Arginase I is constitutively expressed in human granulocytes and participates in fungicidal activity
http://www.100md.com
《血液学杂志》
the Department of Hematology, Oncology and Rheumatology, University Hospital Heidelberg, Heidelberg, Germany
Clinical Cooperation Unit Molecular Hematology and Oncology, German Cancer Research Center, Heidelberg, Germany
Institute of Immunology, University of Heidelberg, Heidelberg, Germany
Centro de Investigacion del Cancer, Instituto de Biología Molecular y Celular del Cancer, CSIC-Universidad de Salamanca, Salamanca, Spain
Division of Cell Biology, Netherlands Cancer Institute, Amsterdam, The Netherlands
Section of Pediatric Pulmonology and Infectious Diseases, IIIrd Department of Pediatrics, Children`s Hospital, University of Heidelberg, Heidelberg, Germany
Departamento de Bioquímica y Biología Molecular, E.U. Enfermería y T.O., Universidad de Extremadura, Caceres, Spain
Departamento de Bioquímica y Biología Molecular, Facultad de Veterinaria, Universidad de Extremadura, Caceres, Spain
Department of Cellular Immunology, Max-Planck Institute for Immunobiology, Freiburg, Germany.
Abstract
The balance of arginine metabolism via nitric oxide synthase (NOS) or arginase is an important determinant of the inflammatory response of murine macrophages and dendritic cells. Here we analyzed the expression of the isoform arginase I in human myeloid cells. Using healthy donors and patients with arginase I deficiency, we found that in human leukocytes arginase I is constitutively expressed only in granulocytes and is not modulated by a variety of proinflammatory and anti-inflammatory stimuli in vitro. We demonstrate that arginase I is localized in azurophil granules of neutrophils and constitutes a novel antimicrobial effector pathway, likely through arginine depletion in the phagolysosome. Our findings demonstrate important differences between murine and human leukocytes with respect to regulation and function of arginine metabolism via arginase.
Introduction
In the mouse, the metabolism of arginine is essential for many immunologic functions. In leukocytes, arginine can be metabolized by inducible nitric oxide synthase (iNOS) to the cytotoxic, tumoricidal, and antimicrobial effector molecule nitric oxide (NO). Myeloid cells also hydrolyze arginine to urea and ornithine with the enzyme arginase, and this leads to the synthesis of polyamines and proline. Two arginase isoenzymes exist (arginase I and II), which differ in subcellular localization, regulation, and possibly function.1 Arginase I is a cytosolic enzyme, which is expressed mainly in the liver as part of the urea cycle, whereas arginase II is a mitochondrial protein found in a variety of tissues. In the murine immune system, arginase and iNOS undergo reciprocal induction in several cell types by a variety of inflammatory and anti-inflammatory agonists. In murine macrophages2-7 and dendritic cells4 the polarization of arginine metabolism is driven by cytokines. T-helper (TH) 1 cytokines induce iNOS, whereas TH2 cytokines up-regulate arginase I.
Several important immunologic functions of arginase have been clarified recently, including a direct and indirect role in inducing T-cell hyporesponsiveness8-10 and the induction of TH2-mediated immunopathology in murine schistosomiasis.11 The TH2-induced arginase I of murine macrophages supports the growth of various microorganisms,12-14 most likely by competing with iNOS for arginine and by providing polyamines, and arginase is also found in murine inflammatory cell infiltrates in experimental autoimmune encephalomyelitis,15 glomerulonephritis,16 herpes simplex virus infection of the eye,17 and asthma.18
In contrast to the wealth of knowledge regarding arginine metabolism in the murine immune system, very little is known about a possible role of arginase in human immunology. Arginase was detected in human mononuclear cells after injury19 and found in inflammatory cells of bronchoalveolar lavage fluid of asthmatic patients18 or in psoriatic lesions.20 We have therefore analyzed the role of arginase in the human immune system. Surprisingly, the expression and function of arginase is strikingly different in human leukocytes when compared with murine leukocytes. Although arginase remains a crucial player in host defense, we demonstrate that within the human immune system arginase I is selectively expressed in granulocytes (polymorphonuclear leukocytes [PMNs]), quantitatively the largest subpopulation of myeloid cells. We show that in human PMNs the enzyme is not cytosolic but localizes to the azurophil granules. Finally, we provide evidence that arginase I is a novel antimicrobial effector mechanism of human PMNs, likely operating through arginine depletion within the phagolysosome.
Materials and methods
Blood donors and mice
Experiments were performed in compliance with the relevant laws and institutional guidelines, and human studies were approved by the ethics committee of the University of Heidelberg. Informed consent was provided according to the Declaration of Helsinki. Mice of strain C57BL/6 were obtained from the specific pathogen-free animal facilities of the Max-Planck Institute for Immunobiology and were used between 6 and 8 weeks of age.
Reagents and cells
If not otherwise stated, chemicals were purchased from Sigma (St Louis, MO). N--hydroxy-nor-L-arginine (nor-NOHA) was from Bachem (Bubendorf, Switzerland). Recombinant murine cytokines interleukin 10 (IL-10) and granulocyte colony-stimulating factor (G-CSF) were from Pepro-Tech (London, United Kingdom), and IL-4 was from R&D Systems (Minneapolis, MN). All recombinant human cytokines were from PromoCell (Heidelberg, Germany). The L-arginine auxotroph strain of Saccharomyces cerevisiae is an isogenic derivative of the laboratory strain FL100 with a deficiency in ornithine transcarbamoylase (OTC, arg3), generously provided by Jacky de Montigny and Ives Lombard (both Universite Louis Pasteur, Strasbourg, France). In some experiments wild-type Candida albicans strain ATCC 90028 was used.
Generation of murine BMDMs and PMNs
Murine bone marrow-derived macrophages (BMDMs) were generated as previously described.3,4 Murine PMNs were generated from bone marrow21 with slight modifications. Briefly, bone marrow cells (1 x 106/mL) were cultured in RPMI 1640 medium (20% horse serum, 5 ng/mL G-CSF, 5 nM hydroxycortisone, 2 mM L-glutamine, 60 μM 2-mercaptoethanol, 1 mM sodium pyruvate, 1 x nonessential amino acids, 100 U/mL penicillin, and 100 μg/mL streptomycin). The cells were fed on day 4 with fresh medium (without G-CSF and hydroxycortisone) and harvested on day 7. After 1 hour of plastic adherence, the nonadherent fraction was harvested. This procedure led to a PMN purity of 90% to 95%, assessed by microscopy and flow cytometry analysis (GR1hi cells with characteristic forward/side-scatter profile).
Generation of human macrophages and isolation of human PBMCs and PMNs
CD14+ monocytes were purified from peripheral blood of healthy human donors with the magnetically activated cell sorting (MACS) system (Miltenyi Biotec, Bergisch-Gladbach, Germany) and cultured for 10 days with M-CSF (5 ng/mL) in hydrophobic Teflon bags3,4 to yield macrophages. To isolate human peripheral blood mononuclear cells (PBMCs) and PMNs, EDTA (ethylenediaminetetraacetic acid)–anticoagulated peripheral blood of healthy human donors was layered on top of Ficoll. After 20 minutes of centrifugation (700g), the PBMCs (= interphase) were harvested. The pellet, containing erythrocytes and PMNs, was resuspended in Hanks balanced salt solution (HBSS, without Ca2+ and Mg2+) and mixed at a ratio of 1:1 with 3% dextran. After sedimentation of erythrocytes (20 minutes) the PMN-rich supernatant was harvested and the remaining erythrocytes were subjected to hypotonic lysis (155 mM NH4Cl, 10 mM KHCO3, 0.1 mM EDTA, pH 7.4) for 15 minutes on ice. After washing, cell purity and viability were checked by flow cytometry and microscopy. The method reproducibly yields PMNs with a purity and viability of more than 95%. When indicated, PMNs were further purified by flow cytometry on a Vantage Flow Cytometer (BD Biosciences, Heidelberg, Germany) according to forward and side scatter, yielding purities greater than 99.5%. PBMCs were separated in CD14+ (monocyte fraction, purity 95%) and CD14- cells (lymphocyte fraction, purity 90%) with the MACS system.
Measurement of PMN fungicidal activity
PMNs were preincubated with inhibitors for 30 minutes at 37°C. Yeast and PMNs (5 x 106 each) were cultured in 1 mL HBSS (+ 10% human serum) for 60 minutes. Viability of yeast was evaluated with the XTT ((2,3)-bis-(2-methoxy-4-nitro-5-sulphenyl)-(2H)-tetrazolium-5-carboxanilide) assay exactly as described22 and with the FUN-1 yeast viability kit (Molecular Probes, Gttingen, Germany). In the latter case, PMNs were lysed in ice-cold water and in case of S cerevisiae the cell pellet was resuspended for 10 minutes at 34°C in 50 μL DNAse I (1 U/μL, Invitrogen, Carlsbad, CA) and streptolysin O (2 mg/mL) for permeabilization. Then, 50 μL FUN-1 cell stain (0.125 mM) plus Calcofluor White M2R (0.05 mM) in HBSS/4% glucose were added. After 30 minutes at 30°C in the dark, at least 200 yeast cells per sample were analyzed by fluorescence microscopy. The percentage of fungal cell damage ("% kill") was defined by the following equation: 100 - (100 x [OD405 of fungi with PMN]/[OD405 of fungi]). The percentage of reduction of fungal cell damage ("% kill inhibition") on treatment with inhibitors was defined by the following equation: 100 - 100 x [% Kill of fungi with PMN with inhibitor]/[% Kill of fungi with PMN]).
Metabolism of 14C-L-arginine and TLC
After washing, 2 x 105 cells were incubated for 2 hours at 37°C with arginine-free Dulbecco modified Eagle medium containing 2% fetal calf serum (FCS) and 0.1 μCi (0.037 MBq) L-(U-14C) arginine (Amersham Biosciences, Freiburg, Germany). Cells were subsequently lysed by 2 freeze-thaw cycles. Then, 20 μL of the lysate was spotted onto thin-layer chromatography (TLC) plates (Merck, Darmstadt, Germany), dried for 1 hour at 42°C, and developed in the solvent system chloroform/methanol/ammonium hydroxide/water 0.5/4.5/2.0/1.0 (vol/vol). Spots were developed with Ninhydrin by heating at 120°C for 5 minutes and analyzed by scintillation counting.
Arginase enzymatic assay
Arginase activity was measured in cell lysates as previously described4 with slight modifications. Briefly, cells were lysed with 0.5% Triton X-100, 25 mM Tris (tris(hydroxymethyl)aminomethane)–HCl, pH 7.5, containing 1 mM phenylmethylsulfonyl fluoride (PMSF) and 2 μg/mL each of aprotinin, leupeptin, and pepstatin (lysis buffer). To 100 μL of this lysate 20 μLof 10 mM MnCl2 was added, and the enzyme was activated by heating for 10 minutes at 56°C. In case of determination of native arginase activity, these steps (addition of exogenous manganese and heating) were omitted. Arginine hydrolysis and measurement of urea concentration were performed exactly as previously described.4 One unit of enzyme activity is defined as the amount of enzyme that catalyzes the formation of 1 μmol urea/min.
SDS-PAGE and Western blot analysis
Cells were lysed for 30 minutes on ice in lysis buffer (see "Arginase enzymatic assay"). Cell debris was spun down at 18 000g for 5 minutes at 4°C. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was done as previously described.4 The proteins were transferred to a Hybond-P polyvinylidene difluoride (PVDF) membrane (Amersham Biosciences). After blocking with 5% nonfat dry milk in TBST-buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 0.05% Tween 20) for 2 hours, the membranes were incubated with the following primary antibodies (1:5000 in TBST, 5% bovine serum albumin): polyclonal rabbit antirat arginase I antiserum,23 which is cross-reactive to mouse and human arginase I and anti-ERK1/2 (extracellular signal regulated kinase, Cell Signaling Technology, Beverly, MA). Antibody reactivity was monitored with horseradish peroxidase (HRP)–conjugated antirabbit IgG (Santa Cruz Biotechnology, Santa Cruz, CA), followed by visualization with the enhanced chemiluminescence (ECL) detection system (Amersham Biosciences).
Subcellular fractionation of neutrophils
Subcellular fractionation of neutrophils was performed as previously described.24 Briefly, resting neutrophils isolated from freshly heparinized human peripheral blood were resuspended in 50 mM Tris-HCl, pH 7.5 containing 2 mM PMSF, and then disrupted by repeated freeze-thaw. Homogenates were centrifuged at 1200 rpm in a Sorvall T 6000D centrifuge (Kendro, Hertfordshire, United Kingdom) for 10 minutes, and the supernatant, representing the postnuclear extract, was saved. After centrifugation of the postnuclear extract at 110 000g in a TLA rotor for 90 minutes at 4°C using an Optima TL Ultracentrifuge (Beckman Instruments, Palo Alto, CA), supernatant (soluble fraction) and pellet (membrane fraction, resuspended in 50 mM Tris-HCl, pH 7.5, containing 2 mM PMSF) were saved. To prepare the distinct subcellular fractions, freshly prepared neutrophils (about 3-5 x 108) were gently disrupted, and the postnuclear fractions were fractionated in 15% to 40% continuous sucrose gradients as described previously.25 Subcellular fractions were analyzed for marker proteins for each organelle as described.25
Immunoelectron microscopy
Resting human neutrophils and exudate neutrophils from skin window chambers after phagocytosis of latex beads26 were fixed for 24 hours in 4% paraformaldehyde in PHEM buffer (60 mM PIPES [piperazine diethanesulfonic acid], 25 mM HEPES [N-2-hydroxyethylpiperazine-N'2-ethanesulfonic acid], 2 mM MgCl2, 10 mM EGTA [ethylene glycol tetraacetic acid], pH 6.9) and then processed for ultrathin cryosectioning as previously described.27 Ultra-thin frozen sections were incubated at room temperature with the indicated antibodies and 10 or 15 nm protein A gold,27 embedded in a mixture of methylcellulose and uranyl acetate, and examined with a Philips CM 10 electron microscope (Eindhoven, The Netherlands; original magnification x 12 500). For controls, the primary antibody was replaced by an irrelevant rabbit antibody. Antibodies used in double immunolabeling were rabbit antihuman lactoferrin (Lf; Cappel Laboratories, Durham, NC) and rabbit antihuman myeloperoxidase (MPO) (Dako, Glostrup, Denmark).
Results
Arginase in human myeloid cells: specific expression in PMNs
We analyzed the expression of arginase in various human myeloid cell types treated or not with several agonists. We stimulated the cell populations with the recombinant human cytokines IL-1, IL-1, interferon (IFN-), tumor necrosis factor (TNF-), IL-4, IL-10, and IL-13 as well as lipopolysaccharide (LPS), prostaglandin E2 (PGE2), dexamethasone, and the cyclic adenosine monophosphate (cAMP)–enhancing agents dibutyryl-cAMP and forskolin, alone and in various combinations. Human monocyte-derived macrophages and dendritic cells showed no arginase activity in the resting state. In contrast to the murine system, we were also unable to induce arginase activity or protein in human macrophages (Figure 1), dendritic cells, and monocytes (data not shown) with TH2 cytokines over a wide range of individual concentrations (0.01-100 ng/mL for each cytokine) or any of the agonists. We also stimulated various immortalized human myeloid cell lines (U937, HL-60, THP-1, with or without induction of differentiation by phorbol myristate acetate or retinoic acid) in the same way with the same negative results (data not shown). We then turned to leukocytes, which we derived directly from the blood of healthy donors. In the PMNs of all the blood donors (n = 31) we consistently found high arginase activity (mean, 1644 ± 423 mU/mg protein; range, 667-2189 mU/mg protein; Figure 2A). In contrast, in 19 of 25 PBMC fractions (involving lymphocytes and monocytes) no arginase activity was detectable, whereas in the remaining 6 PBMC samples we found low arginase activity (mean, 111 ± 46 mU/mg protein; range, 35-156 mU/mg protein; Figure 2A). To rule out any contamination of the PBMC fraction with PMNs or vice versa, we corroborated our findings by fluorescence-activated cell sorting (FACS) that sorted a pure PMN population according to forward/side-scatter pattern from the conventional PMN fraction (purity of FACS sorted PMNs > 99.5%). Alternatively, we eliminated contaminating PMNs from the PBMC fraction by FACS. In 3 independent experiments we only demonstrated arginase activity and protein in the PMN fraction, whereas PBMCs, purified monocytes (CD14+, purity 95%), or purified lymphocytes (CD14-, purity 90%) completely lacked enzymatic activity and protein (Figure 2B). We also analyzed PMNs from 2 patients with complete arginase I deficiency (OMIM 207800 [OMIM] ; patient 1, 18-year-old man; patient 2, 5-year-old girl). The PMNs of these patients had neither arginase activity nor expression of arginase I protein (Figure 2C). Therefore, we concluded that resting human PMNs express solely the hepatic isoform (arginase I) and have no compensatory up-regulation of arginase II even in the face of complete arginase I deficiency.
Because arginase I is a TH2-inducible protein in murine macrophages and dendritic cells,2-7 we wondered if human PMN arginase would be further augmented by TH2 cytokine stimulation. We treated human PMNs with the TH2 cytokine IL-4 or a combination of IL-4 and IL-10, treatments that yield maximal arginase activity in murine myeloid cells3 (Figure 1). In 6 separate experiments (4 with conventionally purified PMNs and 2 with FACS-sorted PMNs) we found no significant change in human PMN arginase activity, protein expression (Figure 3A), or arginase I mRNA levels (data not shown) on TH2 cytokine stimulation. Finally, we analyzed the expression of arginase in murine PMNs. Unlike their human counterparts, unstimulated murine PMNs show no arginase protein or activity (Figure 3B). The TH2 cytokine IL-4 either alone or in cooperation with IL-10, induces the hepatic isoform arginase I and considerable arginase activity (mean of 3 experiments, 1153 mU/mg protein; range, 1162-1290 mU/mg protein) in murine PMNs (Figure 3B). These cells therefore follow the same pattern of arginase I regulation as murine macrophages and dendritic cells. We conclude that murine and human PMNs differ fundamentally in terms of basal expression and cytokine-mediated regulation of arginase I.
PMNs do not metabolize L-arginine despite high arginase activity
On TH2 stimulation, murine macrophages up-regulate arginase I and consume arginine present in the local microenvironment.1,11 Because human PMNs express high arginase activity constitutively, we wondered if these cells were constantly hydrolyzing arginine after uptake from the extracellular milieu. We therefore incubated resting human PMNs with radioactively labeled 14C-L-arginine and analyzed the conversion of this substrate into ornithine and spermine by TLC. As a control we used resting and IL-4–stimulated murine BMDMs. Whereas resting BMDMs showed no arginase activity and no significant hydrolysis of arginine, the TH2-stimulated, arginase-expressing murine BMDMs metabolized arginine and produced ornithine and spermine. Surprisingly, human PMNs were very inefficient in degrading arginine (Table 1). This led us to question whether, in contrast to the cytosolic location of arginase within murine macrophages,1,28 the enzyme might be sequestered in a different compartment in human PMNs with impaired arginine accessibility.
Arginase I is present in azurophil granules of human PMNs
To determine the localization of the enzyme within human PMNs, we performed subcellular fractionation experiments. First, we separated resting human PMNs in a soluble and a membrane fraction and found that arginase I is not a membrane-bound protein in resting human PMNs (Figure 4A). Subsequent subcellular fractionation assays (Figure 4B), which resolved cytosol, plasma membrane, tertiary granules, specific granules, and azurophil granules,24,25 showed that arginase I protein is mainly present in fraction 8 (Figure 4C), the MPO+ azurophil granules of resting human PMNs. Arginase activity assays with aliquots of the various PMN fractions confirmed these results (Figure 4C). Because arginase is enriched in this subcellular compartment, we observed an even higher enzyme activity (mean of 3 experiments: 3650 ± 97 mU/mg protein in fraction 8) compared to whole-cell PMN lysates (Figure 2A).
To confirm the biochemical data, resting human neutrophils were immunolabeled for arginase I and analyzed by immunogold electron microscopy. Arginase I localized to a granular compartment, where it was present in the matrix (Figure 5A). To define further the subtype of granule in which arginase I is present, cryosections of resting human neutrophils were double-labeled with anti–arginase I and anti-MPO (for azurophil granules) or anti-Lf (for specific granules). With this technique, almost all arginase I+ granules were also MPO+ (Figure 5B), whereas only very few arginase I+ granules were positive for Lf (Figure 5B). We found that in human eosinophils arginase I is also a granular enzyme, which is present in crystalloid-containing and in crystalloid-free granules (Figure 5C). We conclude from these morphologic data and the biochemical subcellular fractionation results (Figure 4) that arginase I is present almost exclusively in azurophil granules of human PMNs as well as in human eosinophils.
These findings explain the lack of constitutive arginine degradation by human PMNs because the content of azurophil granules of resting PMNs is kept separate from the extracellular milieu to protect the cell. Indeed, a constant uptake and degradation of L-arginine by human PMNs would deplete the plasma of L-arginine, so that this important amino acid would be unavailable for other cell types. We therefore hypothesized that the function of PMN arginase probably differs from that of arginase-expressing murine macrophages.
Human PMN arginase: a novel fungicidal effector mechanism
The function of neutrophil azurophil granules, which contain a large number of proteolytic and microbicidal enzymes, is mainly to fuse with the phagosome during phagocytosis.29 Following fusion, the contents of the granules enter the phagosome and kill invading microorganisms. We therefore wondered whether arginase I might fulfill an antimicrobial function. If this were the case then the enzyme should localize to the phagosome during phagocytosis. We addressed this by performing immunogold electron microscopy on human PMNs after phagocytosis of latex beads. Arginase I localized to the phagolysosome of human PMNs after uptake of the beads (Figure 6A). It has recently been shown that the phagosome of human PMNs is depleted of arginine during phagocytosis.30 By demonstrating that arginase I is present in the azurophil granule fraction (Figures 4, 5) and in the phagosome during phagocytosis (Figure 6A) we provide a mechanistic explanation for phagosomal arginine depletion. To investigate the relevance of this in defense to infection, we first used an arginine-auxotroph strain of S cerevisiae during phagocytosis and fungicidal killing by human PMNs. In vitro, this mutant depends on the presence of arginine in the medium to grow and survive (data not shown). We hypothesized that it should provide a sensitive test of the effectiveness of arginine depletion in the phagosomal environment. We coincubated human PMNs with S cerevisiae and assessed the fungicidal activity of the phagocytes in the presence or absence of the potent arginase inhibitor nor-NOHA31 using 2 different methods to assess the viability of S cerevisiae after phagocytosis. Both assays demonstrated a small but significant reduction of neutrophil fungicidal activity when arginase activity is blocked during phagocytosis (mean reduction of kill by arginase inhibition: 11.3%, SE 0.07 by XTT assay and 13.9%, SE 0.29 by fluorescence microscopy, Figure 6B-C). PMNs possess multiple oxidative and nonoxidative effector pathways for microbial killing, which cooperate and also yield a certain degree of redundancy.32 To test the role of arginase further, we inhibited the oxidative fungicidal activity of human neutrophils in the presence or absence of arginase inhibition. MPO inhibition by NaN3 alone reduced the killing activity of human neutrophils (mean reduction of kill 9.2%, SE 0.08, and 19.4%, SE 0.09, as assessed by XTT assay and fluorescence microscopy, respectively; Figure 6B-C). But when arginase was also inhibited, the fungicidal activity is further impaired (mean reduction of kill by combined inhibition: 21.3%, SE 0.08 by XTT assay and 29.8%, SE 0.27 by fluorescence microscopy, Figure 6B-C), demonstrating the additive effect of inhibiting both neutrophil effector pathways. We then analyzed the significance and contribution of arginase to the fungicidal activity of PMNs for a wild-type, nonarginine auxotroph strain of C albicans. In contrast to the arginine-auxotroph strain of S cerevisiae we saw no impairment of PMN fungicidal activity when PMN arginase activity was blocked during phagocytosis (Figure 7A). As a control, we again blocked the formation of reactive oxygen species by inhibiting MPO with NaN3 or alternatively nicotinamide adenine dinucleotide phosphate (NADPH) oxidase with diphenylene iodonium (DPI) and significantly inhibited PMN fungicidal activity (Figure 7A). Like other members of the arginase family,1 human PMN arginase activity critically depends on the presence of manganese, most probably as part of the active center of the enzyme. Because human PMN lysates show only about 5% to 10% of arginase activity without addition of exogenous manganese during the enzymatic assay ("native arginase activity," data not shown), we concluded that manganese might be a limiting factor for the activity of the enzyme also in the cellular context. RPMI medium 1640 does not contain manganese by itself. In our phagocytosis experiments the concentration of manganese was therefore only about 10% to 20% of the physiologic situation corresponding to the fraction of human serum used for yeast opsonization. We therefore added various nontoxic concentrations of manganese during phagocytosis of C albicans and saw a pronounced induction of PMN arginase activity in 7 independent experiments (Figure 7B). More importantly, this induction of enzymatic activity always correlated with an increase in PMN fungicidal killing activity (Figure 7C) that could be reversed on blocking with the arginase inhibitor nor-NOHA (mean reduction of kill, 28.7%; Figure 7D). As demonstrated with the arginine-auxotroph strain of S cerevisiae (Figure 6B-C), the reduction of fungicidal kill on blocking of MPO with azide (mean reduction of kill, 29.2%) can be further augmented on additional arginase inhibition (mean reduction of kill, 47.4%), whereas no additive effect is observable when NADPH oxidase is blocked with DPI (without arginase inhibition, 45.5% kill reduction; with arginase inhibition, 47.3%; Figure 7D). All the substances that increased killing activity of PMNs (ie, MnCl2, nor-NOHA, NaN3, and DPI) did not impair microbial growth by themselves when tested at the respective concentrations.
In summary, we have demonstrated that phagosomal arginase I constitutes a novel antimicrobial effector mechanism in the phagolysosome of human PMNs probably by depleting arginine from the local environment.
Discussion
The cytokine-driven regulation of arginine metabolism via iNOS and arginase in the murine immune system is well established.2-7,33 In this study we have analyzed the expression, subcellular localization, and function of arginase in the human immune system. We have uncovered important differences between human and murine leukocyte arginase. In the murine system the enzyme is not expressed in unstimulated leukocytes but is inducible in all major myeloid cells by TH2 stimulation2-4 (Figure 3B). In human leukocytes this is not the case. We found that neither arginase activity nor protein is present in human monocytes, macrophages, or dendritic cells, either resting or activated in vitro by a variety of proinflammatory and anti-inflammatory stimuli. Human resting PMNs, on the other hand, show high constitutive arginase activity (mean, 1644 mU/mg protein), which is of the same order of magnitude as TH2-stimulated murine macrophages (Figure 1) or dendritic cells3,4 and human hepatocytes (Figure 1). This arginase expression is not further modulated by TH2 cytokines (Figure 3A).
Only a few reports have addressed the expression of arginase in cells of the human immune system. Early reports described a protein with arginase activity that was isolated from PMNs of a patient with chronic myeloid leukemia34 and secreted arginase was found in the supernatant of maturing malignant cells from 2 patients with acute myeloid leukemia.35 Arginase was induced in human mononuclear cells after injury19 or by stimulation with an immunomodulatory peptide.36 The enzyme was also detected in a subpopulation of inflammatory cells found in the bronchoalveolar lavage fluid of asthmatic patients18 or isolated from psoriatic lesions.20 Here we have analyzed highly purified human leukocyte subsets and demonstrated that leukocyte arginase activity is mainly, if not exclusively, localized in PMNs.
Whereas apoptosis of human PMNs degrades the various toxic granule constituents in a highly regulated way, this might not be the case under conditions of chronic inflammation or necrosis. We hypothesize that a dysregulated liberation of arginase into the local microenvironment depletes arginine and participates in local immunosuppression via T-cell hyporesponsiveness8 or in fibrosis via enhanced proline synthesis.11
Arginase I deficiency is a rare autosomal recessive genetic defect (incidence, 1/350 000) that leads to hyperargininemia, hyperammonemia, neurologic impairment, and progressive dementia due to the compromised hepatic urea cycle.1,37 The severity of symptoms depends mainly on the degree of enzymatic deficiency (reduction or total absence of arginase I) and the timely diagnosis with institution of dietary treatment. Two patients with arginase I deficiency were assessed and their PMNs showed no arginase activity or protein. This both confirms that arginase activity in human PMNs is only due to the isoenzyme arginase I and demonstrates that a compensatory increase of arginase II, as described for kidney tissue from hyperargininemic patients37,38 and in the arginase I–deficient mouse39 does not occur in human PMNs (Figure 2C).
What is the function of human PMN arginase Myeloid cells selectively deplete the phagosome of essential nutrients for microbial pathogens.40 On phagocytosis, microorganisms respond by inducing and repressing transcription of a variety of genes to adapt to the new environment. The transcriptional profile of the ingested microbe therefore reflects the microenvironment of the phagosome. It was recently demonstrated that S cerevisiae and C albicans up-regulate genes of their endogenous arginine biosynthetic pathways on phagocytosis by human neutrophils.30 This transcriptional response is most likely due to an arginine-deprived surrounding within the phagosome. Interestingly, on phagocytosis by human monocytes no up-regulation of yeast arginine biosynthetic genes was noted. The authors concluded from this discrepancy that, in contrast to human PMNs, the phagosomal environment of monocytes is not depleted of arginine. Our findings of a selective expression of arginase in human PMN phagosomes (and not in monocytes) offer a likely explanation for the arginine deprivation encountered by yeast on PMN phagocytosis.
Numerous reports have clarified that various oxidative and nonoxidative antimicrobial or antitumor effector pathways exist in PMNs.29,32,41 These pathways cooperate and might provide redundancy in PMN effector function. Depending on the specific microorganism, individual effector mechanisms are more important than others.32,41 In this study we have worked with a model organism that is exquisitely sensitive to arginine deprivation due to its inability to synthesize the amino acid itself. We used this strain of S cerevisiae as a functional bioprobe to prove the existence of arginase-mediated arginine deprivation within the human PMN phagosome even under suboptimal conditions of limited manganese availability and consecutive low arginase activity. More importantly, we show that arginase activity and fungicidal potential of human PMNs increase during phagocytosis of wild-type C albicans with the availability of manganese, an essential constituent of active arginase. When manganese is not limiting, arginase and MPO contribute comparably to the killing of C albicans (Figure 7D). Human arginase I, like all mammalian arginases, has a basic (pH 8.5-9) pH optimum.1 The vacuolar pH of human neutrophils initially becomes basic on phagocytosis before it gets acidic after 15 to 30 minutes of ingestion.41,42 The rise in pH from about 6.0 to 7.8/8.0 soon after phagocytosis depends on the activity of NADPH oxidase, because it is due to consumption of protons by the protonation of and . This alkalinization reaches an optimal level for the activity of granule proteases.41 The rise in pH might also be necessary for the enzymatic activity of arginase in the phagosome during the early phase of phagocytosis. We have seen pronounced inhibition of fungicidal activity by arginase inhibition especially during the first 30 to 90 minutes of phagocytosis, whereas later on the effect of arginase inhibition disappeared (data not shown). Similarly, when NADPH oxidase is blocked by DPI, arginase does not further add to the killing of C albicans (Figure 7A,D) and S cerevisiae (data not shown), likely because the necessary phagosome alkalinization is missing.
The antimicrobial effector function of intraphagosomal arginase needs to be evaluated with additional pathogenic microbes. Arginine deprivation might enhance the sensitivity of the parasite toward killing by other effector pathways. We further hypothesize that arginine might be necessary for the synthesis of antiphagocyte defense mechanisms in certain pathogens.
A review of published cases of patients with arginase I deficiency37 as well as the clinical history of the 2 patients analyzed in this study did not reveal an increased incidence of severe infectious problems. Furthermore, we were able to do one phagocytosis experiment with PMNs of a patient with arginase I deficiency and saw no impairment in fungicidal activity of arginase I–deficient PMNs (data not shown). Obviously, human PMNs are able to compensate for lack of the enzyme due to the redundancy of multiple antimicrobial pathways. The immunologic competence of patients with arginase I deficiency is reminiscent of patients with MPO deficiency of PMNs. Although this genetic defect is associated with impairment of in vitro PMN fungicidal activity, it is usually clinically silent.
In summary, we have clarified the cellular distribution, subcellular localization, regulatory aspects, and function of arginase in human leukocytes. We have demonstrated for the first time that only human PMNs constitutively express arginase I, that the enzyme is localized in the azurophil granules, and that it works as a novel fungicidal effector mechanism of human PMNs.
Acknowledgements
We thank J. Mytilineos and V. Daniel (both Institute of Immunology, University of Heidelberg) for providing blood of healthy donors, V. Weber (Max-Planck-Institute for Immunobiology, Freiburg, Germany), Hans Janssen and Nico Ong (both The Netherlands Cancer Institute, Amsterdam) for expert technical assistance and M. Scheuermann (German Cancer Research Center, Heidelberg) for FACS sorting. We appreciate the help of M. Schwarz, U. Wendel (both University Hospital Düsseldorf) and F.-K. Trefz (Children`s Hospital Reutlingen) for help with recruitment of arginase I-deficient patients. We thank L. B. Nicholson (University of Bristol) for critical review of the manuscript.
Footnotes
Prepublished online as Blood First Edition Paper, November 16, 2004; DOI 10.1182/blood-2004-07-2521.
Supported by grants FIS-02/1199 and FIS-01/1048 from the Fondo de Investigacion Sanitaria (F.M.) and by grants 2PR01A079 (G.S.), 2PR02B034, and 03/13 (J.M.F.) from Junta de Extremadura.
The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked "advertisement" in accordance with 18 U.S.C. section 1734.
References
Jenkinson CP, Grody WW, Cederbaum SD. Comparative properties of arginases. Comp Biochem Physiol B Biochem Mol Biol. 1996;114: 107-132.
Modolell M, Corraliza IM, Link F, Soler G, Eichmann K. Reciprocal regulation of the nitric oxide synthase/arginase balance in mouse bone marrow-derived macrophages by TH1 and TH2 cytokines. Eur J Immunol. 1995;25: 1101-1104.
Munder M, Eichmann K, Modolell M. Alternative metabolic states in murine macrophages reflected by the nitric oxide synthase/arginase balance: competitive regulation by CD4+ T cells correlates with Th1/Th2 phenotype. J Immunol. 1998;160: 5347-5354.
Munder M, Eichmann K, Moran JM, Centeno F, Soler G, Modolell M. Th1/Th2-regulated expression of arginase isoforms in murine macrophages and dendritic cells. J Immunol. 1999;163: 3771-3777.
Louis CA, Mody V, Henry WL, Reichner JS, Albina JE. Regulation of arginase isoforms I and II by IL-4 in cultured murine peritoneal macrophages. Am J Physiol 1999;276: R237-R242.
Chang CI, Zoghi B, Liao JC, Kuo L. The involvement of tyrosine kinases, cyclic AMP/protein kinase A, and p38 mitogen-activated protein kinase in IL-13-mediated arginase I induction in macrophages: its implications in IL-13-inhibited nitric oxide production. J Immunol. 2000;165: 2134-2141.
Rutschman R, Lang R, Hesse M, Ihle JN, Wynn TA, Murray PJ. Stat6-dependent substrate depletion regulates nitric oxide production. J Immunol. 2001;166: 2173-2177.
Rodriguez PC, Zea AH, DeSalvo J, et al. L-arginine consumption by macrophages modulates the expression of CD3 zeta chain in T lymphocytes. J Immunol. 2003;171: 1232-1239.
Bronte V, Serafini P, De Santo C, et al. IL-4-induced arginase 1 suppresses alloreactive T cells in tumor-bearing mice. J Immunol. 2003;170: 270-278.
Kusmartsev S, Nefedova Y, Yoder D, Gabrilovich DI. Antigen-specific inhibition of CD8+ T cell response by immature myeloid cells in cancer is mediated by reactive oxygen species. J Immunol. 2004;172: 989-999.
Hesse M, Modolell M, La Flamme AC, et al. Differential regulation of nitric oxide synthase-2 and arginase-1 by type 1/type 2 cytokines in vivo: granulomatous pathology is shaped by the pattern of L-arginine metabolism. J Immunol. 2001;167: 6533-6544.
Granger DL, Hibbs JB, Perfect JR, Durack DT. Metabolic fate of L-arginine in relation to microbiostatic capability of murine macrophages. J Clin Invest. 1990;85: 264-273.
Iniesta V, Gomez-Nieto LC, Corraliza I. The inhibition of arginase by N(omega)-hydroxy-l-arginine controls the growth of Leishmania inside macrophages. J Exp Med. 2001;193: 777-784.
Gobert AP, Daulouede S, Lepoivre M, et al. L-Arginine availability modulates local nitric oxide production and parasite killing in experimental trypanosomiasis. Infect Immun. 2000;68: 4653-4657.
Xu L, Hilliard B, Carmody RJ, et al. Arginase and autoimmune inflammation in the central nervous system. Immunology. 2003;110: 141-148.
Waddington SN, Tam FW, Cook HT, Cattell V. Arginase activity is modulated by IL-4 and HOArg in nephritic glomeruli and mesangial cells. Am J Physiol. 1998;274: F473-F480.
Mistry SK, Zheng M, Rouse BT, Morris SM. Induction of arginases I and II in cornea during herpes simplex virus infection. Virus Res. 2001;73: 177-182.
Zimmermann N, King NE, Laporte J, et al. Dissection of experimental asthma with DNA microarray analysis identifies arginase in asthma pathogenesis. J Clin Invest. 2003;111: 1863-1874.
Ochoa JB, Bernard AC, O'Brien WE, et al. Arginase I expression and activity in human mononuclear cells after injury. Ann Surg. 2001;233: 393-399.
Bruch-Gerharz D, Schnorr O, Suschek C, et al. Arginase 1 overexpression in psoriasis: limitation of inducible nitric oxide synthase activity as a molecular mechanism for keratinocyte hyperproliferation. Am J Pathol. 2003;162: 203-211.
Escribano S, Cuenllas E, Gaitan S, Tejero C. Delayed neutrophil apoptosis after total body irradiation in mice. The role of granulocyte-macrophage colony-stimulating factor in neutrophil function. Exp Hematol. 1998;26: 942-949.
Meshulam T, Levitz SM, Christin L, Diamond RD. A simplified new assay for assessment of fungal cell damage with the tetrazolium dye, (2,3)-bis-(2-methoxy-4-nitro-5-sulphenyl)-(2H)-tetrazolium-5-carboxanilide (XTT). J Infect Dis. 1995;172: 1153-1156.
Diez A, Fuentes JM, Prada F, Campo ML, Soler G. Immunological identity of the two different molecular mass constitutive subunits of liver arginase. Biol Chem Hoppe Seyler. 1994;375: 537-541.
Martin-Martin B, Nabokina SM, Blasi J, Lazo PA, Mollinedo F. Involvement of SNAP-23 and syntaxin 6 in human neutrophil exocytosis. Blood. 2000;96: 2574-2583.
Mollinedo F, Martin-Martin B, Calafat J, Nabokina SM, Lazo PA. Role of vesicle-associated membrane protein-2, through Q-soluble N-ethylmaleimide-sensitive factor attachment protein receptor/R-soluble N-ethylmaleimide-sensitive factor attachment protein receptor interaction, in the exocytosis of specific and tertiary granules of human neutrophils. J Immunol. 2003;170: 1034-1042.
Sorensen OE, Follin P, Johnsen AH, et al. Human cathelicidin, hCAP-18, is processed to the antimicrobial peptide LL-37 by extracellular cleavage with proteinase 3. Blood. 2001;97: 3951-3959.
Calafat J, Janssen H, Stahle-Bckdahl M, Zuurbier AEM, Knol EF, Egesten A. Human monocytes and neutrophils store transforming growth factor-alpha in a subpopulation of cytoplasmic granules. Blood. 1997;90: 1255-1266.
Jost MM, Ninci E, Meder B, et al. Divergent effects of GM-CSF and TGFbeta1 on bone marrow-derived macrophage arginase-1 activity, MCP-1 expression, and matrix metalloproteinase-12: a potential role during arteriogenesis. FASEB J. 2003;17: 2281-2283.
Borregaard N, Cowland JB. Granules of the human neutrophilic polymorphonuclear leukocyte. Blood. 1997;89: 3503-3521.
Rubin-Bejerano I, Fraser I, Grisafi P, Fink GR. Phagocytosis by neutrophils induces an amino acid deprivation response in Saccharomyces cerevisiae and Candida albicans. Proc Natl Acad Sci U S A. 2003;100: 11007-11012.
Cox JD, Cama E, Colleluori DM, et al. Mechanistic and metabolic inferences from the binding of substrate analogues and products to arginase. Biochemistry. 2001;40: 2689-2701.
Gil-Lamaignere C, Roilides E, Lyman CA, et al. Human phagocytic cell responses to Scedosporium apiospermum (Pseudallescheria boydii): variable susceptibility to oxidative injury. Infect Immun. 2003;71: 6472-6478.
Bogdan C. Nitric oxide and the immune response. Nat Immunol. 2001;2: 907-916.
Reyero C, Dorner F. Purification of arginases from human-leukemic lymphocytes and granulocytes: study of their physicochemical and kinetic properties. Eur J Biochem. 1975;56: 137-147.
Palu G, Powles R, Selby P, Summersgill BM, Alexander P. Patterns of maturation in short-term culture of human acute myeloid leukaemic cells. Br J Cancer 1979;40: 719-730.
Rouzaut A, Subira ML, de Miguel C, et al. Co-expression of inducible nitric oxide synthase and arginases in different human monocyte subsets. Apoptosis regulated by endogenous NO. Biochim Biophys Acta 1999;1451: 319-333.
Iyer R, Jenkinson CP, Vockley JG, Kern RM, Grody WW, Cederbaum S. The human arginases and arginase deficiency. J Inherit Metab Dis. 1998;21(suppl 1): 86-100.
Grody WW, Argyle C, Kern RM, et al. Differential expression of the two human arginase genes in hyperargininemia. Enzymatic, pathologic, and molecular analysis. J Clin Invest. 1989;83: 602-609.
Iyer RK, Yoo PK, Kern RM, et al. Mouse model for human arginase deficiency. Mol Cell Biol. 2002;22: 4491-4498.
Jabado N, Jankowski A, Dougaparsad S, Picard V, Grinstein S, Gros P. Natural resistance to intracellular infections: natural resistance-associated macrophage protein 1 (Nramp1) functions as a pH-dependent manganese transporter at the phagosomal membrane. J Exp Med. 2000;192: 1237-1248.
Reeves EP, Lu H, Lortat Jacobs H, et al. Killing activity of neutrophils is mediated through activation of proteases by K+ flux. Nature. 2002;416: 291-297.
Segal AW, Geisow M, Garcia R, Harper A, Miller R. The respiratory burst of phagocytic cells is associated with a rise in vacuolar pH. Nature. 1981;290: 406-409.(Markus Munder, Faustino M)
Clinical Cooperation Unit Molecular Hematology and Oncology, German Cancer Research Center, Heidelberg, Germany
Institute of Immunology, University of Heidelberg, Heidelberg, Germany
Centro de Investigacion del Cancer, Instituto de Biología Molecular y Celular del Cancer, CSIC-Universidad de Salamanca, Salamanca, Spain
Division of Cell Biology, Netherlands Cancer Institute, Amsterdam, The Netherlands
Section of Pediatric Pulmonology and Infectious Diseases, IIIrd Department of Pediatrics, Children`s Hospital, University of Heidelberg, Heidelberg, Germany
Departamento de Bioquímica y Biología Molecular, E.U. Enfermería y T.O., Universidad de Extremadura, Caceres, Spain
Departamento de Bioquímica y Biología Molecular, Facultad de Veterinaria, Universidad de Extremadura, Caceres, Spain
Department of Cellular Immunology, Max-Planck Institute for Immunobiology, Freiburg, Germany.
Abstract
The balance of arginine metabolism via nitric oxide synthase (NOS) or arginase is an important determinant of the inflammatory response of murine macrophages and dendritic cells. Here we analyzed the expression of the isoform arginase I in human myeloid cells. Using healthy donors and patients with arginase I deficiency, we found that in human leukocytes arginase I is constitutively expressed only in granulocytes and is not modulated by a variety of proinflammatory and anti-inflammatory stimuli in vitro. We demonstrate that arginase I is localized in azurophil granules of neutrophils and constitutes a novel antimicrobial effector pathway, likely through arginine depletion in the phagolysosome. Our findings demonstrate important differences between murine and human leukocytes with respect to regulation and function of arginine metabolism via arginase.
Introduction
In the mouse, the metabolism of arginine is essential for many immunologic functions. In leukocytes, arginine can be metabolized by inducible nitric oxide synthase (iNOS) to the cytotoxic, tumoricidal, and antimicrobial effector molecule nitric oxide (NO). Myeloid cells also hydrolyze arginine to urea and ornithine with the enzyme arginase, and this leads to the synthesis of polyamines and proline. Two arginase isoenzymes exist (arginase I and II), which differ in subcellular localization, regulation, and possibly function.1 Arginase I is a cytosolic enzyme, which is expressed mainly in the liver as part of the urea cycle, whereas arginase II is a mitochondrial protein found in a variety of tissues. In the murine immune system, arginase and iNOS undergo reciprocal induction in several cell types by a variety of inflammatory and anti-inflammatory agonists. In murine macrophages2-7 and dendritic cells4 the polarization of arginine metabolism is driven by cytokines. T-helper (TH) 1 cytokines induce iNOS, whereas TH2 cytokines up-regulate arginase I.
Several important immunologic functions of arginase have been clarified recently, including a direct and indirect role in inducing T-cell hyporesponsiveness8-10 and the induction of TH2-mediated immunopathology in murine schistosomiasis.11 The TH2-induced arginase I of murine macrophages supports the growth of various microorganisms,12-14 most likely by competing with iNOS for arginine and by providing polyamines, and arginase is also found in murine inflammatory cell infiltrates in experimental autoimmune encephalomyelitis,15 glomerulonephritis,16 herpes simplex virus infection of the eye,17 and asthma.18
In contrast to the wealth of knowledge regarding arginine metabolism in the murine immune system, very little is known about a possible role of arginase in human immunology. Arginase was detected in human mononuclear cells after injury19 and found in inflammatory cells of bronchoalveolar lavage fluid of asthmatic patients18 or in psoriatic lesions.20 We have therefore analyzed the role of arginase in the human immune system. Surprisingly, the expression and function of arginase is strikingly different in human leukocytes when compared with murine leukocytes. Although arginase remains a crucial player in host defense, we demonstrate that within the human immune system arginase I is selectively expressed in granulocytes (polymorphonuclear leukocytes [PMNs]), quantitatively the largest subpopulation of myeloid cells. We show that in human PMNs the enzyme is not cytosolic but localizes to the azurophil granules. Finally, we provide evidence that arginase I is a novel antimicrobial effector mechanism of human PMNs, likely operating through arginine depletion within the phagolysosome.
Materials and methods
Blood donors and mice
Experiments were performed in compliance with the relevant laws and institutional guidelines, and human studies were approved by the ethics committee of the University of Heidelberg. Informed consent was provided according to the Declaration of Helsinki. Mice of strain C57BL/6 were obtained from the specific pathogen-free animal facilities of the Max-Planck Institute for Immunobiology and were used between 6 and 8 weeks of age.
Reagents and cells
If not otherwise stated, chemicals were purchased from Sigma (St Louis, MO). N--hydroxy-nor-L-arginine (nor-NOHA) was from Bachem (Bubendorf, Switzerland). Recombinant murine cytokines interleukin 10 (IL-10) and granulocyte colony-stimulating factor (G-CSF) were from Pepro-Tech (London, United Kingdom), and IL-4 was from R&D Systems (Minneapolis, MN). All recombinant human cytokines were from PromoCell (Heidelberg, Germany). The L-arginine auxotroph strain of Saccharomyces cerevisiae is an isogenic derivative of the laboratory strain FL100 with a deficiency in ornithine transcarbamoylase (OTC, arg3), generously provided by Jacky de Montigny and Ives Lombard (both Universite Louis Pasteur, Strasbourg, France). In some experiments wild-type Candida albicans strain ATCC 90028 was used.
Generation of murine BMDMs and PMNs
Murine bone marrow-derived macrophages (BMDMs) were generated as previously described.3,4 Murine PMNs were generated from bone marrow21 with slight modifications. Briefly, bone marrow cells (1 x 106/mL) were cultured in RPMI 1640 medium (20% horse serum, 5 ng/mL G-CSF, 5 nM hydroxycortisone, 2 mM L-glutamine, 60 μM 2-mercaptoethanol, 1 mM sodium pyruvate, 1 x nonessential amino acids, 100 U/mL penicillin, and 100 μg/mL streptomycin). The cells were fed on day 4 with fresh medium (without G-CSF and hydroxycortisone) and harvested on day 7. After 1 hour of plastic adherence, the nonadherent fraction was harvested. This procedure led to a PMN purity of 90% to 95%, assessed by microscopy and flow cytometry analysis (GR1hi cells with characteristic forward/side-scatter profile).
Generation of human macrophages and isolation of human PBMCs and PMNs
CD14+ monocytes were purified from peripheral blood of healthy human donors with the magnetically activated cell sorting (MACS) system (Miltenyi Biotec, Bergisch-Gladbach, Germany) and cultured for 10 days with M-CSF (5 ng/mL) in hydrophobic Teflon bags3,4 to yield macrophages. To isolate human peripheral blood mononuclear cells (PBMCs) and PMNs, EDTA (ethylenediaminetetraacetic acid)–anticoagulated peripheral blood of healthy human donors was layered on top of Ficoll. After 20 minutes of centrifugation (700g), the PBMCs (= interphase) were harvested. The pellet, containing erythrocytes and PMNs, was resuspended in Hanks balanced salt solution (HBSS, without Ca2+ and Mg2+) and mixed at a ratio of 1:1 with 3% dextran. After sedimentation of erythrocytes (20 minutes) the PMN-rich supernatant was harvested and the remaining erythrocytes were subjected to hypotonic lysis (155 mM NH4Cl, 10 mM KHCO3, 0.1 mM EDTA, pH 7.4) for 15 minutes on ice. After washing, cell purity and viability were checked by flow cytometry and microscopy. The method reproducibly yields PMNs with a purity and viability of more than 95%. When indicated, PMNs were further purified by flow cytometry on a Vantage Flow Cytometer (BD Biosciences, Heidelberg, Germany) according to forward and side scatter, yielding purities greater than 99.5%. PBMCs were separated in CD14+ (monocyte fraction, purity 95%) and CD14- cells (lymphocyte fraction, purity 90%) with the MACS system.
Measurement of PMN fungicidal activity
PMNs were preincubated with inhibitors for 30 minutes at 37°C. Yeast and PMNs (5 x 106 each) were cultured in 1 mL HBSS (+ 10% human serum) for 60 minutes. Viability of yeast was evaluated with the XTT ((2,3)-bis-(2-methoxy-4-nitro-5-sulphenyl)-(2H)-tetrazolium-5-carboxanilide) assay exactly as described22 and with the FUN-1 yeast viability kit (Molecular Probes, Gttingen, Germany). In the latter case, PMNs were lysed in ice-cold water and in case of S cerevisiae the cell pellet was resuspended for 10 minutes at 34°C in 50 μL DNAse I (1 U/μL, Invitrogen, Carlsbad, CA) and streptolysin O (2 mg/mL) for permeabilization. Then, 50 μL FUN-1 cell stain (0.125 mM) plus Calcofluor White M2R (0.05 mM) in HBSS/4% glucose were added. After 30 minutes at 30°C in the dark, at least 200 yeast cells per sample were analyzed by fluorescence microscopy. The percentage of fungal cell damage ("% kill") was defined by the following equation: 100 - (100 x [OD405 of fungi with PMN]/[OD405 of fungi]). The percentage of reduction of fungal cell damage ("% kill inhibition") on treatment with inhibitors was defined by the following equation: 100 - 100 x [% Kill of fungi with PMN with inhibitor]/[% Kill of fungi with PMN]).
Metabolism of 14C-L-arginine and TLC
After washing, 2 x 105 cells were incubated for 2 hours at 37°C with arginine-free Dulbecco modified Eagle medium containing 2% fetal calf serum (FCS) and 0.1 μCi (0.037 MBq) L-(U-14C) arginine (Amersham Biosciences, Freiburg, Germany). Cells were subsequently lysed by 2 freeze-thaw cycles. Then, 20 μL of the lysate was spotted onto thin-layer chromatography (TLC) plates (Merck, Darmstadt, Germany), dried for 1 hour at 42°C, and developed in the solvent system chloroform/methanol/ammonium hydroxide/water 0.5/4.5/2.0/1.0 (vol/vol). Spots were developed with Ninhydrin by heating at 120°C for 5 minutes and analyzed by scintillation counting.
Arginase enzymatic assay
Arginase activity was measured in cell lysates as previously described4 with slight modifications. Briefly, cells were lysed with 0.5% Triton X-100, 25 mM Tris (tris(hydroxymethyl)aminomethane)–HCl, pH 7.5, containing 1 mM phenylmethylsulfonyl fluoride (PMSF) and 2 μg/mL each of aprotinin, leupeptin, and pepstatin (lysis buffer). To 100 μL of this lysate 20 μLof 10 mM MnCl2 was added, and the enzyme was activated by heating for 10 minutes at 56°C. In case of determination of native arginase activity, these steps (addition of exogenous manganese and heating) were omitted. Arginine hydrolysis and measurement of urea concentration were performed exactly as previously described.4 One unit of enzyme activity is defined as the amount of enzyme that catalyzes the formation of 1 μmol urea/min.
SDS-PAGE and Western blot analysis
Cells were lysed for 30 minutes on ice in lysis buffer (see "Arginase enzymatic assay"). Cell debris was spun down at 18 000g for 5 minutes at 4°C. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was done as previously described.4 The proteins were transferred to a Hybond-P polyvinylidene difluoride (PVDF) membrane (Amersham Biosciences). After blocking with 5% nonfat dry milk in TBST-buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 0.05% Tween 20) for 2 hours, the membranes were incubated with the following primary antibodies (1:5000 in TBST, 5% bovine serum albumin): polyclonal rabbit antirat arginase I antiserum,23 which is cross-reactive to mouse and human arginase I and anti-ERK1/2 (extracellular signal regulated kinase, Cell Signaling Technology, Beverly, MA). Antibody reactivity was monitored with horseradish peroxidase (HRP)–conjugated antirabbit IgG (Santa Cruz Biotechnology, Santa Cruz, CA), followed by visualization with the enhanced chemiluminescence (ECL) detection system (Amersham Biosciences).
Subcellular fractionation of neutrophils
Subcellular fractionation of neutrophils was performed as previously described.24 Briefly, resting neutrophils isolated from freshly heparinized human peripheral blood were resuspended in 50 mM Tris-HCl, pH 7.5 containing 2 mM PMSF, and then disrupted by repeated freeze-thaw. Homogenates were centrifuged at 1200 rpm in a Sorvall T 6000D centrifuge (Kendro, Hertfordshire, United Kingdom) for 10 minutes, and the supernatant, representing the postnuclear extract, was saved. After centrifugation of the postnuclear extract at 110 000g in a TLA rotor for 90 minutes at 4°C using an Optima TL Ultracentrifuge (Beckman Instruments, Palo Alto, CA), supernatant (soluble fraction) and pellet (membrane fraction, resuspended in 50 mM Tris-HCl, pH 7.5, containing 2 mM PMSF) were saved. To prepare the distinct subcellular fractions, freshly prepared neutrophils (about 3-5 x 108) were gently disrupted, and the postnuclear fractions were fractionated in 15% to 40% continuous sucrose gradients as described previously.25 Subcellular fractions were analyzed for marker proteins for each organelle as described.25
Immunoelectron microscopy
Resting human neutrophils and exudate neutrophils from skin window chambers after phagocytosis of latex beads26 were fixed for 24 hours in 4% paraformaldehyde in PHEM buffer (60 mM PIPES [piperazine diethanesulfonic acid], 25 mM HEPES [N-2-hydroxyethylpiperazine-N'2-ethanesulfonic acid], 2 mM MgCl2, 10 mM EGTA [ethylene glycol tetraacetic acid], pH 6.9) and then processed for ultrathin cryosectioning as previously described.27 Ultra-thin frozen sections were incubated at room temperature with the indicated antibodies and 10 or 15 nm protein A gold,27 embedded in a mixture of methylcellulose and uranyl acetate, and examined with a Philips CM 10 electron microscope (Eindhoven, The Netherlands; original magnification x 12 500). For controls, the primary antibody was replaced by an irrelevant rabbit antibody. Antibodies used in double immunolabeling were rabbit antihuman lactoferrin (Lf; Cappel Laboratories, Durham, NC) and rabbit antihuman myeloperoxidase (MPO) (Dako, Glostrup, Denmark).
Results
Arginase in human myeloid cells: specific expression in PMNs
We analyzed the expression of arginase in various human myeloid cell types treated or not with several agonists. We stimulated the cell populations with the recombinant human cytokines IL-1, IL-1, interferon (IFN-), tumor necrosis factor (TNF-), IL-4, IL-10, and IL-13 as well as lipopolysaccharide (LPS), prostaglandin E2 (PGE2), dexamethasone, and the cyclic adenosine monophosphate (cAMP)–enhancing agents dibutyryl-cAMP and forskolin, alone and in various combinations. Human monocyte-derived macrophages and dendritic cells showed no arginase activity in the resting state. In contrast to the murine system, we were also unable to induce arginase activity or protein in human macrophages (Figure 1), dendritic cells, and monocytes (data not shown) with TH2 cytokines over a wide range of individual concentrations (0.01-100 ng/mL for each cytokine) or any of the agonists. We also stimulated various immortalized human myeloid cell lines (U937, HL-60, THP-1, with or without induction of differentiation by phorbol myristate acetate or retinoic acid) in the same way with the same negative results (data not shown). We then turned to leukocytes, which we derived directly from the blood of healthy donors. In the PMNs of all the blood donors (n = 31) we consistently found high arginase activity (mean, 1644 ± 423 mU/mg protein; range, 667-2189 mU/mg protein; Figure 2A). In contrast, in 19 of 25 PBMC fractions (involving lymphocytes and monocytes) no arginase activity was detectable, whereas in the remaining 6 PBMC samples we found low arginase activity (mean, 111 ± 46 mU/mg protein; range, 35-156 mU/mg protein; Figure 2A). To rule out any contamination of the PBMC fraction with PMNs or vice versa, we corroborated our findings by fluorescence-activated cell sorting (FACS) that sorted a pure PMN population according to forward/side-scatter pattern from the conventional PMN fraction (purity of FACS sorted PMNs > 99.5%). Alternatively, we eliminated contaminating PMNs from the PBMC fraction by FACS. In 3 independent experiments we only demonstrated arginase activity and protein in the PMN fraction, whereas PBMCs, purified monocytes (CD14+, purity 95%), or purified lymphocytes (CD14-, purity 90%) completely lacked enzymatic activity and protein (Figure 2B). We also analyzed PMNs from 2 patients with complete arginase I deficiency (OMIM 207800 [OMIM] ; patient 1, 18-year-old man; patient 2, 5-year-old girl). The PMNs of these patients had neither arginase activity nor expression of arginase I protein (Figure 2C). Therefore, we concluded that resting human PMNs express solely the hepatic isoform (arginase I) and have no compensatory up-regulation of arginase II even in the face of complete arginase I deficiency.
Because arginase I is a TH2-inducible protein in murine macrophages and dendritic cells,2-7 we wondered if human PMN arginase would be further augmented by TH2 cytokine stimulation. We treated human PMNs with the TH2 cytokine IL-4 or a combination of IL-4 and IL-10, treatments that yield maximal arginase activity in murine myeloid cells3 (Figure 1). In 6 separate experiments (4 with conventionally purified PMNs and 2 with FACS-sorted PMNs) we found no significant change in human PMN arginase activity, protein expression (Figure 3A), or arginase I mRNA levels (data not shown) on TH2 cytokine stimulation. Finally, we analyzed the expression of arginase in murine PMNs. Unlike their human counterparts, unstimulated murine PMNs show no arginase protein or activity (Figure 3B). The TH2 cytokine IL-4 either alone or in cooperation with IL-10, induces the hepatic isoform arginase I and considerable arginase activity (mean of 3 experiments, 1153 mU/mg protein; range, 1162-1290 mU/mg protein) in murine PMNs (Figure 3B). These cells therefore follow the same pattern of arginase I regulation as murine macrophages and dendritic cells. We conclude that murine and human PMNs differ fundamentally in terms of basal expression and cytokine-mediated regulation of arginase I.
PMNs do not metabolize L-arginine despite high arginase activity
On TH2 stimulation, murine macrophages up-regulate arginase I and consume arginine present in the local microenvironment.1,11 Because human PMNs express high arginase activity constitutively, we wondered if these cells were constantly hydrolyzing arginine after uptake from the extracellular milieu. We therefore incubated resting human PMNs with radioactively labeled 14C-L-arginine and analyzed the conversion of this substrate into ornithine and spermine by TLC. As a control we used resting and IL-4–stimulated murine BMDMs. Whereas resting BMDMs showed no arginase activity and no significant hydrolysis of arginine, the TH2-stimulated, arginase-expressing murine BMDMs metabolized arginine and produced ornithine and spermine. Surprisingly, human PMNs were very inefficient in degrading arginine (Table 1). This led us to question whether, in contrast to the cytosolic location of arginase within murine macrophages,1,28 the enzyme might be sequestered in a different compartment in human PMNs with impaired arginine accessibility.
Arginase I is present in azurophil granules of human PMNs
To determine the localization of the enzyme within human PMNs, we performed subcellular fractionation experiments. First, we separated resting human PMNs in a soluble and a membrane fraction and found that arginase I is not a membrane-bound protein in resting human PMNs (Figure 4A). Subsequent subcellular fractionation assays (Figure 4B), which resolved cytosol, plasma membrane, tertiary granules, specific granules, and azurophil granules,24,25 showed that arginase I protein is mainly present in fraction 8 (Figure 4C), the MPO+ azurophil granules of resting human PMNs. Arginase activity assays with aliquots of the various PMN fractions confirmed these results (Figure 4C). Because arginase is enriched in this subcellular compartment, we observed an even higher enzyme activity (mean of 3 experiments: 3650 ± 97 mU/mg protein in fraction 8) compared to whole-cell PMN lysates (Figure 2A).
To confirm the biochemical data, resting human neutrophils were immunolabeled for arginase I and analyzed by immunogold electron microscopy. Arginase I localized to a granular compartment, where it was present in the matrix (Figure 5A). To define further the subtype of granule in which arginase I is present, cryosections of resting human neutrophils were double-labeled with anti–arginase I and anti-MPO (for azurophil granules) or anti-Lf (for specific granules). With this technique, almost all arginase I+ granules were also MPO+ (Figure 5B), whereas only very few arginase I+ granules were positive for Lf (Figure 5B). We found that in human eosinophils arginase I is also a granular enzyme, which is present in crystalloid-containing and in crystalloid-free granules (Figure 5C). We conclude from these morphologic data and the biochemical subcellular fractionation results (Figure 4) that arginase I is present almost exclusively in azurophil granules of human PMNs as well as in human eosinophils.
These findings explain the lack of constitutive arginine degradation by human PMNs because the content of azurophil granules of resting PMNs is kept separate from the extracellular milieu to protect the cell. Indeed, a constant uptake and degradation of L-arginine by human PMNs would deplete the plasma of L-arginine, so that this important amino acid would be unavailable for other cell types. We therefore hypothesized that the function of PMN arginase probably differs from that of arginase-expressing murine macrophages.
Human PMN arginase: a novel fungicidal effector mechanism
The function of neutrophil azurophil granules, which contain a large number of proteolytic and microbicidal enzymes, is mainly to fuse with the phagosome during phagocytosis.29 Following fusion, the contents of the granules enter the phagosome and kill invading microorganisms. We therefore wondered whether arginase I might fulfill an antimicrobial function. If this were the case then the enzyme should localize to the phagosome during phagocytosis. We addressed this by performing immunogold electron microscopy on human PMNs after phagocytosis of latex beads. Arginase I localized to the phagolysosome of human PMNs after uptake of the beads (Figure 6A). It has recently been shown that the phagosome of human PMNs is depleted of arginine during phagocytosis.30 By demonstrating that arginase I is present in the azurophil granule fraction (Figures 4, 5) and in the phagosome during phagocytosis (Figure 6A) we provide a mechanistic explanation for phagosomal arginine depletion. To investigate the relevance of this in defense to infection, we first used an arginine-auxotroph strain of S cerevisiae during phagocytosis and fungicidal killing by human PMNs. In vitro, this mutant depends on the presence of arginine in the medium to grow and survive (data not shown). We hypothesized that it should provide a sensitive test of the effectiveness of arginine depletion in the phagosomal environment. We coincubated human PMNs with S cerevisiae and assessed the fungicidal activity of the phagocytes in the presence or absence of the potent arginase inhibitor nor-NOHA31 using 2 different methods to assess the viability of S cerevisiae after phagocytosis. Both assays demonstrated a small but significant reduction of neutrophil fungicidal activity when arginase activity is blocked during phagocytosis (mean reduction of kill by arginase inhibition: 11.3%, SE 0.07 by XTT assay and 13.9%, SE 0.29 by fluorescence microscopy, Figure 6B-C). PMNs possess multiple oxidative and nonoxidative effector pathways for microbial killing, which cooperate and also yield a certain degree of redundancy.32 To test the role of arginase further, we inhibited the oxidative fungicidal activity of human neutrophils in the presence or absence of arginase inhibition. MPO inhibition by NaN3 alone reduced the killing activity of human neutrophils (mean reduction of kill 9.2%, SE 0.08, and 19.4%, SE 0.09, as assessed by XTT assay and fluorescence microscopy, respectively; Figure 6B-C). But when arginase was also inhibited, the fungicidal activity is further impaired (mean reduction of kill by combined inhibition: 21.3%, SE 0.08 by XTT assay and 29.8%, SE 0.27 by fluorescence microscopy, Figure 6B-C), demonstrating the additive effect of inhibiting both neutrophil effector pathways. We then analyzed the significance and contribution of arginase to the fungicidal activity of PMNs for a wild-type, nonarginine auxotroph strain of C albicans. In contrast to the arginine-auxotroph strain of S cerevisiae we saw no impairment of PMN fungicidal activity when PMN arginase activity was blocked during phagocytosis (Figure 7A). As a control, we again blocked the formation of reactive oxygen species by inhibiting MPO with NaN3 or alternatively nicotinamide adenine dinucleotide phosphate (NADPH) oxidase with diphenylene iodonium (DPI) and significantly inhibited PMN fungicidal activity (Figure 7A). Like other members of the arginase family,1 human PMN arginase activity critically depends on the presence of manganese, most probably as part of the active center of the enzyme. Because human PMN lysates show only about 5% to 10% of arginase activity without addition of exogenous manganese during the enzymatic assay ("native arginase activity," data not shown), we concluded that manganese might be a limiting factor for the activity of the enzyme also in the cellular context. RPMI medium 1640 does not contain manganese by itself. In our phagocytosis experiments the concentration of manganese was therefore only about 10% to 20% of the physiologic situation corresponding to the fraction of human serum used for yeast opsonization. We therefore added various nontoxic concentrations of manganese during phagocytosis of C albicans and saw a pronounced induction of PMN arginase activity in 7 independent experiments (Figure 7B). More importantly, this induction of enzymatic activity always correlated with an increase in PMN fungicidal killing activity (Figure 7C) that could be reversed on blocking with the arginase inhibitor nor-NOHA (mean reduction of kill, 28.7%; Figure 7D). As demonstrated with the arginine-auxotroph strain of S cerevisiae (Figure 6B-C), the reduction of fungicidal kill on blocking of MPO with azide (mean reduction of kill, 29.2%) can be further augmented on additional arginase inhibition (mean reduction of kill, 47.4%), whereas no additive effect is observable when NADPH oxidase is blocked with DPI (without arginase inhibition, 45.5% kill reduction; with arginase inhibition, 47.3%; Figure 7D). All the substances that increased killing activity of PMNs (ie, MnCl2, nor-NOHA, NaN3, and DPI) did not impair microbial growth by themselves when tested at the respective concentrations.
In summary, we have demonstrated that phagosomal arginase I constitutes a novel antimicrobial effector mechanism in the phagolysosome of human PMNs probably by depleting arginine from the local environment.
Discussion
The cytokine-driven regulation of arginine metabolism via iNOS and arginase in the murine immune system is well established.2-7,33 In this study we have analyzed the expression, subcellular localization, and function of arginase in the human immune system. We have uncovered important differences between human and murine leukocyte arginase. In the murine system the enzyme is not expressed in unstimulated leukocytes but is inducible in all major myeloid cells by TH2 stimulation2-4 (Figure 3B). In human leukocytes this is not the case. We found that neither arginase activity nor protein is present in human monocytes, macrophages, or dendritic cells, either resting or activated in vitro by a variety of proinflammatory and anti-inflammatory stimuli. Human resting PMNs, on the other hand, show high constitutive arginase activity (mean, 1644 mU/mg protein), which is of the same order of magnitude as TH2-stimulated murine macrophages (Figure 1) or dendritic cells3,4 and human hepatocytes (Figure 1). This arginase expression is not further modulated by TH2 cytokines (Figure 3A).
Only a few reports have addressed the expression of arginase in cells of the human immune system. Early reports described a protein with arginase activity that was isolated from PMNs of a patient with chronic myeloid leukemia34 and secreted arginase was found in the supernatant of maturing malignant cells from 2 patients with acute myeloid leukemia.35 Arginase was induced in human mononuclear cells after injury19 or by stimulation with an immunomodulatory peptide.36 The enzyme was also detected in a subpopulation of inflammatory cells found in the bronchoalveolar lavage fluid of asthmatic patients18 or isolated from psoriatic lesions.20 Here we have analyzed highly purified human leukocyte subsets and demonstrated that leukocyte arginase activity is mainly, if not exclusively, localized in PMNs.
Whereas apoptosis of human PMNs degrades the various toxic granule constituents in a highly regulated way, this might not be the case under conditions of chronic inflammation or necrosis. We hypothesize that a dysregulated liberation of arginase into the local microenvironment depletes arginine and participates in local immunosuppression via T-cell hyporesponsiveness8 or in fibrosis via enhanced proline synthesis.11
Arginase I deficiency is a rare autosomal recessive genetic defect (incidence, 1/350 000) that leads to hyperargininemia, hyperammonemia, neurologic impairment, and progressive dementia due to the compromised hepatic urea cycle.1,37 The severity of symptoms depends mainly on the degree of enzymatic deficiency (reduction or total absence of arginase I) and the timely diagnosis with institution of dietary treatment. Two patients with arginase I deficiency were assessed and their PMNs showed no arginase activity or protein. This both confirms that arginase activity in human PMNs is only due to the isoenzyme arginase I and demonstrates that a compensatory increase of arginase II, as described for kidney tissue from hyperargininemic patients37,38 and in the arginase I–deficient mouse39 does not occur in human PMNs (Figure 2C).
What is the function of human PMN arginase Myeloid cells selectively deplete the phagosome of essential nutrients for microbial pathogens.40 On phagocytosis, microorganisms respond by inducing and repressing transcription of a variety of genes to adapt to the new environment. The transcriptional profile of the ingested microbe therefore reflects the microenvironment of the phagosome. It was recently demonstrated that S cerevisiae and C albicans up-regulate genes of their endogenous arginine biosynthetic pathways on phagocytosis by human neutrophils.30 This transcriptional response is most likely due to an arginine-deprived surrounding within the phagosome. Interestingly, on phagocytosis by human monocytes no up-regulation of yeast arginine biosynthetic genes was noted. The authors concluded from this discrepancy that, in contrast to human PMNs, the phagosomal environment of monocytes is not depleted of arginine. Our findings of a selective expression of arginase in human PMN phagosomes (and not in monocytes) offer a likely explanation for the arginine deprivation encountered by yeast on PMN phagocytosis.
Numerous reports have clarified that various oxidative and nonoxidative antimicrobial or antitumor effector pathways exist in PMNs.29,32,41 These pathways cooperate and might provide redundancy in PMN effector function. Depending on the specific microorganism, individual effector mechanisms are more important than others.32,41 In this study we have worked with a model organism that is exquisitely sensitive to arginine deprivation due to its inability to synthesize the amino acid itself. We used this strain of S cerevisiae as a functional bioprobe to prove the existence of arginase-mediated arginine deprivation within the human PMN phagosome even under suboptimal conditions of limited manganese availability and consecutive low arginase activity. More importantly, we show that arginase activity and fungicidal potential of human PMNs increase during phagocytosis of wild-type C albicans with the availability of manganese, an essential constituent of active arginase. When manganese is not limiting, arginase and MPO contribute comparably to the killing of C albicans (Figure 7D). Human arginase I, like all mammalian arginases, has a basic (pH 8.5-9) pH optimum.1 The vacuolar pH of human neutrophils initially becomes basic on phagocytosis before it gets acidic after 15 to 30 minutes of ingestion.41,42 The rise in pH from about 6.0 to 7.8/8.0 soon after phagocytosis depends on the activity of NADPH oxidase, because it is due to consumption of protons by the protonation of and . This alkalinization reaches an optimal level for the activity of granule proteases.41 The rise in pH might also be necessary for the enzymatic activity of arginase in the phagosome during the early phase of phagocytosis. We have seen pronounced inhibition of fungicidal activity by arginase inhibition especially during the first 30 to 90 minutes of phagocytosis, whereas later on the effect of arginase inhibition disappeared (data not shown). Similarly, when NADPH oxidase is blocked by DPI, arginase does not further add to the killing of C albicans (Figure 7A,D) and S cerevisiae (data not shown), likely because the necessary phagosome alkalinization is missing.
The antimicrobial effector function of intraphagosomal arginase needs to be evaluated with additional pathogenic microbes. Arginine deprivation might enhance the sensitivity of the parasite toward killing by other effector pathways. We further hypothesize that arginine might be necessary for the synthesis of antiphagocyte defense mechanisms in certain pathogens.
A review of published cases of patients with arginase I deficiency37 as well as the clinical history of the 2 patients analyzed in this study did not reveal an increased incidence of severe infectious problems. Furthermore, we were able to do one phagocytosis experiment with PMNs of a patient with arginase I deficiency and saw no impairment in fungicidal activity of arginase I–deficient PMNs (data not shown). Obviously, human PMNs are able to compensate for lack of the enzyme due to the redundancy of multiple antimicrobial pathways. The immunologic competence of patients with arginase I deficiency is reminiscent of patients with MPO deficiency of PMNs. Although this genetic defect is associated with impairment of in vitro PMN fungicidal activity, it is usually clinically silent.
In summary, we have clarified the cellular distribution, subcellular localization, regulatory aspects, and function of arginase in human leukocytes. We have demonstrated for the first time that only human PMNs constitutively express arginase I, that the enzyme is localized in the azurophil granules, and that it works as a novel fungicidal effector mechanism of human PMNs.
Acknowledgements
We thank J. Mytilineos and V. Daniel (both Institute of Immunology, University of Heidelberg) for providing blood of healthy donors, V. Weber (Max-Planck-Institute for Immunobiology, Freiburg, Germany), Hans Janssen and Nico Ong (both The Netherlands Cancer Institute, Amsterdam) for expert technical assistance and M. Scheuermann (German Cancer Research Center, Heidelberg) for FACS sorting. We appreciate the help of M. Schwarz, U. Wendel (both University Hospital Düsseldorf) and F.-K. Trefz (Children`s Hospital Reutlingen) for help with recruitment of arginase I-deficient patients. We thank L. B. Nicholson (University of Bristol) for critical review of the manuscript.
Footnotes
Prepublished online as Blood First Edition Paper, November 16, 2004; DOI 10.1182/blood-2004-07-2521.
Supported by grants FIS-02/1199 and FIS-01/1048 from the Fondo de Investigacion Sanitaria (F.M.) and by grants 2PR01A079 (G.S.), 2PR02B034, and 03/13 (J.M.F.) from Junta de Extremadura.
The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked "advertisement" in accordance with 18 U.S.C. section 1734.
References
Jenkinson CP, Grody WW, Cederbaum SD. Comparative properties of arginases. Comp Biochem Physiol B Biochem Mol Biol. 1996;114: 107-132.
Modolell M, Corraliza IM, Link F, Soler G, Eichmann K. Reciprocal regulation of the nitric oxide synthase/arginase balance in mouse bone marrow-derived macrophages by TH1 and TH2 cytokines. Eur J Immunol. 1995;25: 1101-1104.
Munder M, Eichmann K, Modolell M. Alternative metabolic states in murine macrophages reflected by the nitric oxide synthase/arginase balance: competitive regulation by CD4+ T cells correlates with Th1/Th2 phenotype. J Immunol. 1998;160: 5347-5354.
Munder M, Eichmann K, Moran JM, Centeno F, Soler G, Modolell M. Th1/Th2-regulated expression of arginase isoforms in murine macrophages and dendritic cells. J Immunol. 1999;163: 3771-3777.
Louis CA, Mody V, Henry WL, Reichner JS, Albina JE. Regulation of arginase isoforms I and II by IL-4 in cultured murine peritoneal macrophages. Am J Physiol 1999;276: R237-R242.
Chang CI, Zoghi B, Liao JC, Kuo L. The involvement of tyrosine kinases, cyclic AMP/protein kinase A, and p38 mitogen-activated protein kinase in IL-13-mediated arginase I induction in macrophages: its implications in IL-13-inhibited nitric oxide production. J Immunol. 2000;165: 2134-2141.
Rutschman R, Lang R, Hesse M, Ihle JN, Wynn TA, Murray PJ. Stat6-dependent substrate depletion regulates nitric oxide production. J Immunol. 2001;166: 2173-2177.
Rodriguez PC, Zea AH, DeSalvo J, et al. L-arginine consumption by macrophages modulates the expression of CD3 zeta chain in T lymphocytes. J Immunol. 2003;171: 1232-1239.
Bronte V, Serafini P, De Santo C, et al. IL-4-induced arginase 1 suppresses alloreactive T cells in tumor-bearing mice. J Immunol. 2003;170: 270-278.
Kusmartsev S, Nefedova Y, Yoder D, Gabrilovich DI. Antigen-specific inhibition of CD8+ T cell response by immature myeloid cells in cancer is mediated by reactive oxygen species. J Immunol. 2004;172: 989-999.
Hesse M, Modolell M, La Flamme AC, et al. Differential regulation of nitric oxide synthase-2 and arginase-1 by type 1/type 2 cytokines in vivo: granulomatous pathology is shaped by the pattern of L-arginine metabolism. J Immunol. 2001;167: 6533-6544.
Granger DL, Hibbs JB, Perfect JR, Durack DT. Metabolic fate of L-arginine in relation to microbiostatic capability of murine macrophages. J Clin Invest. 1990;85: 264-273.
Iniesta V, Gomez-Nieto LC, Corraliza I. The inhibition of arginase by N(omega)-hydroxy-l-arginine controls the growth of Leishmania inside macrophages. J Exp Med. 2001;193: 777-784.
Gobert AP, Daulouede S, Lepoivre M, et al. L-Arginine availability modulates local nitric oxide production and parasite killing in experimental trypanosomiasis. Infect Immun. 2000;68: 4653-4657.
Xu L, Hilliard B, Carmody RJ, et al. Arginase and autoimmune inflammation in the central nervous system. Immunology. 2003;110: 141-148.
Waddington SN, Tam FW, Cook HT, Cattell V. Arginase activity is modulated by IL-4 and HOArg in nephritic glomeruli and mesangial cells. Am J Physiol. 1998;274: F473-F480.
Mistry SK, Zheng M, Rouse BT, Morris SM. Induction of arginases I and II in cornea during herpes simplex virus infection. Virus Res. 2001;73: 177-182.
Zimmermann N, King NE, Laporte J, et al. Dissection of experimental asthma with DNA microarray analysis identifies arginase in asthma pathogenesis. J Clin Invest. 2003;111: 1863-1874.
Ochoa JB, Bernard AC, O'Brien WE, et al. Arginase I expression and activity in human mononuclear cells after injury. Ann Surg. 2001;233: 393-399.
Bruch-Gerharz D, Schnorr O, Suschek C, et al. Arginase 1 overexpression in psoriasis: limitation of inducible nitric oxide synthase activity as a molecular mechanism for keratinocyte hyperproliferation. Am J Pathol. 2003;162: 203-211.
Escribano S, Cuenllas E, Gaitan S, Tejero C. Delayed neutrophil apoptosis after total body irradiation in mice. The role of granulocyte-macrophage colony-stimulating factor in neutrophil function. Exp Hematol. 1998;26: 942-949.
Meshulam T, Levitz SM, Christin L, Diamond RD. A simplified new assay for assessment of fungal cell damage with the tetrazolium dye, (2,3)-bis-(2-methoxy-4-nitro-5-sulphenyl)-(2H)-tetrazolium-5-carboxanilide (XTT). J Infect Dis. 1995;172: 1153-1156.
Diez A, Fuentes JM, Prada F, Campo ML, Soler G. Immunological identity of the two different molecular mass constitutive subunits of liver arginase. Biol Chem Hoppe Seyler. 1994;375: 537-541.
Martin-Martin B, Nabokina SM, Blasi J, Lazo PA, Mollinedo F. Involvement of SNAP-23 and syntaxin 6 in human neutrophil exocytosis. Blood. 2000;96: 2574-2583.
Mollinedo F, Martin-Martin B, Calafat J, Nabokina SM, Lazo PA. Role of vesicle-associated membrane protein-2, through Q-soluble N-ethylmaleimide-sensitive factor attachment protein receptor/R-soluble N-ethylmaleimide-sensitive factor attachment protein receptor interaction, in the exocytosis of specific and tertiary granules of human neutrophils. J Immunol. 2003;170: 1034-1042.
Sorensen OE, Follin P, Johnsen AH, et al. Human cathelicidin, hCAP-18, is processed to the antimicrobial peptide LL-37 by extracellular cleavage with proteinase 3. Blood. 2001;97: 3951-3959.
Calafat J, Janssen H, Stahle-Bckdahl M, Zuurbier AEM, Knol EF, Egesten A. Human monocytes and neutrophils store transforming growth factor-alpha in a subpopulation of cytoplasmic granules. Blood. 1997;90: 1255-1266.
Jost MM, Ninci E, Meder B, et al. Divergent effects of GM-CSF and TGFbeta1 on bone marrow-derived macrophage arginase-1 activity, MCP-1 expression, and matrix metalloproteinase-12: a potential role during arteriogenesis. FASEB J. 2003;17: 2281-2283.
Borregaard N, Cowland JB. Granules of the human neutrophilic polymorphonuclear leukocyte. Blood. 1997;89: 3503-3521.
Rubin-Bejerano I, Fraser I, Grisafi P, Fink GR. Phagocytosis by neutrophils induces an amino acid deprivation response in Saccharomyces cerevisiae and Candida albicans. Proc Natl Acad Sci U S A. 2003;100: 11007-11012.
Cox JD, Cama E, Colleluori DM, et al. Mechanistic and metabolic inferences from the binding of substrate analogues and products to arginase. Biochemistry. 2001;40: 2689-2701.
Gil-Lamaignere C, Roilides E, Lyman CA, et al. Human phagocytic cell responses to Scedosporium apiospermum (Pseudallescheria boydii): variable susceptibility to oxidative injury. Infect Immun. 2003;71: 6472-6478.
Bogdan C. Nitric oxide and the immune response. Nat Immunol. 2001;2: 907-916.
Reyero C, Dorner F. Purification of arginases from human-leukemic lymphocytes and granulocytes: study of their physicochemical and kinetic properties. Eur J Biochem. 1975;56: 137-147.
Palu G, Powles R, Selby P, Summersgill BM, Alexander P. Patterns of maturation in short-term culture of human acute myeloid leukaemic cells. Br J Cancer 1979;40: 719-730.
Rouzaut A, Subira ML, de Miguel C, et al. Co-expression of inducible nitric oxide synthase and arginases in different human monocyte subsets. Apoptosis regulated by endogenous NO. Biochim Biophys Acta 1999;1451: 319-333.
Iyer R, Jenkinson CP, Vockley JG, Kern RM, Grody WW, Cederbaum S. The human arginases and arginase deficiency. J Inherit Metab Dis. 1998;21(suppl 1): 86-100.
Grody WW, Argyle C, Kern RM, et al. Differential expression of the two human arginase genes in hyperargininemia. Enzymatic, pathologic, and molecular analysis. J Clin Invest. 1989;83: 602-609.
Iyer RK, Yoo PK, Kern RM, et al. Mouse model for human arginase deficiency. Mol Cell Biol. 2002;22: 4491-4498.
Jabado N, Jankowski A, Dougaparsad S, Picard V, Grinstein S, Gros P. Natural resistance to intracellular infections: natural resistance-associated macrophage protein 1 (Nramp1) functions as a pH-dependent manganese transporter at the phagosomal membrane. J Exp Med. 2000;192: 1237-1248.
Reeves EP, Lu H, Lortat Jacobs H, et al. Killing activity of neutrophils is mediated through activation of proteases by K+ flux. Nature. 2002;416: 291-297.
Segal AW, Geisow M, Garcia R, Harper A, Miller R. The respiratory burst of phagocytic cells is associated with a rise in vacuolar pH. Nature. 1981;290: 406-409.(Markus Munder, Faustino M)