Published online before print December 15, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Cambridge Institute for Medical Research and Department of Medicine, Wellcome Trust/MRC Building, University of Cambridge School of Clinical Medicine, Addenbrookes Hospital, Hills Road, United Kingdom; and
Centre for Veterinary Science, Department of Clinical Veterinary Medicine, University of Cambridge, United Kingdom
3 Correspondence: Cambridge Institute for Medical Research, Wellcome Trust/MRC Building, University of Cambridge School of Clinical Medicine, Addenbrookes Hospital, Hills Road, Cambridge CB2 2XY, UK. E-mail: jennie.blackwell{at}cimr.cam.ac.uk
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: Nramp1 • divalent cation transporter • macrophage activation
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) [5
], interleukin-1ß [6
], and tumor necrosis factor [7
] at protein and/or mRNA levels, as well as iron- [8
] and interferon-
/lipopolysaccharide [8
, 9
]-regulated differences in expression and stability of mRNA for inducible nitric oxide synthase (iNOS) encoded by Nos2A. Differences in NADPH oxidase activity have also been observed in response to phorbol myristate acetate in RAW264.7 macrophages, stably transfected with constructs carrying wild-type or mutant (mt) Slc11a1 [10
]. In vitro, studies of iNOS induction and use of inhibitors have shown that NO generated by iNOS activity is responsible for macrophage antimicrobial activity against numerous intracellular pathogens (reviewed in refs. [11
12
13
]), including Salmonella enterica serovar Typhimurium [14
], Mycobacterium bovis [15
], and Leishmania donovani [16
], which come under innate Slc11a1 control in the early phases of infection. It was assumed, therefore, that iNOS activity might contribute to early Slc11a1-mediated resistance to infection. Similarly, reactive oxygen intermediates (ROI), generated by NADPH oxidase, have antimicrobial activity against all three groups of the pathogen in vitro [17
18
19
], and it had been proposed [20
] that the interaction of ferrous ions (Fe2+) transported by Slc11a1 with ROI via the Fenton and Haber-Weiss reactions might contribute to the generation of more toxic radicals in Slc11a1 wild-type but not mt macrophages. By intercrossing mice with gene disruptions at Nos2A and Cybb (encoding gp91phox, the heavy chain subunit of cytochrome b-245 and an essential component of phagocyte NADPH oxidase) onto equivalent Slc11a1 wild-type and mt genetic backgrounds, we demonstrate that neither iNOS nor gp91phox activity is required for Slc11a1-mediated innate resistance to S. enterica serovar Typhimurium or L. donovani infection. We show that functional gp91phox and iNOS activities are required for non-Slc11a1-regulated early and late phase of S. enterica serovar Typhimurium infection, respectively. However, only an organ-specific requirement for iNOS was observed in clearance of parasites from spleen but not liver in the later, non-Slc11a1-regulated phases of L. donovani infection. This contrasts with rapid lesion growth and death in Nos2A–/– knockout mice following Leishmania major infection, with some exacerbation of disease also observed with Cybb–/– gene disruption. This highlights the adaptive differences in tissue and cellular tropisms between L. donovani and L. major and the different genes and mechanisms that regulate visceral versus cutaneous forms of the disease. |
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
10 Mb interval distal to D1Mit80 and proximal to D1Mit44 on chromosome 1, which carries the Slc11a1 gene. To produce new lines that were homozygous Slc11a1 wild-type (=Slc11a1+/+) but carried gene disruptions at Nos2A (=Nos2A–/–) or Cybb (=Cybb–/–), founder Nos2A–/– [22
] and Cybb–/– [23
] were intercrossed with N20 mice. The Nos2A–/– produced by Laubach and co-workers [22
] was reported to have undergone six generations of backcrossing onto a C57BL/6J (=B6) background prior to our acquiring it. As B6 mice are Slc11a1 mt, we expected to obtain double heterozygous mice for Slc11a1+/mt and Nos2A+/– at the F1 of the first intercross. In the event, all mice were homozygous Slc11a1+/+, indicating that the region of chromosome 1 carrying Slc11a1 in the original line [19
] was still homozygous for the 129 background in which the Nos2A knockout was generated. These F1 mice were further intercrossed, and Nos2A+/+ and Nos2A–/– mice were selected to establish breeding colonies (by brother-sister matings of Nos2A+/+xNos2A+/+ or Nos2A–/–xNos2A–/– F2 progeny from the same F1 intercross) of double homozygous Slc11a1+/+.Nos2A+/+ and Slc11a1+/+.Nos2A–/– on the intercrossed B6/B10 background at F3. The founder Cybb–/– mice were also on a B6 background, and in this case, the expected double heterozygous mice for Slc11a1+/mt and Cybb+/– were obtained at the F1 of the first intercross. Slc11a1 and Cybb genotyping was performed using tail DNA from all mice and selective breeding used to establish double homozygous Slc11a1+/+.Cybb+/+ and Slc11a1+/+.Cybb–/– lines at F3 on the intercrossed B6/B10 background. For both sets of mice, infections were compared with Slc11a1mt/mt, Nos2A+/+, and Cybb+/+ B10 or B6 mice as indicated. BALB/c mice (Harlen Olac, Bicester, UK) were used as susceptible controls for L. major infection experiments. All breeding and procedures were carried out under license and UK Government Home Office regulations. Mice were between 6 and 16 weeks old over all infection experiments but age-matched ±2 weeks and sex-matched within each experiment as indicated below.
Genotyping
During the breeding protocols, the Nos2A–/– gene-disrupted allele for Nos2A was genotyped on BamHI-digested, genomic DNA Southern blots hybridized with a Nos2A intron 10 probe amplified by polymerase chain reaction (PCR) from genomic DNA using forward (5'-CTGGACCAGGCCACTTTG-3') and reverse (5'-CCTCCACTTGTCCATCCATT-3') primers. Diagnostic bands of 14 kb and 5 kb distinguished Nos2A wild-type from gene-disrupted alleles, respectively. Similarly, the Cybb–/– gene-disrupted allele for Cybb was genotyped on EcoRI-digested, genomic DNA Southern blots, hybridized with a Cybb exon 2 probe, amplified by PCR from genomic DNA using forward (5'-CAAGATTGCATGAGGCAGAA-3') and reverse (5'-CAACTTTTCCTGTTTGTGCAG-3') primers. Diagnostic bands of 13 kb and 6.8 kb distinguished Cybb wild-type from gene-disrupted alleles, respectively. Slc11a1 genotyping was undertaken by sequence analysis across the mutation site following amplification of a 492-bp product from genomic DNA using exon 5 forward (5'-ATATGCAGGAAGTCATCGGG-3') and intron 6 reverse (5'-CAAGATGAGGGGCTTTCTCTC-3') primers, with the forward primer used as the sequencing primer.
Salmonella infection and monitoring
For intravenous (i.v.) inoculation into mice, virulent S. enterica serovar Typhimurium C5 was grown at 37°C as stationary, overnight cultures in Luria-Bertani (LB) broth (Difco, Becton Dickinson, Cowley, Oxford, UK). Aliquots were snap-frozen and stored in liquid nitrogen. The inoculum was diluted in phosphate-buffered saline (PBS) and injected into the lateral tail vein. The number of viable bacteria in each inoculum was checked by dilution and pour-plating onto LB agar plates. To enumerate bacterial load in the organs, spleens and livers of mice (four per time-point per strain) were removed aseptically and homogenized in a Colworth Stomacher in 10 ml cold, distilled water [24
]. Viable counts were determined using pour plates of LB agar. For 50% lethal dose (LD50) determinations, groups of mice (four to six per dose per strain) were injected i.v. with 10- or 20-fold decreasing doses of S. enterica serovar Typhimurium C5 over the range 105–101, and mortality scored over a 30-day period. LD50 values were calculated according to the method of Reed and Muench [25
]. Age-matched (±2 weeks) male mice were used for S. enterica serovar Typhimurium experiments as indicated.
Leishmania infection and monitoring
To determine response to cutaneous leishmaniasis, mice (five to six per strain) were inoculated subcutaneously (s.c.) into the right-hind footpad with 2 x 106 stationary-phase, metacyclic promastigotes of L. major strain LV39, cultured in Schneider’s insect medium (Sigma, Poole, UK), supplemented with 10% fetal calf serum (Invitrogen, Paisley, UK) and 100 U/ml penicillin/100 µg/ml streptomycin (Invitrogen). Infection was monitored as increase in footpad depth, measured using digital Vernier callipers. To conform to UK Home Office regulations, euthanasia of mice was carried out when footpad lesions became ulcerated and necrotic. Experience shows that this level of pathology is always lethal. To determine response to visceral leishmaniasis, mice (three to four per time-point per strain) were inoculated i.v. into the lateral tail vein with 2 x 107 or 5 x 107 amastigotes freshly harvested from spleens of hamsters infected with L. donovani strain LV9 as described [26
]. Groups of mice were killed on days 1, 4, 7, and 15 and 50 days post-infection as indicated. Parasite loads were enumerated microscopically from KWIKTMDIFF (Thermo Shandon, Pittsburgh, PA)-stained impression smears of spleen and liver. Results are expressed as Leishman Donovan units (LDU) [26
], which equate to the number of parasites per 1000 organ cell nuclei multiplied by the organ weight in mg. Age-matched (±2 weeks), female mice were used for Leishmania experiments as indicated.
Immunohistology
Liver segments from N20 mice harvested 2–4 days post L. donovani infection were mounted in optimal cutting temperature compound (CellPath, Powys, UK) and snap-frozen. For low-power fluorescence microscopy, cryostat sections (8 µm thick) were warmed to room temperature, fixed for 10 min in 4% w/v paraformaldehyde (BDH, Poole, UK) in PBS (pH 7.4), and washed 2x in PBS for 5 min. Sections were post-fixed by immersion in 2:1 ethanol:acetic acid at 4°C and washed 2x in PBS for 5 min. Nonspecific staining was reduced by incubation for 1 h with 3% fatty acid-free bovine serum albumin in Triton–Tris-buffered saline (154 mM NaCl, 10 mM Tris, 0.01% sodium azide, 0.1% Triton X-100, pH 7.4) prior to incubation with pooled, primary monoclonal antibody (Ab) rat anti-mouse M1/70 Ab (Harlan Sera-Lab, Loughborough, UK, 1:10) and affinity-purified polyclonal rabbit anti-iNOS Ab (Affiniti Research Products, Exeter, UK, 1:500). M1/70 recognizes CD11b, the ß-chain of the type-3 complement receptor [27
]. The anti-iNOS Ab was raised against a 21-kDa protein fragment corresponding to amino acids 961–1144 of the mouse macrophage iNOS protein [28
] and detects the 130-kDa full-length protein in macrophages from Nos2A+/+ mice, which was completely absent in macrophages from Nos2A–/– mice (data not shown). After washing, slides were coincubated in fluorescein isothiocyanate (FITC)-conjugated, affinity-purified donkey anti-rat immunoglobulin G (IgG; Jackson Immuno Research Laboratories, West Grove, PA, 1:50) and affinitiy-purified tetramethyl rhodamine isothiocyanate (TRITC)-conjugated goat anti-rabbit IgG (Sigma, 1:20) for 5 h at ambient temperature. Sections were washed 3x in PBS, rinsed in distilled water, and mounted in Citifluor AFI (Agar Scientific, Kent, UK). Staining was viewed with an Olympus Provis AX70 fluorescent microscope (London, UK), mounted with a Hamamatsu C47-42-95 digital camera (Photonics KK, Japan), and images were captured with Openlab software. Exposures for FITC and TRITC channels varied between 75 and 400 ms. The microscope, camera, and printer settings were standardized within experiments for stained and control sections, all of which had been incubated in parallel. For high-power confocal microscopy, 6 µm cryostat sections were incubated in primary Ab simultaneously. The first primary Ab was an affinity-purified polyclonal anti-Slc11a1 Ab (1:1000) raised against the N-terminal 82 amino acids of the Slc11a1 protein [29
] and was a gift from Dr. C. Howard Barton (University of Southampton, UK). This Ab recognizes the mature, fully glycosylated, 90–100 kDA Slc11a1 band on Western blots of Slc11a1 wild-type macrophages [30
]. The second primary Ab was M1/70 (1:800) or monoclonal rat anti-mouse F4/80 (Serotec, Oxford, UK, 1:50). Secondary Ab were AlexaTM594-coupled goat anti-rat IgG (1:200) and AlexaTM488-coupled goat anti-rabbit IgG (1:400, Molecular Probes, Leiden, Netherlands). Slides with cryostat sections were fixed for 10 min at 4°C with 100% acetone, air-dried, and then rinsed in PBS and blocked for 1 h at ambient temperature in 0.1% Tween-20/PBS supplemented with normal goat serum (1:40). Primary Ab, diluted in blocking buffer, were incubated for 1 h at ambient temperature. Slides were washed 3x for 5 min in 0.1% Tween-20/PBS. Secondary Ab, diluted in blocking buffer, were incubated for 1 h at ambient temperature. Slides were washed as above, rinsed in PBS, and then mounted in Prolong Antifade (Molecular Probes) and viewed with a Nikon optiphot-2 epifluorescence microscope coupled to a Bio-Rad MRC 1000 confocal laser-scanning attachment (Bio-Rad Laboratories Ltd., Hemel Hempsted, UK). Images were collected using Lasersharp 2000 software.
Statistics
All footpad measurements and parasite loads were determined blind to the experimental group. Unpaired, two-tailed Student’s t-tests were used to compare bacterial or parasite loads between mouse strains. Differences between experimental groups were considered significant for P values <0.05.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (18K): [in a new window] |
Figure 1. S. enterica serovar Typhimurium infection in B6/B10 Slc11a1+/+. Nos2A+/+ and B6/B10 Slc11a1+/+.Nos2A–/– compared with B10 Slc11a1mt/mt. Nos2A+/+ mice. Mice (four to six per dose per strain) were infected with tenfold decreasing doses (104, 103, 102, 101) of S. enterica serovar Typhimurium C5. (A) Percent survival over 30 days following inoculation of four to five mice per strain with 103 S. enterica serovar Typhimurium C5. The same rank order in mortality curves was observed for the other bacterial doses. Bacterial counts (mean±SEM; four mice per strain per time-point) in liver (B) and spleen (C) over 8 days post-infection are shown for the 103 inoculum. P values for significant differences between strains are indicated in the text. Similar results were obtained for organ bacterial loads in a replicate experiment.
|
 View larger version (21K): [in a new window] |
Figure 2. L. donovani infection in B6/B10 Slc11a1+/+.Nos2A+/+ and B6/B10 Slc11a1+/+.Nos2A–/– compared with B10 Slc11a1mt/mt.Nos2A+/+ mice. Parasite loads (LDU; mean±SEM; three or four mice per strain per time-point) were measured in liver (A) and spleen (B) at intervals over 50 days post-infection with 5 x 107 L. donovani LV9 amastigotes. P values for significant differences between strains are indicated in the text. Similar results were obtained for organ parasite loads in a replicate experiment.
|
 View larger version (25K): [in a new window] |
Figure 3. L. major infection in B6/B10 Slc11a1+/+.Nos2A+/+ and B6/B10 Slc11a1+/+.Nos2A–/– compared with BALB/c mice. Infection was measured as footpad depth (mean±SEM; five or six mice per strain) at intervals over 86 days post-infection with 2 x 106 L. major LV39 metacyclic promastigotes. The inset shows parallel data for congenic B10 (Slc11a1mt/mt.Nos2A+/+) and N20 (Slc11a1+/+.Nos2A+/+) mice. P values for significant differences between strains are indicated in the text. This experiment replicated previous data [38
] and was not repeated.
|
 View larger version (22K): [in a new window] |
Figure 4. S. enterica serovar Typhimurium infection in B6/B10 Slc11a1+/+. Cybb+/+ and B6/B10 Slc11a1+/+.Cybb–/– compared with B6 Slc11a1mt/mt. Cybb+/+ and B6 Slc11a1mt/mt.Cybb–/– mice (four to six per dose per strain), which were infected with 20-fold decreasing doses (8x104, 4x103, 2x102, 101) of S. enterica serovar Typhimurium C5. (A) Percent survival over 30 days following inoculation of 101 S. enterica serovar Typhimurium C5. The same rank order in mortality curves was observed for the 2 x 102 bacterial dose. Higher doses could not distinguish rapid mortality rates between the two Cybb–/– strains. Bacterial counts (mean±SEM; four mice per strain per time-point) in liver (B) and spleen (C) are shown over 6 days post-infection with 1.5 x 102 S. enterica serovar Typhimurium C5. (B and C) Inset graphs show the same data plotted as percent of the day 1 count for each mouse strain. P values for significant differences between strains are indicated in the text. Similar results were obtained for organ bacterial loads in a replicate experiment.
|
 View larger version (16K): [in a new window] |
Figure 5. L. donovani infection in B6/B10 Slc11a1+/+.Cybb+/+ and B6/B10 Slc11a1+/+.Cybb–/– compared with B6 Slc11a1mt/mt.Cybb+/+ and B6 Slc11a1mt/mt. Cybb–/– mice. Parasite loads (LDU; mean±SEM; four mice per strain per time-point) were measured in liver (A) and spleen (B) at intervals over 50 days post-infection with 2 x 107 L. donovani LV9 amastigotes. P values for significant differences between strains are indicated in the text. Similar results were obtained for organ parasite loads in a replicate experiment.
|
 View larger version (13K): [in a new window] |
Figure 6. L. major infection in (A) B6 Slc11a1mt/mt.Cybb+/+ and B6 Slc11a1mt/mt.Cybb–/– mice or (B) B6/B10 Slc11a1+/+.Cybb+/+ and B6/B10 Slc11a1+/+. Cybb–/– mice, compared in each case with BALB/c mice. Infection was measured as footpad depth (mean±SEM; five mice per strain) at intervals over 94 days post-infection with 2 x 106 L. major LV39 metacyclic promastigotes. P values for significant differences between strains are indicated in the text. This experiment was only performed once.
|
 View larger version (17K): [in a new window] |
Figure 7. Immunofluorescence staining of liver crystat sections from N20 mice 2–4 days after infection. (A) Representative field examined under high power with confocal laser attachment demonstrating no colocalization of Slc11a1 staining in resident Kupffer cells (green) with M1/70 staining on fresh infiltrating monocytes (red). Original bar = 10 µm. (B and C) Complete colocalization of M1/70 (green) and iNOS (red) in infiltrating monocytes with no iNOS staining of the resident Kupffer cell population. Original bars = 100 µm.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
7 days of infection [44
]. This suggestion is confirmed here by our demonstration that Slc11a1+/+ maintain equivalent advantage over Slc11a1mt/mt mice with or without the Cybb–/– gene disruption and Cybb–/– exerting its influence independently and earlier than Slc11a1. Similarly, earlier studies of S. enterica serovar Typhimurium [43
] had suggested that iNOS activity was important later (post-day 7) in infection, again, independent of Slc11a1 control, although this conclusion was, to some extent, serendipitous, as we show here that the B6 Nos2A–/– mice used [22
], in fact, retained the Slc11a1 wild-type allele from the 129 progenitor (Mastroeni and co-workers [43
] did not Slc11a1-genotype the mice used in their experiments), and the 129Sv Nos2A–/– strain used [36
] in that study is not a full, functional null but a hypomorph, as demonstrated by the presence of a full-length protein in Western blots using control-activated macrophage protein preparations (J. K. White and J. M. Blackwell, unpublished data). This is consistent with functional studies [45
] demonstrating that this 129Sv Nos2A–/– line was leaky for NO production. Nevertheless, results obtained here using the full, functional Nos2A null, backcrossed onto equivalent B6/B10 Slc11a1 wild-type and mt genetic backgrounds, support the conclusion that functional iNOS is not required for Slc11a1-mediated resistance over the first 7 days of a virulent S. enterica serovar Typhimurium C5 infection. One problem in attempting to clarify the possible influence of gp91phox and iNOS on Slc11a1-mediated resistance in the context of S. enterica serovar Typhimurium infection is the acute and lethal nature of the infection and the short time-frame in which to examine the different phases of infection following administration of a virulent wild-type strain of the pathogen. In addition, the pleiotropic effects that Slc11a1 has on macrophage activation phenotypes in vitro mean that the same mechanisms might not be important in the in vivo regulation of different pathogens, which come under Slc11a1 control. In particular, we were keen to determine the influence of gp91phox and iNOS in Slc11a1 regulation of L. donovani infection, as both had been shown to be important in killing L. donovani in macrophages in vitro [16 , 17 ]. Unlike Salmonella infection, L. donovani infection is not lethal in mice, and the more prolonged course of infection makes it a more tractable, experimental model in which to analyze the roles of gp91phox and iNOS in Slc11a1- and non-Slc11a1-regulated phases of infection. Using this model, we obtained clear evidence that neither gp91phox nor iNOS activity contributed to resistance to L. donovani infection over the first 15 days when Slc11a1 is known to exert maximal influence in liver and spleen [26 , 33 ]. Our results differed from the earlier study of L. donovani infection using the same B6 (Slc11a1mt/mt) background mice [23 ] in which an increase in parasite loads in the livers of Cybb–/– compared with Cybb+/+ mice was reported at 2 weeks post-infection [46 ]. However, our study and the earlier study [46 ] concur on the observation that gp91phox has no influence over the acquired phase (>2 weeks) of L. donovani infection. Using a different source of Nos2A–/– mice on a mixed B6 x 129/Sv background [47 ], these authors also report [46 ] a more dramatic effect of iNOS deficiency on early (2 weeks) and acquired (>2 weeks) phases of L. donovani compared with our study in which iNOS had only moderate, organ-specific effects in the spleen at 50 days post-infection. Although the Slc11a1 status of these mixed-background mice was not reported [46 ], it is possible, in this case, that the line had become fixed for the B6 Slc11a1mt allele and that the requirement for iNOS activity during the acquired phase of immunity is more critical in Slc11a1mt/mt mice than we have observed here in Slc11a1+/+ mice. The results we obtained in our study with L. donovani were in stark contrast to L. major infection, where as reported previously [38 ], the fully functional Nos2A–/– gene disruption on a normally L. major-resistant B6/B10 background led to dramatic, early susceptibility and necessity for euthanasia within the first 21 days post-infection and where Cybb–/– also showed some influence on lesion size over the course of L. major infection. The smaller influence of Cybb–/– on L. major infection compared with Nos2A–/– is consistent with previous work [48 ], showing no effect of gp91phox on resolution of acute skin lesions and only a limited effect on containment of parasite in the draining lymph node. It is intriguing, however, that gp91phox was essential for clearance of L. major from the spleen [48 ]. These species-specific differences reflect the different cellular tropisms and pathologies caused by L. donovani and L. major. Studies in vitro [49 ] and in vivo [35 ] show that L. major has a preference for young, M1/70-positive monocytes, even when injected as amastigotes into mice i.v. and monitored in the liver [35 ], placing them in a cell population readily activated to express toxic antimicrobial iNOS activity. L. donovani amastigotes, conversely, preferentially infect resident tissue macrophages [35 ] and are found in the bone marrow in long-lived, sialoadhesin-positive, resident, stromal bone marrow macrophages [50 ]. This ability to target macrophage populations, which do not normally express potent antimicrobial iNOS or gp91phox activities, means that other mechanisms, such as those regulated more directly by Slc11a1, must be deployed by the host to control L. donovani infection. The ability of L. donovani to sequester into resident macrophage populations likely plays a major role in parasite persistence, with Slc11a1 possibly playing an important, long-term role in keeping this persistent parasite population in check. The species-specific differences in the behavior of leishmanial parasites in vivo and, in particular, the different mechanisms of antimicrobial activity associated with resistance and susceptibility to L. donovani and L. major have important implications for our understanding of human disease and the development of appropriate therapeutics.
To understand why neither gp91phox nor iNOS played a role in Slc11a1-mediated resistance in vivo, despite studies showing differential regulation of expression of activity of both in macrophages in vitro, we considered whether they were expressed in the same or different macrophage populations following infection in vivo. Although not formally re-examined here, we knew from earlier studies [40 ] that oxidative burst activity is not observed in the resident Kupffer cell population in the liver. What we examined specifically was which subpopulations of macrophage express Slc11a1 in the liver following L. donovani infection and whether this population expressed iNOS. Although we were unable to colocalize Slc11a1 and iNOS directly (as both were probed with rabbit Ab), we did demonstrate that iNOS and CD11b (recognized by the M1/70 Ab) were exclusively coexpressed in the fresh monocytes entering the liver after infection and that the M1/70 Ab showed no colocalization with the Slc11a1-positive, F4/80-positive, resident Kupffer cell population. Hence, we conclude that iNOS is not coexpressed in the same population of macrophages in the liver as Slc11a1 and similarly, by inference, that gp91phox and Slc11a1 are also unlikely to be coexpressed in the same macrophages in the liver. This is consistent with early studies [51 52 53 ] demonstrating that Lsh/Ity (=Slc11a1)-mediated resistance to infection was expressed in resident tissue macrophages, in Kupffer cells isolated ex vivo from livers of infected mice [51 ], or in liver, lung, or splenic tissue macrophages isolated and infected in vitro [52 , 53 ] but not in less mature macrophages from the peritoneal cavity [53 ]. Similarly, mature bone marrow-derived macrophages expressed Lsh-mediated resistance to L. donovani infection in vitro, but immature macrophages kept cycling with growth factors did not [53 ]. This is consistent now with RNA and Ab studies demonstrating that bone marrow macrophages have to be matured and activated to express high levels of Slc11a1 [54 ] and our data here showing that Kupffer cells in the livers of L. donovani-infected mice express Slc11a1 but not iNOS or by inference, gp91phox. Overall, our results suggest that although Slc11a1 may influence iNOS and NADPH oxidase activity under conditions where macrophages are activated to coexpress Slc11a1/iNOS or Slc11a1/NADPH oxidase in vitro, this does not occur for all populations of macrophages examined in vitro and does not appear to occur to influence the course of infection in vivo.
Our failure to demonstrate a role for NADPH oxidase or iNOS activity in Slc11a1-regulated S. enterica serovar Typhimurium or L. donovani infection leads us to conclude that interaction between these molecular pathways is not involved in Slc11a1-mediated resistance to infection in vivo. Other explanations must be found to account for Slc11a1-mediated antimicrobial activity. Early studies concluded that Slc11a1 limits bacterial [55 ] or leishmanial [56 ] replication in vivo, rather than enhancing bacterial killing. More recent studies have focused on the influence of Slc11a1 on phagosome-lysosome fusion and maturation in macrophages in vitro [54 , 57 58 59 60 ]. Functional Slc11a1 promotes phagosome-lysosome fusion and maturation, leading to acidification of the pathogen-containing phagosome and creating conditions under which reduced pathogen replication can occur. This appears to be related directly to the divalent cation-transporting function of Slc11a1, as iron chelators restore recruitment of the mannose-6-phosphate receptor to Salmonella containing vacuoles in Slc11a1 mt macrophages, overcoming the ability of Salmonella to alter phagosome maturation [61 ]. This suggests a role for Slc11a1 in regulating the recently described, Toll-like receptor-triggered, inducible, phagosome-lysosome maturation pathway [62 ], perhaps accounting for the influence of Slc11a1 on antigen processing and presentation [63 ] and T helper cell type 1 (Th1)/Th2 bias in immune response to infection [64 , 65 ] or vaccination [66 ].
In conclusion, our studies are important in demonstrating that observations made in vitro do not necessarily reflect the course of events in vivo. We demonstrate that iNOS and gp91phox are not important players in Slc11a1-mediated resistance to infection in vivo, fueling further research into the mechanisms by which this important gene influences susceptibility to infectious and autoimmune diseases.
2 Current address: Department of Infectious Diseases, Imperial College London, Hammersmith Campus, Hammersmith Hospital, Du Cane Road, London W12 0NN, UK.
Received September 24, 2004; revised November 10, 2004; accepted November 17, 2004.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and nitrite release Immunology 82,42-50[Medline] genes in nitric oxide production and their effect on the growth of Salmonella typhimurium in macrophages from Nramp1 congenic and tumor necrosis factor-–/– mice J. Interferon Cytokine Res. 21,53-62[CrossRef][Medline] and infection with BCG Immunology 82,457-464[Medline] in killing Leishmania donovani amastigotes in interferon-lipopolysaccharide-activated macrophages from Lshs and Lshr congenic mouse strains Infect. Immun. 59,3935-3944
se. Applications, Co-, Pid, Miologiques ,211-219 IMEEE Montpellier, France. /ß) and type 2 nitric oxide synthase regulate the innate immune response to a protozoan parasite Immunity 8,77-87[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  T. J. Bourret, M. Song, and A. Vazquez-Torres Codependent and Independent Effects of Nitric Oxide-Mediated Suppression of PhoPQ and Salmonella Pathogenicity Island 2 on Intracellular Salmonella enterica Serovar Typhimurium Survival Infect. Immun., November 1, 2009; 77(11): 5107 - 5115. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-R. Jiang, D. S. Gilchrist, J.-F. Popoff, S. E. Jamieson, M. Truscott, J. K. White, and J. M. Blackwell Influence of Slc11a1 (formerly Nramp1) on DSS-induced colitis in mice J. Leukoc. Biol., April 1, 2009; 85(4): 703 - 710. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Alter-Koltunoff, S. Goren, J. Nousbeck, C. G. Feng, A. Sher, K. Ozato, A. Azriel, and B.-Z. Levi Innate Immunity to Intraphagosomal Pathogens Is Mediated by Interferon Regulatory Factor 8 (IRF-8) That Stimulates the Expression of Macrophage-specific Nramp1 through Antagonizing Repression by c-Myc J. Biol. Chem., February 1, 2008; 283(5): 2724 - 2733. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Techau, J. Valdez-Taubas, J.-F. Popoff, R. Francis, M. Seaman, and J. M. Blackwell Evolution of Differences in Transport Function in Slc11a Family Members J. Biol. Chem., December 7, 2007; 282(49): 35646 - 35656. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. W. MURRAY, Z. XIANG, and X. MA RESPONSES TO LEISHMANIA DONOVANI IN MICE DEFICIENT IN BOTH PHAGOCYTE OXIDASE AND INDUCIBLE NITRIC OXIDE SYNTHASE Am J Trop Med Hyg, June 1, 2006; 74(6): 1013 - 1015. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
