171(9): 4561 - 4568. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
