BCG Infection Mycobacterium bovis Macrophages in the Context of ...

2006, 188(13):4830. DOI: 10.1128/JB.01687-05. J. Bacteriol.

Roberto Colangeli, David Alland and Nancy D. ConnellHernando Hazbón, Marcy Peteroy-Kelly, Anjali Seth, Meliza T. Talaue, Vishwanath Venketaraman, Manzour

BCG InfectionMycobacterium bovisMacrophages in the Context of Arginine Homeostasis in J774.1

http://jb.asm.org/content/188/13/4830Updated information and services can be found at:

These include:

REFERENCEShttp://jb.asm.org/content/188/13/4830#ref-list-1at:

This article cites 45 articles, 22 of which can be accessed free

CONTENT ALERTS more»articles cite this article),

Receive: RSS Feeds, eTOCs, free email alerts (when new

http://journals.asm.org/site/misc/reprints.xhtmlInformation about commercial reprint orders: http://journals.asm.org/site/subscriptions/To subscribe to to another ASM Journal go to:

on February 22, 2013 by P

EN

N S

TA

TE

UN

IVhttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/cgi/alerts

http://jb.asm.org/

JOURNAL OF BACTERIOLOGY, July 2006, p. 4830–4840 Vol. 188, No. 130021-9193/06/$08.00�0 doi:10.1128/JB.01687-05Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Arginine Homeostasis in J774.1 Macrophages in the Context ofMycobacterium bovis BCG Infection

Meliza T. Talaue,1,2 Vishwanath Venketaraman,2 Manzour Hernando Hazbon,2 Marcy Peteroy-Kelly,4Anjali Seth,3 Roberto Colangeli,2 David Alland,2 and Nancy D. Connell1,2*

Department of Microbiology and Molecular Genetics,1 Department of Medicine,2 and Department of Biochemistry andMolecular Biology,3 UMDNJ/New Jersey Medical School, 185 South Orange Avenue, Newark,

New Jersey 07103-2714, and Pace University, 1 Pace Plaza, New York, New York 100384

Received 5 November 2005/Accepted 12 April 2006

The competition for L-arginine between the inducible nitric oxide synthase and arginase contributes to theoutcome of several parasitic and bacterial infections. The acquisition of L-arginine, however, is important notonly for the host cells but also for the intracellular pathogen. In this study we observe that strain AS-1, theMycobacterium bovis BCG strain lacking the Rv0522 gene, which encodes an arginine permease, perturbsL-arginine metabolism in J774.1 murine macrophages. Infection with AS-1, but not with wild-type BCG,induced L-arginine uptake in J774.1 cells. This increase in L-arginine uptake was independent of activation withgamma interferon plus lipopolysaccharide and correlated with increased expression of the MCAT1 andMCAT2 cationic amino acid transport genes. AS-1 infection also enhanced arginase activity in resting J774.1cells. Survival studies revealed that AS-1 survived better than BCG within resting J774.1 cells. Intracellulargrowth of AS-1 was further enhanced by inhibiting arginase and ornithine decarboxylase activities in J774.1cells using L-norvaline and difluoromethylornithine treatment, respectively. These results suggest that thearginine-related activities of J774.1 macrophages are affected by the arginine transport capacity of the infectingBCG strain. The loss of Rv0522 gene-encoded arginine transport may have induced other cationic amino acidtransport systems during intracellular growth of AS-1, allowing better survival within resting macrophages.

Arginine is an essential modulator of the cellular immuneresponse during infection. The generation of nitric oxide (NO)from arginine by the inducible nitric oxide synthase (iNOS) isresponsible for the cytotoxicity of macrophages against bacte-rial and parasitic pathogens (10, 25, 37), while the conversionof arginine to ornithine and urea via the arginase pathway hasbeen shown to support the intracellular survival of Helicobacterpylori, Leishmania spp., Trypanosoma spp., and Schistosomaspp. (1, 9, 20, 21, 28, 33, 44). Indeed, competition betweeniNOS and arginase for arginine has been suggested to contrib-ute to the outcome of infection (11, 35, 36).

Access to arginine is important not only for infected mac-rophages but also for the infecting organism. A number ofpathogens have been shown to alter their arginine-dependentmetabolic activities when they are inside their host cells. Forexample, Listeria monocytogenes preferentially upregulatesarpJ, a gene encoding an arginine permease, during intracel-lular growth (32), Mycobacterium marinum induces the argSgene, encoding arginyl-tRNA synthetase, when inside macro-phages (4), and Giardia lamblia inhibits NO synthesis by mac-rophages by directly consuming arginine using a highly efficientarginine transport system (15). Therefore, arginine availabilityand the capacity of the host cells and the infecting organism toacquire arginine and/or its derived substrates are importantfactors that can influence the course of an infection.

The genes for L-arginine biosynthesis in Mycobacterium tu-

berculosis and Mycobacterium bovis BCG are found in theoperon argCJBDFRGH (22). Although mycobacteria can syn-thesize L-arginine from glutamate, carbamoyl phosphate, andaspartate, L-arginine biosynthesis is energetically expensive asit requires at least six ATP equivalents; less energy is requiredfor mycobacteria to acquire L-arginine from the external envi-ronment. Therefore, it is not surprising that there are severalgenes encoding putative L-arginine uptake systems (Rv0522,Rv2320c, Rv3253c, Rv1999c, and Rv1979c) in M. tuberculosisand M. bovis, most likely with different substrate affinities andcapacities (13). Depending on the environment in which theyreside, the ability of mycobacteria to acquire L-arginine may bereflected by the amounts and types of L-arginine permeases(44). Our laboratory is studying these permeases and relatedarginine metabolic enzymes to understand their role in infec-tion by mycobacteria.

Previous work from our laboratory described L-argininetransport in M. bovis BCG by examining the phenotypes of amutant strain, AS-1, lacking a single gene encoding a cationicamino acid transporter, gabP (Rv0522), and the mutant’s com-plemented derivative, AS-2 (44). L-Arginine transport by AS-1is reduced by 70% when this strain is grown in 7H9 culturemedium. However, AS-1 exhibited undiminished uptake ofL-lysine and L-ornithine. These results suggest that the Rv0522-encoded permease does not significantly contribute to thetransport of other cationic amino acids (L-ornithine and L-lysine) in BCG grown in culture. Thus, the remaining 30% ofL-arginine transport activity and the uptake of other cationicamino acids by AS-1 are mediated by other amino acid per-meases.

In another study we demonstrated that intracellular BCG

* Corresponding author. Mailing address: Department of Medicine,UMDNJ/New Jersey Medical School, 185 South Orange Avenue,Newark, NJ 07103-2714. Phone: (973) 972-3759. Fax: (973) 972-3644.E-mail: [email protected].

4830


EN

N S

TA

TE

UN

IVhttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

was able to incorporate 3H label while inside J774.1 macro-phages grown in culture medium that is supplemented with3H-labeled L-arginine (41). Surprisingly, the AS-1 mutant ac-cumulated twofold more 3H label than wild-type BCG insideJ774.1 macrophages. These results suggest that intracellularAS-1, in contrast to AS-1 grown in culture, has an increased,not decreased, capacity to transport arginine or other arginine-derived substrates when inside the macrophages.

It has long been known that classical activation of murinemacrophages leads to increased arginine transport (6, 7, 26,31). We showed that infection by BCG without activation doesnot stimulate arginine transport (40). We also found that AS-1,but not BCG, induced L-arginine uptake in infected J774.1cells in the absence of activation with cytokines or lipopolysac-charide (LPS) (41).

We hypothesized that intracellular AS-1, having an inducedcapacity to acquire intracellular substrates derived from mac-rophage arginine, perturbs arginine homeostasis of the hostmacrophages. Thus, the aim of the present work was to studythe response of resting and activated (gamma interferon [IFN-�]-plus-LPS-treated) J774.1 macrophages to AS-1 and BCGinfection by evaluating several L-arginine-dependent activities,such as L-arginine uptake and transporter expression, iNOSexpression, NO production, and arginase activity. Further, weinvestigated the effect of modulating arginine and ornithineavailability on the survival of AS-1 and BCG during growth inJ774.1 macrophages. Altogether, our results show that intra-cellular survival of BCG depends on its arginine transportcapacity and arginine availability within the host macrophages.

MATERIALS AND METHODS

Bacterial strains and growth conditions. M. bovis BCG and its derivativesAS-1 and AS-2 were grown in Middlebrook medium (Difco). Middlebrook 7H9(liquid) and 7H11 (1.5% agar) media were supplemented with 0.5% (vol/vol)glycerol and ADC supplement (0.5% [wt/vol] bovine serum albumin, fraction V,0.2% [wt/vol] dextrose, and 0.85% [wt/vol] NaCl). All liquid cultures of BCGwere supplemented with 0.05% (vol/vol) Tween 80 (Sigma). All cultures wereincubated at 37°C.

Maintenance of J774.1 murine macrophages. The J774.1 murine macrophagecell line was maintained at 37°C, 5% CO2 in Dulbecco’s modified Eagle’s me-dium (DMEM; Sigma) containing 10% (vol/vol) fetal bovine serum (Sigma), 2mM glutamine (Sigma), and essential amino acids (Gibco BRL).

Processing of BCG for infection. BCG cultures (optical density at 600 nm of0.7 to 0.8, approximating that of the logarithmic growth phase) were used forinfection of macrophages. Ten milliliters of bacterial suspension was diluted to50-ml volume in phosphate-buffered saline (PBS) and centrifuged at 5,000 � gfor 15 min at room temperature. The bacterial pellet was resuspended in DMEM(above). Bacterial clumps were disaggregated by vortexing five times with 3-mmsterile glass beads (Fisher Scientific) for 2 min. The bacterial suspension waspassed through a 5-�m filter (Micron Separation Inc.) to remove remainingclumps and then aliquoted (1 ml) and stored at �80°C. Representative vials werethawed, serially diluted, and then plated on 7H11 medium to determine thenumber of viable bacilli by counting CFU.

Mycobacterial infection of J774.1 murine macrophages. J774.1 macrophageswere detached from the tissue culture flasks by gentle scraping with a cell scraper(Costar). J774.1 cells (1 � 105/well) were distributed in 24-well tissue cultureplates and allowed to adhere overnight prior to infection. Frozen stocks of theBCG strains were thawed and prewarmed to 37°C and were added to themacrophages at 5:1 mycobacteria per macrophage (multiplicity of infection, 5:1).The macrophages were allowed to phagocytose the bacteria for 2 h, after whichtime nonphagocytosed organisms were removed by washing three times withprewarmed serum-free medium. Infected J774.1 cell lines were maintained inculture for 3 days.

Uptake of L-[3H]arginine in J774.1 macrophages. J774.1 macrophages wereexposed to the following experimental conditions: (i) stimulation with IFN-� plusLPS (100 U/ml and 1 �g/ml, respectively); (ii) infection with BCG, AS-1, or AS-2

strains; or (iii) BCG, AS-1, or AS-2 infection plus stimulation with IFN-� plusLPS. Uninfected J774.1 cells served as baseline controls for all experiments.

Cells were subsequently harvested at 24 h and 72 h to determine the degree ofL-[3H]arginine uptake by washing three times in HEPES-buffered Kreb’s solution(131 mM NaCl, 5.5 mM KCl, 1 mM MgCl2, 2.5 mM CaCl2, 25 mM NaHCO3, 1mM Na2H2PO4, 5.5 mM D-glucose, and 20 mM HEPES [pH 7.4]). Uptake wasinitiated by adding 100 �l of 1 �M L-[3H]arginine (1.0 mCi/ml; NEN) diluted inHEPES-buffered Kreb’s solution per well for 10 min. A 10-min incubation waschosen because our initial experiments showed maximum uptake of L-[3H]argi-nine at this time point (40).

Macrophage cultures were washed three times with 200 �l of ice-cold PBScontaining 10 mM unlabeled L-arginine. After washing, 100 �l of 0.024 M formicacid was added to each well to lyse the macrophage monolayer. The macrophagelysate was mixed with scintillation fluid and the 3H content was determined in aliquid scintillation counter. The radioactive counts were normalized to macro-phage protein content. Protein levels in the formic acid lysates of macrophagecultures were determined spectrophotometrically using the detergent-compati-ble protein assay (DC assay) (Bio-Rad). L-Arginine uptake values were expressedas pmol L-[3H]arginine uptake per mg protein/10 min.

Determination of free amino acid levels by HPLC. J774.1 macrophages wereprocessed according to the method described by Fisher et al. (18). Briefly, after24 h and 72 h postinfection or -treatment, the macrophages were washed threetimes with PBS. Then 400 �l of 1.5 M perchloric acid (Sigma) was added to eachwell. The lysed macrophages were scraped off the wells and transferred into a1.5-ml microcentrifuge tube and the lysates were neutralized by adding 100 �l of2 M potassium hydroxide (Sigma). The samples were cooled on ice for 10 minand then centrifuged at 15,000 � g for 10 min. The supernatants were collectedand then filtered through 0.2-�m filter units (Millipore) into new 1.5-ml micro-centrifuge tubes. Thirty microliters of each sample was then hydrolyzed with 200�l 6 N HCl. The resultant hydrolysate was derivatized to phenylthiocarbamyl(PTC)-amino acids using a precolumn PTC derivatization and analyzed on anApplied Biosystems 420A amino acid analyzer. High-pressure liquid chromatog-raphy (HPLC) runs of PTC-labeled amino acid standards were recorded andpeak areas calculated to make calibration curves to quantify amino acids in eachsample.

Nitrite assay. J774.1 macrophages were cultured in phenol red-free DMEM,supplemented as above. As a measure of NO synthesis, nitrite that accumulatedin the culture medium was determined. Nitrite levels were determined by aspectrophotometric assay based on the Greiss reaction (16). Briefly, 100 �l ofGreiss reagent (1% sulfanilamide, 0.1% naphthylethylenediamine dihydrochlo-ride in 5% H3PO4) was added to a 100-�l sample and mixed immediately. After5 min of incubation at room temperature, absorbance was measured at 540 nm.The nitrite content for each 100-�l sample was determined by comparison witha standard curve derived from absorbance readings of freshly made sodiumnitrate (Sigma) standards (0 to 1,000 �M) dissolved in DMEM, normalized toprotein content using the DC protein assay (Bio-Rad), and then expressed asnmol NO per mg protein.

FACS analysis for iNOS protein. J774.1 cells were cultured in 12-well tissueculture plates. Macrophages were infected and treated as described above. Threedays after infection or treatment, J774.1 cells were scraped from the bottom ofthe wells and centrifuged at 800 � g for 10 min. The medium was then removed,and the cells were fixed for 30 min at room temperature with freshly prepared3.8% paraformaldehyde (Sigma) in PBS. Fixed macrophages were permeabilizedwith 0.1% Triton X for 1 min and blocked for 15 min in blocking solution (PBScontaining 5% [wt/vol] sucrose and 2% [vol/vol] goat serum). Cells were incu-bated with polyclonal rabbit anti-mouse iNOS antibody for 1 h at 37°C, washedthree times in blocking solution, and incubated with fluorescein isothiocyanate-conjugated goat anti-rabbit immunoglobulin secondary antibody (MolecularProbes) for 1 h at 37°C. The cells were then washed three times with PBScontaining 5% (wt/vol) sucrose. Cells were then resuspended in 500 �l PBS andanalyzed by fluorescence-activated cell sorting (FACS) using a FACSCalibur(Becton Dickinson) equipped with a 488-nm argon ion laser and a 530-nmband-pass filter. The mean fluorescence intensity for each sample was acquiredand analyzed using CellQuest software.

Determination of arginase activity. Arginase activity was determined by mea-suring the total production of urea in both J774.1 macrophage cell lysates andculture supernatants. After 24 h and 72 h postinfection or -treatment, the mac-rophage culture supernatants from each well were collected and transferred into1.5-ml microcentrifuge tubes. The macrophages were then washed three timeswith PBS, and 500 �l of sterile distilled water was added into each well to lyse thecells. Both culture supernatants and macrophage lysates were then filteredthrough 0.2-�m filters.

Urea levels were determined using the blood urea nitrogen reagent kit from

VOL. 188, 2006 ARGININE METABOLISM IN BCG-INFECTED MURINE MACROPHAGES 4831


EN

N S

TA

TE

UN

IVhttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

Biotron Diagnostics, Inc. (Hernet, CA). Based on the colorimetric method thatwas first developed by Jung et al. (29), urea levels were measured in 50 �l offiltered culture supernatant or macrophage lysate by adding the following pre-made reagents into a 5-ml reaction tube: 1 ml of blood urea nitrogen reagent(containing 0.6% [vol/vol] o-phthalaldehyde, 9.5% [vol/vol] concentrated sulfuricacid, and surfactant) and 0.5 ml color developer solution (containing 0.5%[vol/vol] naphthylethylenediamine, 6% [wt/vol] boric acid, and surfactant). Thecontents were then mixed and incubated at 37°C for 5 min. Two milliliters of thediluent (0.6% [vol/vol] concentrated sulfuric acid) was added and mixed intoeach tube and the absorbance was read at 540 nm. The urea levels in the sampleswere derived from comparison with urea standards and then normalized toprotein content using the DC protein assay (Bio-Rad) and expressed as nmolurea per mg protein.

Total RNA extraction and cDNA synthesis. To further characterize the mo-lecular events leading to arginine metabolism in the J774.1 murine macrophagecell line we measured the RNA expression of four genes known to encodeenzymes that utilize arginine as their primary substrates. Using real-time PCR,expression levels for MCAT1 and MCAT2 were determined at 6 and 24 hpostinfection in J774.1 murine macrophages.

Total RNA from 3 � 106 cells (either resting or IFN-� plus LPS treated; BCGor AS-1 infected) was isolated using TRIzol reagent (Life Technologies) asspecified by the manufacturer, at 6 and 24 h postinfection. Antisense primerswere designed using the Primer Express software (version 1.5a; Applied Biosys-tems) to target specific mRNAs (A/SmCAT1, A/SmCAT2, and A/S�-actin;Table 1). These primers were used for cDNA synthesis at 60°C using C. therm.polymerase (Roche Diagnostics) as specified by the manufacturer.

Real-time PCR. PCR primers were designed to anneal to their targets at 60°C.In order to lower the background during PCR amplification, a tail that hybridizesto the 5� end of the primer (bases that are underlined in Table 1) was added tothe 3� end yielding hairpin primers (24). The specific PCR primers used in thisstudy are the following: FHPmCAT1, RHPmCAT1, FHPmCAT2, RHPmCAT2,FHP�-actin, and RHP�-actin (Table 1).

Plates (384-well) and the Applied Biosystems 7900HT sequence detector sys-tem were used for real-time PCR and analysis. Each 5-�l reaction mixturecontained 1� AmpliTaq Gold polymerase buffer (Perkin-Elmer); 4 mM MgCl2;2.5 pmol of each primer; 200 �M each of dATP, dCTP, dGTP, and dTTP; 0.25U of AmpliTaq Gold polymerase; 0.5� SYBR green I fluorescent dye (Molec-ular Probes); and 1.75 ng ROX (6-carboxy-X-rhodamine, succinimidyl ester) asa reference dye. PCR conditions and real-time PCR performance were as pre-viously recommended (24). External standards for quantitative analysis included10-fold dilutions (between 107 and 103 molecules) of single-stranded artificialtemplates (custom-made oligonucleotides; Invitrogen) with the sequence of theamplicon yielded during the PCR amplification (FTmCAT1, FTmCAT2, FT�-actin; Table 1). Quantitative results for MCAT1 and MCAT2 cDNA were nor-malized to the number of �-actin cDNA molecules measured in parallel. Chro-mosomal DNA contamination was measured by performance of real-time PCRof “control cDNA” synthesized without the addition of reverse transcriptase. Thenumber of contaminating genomic-DNA molecules was then subtracted from

each sample when necessary. Analysis was based on at least three replicates persample.

Mycobacterial survival studies. BCG-, AS-1-, or AS-2-infected J774.1 cellswere harvested at 1 h and 72 h after infection to determine the intracellularviability of BCG. Infected macrophages were harvested by lysis with steriledistilled water. The lysate was serially diluted and plated on 7H11 (Difco)enriched with ADC. Inoculated plates were incubated at 37°C for 3 weeks toallow counting of mycobacterial CFU. For difluoromethylornithine (DFMO)and L-norvaline studies, J774.1 macrophages were pretreated overnight with 1mM DFMO (Sigma) or 10 mM L-norvaline (Sigma). The next day, cells werewashed three times with PBS and incubated with regular DMEM. The cells werethen infected with BCG or the AS-1 strain at a multiplicity of infection of 5:1.After 2 h, the infected cells were washed three times with PBS to removeextracellular mycobacteria. IFN-� plus LPS and L-ornithine (1 mM) were thenadded to the appropriate wells. CFU count was determined at an initial timepoint of 1 or 2 h postinfection and a final time point of 72 h postinfection.

Statistical analysis. Statistical analysis of the data was carried out using Stat-View software (v. 5), and the statistical significance (P) was determined using theunpaired t test and, for multiple comparisons, Fisher’s protected least-signifi-cant-difference (PLSD) post hoc test. Differences were considered significantwhen P was at a level of �0.05.

RESULTS

An arginine permease mutant strain of BCG (AS-1) inducedL-arginine uptake in nonactivated J774.1 murine macro-phages. A central response of murine macrophages to activa-tion is the increased accumulation of arginine for the produc-tion of NO (5, 6, 7, 26, 31). We have previously shown thatL-arginine uptake by J774.1 murine macrophages is induced 4 hpostinfection with AS-1, a BCG strain lacking the L-argininepermease gene Rv0522 (41). To determine whether increasedL-arginine uptake is sustained at a later time point, we mea-sured transport activity by J774.1 cells at 72 h after mycobac-terial infection with and without IFN-�-plus-LPS stimulation.Figure 1 shows that infection with wild-type BCG resulted in aslight (twofold) increase in macrophage L-arginine uptake, butthis increase was not significantly different from the uptake ofuninfected macrophages (P, 0.1118). In contrast to BCG in-fection, and consistent with our previous observations, AS-1infection increased L-arginine uptake of J774.1 cells at 72 hpostinfection by approximately 4.4-fold, compared to restingmacrophages (P, �0.0001). As a control, infection of J774.1cells with AS-2 (the AS-1 strain complemented with a copy ofthe wild-type Rv0522 gene) resulted in levels of arginine up-take similar to those of BCG-infected macrophages (P,0.3255). Treatment with IFN-� plus LPS in the absence ofinfection resulted in a fivefold increase in arginine uptake overthat of unstimulated macrophages (P, �0.0001), confirmingearlier observations of our and other laboratories (7, 31, 40, 41,46). Uptake by AS-1-infected cells showed similar levels ofL-arginine uptake to J774.1 macrophages treated with IFN-�plus LPS (P, 0.1140). Activation of BCG-, AS-1-, and AS-2-infected macrophages with IFN-� plus LPS also resulted insignificant increases in arginine uptake (all three conditionshave P values of �0.0001).

AS-1-induced L-arginine uptake was independent of L-argi-nine–NO pathway. Several studies have shown that increases inL-arginine transport correlate with iNOS induction in macro-phage cells (5, 6, 7, 26, 23, 27, 31). To determine whether theincreased cellular uptake of arginine in J774.1 macrophagesinfected with AS-1 correlates with NO generation, we mea-sured iNOS protein expression by flow cytometry (FACS anal-ysis) and NO production by Greiss assay.

TABLE 1. List of primers used for cDNA synthesis and RT-PCRa

Name Length(bases) Sequence (5�–3�)

A/SmCAT1 16 GGGGTTTGGGCCAGCAA/SmCAT2 20 GCATAAGCACACAGGCTGCCA/S�-actin 16 CTAGGGCGGCCCACGAFHPmCAT1 30 GTGCTGACCGGAGAGTTCTCACGTCAGCACRHPmCAT1 24 GCCCTGCCCAGGAGCATTCAGGGCFHPmCAT2 27 ACGAGGGTTTGACCTGAAGGCCCTCGTRHPmCAT2 26 GGCACCGAGTAGGCCATGAGGGTGCCFHP�-actin 22 GCCGGGCATGTGCAAAGCCGGCRHP�-actin 24 GCTCCCGAATACAGCCCGGGGAGCFTmCAT1 60 CGGAGAGTTCTCACGTCAGCACATGGCCCTG

AATGCTCCTGGGGTGCTGGCCCAAACCCCFTmCAT2 82 TTTGACCTGAAGGCCCTCGTGGACATGATGT

CTATTGGCACCCTCATGGCCTACTCTCTGGTGGCAGCCTGTGTGCTTATGC

FT�-actin 73 GCATGTGCAAAGCCGGCTTCGCGGGCGACGATGCTCCCCGGG CTGTATTCCCCTCCATCGTGGGCCGCCCTAG

a Underlined bases represent a tail that hybridizes to the 5� end of the primer.RT-PCR, reverse transcriptase PCR.

4832 TALAUE ET AL. J. BACTERIOL.


EN

N S

TA

TE

UN

IVhttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

Treatment of J774.1 macrophages with IFN-� plus LPS in-duced a 17-fold increase in iNOS protein expression and a21-fold increase in NO production at 72 h postactivation (Fig.2). These increases in iNOS protein expression and activity inIFN-�-plus-LPS-treated macrophages correlated well with theincreased L-arginine uptake shown previously in Fig. 1, sup-porting previous observations that L-arginine transport plays akey role in sustaining NO synthesis in activated macrophages(7, 26, 27).

Although infection with BCG or AS-1 increased iNOS pro-tein expression by 4.6- or 2.4-fold, respectively, there was neg-ligible NO production in these infected cells. Thus, unlike inIFN-�-plus-LPS-treated macrophages, the induction of L-argi-nine uptake appears to be uncoupled from iNOS pathwayactivation.

Infection with both BCG and AS-1 was associated with sig-nificant reduction in iNOS protein levels (41%) and NO pro-duction (43%) in macrophages activated with IFN-� plus LPS,compared to uninfected macrophages treated with IFN-� plusLPS. These results suggest that mycobacterial infection sup-presses the full induction of iNOS protein expression, leadingto lower NO production in IFN-�-plus-LPS-activated macro-phages, as described by Fortune et al. (19).

AS-1 infection led to increased expression of both MCAT1and MCAT2 mRNAs in J774.1 macrophages. To investigatethe mechanism behind the enhanced uptake of arginine byAS-1-infected macrophages, we quantified the expression oftwo genes that encode macrophage arginine transportersMCAT1 and MCAT2. Both transporters belong to the murinecationic amino acid family of transporters that are responsiblefor the transport of arginine, lysine, and ornithine. The MCAT1transporter is ubiquitously expressed, is found in all tissuesexcept the liver (12), and is responsible for the uptake ofarginine for general macrophage metabolism (2, 14). TheMCAT2 product has been shown to be induced in murinemacrophages upon stimulation with IFN-� plus LPS (17, 30,

FIG. 1. Uptake of L-[3H]arginine by J774.1 macrophages at 72 h postinfection. Uptake is expressed as pmol of L-arginine per mg total protein in 10min. Change (n-fold) is equivalent to uptake value normalized to uptake value of control cells. The results shown are the averages of three independentexperiments performed in triplicate. *, PFisher’s PLSD � 0.05 (compared to uptake of control cells). (Some relevant P values include the following: controlversus IFN-� [IFN-g] plus LPS, P � 0.0001; control versus BCG, P 0.1118; control versus AS-1, P � 0.0001; BCG versus AS-1, P 0.0003 [#]; IFN-�plus LPS versus AS-1, P 0.1140; AS-1 versus AS-2, P 0.0001; and BCG versus AS-2, P 0.3255).

FIG. 2. iNOS protein expression (A) and NO production (B) inJ774.1 macrophages at 72 h postinfection. iNOS protein levels (num-ber of iNOS-positive cells in 10,000 J774.1 cells) as determined byFACS analysis were normalized to those of resting macrophages NOconcentration in 100-�l culture supernatants was measured by Greissassay. The results shown are the averages of two independent experi-ments performed in triplicate. *, PFisher’s PLSD � 0.05 compared to NOlevels in resting macrophages; #, P � 0.05 compared to NO levels inIFN-�-plus-LPS-treated macrophages.



EN

N S

TA

TE

UN

IVhttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

34). Kakuda et al. (31) demonstrated that MCAT2 is solelyresponsible for the increase in arginine uptake detected inIFN-�-plus-LPS-treated macrophages, leading to the produc-tion of NO.

We examined the transcriptional expression of MCAT1, asshown in Fig. 3. Treatment with IFN-� and LPS significantlydecreased the expression of MCAT1 at 24 h compared toresting macrophages (P 0.0015). This result is consistentwith our previous studies (40, 41) and those of MacLeod et al.(34), where downregulation of MCAT1 mRNA was observedin IFN-�-plus-LPS-treated macrophages. BCG infection didnot significantly change the levels of MCAT1 mRNA in J774.1macrophages at either time point. MCAT1 mRNA in AS-1-infected macrophages was significantly elevated at 6 h postin-fection compared to the levels for resting macrophages (P,0.0302) and BCG-infected macrophages (P, 0.0068). Theseresults suggest that the enhanced L-arginine uptake detected inAS-1-infected macrophages may be due in part to MCAT1upregulation. In view of MCAT1’s known role in providingarginine for cellular metabolism, these data suggest that AS-1infection alters the arginine metabolism of the macrophage.

Consistent with previous studies (30, 31, 34), we observed anapproximately sixfold increase in expression of MCAT2mRNA at 6 h and 24 h posttreatment in IFN-�-plus-LPS-treated macrophages compared to resting macrophages (P,�0.0001 at both time points) as shown in Fig. 4. Infection ofJ774.1 macrophages with either BCG or AS-1 induced a two-fold increase in MCAT2 mRNA levels at 6 h postinfectioncompared to resting macrophages. Therefore, although BCGinfection did not significantly induce L-arginine uptake inJ774.1 macrophages, it resulted in increased MCAT2 mRNAexpression. In contrast, AS-1 infection induced a significantincrease in L-arginine uptake and increased expression of

both MCAT1 and MCAT2 mRNA levels in J774.1 macro-phages. These results suggest that the increased L-arginineuptake in AS-1-infected macrophages is due to the com-bined upregulation of both MCAT1 and MCAT2 genes.

Intracellular arginine levels were maintained in J774.1 mac-rophages. We next examined whether intracellular L-argininelevels actually increase due to uptake of exogenous L-argininein AS-1-infected cells and in IFN-�-plus-LPS-treated cells. In-tracellular free amino acids were extracted using perchloricacid to separate acid-soluble amino acids from precipitatedmacrophage proteins and then were quantitatively analyzed byHPLC. Figure 5 shows the intracellular levels of arginine andother amino acids in J774.1 macrophage lysates. Among theamino acids analyzed, arginine was the most abundant freeamino acid found inside macrophages, with intracellular levelsranging from 350 to 700 pmol per mg total protein, while theintracellular levels of other free amino acids such as lysine andphenylalanine ranged only from 40 to 65 pmol and from 2 to 23pmol, respectively.

Infection with BCG or AS-1 slightly decreased the intracel-lular arginine levels, compared to resting macrophages at 24 hpostinfection. Treatment with IFN-� plus LPS slightly in-creased intracellular levels of arginine and lysine. However, theoverall trend was that BCG or AS-1 infection or treatment withIFN-� plus LPS did not significantly alter intracellular levels offree amino acids in J774.1 macrophages. These results suggestthat macrophages adapt to maintain arginine levels, as well asthose of other amino acids, inside their cells.

Arginase activity was induced by infection and treatmentwith IFN-� plus LPS. Since we did not find significant changesin the free arginine pools in J774.1 cells during infection withAS-1, we next determined whether AS-1 infection was altering

FIG. 3. Transcriptional levels of MCAT1 in J774.1 macrophages at6 and 24 h postinfection. Artificial MCAT1 templates of known con-centrations were used as standards (see Materials and Methods). Ex-pression levels were expressed as the number of molecules of MCAT1RNA per 1,000 molecules of �-actin RNA. The results shown are theaverages of two independent experiments performed in triplicate. *,PFisher’s PLSD � 0.05 compared to expression levels in resting macro-phages. (Some relevant P values include the following. At 6 h: restingversus AS-1, P 0.0302 [*]; BCG versus AS-1, P 0.0068 [#]; restingversus IFN-� plus LPS, P 0.1035. At 24 h: resting versus AS-1, P 0.0724; BCG versus AS-1, P 0.4419; resting versus IFN-� plus LPS,P 0.0015 [*]).

FIG. 4. Transcriptional levels of MCAT2 RNA in J774.1 macro-phages at 6 and 24 h postinfection. Artificial MCAT2 templates ofknown concentrations were used as standards (see Materials andMethods). Expression levels are expressed as the number of moleculesof MCAT2 RNA per 1,000 molecules of �-actin RNA. The resultsshown are the averages of two independent experiments performed intriplicate. *, PFisher’s PLSD � 0.05 compared to expression levels inresting macrophages. (Some relevant P values include the following.At 6 h: resting versus AS-1, P 0.0369 [*]; BCG versus AS-1, P 0.8635; resting versus IFN-� (IFN-g) plus LPS, P � 0.0001. At 24 h:resting versus AS-1, P 0.1277; resting versus BCG, P 0.0091 [*];BCG versus AS-1, P 0.1239; resting versus IFN-� plus LPS, P �0.0001 [*]).



EN

N S

TA

TE

UN

IVhttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

the activities of other arginine-dependent enzymes. In macro-phages, arginine is a substrate not only for iNOS but also forarginase, and the two enzymes demonstrate reciprocal regula-tion (27, 36). Arginase hydrolyzes arginine to ornithine andurea: urea production is thus an indicator of arginase activity.To determine the total arginase activity in J774.1 macrophages,total urea production was measured in culture supernatants andmacrophage lysates by using a colorimetric assay (29). Figure 6shows total urea production in J774.1 macrophages at 24 and72 h postinfection. Previous studies have shown treatment ofmurine macrophages with LPS induced arginase expressionand activity (42, 43). Consistent with these studies, we foundthat urea production was significantly enhanced by three- tofourfold in IFN-�-plus-LPS-treated macrophages compared to

resting macrophages. Infection with BCG and AS-1 also in-creased urea levels by 0.3- and 0.4-fold, respectively, at 24 hpostinfection. Total urea production was further increased by1.6- and 2.2-fold at 72 h postinfection with BCG and the AS-1strain, respectively. These results suggest that mycobacterialinfection increases arginase activity in J774.1 macrophages.

Urea production by AS-1-infected macrophages was higherthan that by BCG-infected macrophages at 72 h postinfection.Since AS-1-infected macrophages induced more L-arginine up-take in J774.1 macrophages than did BCG infection, it is likelythat L-arginine taken up in AS-1-infected macrophages is me-tabolized by arginase for ornithine and urea production. Treat-ment of BCG- or AS-1-infected macrophages with IFN-� plusLPS further increased total urea production up to 4.3- to 4.8-fold compared with IFN-�-plus-LPS-treated macrophages (un-infected) (P, �0.0001 and 0.0001, respectively). These resultssuggest that intracellular BCG or AS-1 can further stimulatearginase activity in IFN-�-plus-LPS-activated macrophages.

Survival studies. We determined whether the changes inL-arginine-dependent pathways have either detrimental orbeneficial influence on intracellular survival of BCG or AS-1 inJ774.1 macrophages.

To determine the survival of BCG and AS-1 in J774.1 mac-rophages, intracellular mycobacteria from lysed macrophageswith and without IFN-�-plus-LPS treatment were plated on7H11 agar. Infection of resting J774.1 macrophages with BCGresulted in a 24% increase in viable colony counts after 72 h, asshown in Fig. 7 (P, 0.4527). In contrast, AS-1 colony countssignificantly increased by 55% after 72 h (P, 0.0321). Stimula-tion of J774.1 macrophages with IFN-� plus LPS resulted in areduction in viable colony counts of both BCG and AS-1. Thus,both strains were susceptible to IFN-�-plus-LPS-inducedgrowth inhibition in J774.1 macrophages. However, in the ab-

FIG. 5. Intracellular levels of free amino acids in J774.1 macro-phage lysates at 24 h postinfection. The concentration of each aminoacid was determined by HPLC analysis (see Materials and Methods)and then normalized to the total protein content of the lysate and isexpressed as nmol amino acid per mg total protein. (Some relevantPFisher’s PLSD values for arginine levels include the following: restingversus IFN-� plus LPS, P 0.1666; BCG versus AS-1, P 0.8089;BCG versus IFN-� plus LPS, P 0.2842). a.a., amino acids.

FIG. 6. Total urea production by J774.1 macrophages at 24 and 72 hpostinfection. Urea levels were determined by spectrophotometric assaywith ophthalaldehyde and naphthylethylenediamine (see Materials andMethods). The results shown are from two independent experimentsperformed in triplicate for each condition. *, PFisher’s PLSD � 0.05 com-pared to urea production of resting macrophages. (Some relevant P valuesinclude the following. At 24 h: resting versus BCG, P 0.2122; restingversus AS-1, P 0.0636; resting versus IFN-� (IFNg) plus LPS, P �0.0001. At 72 h: resting versus BCG, P 0.0007; resting versus AS-1, P �0.0001; BCG versus AS-1, P 0.0007 [#]; resting versus IFN-� plus LPS,P � 0.0001; resting versus BCG plus IFN-� plus LPS, P � 0.0001.)

FIG. 7. Number of CFU of intracellular BCG and AS-1 fromJ774.1 macrophage lysates at 1 and 72 h postinfection. CFU count wasdetermined on 7H11 agar plates after 2 to 3 weeks. The results shownare the averages of four independent experiments performed in trip-licate for each condition. *, Pt test � 0.05 compared to number of CFUat the initial time point (1 h). (P values include the following: BCG [1h versus 72 h], P 0.4527; BCG plus IFN-� [IFNg] plus LPS [1 hversus 72 h], P 0.0101; AS-1 [1 h versus 72 h], P 0.0321; AS-1 plusIFN-� plus LPS [1 h versus 72 h], P 0.0591.)



EN

N S

TA

TE

UN

IVhttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

sence of IFN-�-plus-LPS activation, AS-1 survived significantlybetter inside resting macrophages than wild-type BCG.

DFMO, an inhibitor of ornithine decarboxylase (ODC), pro-moted mycobacterial survival in resting cells but enhancedmycobacterial growth inhibition in IFN-�-plus-LPS-activatedJ774.1 macrophages. In our survival studies we found thatAS-1 survives better than wild-type BCG inside resting J774.1macrophages. We hypothesized that the altered metabolic ac-tivities in J774.1 macrophages during infection with AS-1 pro-moted intracellular survival of AS-1. We have previouslyshown that there was enhanced macrophage arginase activityin AS-1-infected macrophages compared to BCG-infectedmacrophages, suggesting an increased diversion of L-arginineinto ornithine and urea synthesis. These L-arginine-derivedbiomolecules are important not only for the host macrophage,but also for intracellular mycobacteria.

DFMO was used to inhibit ODC, blocking ornithine turn-over to polyamine synthesis, resulting in buildup of ornithinestores. Ornithine, in turn, inhibits the activity of arginase byfeedback inhibition, and therefore, more arginine and orni-thine accumulate in the cell. ODC is the first enzyme involvedin metabolizing ornithine for polyamine synthesis. ODC activ-ity has also been described in mycobacteria (3, 39), so it wasimportant to treat J774.1 cells with DFMO before infection.To test the effect of changes in arginine metabolism on thegrowth of intracellular mycobacteria, we pretreated the J774.1macrophages with DFMO and then infected the macrophageswith BCG or AS-1 and determined the viable colony counts.

Figure 8 shows viable colony counts of intracellular BCGand AS-1 from J774.1 macrophages that were treated with andwithout DFMO and with DFMO plus IFN-� plus LPS. Treat-ment with DFMO greatly enhanced the survival of both BCGand AS-1, resulting in two- and threefold increases in viablecolony counts at 72 h, respectively (P values of 0.0009 and0.0128, respectively).

However, when infected macrophages were treated withDFMO and IFN-� plus LPS, there was enhanced killing ofintracellular BCG and AS-1 (72% and 77% growth inhibition,respectively). Baydoun et al. (5, 6) showed that treatment ofLPS-activated cells with DFMO, which decreased endogenouslevels of polyamines, potentiated NO synthesis. Therefore, theenhanced mycobacterial growth inhibition in DFMO- andIFN-�-plus-LPS-treated macrophages was probably due to am-plified NO-dependent growth inhibition.

Effects of exogenous L-ornithine and L-norvaline treatmenton intracellular survival of BCG and AS-1 in J774.1 macro-phages. To determine whether ornithine levels affect intracel-lular survival of BCG or AS-1 in J774.1 macrophages, wedetermined the number of intracellular BCG and AS-1 organ-isms that survived in J774.1 macrophages growing in culturemedium supplemented with L-ornithine. Addition of ornithineto macrophage DMEM did not promote mycobacterial growthof either BCG or AS-1 inside J774.1 macrophages (Fig. 9).

Since L-ornithine supplementation to J774.1 macrophagesdid not affect intracellular mycobacterial survival in J774.1macrophages, we used an arginase inhibitor, L-norvaline, toperturb the intracellular levels of arginine and ornithine insideJ774.1 cells. Chang et al. (11) have shown that L-norvalineinhibits arginase activity, reducing urea production by 55% inLPS-activated macrophages. L-Norvaline can directly inhibit

arginase because of its structural similarity to ornithine (30).Ornithine is one of the products of the arginase pathway and ithas been shown to exert feedback inhibition of arginase (8, 38).Inhibition of arginase activity in J774.1 cells has two outcomes:the increase of endogenous arginine levels and the decrease ofendogenous ornithine and urea levels. Therefore, L-norvalinetreatment may indicate which of these biomolecules contrib-utes to mycobacterial survival within J774.1 macrophages. Wepredicted that if arginine is more important for mycobacterialsurvival than ornithine and urea, then intracellular BCG orAS-1 will grow better in L-norvaline-treated cells than in un-treated macrophages. On the other hand, if ornithine and ureacontribute more to intracellular survival of BCG or AS-1, thengrowth of BCG or AS-1 should be inhibited in L-norvaline-treated cells.

Treatment of J774.1 macrophages with L-norvaline signifi-cantly increased the bacterial numbers of both BCG and AS-1organisms by five- and sixfold (P, 0.0005 and 0.0042, respec-tively) after 72 h (Fig. 9). These results strongly suggest thatintracellular survival of mycobacteria might be enhanced by

FIG. 8. CFU counts of BCG (A) and AS-1 (B) from J774.1 mac-rophage lysates at 1 and 72 h postinfection with and without DFMOtreatment (5 mM). *, Pt test � 0.05 compared to number of CFU at theinitial time point (1 h). (P values include the following: BCG [1 hversus 72 h], P 0.3106; BCG plus DFMO [1 h versus 72 h], P 0.0009; BCG plus IFN-� [IFNg] plus LPS plus DFMO [1 h versus 72h], P 0.0002; AS-1 [1 h versus 72 h], P 0.0185; AS-1 plus DFMO[1 h versus 72 h], P 0.0128; AS-1 plus IFN-� plus LPS plus DFMO[1 h versus 72 h], P 0.0138.)



EN

N S

TA

TE

UN

IVhttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

increasing the availability of endogenous arginine inside thehost macrophage.

L-Norvaline was used to inhibit arginase activity to decreaseendogenous ornithine levels inside J774.1 macrophages. Wethen determined whether exogenous L-ornithine abrogates theeffects of L-norvaline treatment on BCG survival within J774.1macrophages. Supplementation of L-ornithine in L-norvaline-treated J774.1 macrophages was synergistic in promoting in-tracellular growth of AS-1 (increased from 6- to 8.7-fold) butwas antagonistic for BCG (decreased from 5- to 3.7-fold).These results suggest that intracellular growth of wild-typeBCG was enhanced by increasing L-arginine availability (nor-valine treatment) but not by increasing L-ornithine availability(ornithine supplementation). In contrast, intracellular growthof AS-1 inside J774.1 macrophages was further enhanced byincreasing the availability of both L-arginine and L-ornithine.Although the only genotypic difference between AS-1 andBCG was that AS-1 was lacking an L-arginine permease en-coded by the Rv0522 gene, it seems that AS-1 has more ca-pacity to metabolize L-ornithine inside J774.1 macrophagesthan does BCG.

DISCUSSION

Nitric oxide production by murine macrophages is essentialfor control of mycobacterial infection, and the control of iNOSexpression is a hallmark of the success or failure of this process(10, 37). The amino acid arginine, the sole substrate of iNOS,might be expected to play an influential role during the mac-rophage’s response to infection. In the present study, we reportthat several macrophage activities related to L-arginine metab-olism were altered in response to M. bovis BCG infection andthat these alterations are related to the transport capacity ofthe infecting BCG strain. Infection with the L-arginine per-mease mutant, AS-1, which lacks the gabP gene (Rv0522),significantly induced L-arginine uptake in J774.1 macrophages.In contrast, infection with wild-type BCG did not alter L-argi-nine uptake in J774.1 cells. Unlike IFN-�-plus-LPS stimula-tion, infection with AS-1 induced L-arginine uptake in J774.1macrophages independent of NO production. The combinedtranscriptional upregulation of the MCAT1 and MCAT2 genesencoding cationic amino acid transporters in AS-1-infectedmacrophages may account for L-arginine uptake greater thanthat of BCG-infected cells. Furthermore, AS-1-infected mac-rophages exhibited greater arginase activity than BCG-infectedcells. These data suggest that in vitro AS-1 infection upregu-lates L-arginine consumption toward L-ornithine and urea for-mation within resting macrophages.

Increased arginine uptake in macrophages would lead toincreased substrate availability not only for macrophage me-tabolism but also for intracellular survival of mycobacteria.Interestingly, we found that J774.1 macrophages have the ca-pacity to maintain high concentrations of free L-arginine eitherfrom exogenous or endogenous sources (Fig. 5). J774.1 mac-rophages that were either activated by IFN-� plus LPS orinfected with BCG or AS-1 showed only slight differences inintracellular arginine levels, suggesting that murine macro-phages maintain homeostatic levels of intracellular arginine.The capacity of macrophages to metabolize arginine via thearginase and iNOS pathways plays an important role in theregulation of macrophage proliferation and functions. The ar-ginase pathway, as part of the host constructive and reparativeprogram, is important for macrophage proliferation and tissuerepair functions. The iNOS pathway, as part of the host de-fense immune response, is key to the production of NO, apotent antimycobacterial agent and signaling molecule (10,37). Granger et al. (23) reported that 96% of total L-arginineconsumed by resident peritoneal macrophages is utilizedthrough the arginase pathway. The iNOS pathway utilized only0.7% of total L-arginine in resting macrophages and 30% inactivated cells. Consistent with their study, we found that rest-ing J774.1 macrophages exhibited high arginase activity andalmost negligible NO synthesis. Activation of J774.1 cells withIFN-� plus LPS not only increased NO production by 21-foldbut also increased arginase activity by 3- to 4-fold (Fig. 6).

In resting J774.1 macrophages, intracellular AS-1 survivedbetter than intracellular BCG (Fig. 7), perhaps because AS-1infection induced greater arginase activity than BCG infection.In IFN-�-plus-LPS-activated cells, although infection with ei-ther BCG or AS-1 attenuated iNOS protein expression andNO production and enhanced arginase activity, both strainsfailed to multiply intracellularly within these cells, perhaps

FIG. 9. CFU counts of BCG (A) and AS-1 (B) from J774.1 mac-rophage lysates at 1 and 72 h postinfection with and without L-norva-line (Nval) treatment (10 mM) and/or L-ornithine (Orn) supplemen-tation (1 mM). *, Pt test � 0.05 compared to number of CFU at theinitial time point (1 h). (P values include the following: BCG plus Orn[1 h versus 72 h], P 0.2002; BCG plus Nval [1 h versus 72 h], P 0.0005; BCG plus Orn plus Nval [1 h versus 72 h], P 0.0097; AS-1plus Orn [1 h versus 72 h], P 0.7655; AS-1 plus Nval [1 h versus 72h], P 0.0042; AS-1 plus Orn plus Nval [1 h versus 72 h], P 0.0188.)



EN

N S

TA

TE

UN

IVhttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

because the small amount of NO expressed is sufficient toinhibit the growth of intracellular mycobacteria or becauseother host defense mechanisms (production of reactive oxygenintermediates, phagosome-lysosome fusion, induction of theLRG-47 pathway, or apoptosis) have also been activated.

For other pathogens, such as Trypanosoma cruzi, Leishmaniaspp., and Schistosoma mansoni, the induction of host arginaseactivity promotes intracellular survival, not only by attenuatingthe NO-mediated immune response mounted against them,but also by providing these pathogens with polyamines, whichare essential for their growth (1, 20, 28). Interestingly, in T.cruzi (45) and in Leishmania major (33), pretreatment ofJ774.1 cells with N-omega-hydroxy-L-arginine, an arginase in-hibitor, resulted in a dramatic decrease in the number of in-tracellular amastigotes. In contrast, intracellular BCG andAS-1 increased when macrophage arginase activity was inhib-ited by L-norvaline (Fig. 9). Furthermore, when macrophageODC activity (or polyamine synthesis) was inhibited by DFMO,intracellular growth of both BCG and AS-1 was also enhanced(Fig. 8). These results indicate that intracellular growth of BCG,unlike that of the parasitic pathogens, depends more on the avail-ability of L-arginine and L-ornithine than on polyamines inside themacrophages, as found with the parasites.

The enhanced ability of the AS-1 strain to survive intracel-lularly within resting J774.1 macrophages may be in part due toits increased capacity to accumulate L-arginine and/or L-argi-nine-derived substrates while growing within J774.1 macro-phages compared to BCG (41). Since the arginase pathway isthe major L-arginine catabolic pathway in resting macrophages,the arginase by-products, L-ornithine and urea, are the majorL-arginine-derived biomolecules that intracellular mycobacte-ria are likely to accumulate inside J774.1 macrophages. Be-cause AS-1 is defective in transporting the cationic amino acid

L-arginine but not L-ornithine (44), perhaps L-ornithine may bepreferentially taken up by intracellular AS-1.

Several in vitro observations support our hypothesis thatAS-1 has an increased capacity to metabolize ornithine intra-cellularly compared to BCG. First, AS-1 infection inducedgreater arginase activity in J774.1 macrophages than BCG in-fection. This result suggests that there is a greater metabolicdemand for L-ornithine and/or urea in AS-1-infected macro-phages than in BCG-infected cells. Note that the macrophageculture medium, DMEM, does not contain L-ornithine, andtherefore the only source of L-ornithine for J774.1 macro-phages is through exogenous L-arginine metabolized via thearginase pathway. We hypothesize that to replenish the intra-cellular ornithine pools that have been diminished by intracel-lular AS-1, macrophages take up more L-arginine from theextracellular medium, which can be utilized through the argi-nase pathway.

Next, when L-norvaline-treated macrophages were supple-mented with L-ornithine, intracellular growth of AS-1 increasedby 8.7-fold while that of BCG only increased by 3.7-fold at 72 hpostinfection. These results suggest that the intracellular growthof AS-1 in J774.1 macrophages can be further enhanced by in-creasing the availability of ornithine, while survival of BCGdoes not dramatically change in response to increasing or-nithine availability under conditions in which arginine avail-ability is already high.

L-Ornithine can be utilized by intracellular mycobacteria formetabolic processes, such as arginine biosynthesis and poly-amine and proline synthesis, that will support mycobacterialgrowth. The substrate demand for these ornithine-dependentactivities is perhaps supported by other cationic amino acidpermeases. There are several putative cationic amino acid per-meases (encoded by Rv2320c, Rv1999c, and Rv3253c) that

FIG. 10. Proposed model for arginine homeostasis in BCG- or AS-1-infected macrophages. AS-1 infection induced L-arginine uptake in J774.1macrophages (independent of the iNOS pathway) to replenish the L-arginine (and its derived substrates) that was used up by intracellular AS-1.In contrast to BCG-infected macrophages, increased macrophage L-arginine transport in AS-1-infected macrophages correlated with upregulationof the MCAT1 and MCAT2 genes encoding cationic amino acid transporters and increased arginase activity. The altered L-arginine metabolic stateof infected J774.1 macrophages allowed survival and multiplication of intracellular AS-1. WT, wild-type.



EN

N S

TA

TE

UN

IVhttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

may compensate for loss of Rv0522-mediated transport inAS-1, and the best candidate is the permease encoded by theRv2320c or rocE gene. Based on protein homology studies(13), Rv2320c encodes a putative cationic amino acid trans-porter with possible affinity for the substrates arginine andornithine. Interestingly, unlike the Rv0522 gene, Rv2320c isfound within an operon, rocD1D2E (Rv2320-Rv2322c), withtwo rocD gene homologs, which encode ornithine aminotrans-ferases. Such a genetic arrangement suggests that this trans-porter, since it is coexpressed with the ornithine aminotrans-ferases, mediates the uptake of their substrate, ornithine.Preliminary studies in our laboratory using promoter fusionshave suggested that the promoter activity of Rv2320c did notchange in the AS-1 mutant during the infection of J774.1macrophages (40). However, we cannot exclude the possibilitythat the activity of this permease is modulated at the posttran-scriptional level. Our laboratory is now in the process of cre-ating a Rv2320c deletion mutant strain in both M. bovis BCGand the virulent lab strain M. tuberculosis H37Rv, which willallow us to characterize the specificity and affinity of the mu-tant strains for transporting arginine, ornithine, and other re-lated amino acids. The strains will also allow us to explorefurther whether ornithine is a major substrate that is requiredfor intracellular survival in macrophages.

Finally, these studies describe an intimate relationship be-tween intracellular mycobacteria and J774.1 macrophages withrespect to metabolic currency in the infection process. Theresults shed light on the complexities inherent in the host-pathogen interaction quite apart from those processes identi-fied by classical searches for virulence factors. Figure 10 showsa more detailed model of how macrophages maintain argininehomeostasis during mycobacterial infection. It summarizes thechanges in L-arginine metabolism of murine macrophages in-fected with either BCG or AS-1. Our studies with this argininetransport mutant have unmasked the importance of L-orni-thine in the survival of intracellular mycobacteria, especiallyunder conditions when the bioavailability of L-arginine withinthe host macrophages is limiting, and thus, access to ornithinemay be a regulatory approach to the control of mycobacterialintracellular growth.

ACKNOWLEDGMENTS

This work was supported by NIH grant RO1A134436 and AmericanHealth Association grant 033530T.

We thank Renee Robinson, Liam McCallum, and William Laneat the Harvard Microchemistry facility for the amino acid analysis.We also thank Vivian Bellofatto, Grant Gallagher, Yaswant KumarDayaram, Jessica Mann, and John David Muth for valuable discussionsand useful advice during the course of this study and for help inpreparing the manuscript.

REFERENCES

1. Abdallahi, O. M., H. Bensalem, R. Augier, M. Diagana, M. De Reggi, and B.Gharib. 2001. Arginase expression in peritoneal macrophages and increasein circulating polyamine levels in mice infected with Schistosoma mansoni.Cell. Mol. Life Sci. 58:1350–1357.

2. Aulak, K. S., R. Mishra, L. Zhou, S. L. Hyatt, W. de Jounge, W. Lamers, M.Snider, and M. Hatzoglou. 1999. Post-transcriptional regulation of the argi-nine transporter Cat-1 by amino acid availability. J. Biol. Chem. 274:30424–30432.

3. Balasundaram, D., and A. K. Tyagi. 1988. Regulation of ornithine de-carboxylase from Mycobacterium smegmatis. Arch. Biochem. Biophys.264:288–294.

4. Barker, L. P., D. M. Brooks, and P. L. C. Small. 1998. The identification of

Mycobacterium marinum genes differentially expressed in macrophagephagosomes using promoter fusions to green fluorescent protein. Mol. Mi-crobiol. 29:1167–1177.

5. Baydoun, A., and D. Morgan. 1998. Inhibition of ornithine decarboxylasepotentiates nitric oxide production in LPS-activated J774 cells. Br. J. Phar-macol. 125:1511–1516.

6. Baydoun, A. R., R. G. Bogle, J. D. Pearson, and G. E. Mann. 1993. Arginineuptake and metabolism in cultured murine macrophages. Agents Actions38(Suppl. 2):C127–C129.

7. Bogle, R. G., A. R. Baydoun, J. D. Pearson, S. Moncada, and G. E. Mann.1992. L-arginine transport is increased in macrophages generating nitricoxide. Biochem. J. 284:15–18.

8. Bommarius, A. S., and K. Drauz. 1994. An enzymatic route to L-ornithinefrom arginine—activation, selectivity and stabilization of L-arginase. Bioorg.Med. Chem. 2:617–626.

9. Bussiere, F. I., R. Chaturvedi, Y. Cheng, A. P. Gobert, M. Asim, D. R.Blumberg, H. Xu, P. Y. Kim, A. Hacker, R. A. Casero, Jr., and K. T. Wilson.2005. Spermine causes loss of innate immune response to Helicobacter pyloriby inhibition of inducible nitric-oxide synthase translation. J. Biol. Chem.280:2409–2412.

10. Chan, E. D., J. Chan, and N. W. Schluger. 2001. What is the role of nitricoxide in murine and human host defense against tuberculosis? Currentknowledge. Am. J. Respir. Cell Mol. Biol. 25:606–612.

11. Chang, C.-I., J. C. Liao, and L. Kuo. 1998. Arginase modulates nitric oxideproduction in activated macrophages. Am. J. Physiol. Heart Circ. Physiol.274:H342–H348.

12. Closs, E. I., L. M. Albritton, J. W. Kim, and J. M. Cunningham. 1993.Identification of a low affinity, high capacity transporter of cationic aminoacids in mouse liver. J. Biol. Chem. 268:7538–7544.

13. Cole, S. T., R. Brosch, J. Parkhill, T. Garnier, C. Churcher, D. Harris, S. V.Gordon, K. Eiglmeier, S. Gas, C. E. Barry III, F. Tekaia, K. Badcock, D.Basham, D. Brown, T. Chillingworth, R. Connor, R. Davies, K. Devlin, T.Feltwell, S. Gentles, N. Hamlin, S. Holroyd, T. Hornsby, K. Jagels, B. G.Barrell, et. al. 1998. Deciphering the biology of Mycobacterium tuberculosisfrom the complete genome sequence. Nature 393:537–544.

14. Deves, R., and C. A. Boyd. 1998. Transporters for cationic amino acids inanimal cells: discovery, structure, and function. Physiol. Rev. 78:487–545.

15. Eckmann, L., F. Laurent, T. D. Langford, M. L. Hetsko, J. R. Smith, M. F.Kagnoff, and F. D. Gillin. 2000. Nitric oxide production by human intestinalepithelial cells and competition for arginine as potential determinants of hostdefense against the lumen-dwelling pathogen Giardia lamblia. J. Immunol.164:1478–1487.

16. Fiddler, R. N. 1977. Collaborative study of modified AOAC method ofanalysis for nitrite in meat and meat products. J. Assoc. Off. Anal. Chem.60:594–599.

17. Finley, K. D., D. K. Kakuda, A. Barrieux, J. Kleeman, P. D. Huynh, and C. L.MacLeod. 1995. A mammalian arginine/lysine transporter uses multiple pro-moters. Proc. Natl. Acad. Sci. USA 92:9378–9382.

18. Fisher, G. H., I. Arias, I. Quesada, S. D’Aniello, F. Errico, M. M. Di Fiore,and A. D’Aniello. 2001. A fast and sensitive method for measuring picomolelevels of total free amino acids in very small amounts of biological samples.Amino Acids 20:163–173.

19. Fortune, S. M., A. Solache, A. Jaeger, P. J. Hill, J. T. Belisle, B. R. Bloom,E. J. Rubin, and J. D. Ernst. 2004. Mycobacterium tuberculosis inhibitsmacrophage responses to IFN-� through myeloid differentiation factor 88-dependent and -independent mechanisms. J. Immunol. 172:6272–6280.

20. Gobert, A. P., S. Daulouede, M. Lepoivre, J. L. Boucher, B. Bouteille, A.Buguet, R. Cespuglio, B. Veyret, and P. Vincendeau. 2000. L-Arginine avail-ability modulates local nitric oxide production and parasite killing in exper-imental trypanosomiasis. Infect. Immun. 68:4653–4657.

21. Gobert, A. P., Y. Cheng, J.-Y. Wang, J.-L. Boucher, R. K. Iyer, S. D.Cederbaum, R. A. Casero, Jr., J. C. Newton, and K. T. Wilson. 2002. Heli-cobacter pylori induces macrophage apoptosis by activation of arginase II.J. Immunol. 168:4692–4700.

22. Gordhan, B. G., D. A. Smith, H. Alderton, R. A. McAdam, G. J. Bancroft,and V. Mizrahi. 2002. Construction and phenotypic characterization of anauxotrophic mutant of Mycobacterium tuberculosis defective in L-argininebiosynthesis. Infect. Immun. 70:3080–3084.

23. Granger, D. L., J. B. Hibbs, Jr., J. R. Perfect, and D. T. Durack. 1990.Metabolic fate of L-arginine in relation to microbiostatic capability of murinemacrophages. J. Clin. Investig. 85:264–273.

24. Hazbon, M. H., and D. Alland. 2004. Hairpin primers for simplified single-nucleotide polymorphism analysis of Mycobacterium tuberculosis and otherorganisms. J. Clin. Microbiol. 42:1236–1242.

25. Hibbs, J. B., Jr. 2002. Infection and nitric oxide. J. Infect. Dis. 185(Suppl.1):S9–S17.

26. Hrabak, A., M. Idei, and A. Temesi. 1994. Arginine supply for nitric oxidesynthesis and arginase is mainly exogenous in elicited murine and rat mac-rophages. Life Sci. 55:797–805.

27. Hrabak, A., T. Bajor, and A. Temesi. 1996. Computer-aided comparison ofthe inhibition of arginase and nitric oxide synthase in macrophages by aminoacids not related to arginine. Comp. Biochem. Physiol. Part B 113:375–381.



EN

N S

TA

TE

UN

IVhttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

28. Iniesta, V., L. C. Gomez-Nieto, and I. Corraliza. 2001. The inhibition ofarginase by N-hydroxy-L-arginine controls the growth of Leishmania insidemacrophages. J. Exp. Med. 193:777–783.

29. Jung, D., H. Biggs, J. Erikson, and P. U. Ledyard. 1975. New colorimetricreaction for end-point, continuous-flow, and kinetic measurement of urea.Clin. Chem. 21:1136–1140.

30. Kakuda, D. K., K. D. Finley, M. Maruyama, and C. L. MacLeod. 1998. Stressdifferentially induces cationic amino acid transporter gene expression. Bio-chim. Biophys. Acta 1414:75–84.

31. Kakuda, D. K., M. J. Sweet, C. L. MacLeod, D. A. Hume, and D. Markovich.1999. CAT2-mediated L-arginine transport and nitric oxide production inactivated macrophages. Biochem. J. 340:549–553.

32. Klarsfeld, A. D., P. L. Goossens, and P. Cossart. 1994. Five Listeria mono-cytogenes genes preferentially expressed in infected mammalian cells: plcA,purH, purD, pyrE and an arginine ABC transporter gene, arpJ. Mol. Micro-biol. 13:585–597.

33. Kropf, P., J. M. Fuentes, E. Fahnrich, L. Arpa, S. Herath, V. Weber, G.Soler, A. Celada, M. Modolell, and I. Muller. 2005. Arginase and polyaminesynthesis are key factors in the regulation of experimental leishmaniasis invivo. FASEB J. 19:1000–1002.

34. MacLeod, C. L., K. D. Finley, and D. K. Kakuda. 1994. y(�)-type cationicamino acid transport: expression and regulation of the MCAT genes. J. Exp.Biol. 196:109–121.

35. Mills, C. D. 2001. Macrophage arginine metabolism to ornithine/urea ornitric oxide/citrulline: a life or death issue. Crit. Rev. Immunol. 21:399–425.

36. Modolell, M., I. M. Corraliza, F. Link, G. Soler, and K. Eichmann. 1995.Reciprocal regulation of the nitric oxide synthase/arginase balance in mousebone marrow-derived macrophages by TH1 and TH2 cytokines. Eur. J. Im-munol. 25:1101–1104.

37. Nathan, C., and M. U. Shiloh. 2000. Reactive oxygen and nitrogen interme-

diates in the relationship between mammalian hosts and microbial patho-gens. Proc. Natl. Acad. Sci. USA 97:8841–8848.

38. Pace, C. N., and R. A. Landers. 1981. Arginase inhibition. Biochim. Biophys.Acta 658:410–412.

39. Paulin, L., E. Brander, and H. Poso. 1987. Ornithine decarboxylase, S-adenosyl-L-methionine decarboxylase and arginine decarboxylase from My-cobacterium bovis (BCG). Experientia 43:174–176.

40. Peteroy-Kelly, M., V. Venketaraman, and N. D. Connell. 2001. Effects ofMycobacterium bovis BCG infection on regulation of L-arginine uptake andsynthesis of reactive nitrogen intermediates in J774.1 murine macrophages.Infect. Immun. 69:5823–5831.

41. Peteroy-Kelly, M. A., V. Venketaraman, M. Talaue, A. Seth, and N. D.Connell. 2003. Modulation of J774.1 macrophage L-arginine metabolism byintracellular Mycobacterium bovis BCG. Infect. Immun. 71:1011–1015.

42. Salimuddin, A. Nagasaki, T. Gotoh, H. Isobe, and M. Mori. 1999. Regulationof the genes for arginase isoforms and related enzymes in mouse macro-phages by lipopolysaccharide. Am. J. Physiol. 277:E110–E117.

43. Sato, H., M. Fujiwara, and S. Bannai. 1992. Effect of lipopolysaccharide ontransport and metabolism in mouse peritoneal macrophages. J. Leukoc. Biol.52:161–164.

44. Seth, A., and N. D. Connell. 2000. Amino acid transport and metabolism inmycobacteria: cloning, interruption, and characterization of an L-arginine/�-aminobutyric acid permease in Mycobacterium bovis BCG. J. Bacteriol. 182:919–927.

45. Stempin, C. C., T. B. Tanos, O. A. Coso, and F. M. Cerban. 2004. Arginaseinduction promotes Trypanosoma cruzi intracellular replication in cruzipain-treated J774 cells through the activation of multiple signaling pathways. Eur.J. Immunol. 34:200–209.

46. Venketaraman, V., M. T. Talaue, Y. K. Dayaram, M. A. Peteroy-Kelly, W. Bu,and N. D. Connell. 2003. Nitric oxide regulation of L-arginine uptake inmurine and human macrophages. Tuberculosis 83:311–318.



EN

N S

TA

TE

UN

IVhttp://jb.asm

.org/D

ownloaded from

http://jb.asm.org/

BCG Infection Mycobacterium bovis Macrophages in the Context of ...

Documents

Transcript of BCG Infection Mycobacterium bovis Macrophages in the Context of ...