Dynamics of lung macrophage activation in response to helminth infection

Abstract

Most of our understanding of the development and phenotype of alternatively activated macrophages (AAMs) has been obtained from studies investigating the response of bone marrow- and peritoneal-derived cells to IL-4 or IL-13 stimulation. Comparatively little is known about the development of AAMs in the lungs, and how the complex signals associated with pulmonary inflammation influence the AAM phenotype. Here, we use Nippostrongylus brasiliensis to initiate AAM development and define the dynamics of surface molecules, gene expression, and cell function of macrophages isolated from lung tissue at different times postinfection (PI). Initially, lung macrophages take on a foamy phenotype, up-regulate MHC and costimulatory molecules, express reduced levels of TNF and IL-12, and undergo proliferation. Cells isolated between days 8 and 15 PI adopt a dense, granular phenotype and exhibit reduced levels of costimulatory molecules and elevated levels of programmed death ligand-1 (PDL-1) and PDL-2 and an increase in IL-10 expression. Functionally, AAMs isolated on days 13–15 PI demonstrate an enhanced capacity to take up and sequester antigen. However, these same cells did not mediate antigen-specific T cell proliferation and dampened the proliferation of CD3/CD28-activated CD4⁺ T cells. These data indicate that the alternative activation of macrophages in the lungs, although initiated by IL-4/IL-13, is a dynamic process that is likely to be influenced by other immune and nonimmune factors in the pulmonary environment.

INTRODUCTION

The mucosal immune environments have developed the capacity to discriminate between potential pathogens and the vast array of harmless antigens to which they are constantly exposed [1]. The ability to distinguish between dangerous and nondangerous challenges is carried out, in large part, by specialized subsets of dendritic cells (DC) and macrophages [2, 3]. Although many studies have focused on defining the specialized activation states of DC within mucosal surfaces, much less attention has been paid to the more abundant macrophage populations [2, 4].

In contrast to IFN-γ-mediated, classical activation of macrophages, alternative activation is induced by IL-4 and/or IL-13 signaling through STAT6 [5, 6]. Alternatively activated macrophages (AAMs) are typically characterized as anti-inflammatory in nature and are thought to provide protective qualities by contributing to repair and remodeling of tissues [5, 7]. The majority of the studies characterizing the molecular and functional phenotype of AAMs has used peritoneal- or bone marrow-derived cells exposed to IL-4 and/or IL-13 in vitro [8, 9]. Although informative and important for establishing the basic tenets of alternative activation, these studies have not accounted for the additional signals that macrophages receive in an inflammatory environment, which are likely to modulate the IL-4/IL-13-induced phenotype. In addition, for those studies that have used cells alternatively activated in situ during strongly polarized, Th2-immune responses, the majority has inferred activation within tissues by measuring the increased transcription of genes encoding the AAM signature molecules arginase I (arg1), resistin-like α (fizz1), and the chitinase-like molecules YM1 and AMCase [10, 11]. Thus, our understanding of AAMs is likely to be somewhat limited by the experimental approaches used to study them.

In this study, we define the dynamics of the alternate activation of macrophages within the lungs of mice as a result of an infection with the rodent hookworm Nippostrongylus brasiliensis (Nb). We describe the morphological, surface-associated, transcriptional, and phenotypic changes that take place as a result of a dynamic transition of macrophages into the alternatively activated phenotype. These studies offer the first insight into the macrophage populations during alternative activation in the context of a complex Th2, inflammatory environment.

MATERIALS AND METHODS

Animals and infection

Six- to 8-week-old male wild-type (WT) and OVA-specific DO.11.10 T cell-transgenic mice, all on a BALB/c background, were obtained from Jackson Laboratories (Bar Harbor, ME, USA). Mice were housed in barrier filter-top cages, maintained on a 12-h light/dark cycle, and provided food and water ad libitum. All animal procedures described in this study were performed under the approval of the Johns Hopkins Animal Care and Use Committee (Baltimore, MD, USA), in accord with the guidelines set by the National Research Council’s Guide for the Care and Use of Laboratory Animals. Infectious, third-stage larvae of Nb were harvested from fecal culture via a Baermann apparatus [12]. The larvae were washed in PBS and counted, and 500 parasites were introduced to a s.c. site in the ruff of the neck.

Antibodies

The following anti-mouse antibodies were purchased from eBioscience (San Diego, CA, USA): CD80-PE (12-0801-81), MHCI-PE (12-5998-81), MHCII-PE (12-5321-81), CD40-PE (12-0401-81), CD86-PE (12-0861-81), programmed death ligand-1 (PDL-1)-PE (12-5982-81), PDL-2-PE (12-5986-81), Thy1.2-allophycocyanin (APC; 17-0902-83), Ter-119-APC (17-0902-83), CD3-APC (17-0031-82), CD28 (16-0281-85), and CD3ε (16-0031-81). PE-conjugated mouse IgG1κ (555-749), Imag APC particles DM (E30-221), and FITC-conjugated BrdU antibody set (556028) were obtained from BD PharMingen (San Jose, CA, USA). Anti-mouse CD11c-APC (120-002-001), B220-APC (130-091-843), and GR1-APC (130-091-931) were obtained from Miltenyi Biotec (Auburn, CA, USA). Anti-F4/80 (ab6440) was obtained from Abcam (Cambridge, MA, USA), and Alexa Fluor 488 goat anti-rat IgG (A11006) was obtained from Invitrogen (Carlsbad, CA, USA).

Bronchial alveolar lavage (BAL)

Surgeries were performed as described previously [13]. In short, animals were anesthetized with 300 mg/kg body weight with Avertin (2,2,2-tribromoethanol, Acros Organics, Morris Plains, NJ, USA), administered i.p. Pericardium and trachea were exposed by dissection. A lateral incision was made through the trachea, and a 19-gauge, flat-tipped needle was inserted and tied off. BAL was performed by flushing the lungs (3×) with 1 ml sterile PBS at 24°C. BAL fluid was pooled and spun at 1500 rpm for 3 min at 4°C. Cells were incubated in 0.5 ml RBC lysing buffer (Quality Biological, Gaithersburg, MD, USA) at 24°C for 2 min and washed twice with PBS at 4°C. Cytology slides were processed using a Shandon Cytospin 3 (Shandon, Pittsburgh, PA, USA), and slides were stained with Hemacolor (Harleco, Gibbstown, NJ, USA).

Lung monocyte preparations

Lungs were perfused with 10 ml sterile PBS (24°C) prior to removal by inserting a 25-guage needle into the right ventricle of the heart. Lungs were removed carefully to assure the exclusion of contaminating tissue. Lung tissue was suspended in 5 ml RPMI 1640 containing 1 mg/ml collagenase type II (1701-015, Gibco, Carlsbad, CA, USA) and 30 μg/ml DNase I (10104159001, Roche, Indianapolis, IN, USA), minced thoroughly, and incubated at 37°C for 30 min. After dispersing the remaining tissue fragments with repeated pipetting, 3 ml fresh collagense/DNase solution was added, and the samples were incubated for 15 min at 37°C. Suspensions were then passed through a 100-μm mesh nylon cell strainer (352360, BD Falcon, Bedford, MA, USA), and the resulting cells were pelleted at 1500 g (4°C). Pellets were resuspended in 5 ml of ACK (150 mm NH₄Cl, 10 mm KHCO₃, 0.1 mm EDTA) lysing buffer for 2 min at room temperature, passed through a fresh strainer, and washed twice in FACS-staining buffer (PBS, 2% heat-inactivated FCS). Pellets were resuspended in 200 μl staining buffer containing Fc Block (553142, BD PharMingen) and incubated on ice for 5 min. Anti-CD11c was then added, and cells were incubated on ice for 30 min. Cells were washed twice in staining buffer, and resulting pellets were resuspended in Imag APC particles DM and incubated on ice for 30 min. Isolation of CD11c⁺ cells was performed according to the manufacturer’s protocol using the BD Imagnet System (BD PharMingen). Cell counts were performed on a standard hemacytometer with dead cells excluded by trypan blue stain (15250-061, Gibco).

Flow cytometry

Cells were incubated with the appropriate fluorochrome-conjugated antibodies in FACS-staining buffer for 20 min on ice in the dark. Data were acquired by running samples on a FACSCalibur flow cytometer using CellQuest software (BD Biosciences, Mountain View, CA, USA). Data were analyzed using FloJo software (Tree Star, Inc., Ashland, OR, USA).

Histology

A 19-gauge, flat-tipped needle was inserted into the trachea, and the lungs were inflated slowly with zinc-buffered formalin fixative (Z-fix, Anatech Ltd., Battle Creek, MI, USA). The lungs were removed and incubated in a 40-fold volume of Z-fix. Inflated and fixed lungs were embedded in paraffin, and sagittal, 10 μm sections were obtained from four different levels of lung. Lung sections and cytology slides were examined using a Nikon Eclipse E800 light microscope (Nikon Inc., Melville, NY, USA), and images were acquired using SPOT RT charge-coupled device imager and software (Diagnostic Instruments, Inc., Sterling Heights, MI, USA).

Electron microscopy

Lung-derived cells were fixed with 2% glutaraldehyde for 2 h at 24°C, followed by postfixation with 2% osmonium tetroxide for 2 h at 24°C. The cells were embedded in Epon 812, sectioned, poststained with 2% uranyl acetate and lead citrate, and viewed on a Philips CM 120 electron microscope at 100 kV.

Immunohistochemistry

For epitope retrieval, slides were incubated in 20 μg/ml proteinase K solution for 15 min at 37°C. Sections were allowed to cool to room temperature for 10 min and washed (2×) for 2 min with PBS. Surfactant protein A (SP-A) staining was performed using (1:4000) polyclonal rabbit anti-SP-A (AB3420, Chemicon International, Temecula, CA, USA), which was detected using a biotinylated goat anti-rabbit IgG (Vector Laboratories, Burlingame, CA, USA), peroxidase strepavidin, and 3,3′ diaminobenzidine (Vector Laboratories). For dual fluorescence staining of macrophages, SP-A was resolved with Fluorescein/Avidin cell sorting grade Avidin D (DCS; Vector Laboratories), and after an avidin/biotin-blocking step (Vector Laboratories), biotin-conjugated CD11c (13-0114-85, eBioscience) binding was detected with Texas Red/Avidin DCS. Stained sections were mounted in Vectashield containing 4′-6-diamidino-2-phenylindole (DAPI; Vector Laboratories).

Gene expression analysis

CD11c-positive cells were harvested, flash-frozen in liquid nitrogen, and stored at –80°C. RNA was isolated as outlined by Reece et al. [13]. RNA quality was determined by RNA Nano LabChip analysis on an Agilent Bioanalyzer 2100 (Agilent Technologies, Palo Alto, CA, USA). An RNeasy total RNA cleanup protocol (Qiagen, Valencia, CA, USA) was performed, followed by spectrophotometric assessment of RNA concentration. CD11c⁺ cells from five animals were processed independently for each day of infection.

Processing of templates and the hybridization protocol for the 430 2.0 Array GeneChip (Affymetrix Inc., Santa Clara, CA, USA) were in accordance with methods described in the Affymetrix GeneChip Expression Analysis Technical Manual, Revision Three [14]. The hybridization results were assessed using the GCS3000 laser scanner (Affymetrix Inc.) at an emission wavelength of 570 nm at 2.5 μm resolution. Intensity of hybridization for each probe pair was computed by GCOS 1.2 software (Affymetrix Inc.). For more detailed methods, please refer to the website of the Malaria Research Institute, Gene Array Core Facility, at the Johns Hopkins Bloomberg School of Public Health (http://jhmmi.jhsph.edu). Affymetrix CEL file data were preprocessed for use with the probe level GeneChip Robust Multi-Array analysis (GC-RMA) [15] option in GeneSpring 7 software (Agilent Technologies). The data discussed in this publication have been deposited in the National Center for Biotechnology Information Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo).

Real-time RT-PCR

Lung RNA (1 μg) was reverse-transcribed using the SuperScript first-strand synthesis system for RT-PCR (Invitrogen) using an oligo(dT) primer. Quantitative real-time RT-PCR was performed using Applied Biosystems 7500 real-time PCR system, TaqMan® Gene Expression Assays-On-Demand, and TaqMan Universal Master Mix (Applied Biosystems, Foster City, CA, USA). The following assays (Applied Biosystems) were used: Arg1 (Mm00475988), Ym1 (Mm00657889), Fizz1 (Mm00445109), IL-12p40 (Mm00434165_m1), TNF-α (Mm00443558_m1), IL-6 (Mm00446190_m1), IL-1β (Mm00434228_m1), IL-10 (Mm00438616_m1), TGF-ß (Mm00441724_m1), IFN-α (Mm00833861_s1), and IFN-β (Mm00439546__s1). Reactions were performed using 1 μl cDNA in a 25-μl sample vol and the following thermal cycler profile: 10 min denaturation at 95°C, 50 cycles of 15 s denaturation at 95°C, followed by 1 min extension at 60°C. Analysis was performed using the 7500 system SDS software package (Applied Biosystems).

BrdU staining

Animals were given BrdU (B5002, Sigma Chemical Co., St. Louis, MO, USA) via drinking water (0.8 mg/ml) from day 0 of infection until they were sacrificed. CD11c⁺ mononuclear cells were isolated from WT and Nb-infected mouse lungs as described above. Isolated cells were resuspended in 500 μl 0.15 NaCl, fixed with 1.2 ml 95% ethanol, resuspended in 1 ml 1% paraformaldehyde and 0.01% Tween-20 solution, and kept overnight at 4°C. Isolated cells were then resuspended in a DNase digestion solution for 30 min at 35°C and subsequently stained with 20 μl anti-BrdU-FITC or an isotype control antibody. BrdU incorporation was determined by flow cytometry.

T cell proliferation

CD11c⁺ mononuclear cells were isolated from WT and Nb-infected mouse lungs as described above. CD4⁺ T cells were isolated using a CD4 T cell isolation kit (Miltenyi Biotec), according to the manufacturer’s protocol. Briefly, a single-cell suspension of splenocytes was incubated in ACK lysing buffer, washed, and stained with a biotin-labeled, anti-CD4 T cell antibody cocktail. Cells were then washed and incubated with the recommended concentration of antibiotin microbeads. Stained and bead-labeled splenocytes were then washed and passed through an LS column (130-042-401). Isolated CD4 T cells were stained with Cell Trace CFSE (Molecular Probes, Eugene, OR, USA) by placing 2.0 × 10⁷ isolated T cells in 2 ml PBS containing 3 μg CFSE for 6 min at room temperature, and unbound CFSE was quenched by washing cells in supplemented DMEM. CFSE-labeled CD4 T cells (1×10⁵) were incubated with CD11c-positive lung cells (1×10⁵) in a 96-well flat-bottom plate. T cells were stimulated with 20 ng/ml recombinant (r)IL-2 (34-8021-82, eBioscience), 5 μg/ml plate-bound anti-CD3ε, and 5 μg/ml soluble anti-CD28. Cells were incubated for 4 days at 37°C. CFSE proliferation profiles were analyzed by flow cytometry after staining with CD4-APC and 7-amino-actinomycin (7-AAD).

Antigen uptake

Antigen uptake was evaluated using the Vybrant phagocytosis assay (Molecular Probes). CD11c⁺ lung macrophages were isolated from the lungs of mice on days 0, 4, and 13–15 postinfection (PI). Briefly, 1 × 10⁶ lung macrophages/well were plated and allowed to adhere to 96-well flat-bottom plates. Fluorescent Bioparticles were added to experimental wells and allowed to incubate for 2 h at 37°C. Trypan blue was added to quench Bioparticles that had not been taken up by macrophages. Plates were read at 520 nm emission and 480 nm excitation on a HTS 7000 Plus Bio Assay Reader with HT Soft 2.0 software (Perkin Elmer, Waltham, MA, USA).

Antigen-specific proliferation

On days 0 and 13–15 PI, lung macrophages were isolated as described above. DC were also isolated from the spleens of normal BALB/c animals as described previously [16]. Briefly, spleens were minced into fine pieces and digested in collagenase. Liberated cells were spun through FCS supplemented with EDTA. Spleen cells were then separated over a Nicodenz gradient (density, 1.075 g/ml), and the DC-rich, low-density fraction was kept. Contaminating cells were removed by staining with APC-conjugated anti-CD3, anti-Thy1.2, anti-GR1, anti-B220, and anti-Ter119 and then incubated with Imag APC particles DM. Depletion of APC-stained cells was performed, according to the manufacturer’s protocol, using the BD Imagnet System. Isolated macrophages and DC were pulsed with 10 μg/ml OVA peptide 323–339 (27024, AnaSpec, San Jose, CA, USA) for 60 min at 37°C. Pulsed cells were washed (3×) in supplemented DMEM. OVA-specific CD4 T cells were isolated from the spleens of DO11.10-transgenic mice and loaded with CFSE as described above. Pulsed macrophages and DC (1×10⁵) were incubated with CD4 T cells (1×10⁵) in the presence of 20 ng/ml rIL-2 in a 96-well v-bottom plate at 37°C for 4 days. CFSE proliferation profiles were analyzed by flow cytometry after staining with CD4-APC and 7-AAD.

RESULTS

BAL-derived lung macrophages

To gain an initial insight into the cellular changes induced by the passage of Nb larvae through the lungs, the cellular component of the BAL was analyzed during the lung (days 2 and 3), intestinal (days 4 and 8), and postexpulsion (days 13 and 15) phases of infection. As described previously [17, 18], macrophages were the dominant cell type found in the BAL fluid during the first 9–12 days PI, after which, eosinophils become more numerous (data not shown). There was a demonstrable increase in the number of BAL macrophages at day 2 PI, the day larvae arrive in the lungs, but the increase did not reach statistical significance until day 4 PI (P<0.05; Fig. 1A). BAL macrophage numbers returned to near-preinfection levels by days 8 and 13 PI. Consistent with previous reports [19, 20], flow cytometric analysis of the BAL-derived mononuclear cells indicated that a vast majority of the cells was CD11c⁺, MHCII⁺, F4/80⁺ macrophages (data not shown).

Fig. 1.

Nb infection results in an increase in the number of lung macrophages and induces a change in their morphology. (A) BAL-derived macrophages were counted on days 0, 2, 4, 8, and 13 post-Nb infection. (B) Cytospin preparations of BAL-derived cells were evaluated for the percentage of vacuolated/foamy (open bars) or granular (closed bars) macrophages on days 0, 2, 4, 8, 13, and 100 post-Nb infection. (C) Changes in the number of CD11c⁺ macrophages isolated from whole lungs using CD11c-magnetic beads on days 0, 2, 4, 8, and 13 post-Nb infection. (D) Evaluation of the percentage of vacuolated/foamy (open bars) or granular (closed bars) macrophages in the CD11c⁺ cells isolated from whole lungs on days 0, 2, 4, 8, and 13 after Nb infection. (E–G) Representative light microscopy images of the morphologies of BAL-derived macrophages from days (E) 0, (F) 2 (vacuolated/foamy), and (G) 13 (granular) PI lungs. (H and I) Transmission electron microscopy of BAL-derived macrophages collected on (H) day 2 or (I) day 13 PI. Multilaminate large granules (large arrowheads) and smaller granules (small arrowheads) are indicated. (J and K) Representative histological sections from days 2 and 13 PI, illustrating the distribution of (J) vacuolated/foamy and (K) granular macrophages, respectively, in the parenchyma of the lung, and lung macrophages are indicate by arrowheads (original, 40×) and in the magnified insets. (L) Immunofluorescence staining of lung macrophages for CD11c (red) and SP-A (green). Histological sections of lungs isolated at days 0 and 13 PI were immunostained for CD11c and SP-A and counterstained with DAPI (blue; original, 40×). *, P < 0.05; **, P< 0.01.

Accompanying the changes in numbers, a significant proportion of the BAL-derived macrophages adopted distinct morphological changes. As early as day 2 PI, ∼50% of the macrophages in the BAL exhibited a highly vacuolated, foamy phenotype (Fig. 1, B and F) that was not observed in the uninfected controls (Fig. 1, B and E). The proportion of extensively vacuolated macrophages decreased to 20% at day 3 (data not shown), and by days 4–8 PI, only ∼10% exhibited a less-pronounced, vacuolated appearance. On day 13 PI, 2–3 days after the adult parasites had been expelled from the intestine, although the number of macrophages in the BAL had returned to near preinfection levels, ∼50% of the BAL macrophages were enlarged cells containing dense granular material in the cytoplasm (Fig. 1, B and G). In contrast to the transient nature of the foamy phenotype, the granular phenotype persisted with ∼50% of the macrophage population exhibiting the granular phenotype through day 100 PI, 89 days after the animals had cleared the infection and 97 days after the larvae were resident in the lung (Fig. 1B). To rule out the possibility that an unapparent infection was contributing to the changes in macrophage phenotype, uninfected littermates housed in the same facilities were screened for a panel of pathogens by serology and monitored by histology over time. The serology was negative, and there were no changes in macrophage phenotype in any of the control animals (data not shown).

Electron microscopic analysis of the foamy macrophages isolated from day 2 PI showed the cytoplasm to be dominated by numerous large vacuoles that contained material of varying densities (Fig. 1H). This ultrastructural appearance was similar to the macrophages identified in the lungs as a result of several infectious agents including HIV and tuberculosis [21, 22]. Ultrastructural analysis of days 13–15 PI granular macrophages (Fig. 1I) revealed the presence of several forms of dense granules. Some of the granules were composed of large multilamellar structures (large arrowhead) that resemble surfactant protein (SP), reported to accumulate in macrophages during acute lung injury [23] and in disease states such as pulmonary alveolar proteinosis [24], and others that contained membrane-like structures but were more dense and did not resemble typical SP accumulation (small arrowheads). Immunohistochemical analysis of day 13 PI macrophages showed that the contents of some of the granules were comprised of SP-A (Fig. 1L) and SP-D (data not shown).

Macrophages in the parenchyma of the lung

To determine the extent to which the profiles derived from the BAL reflected the number and phenotype of macrophages in the parenchyma of the lung, histological sections were prepared from lungs isolated on days 2 and 13 PI. Foamy (Fig. 1J) and granular macrophages (Fig. 1K) were readily identified in histological sections on days 2 and 13, respectively. Foamy and granular cells were found to be distributed uniformly throughout the parenchyma of the lung (arrowheads). A systematic examination of the histological sections strongly suggested that the extent and duration of the changes in the lung macrophage population were underestimated significantly by BAL sampling. For example, on day 13 PI, when BAL results suggested that the response was waning and the macrophage numbers were near-preinfection levels (Fig. 1A), the histological analysis indicated that macrophage levels were actually elevated (data not shown). Thus, to obtain a more accurate assessment of the cellular dynamics, CD11c⁺ cells were isolated from whole lung at 0, 2, 4, 8, and 13 days PI.

Taking advantage of the elevated levels of CD11c present on lung macrophages [13, 20, 25, 26], a protocol using CD11c magnetic beads was developed to isolate cells highly enriched for macrophages from lung digests. On day 2 PI, the number of macrophages in the lungs was not significantly different from preinfection levels (Fig. 1C), but the composition of cells at day 2 PI differed from that seen in the BAL in that only ∼20% exhibited the foamy phenotype (vs. ∼50% in BAL; Fig. 1D). On day 4 PI, the number of lung macrophages increased significantly (P<0.01) to ∼3 × 10⁶, and ∼20% of the cells exhibited a foamy phenotype. There was a decrease in the number of macrophages on day 8 PI that coincided with an unambiguous transition from the foamy to the granular phenotype. The level of granular cells was maintained at between 15% and 20% of the total macrophage population through day 13 PI. In contrast to the BAL results, where peak macrophage numbers were seen at day 4 PI (Fig. 1A), the highest number of macrophages was isolated from day 13 PI lungs at ∼6 × 10⁶. It was concluded that the BAL-based analysis faithfully represented neither the quantitative nor the qualitative changes taking place within the parenchyma of the lung post-Nb infection. Consequently, subsequent studies to define surface phenotypes and expression profiles were carried out using cells isolated from whole lungs.

AAM surface phenotype

A surface expression profile of the CD11c⁺ macrophages isolated from uninfected lungs revealed constitutive production of F4/80, MHCI, MHCII, PDL-1, CD80, ICAM-1 and CD1d but no significant expression of PDL-2, CD40 or CD86 (Fig. 2). Although the CDllc⁺ cells from uninfected lungs could be divided into populations expressing low and high levels of MHCII, at days 2, 4, 8, and 13 PI, a vast majority of the cells was in the MHCII^hi population. Surface expression of MHCI, ICAM-1, and CD1d was elevated during infection and remained elevated after the worms were expelled from the intestine. The costimulatory molecules CD80, CD86, and CD40 as well the classic macrophage marker F4/80 showed a marked increase in expression on day 4 PI, followed by a reduction to near-preinfection levels on days 8 and 13 PI. In contrast, the levels of PDL-1 and PDL-2 were low or remained near baseline during the first few days PI but showed elevated expression on days 8 and 13 PI.

Fig. 2.

Surface phenotype of lung macrophages. CD11c⁺ macrophages were isolated from whole lung tissue on days 0, 2, 4, 8, and 13 post-Nb infection and examined for changes in the levels of key surface-associated molecules by flow cytometry as outlined in Materials and Methods. A representative isotype control is illustrated in white, day 0 surface levels are represented in gray, and the PI surface levels are represented in black, according to the day noted. The profiles are representative of five biological replicates.

Dynamics of AAM gene expression

Expression analysis was performed on CD11c⁺ macrophages isolated from the lungs on days 0, 2, 4, and 8 PI. Although multiple attempts were undertaken to obtain RNA from macrophages isolated on day 13 PI lungs, we were unable to isolate RNA of sufficient quality to use in expression analysis. In total, the transcription of 2319 distinct genes was up-regulated significantly in the CD11c⁺ lung macrophages over the three time-points tested, with a peak of >1600 genes on day 4 PI (Fig. 3A). Between 31% and 48% of the up-regulated genes were expressed uniquely at each time PI. Overall, ∼40% of the up-regulated genes were expressed at two or more time-points, and 277 were common to all three time-points. A total of 1383 genes was found to be down-regulated at least twofold during days 2, 4, and 8 PI (Fig. 3B) compared with macrophages isolated from uninfected lungs, with 61 common to all 3 days PI.

Fig. 3.

Gene expression analysis of Nb-induced lung macrophages. CD11c⁺ macrophages were isolated from whole lung tissue on days 0, 2, 4, and 8 PI and processed for Affymetrix GeneChip analysis as outlined in Materials and Methods. The hybridization results were processed using GC-RMA, and a minimum probe fluorescence intensity threshold was set at 150 for the significantly regulated genes. (A and B) Ven diagrams of up-regulated and down-regulated genes on days 2, 4, and 8 PI. (C) Fold-changes (FC) in the expression of selected genes by CD11c⁺ lung macrophages isolated at days 2, 4, and 8 PI compared with cells isolated from the lungs of uninfected animals; n= 3 biological replicates per time-point. FC of <2, X; FC of 2–3, +; FC of 3–5, ++; FC of 5–10, +++; FC of 10–20, ++++; FC of >20, +++++.

Consistent with the results from a previous study [13], CD11c⁺ macrophages from the post-Nb lungs robustly up-regulated genes encoding proteins associated with alternative activation including ARG1, FIZZ1, YM1, YM2 (ym1, ym2), and AMCase (chia), triggering receptor expressed on myeloid cells 2 (trem2b), and the scavenger receptors Marco (marco) and Msr1 (msr1; Fig. 3C) [13, 27,28,29,30]. Interestingly, argII, which has been shown to negatively regulate proinflammatory qualities in activated macrophages by limiting NO production [31, 32], was also strongly up-regulated on days 4 and 8 PI.

There was also a rapid and sustained induction in genes encoding surface lectins, scavenger receptors, enzymes, and other proteins involved in lipid metabolism. Included among these were the class D scavenger receptor CD68 and a receptor for oxidized low-density lipoproteins (oldr1), which play key roles in the uptake of lipids and apoptotic cells from the environment [33,34,35], leukotriene B4 12-hydroxydehydrogenase (ltb4dh) and acyl-CoA oxidase 1 (acox1), enzymes known to be involved in the catabolism of lipids. This pattern of gene expression is consistent with the development of the foamy macrophage phenotype typically associated the incorporation of lipids during conditions such as atherosclerosis and pulmonary alveolar proteinosis [36, 37]. On day 8 PI, macrophages dramatically up-regulated genes encoding the complement component C1q, Oldr1, and 15-Lipoxygenase (alox15), all of which enhance the uptake of apoptotic bodies along with SP-A and SP-D in the airways [33, 34, 38]. Also of interest was the significantly enhanced transcription of genes encoding CCL4, CCL8, CCL9, CCL11, CCL12, CCL17, and CCL24.

The increase in the CD11c⁺ population in the post-Nb lung (Fig. 1, A and H) could be a result of proliferation of the resident lung macrophages, cell recruitment, or a combination of proliferation and recruitment. Although the source of the increase in lung macrophages remains to be defined, the cells isolated from days 2 and 4 PI showed significant increases in genes encoding proteins that regulate the cell cycle, suggesting that early after infection, proliferation contributes, at least in part, to the increase in macrophage numbers. The capacity of the lung macrophages to proliferate was confirmed by demonstrating that the CD11c⁺ cells isolated from lungs on days 2 and 4 PI incorporated more BrdU than the cells isolated from uninfected lungs (Fig. 4). At day 2 PI, over 6% of the CD11c-positive cells had incorporated BrdU, and that number increased to over 9% by day 4 PI.

Fig. 4.

BrdU incorporation into CD11c⁺ lung macrophages isolated from whole lung tissue on days 0, 2, and 4 post-Nb infection. BrdU was administered to mice via drinking water (0.8 mg/ml) on day 0 and continued until the day of sacrifice. Cells were immunostained with anti-BrdU or an isotype control, and the level of BrdU incorporation was evaluated by flow cytometry.

Among the significantly down-regulated genes were multiple genes known to be involved in the classical activation of macrophages, including IFN-γR2 (ifngr2), the immunity-related GTPase (Irgm1), inducible NO synthase (inos), Kruppel-like factor 4 (klf4), and heat shock protein 1b (hspa1b) [39,40,41].

Changes in cytokine expression

Real-time RT-PCR was performed on CD11c⁺ lung macrophages isolated from days 0, 2, 4, and 8 PI to validate the chip-based expression levels of the macrophage-associated cytokine genes IL-12, TNF-α, IL-6, IL-1β, IL-10, TGF-β, IFN-α, and IFN-β (Fig. 5). There were modest but statistically significant decreases in the expression of TNF-α at days 2, 4, and 8 PI and of IL-12, IL-6, and TGF-β at day 4 PI. IL-1β and IL-10 transcription levels increased as the Nb infection progressed. There was no detectable transcription of IFN-α on day 2 PI.

Fig. 5.

Cytokine gene expression in CD11c⁺ lung macrophages isolated from whole lung tissue on days 0, 2, 4, and 8 post-Nb infection. RNA isolation and processing for real-time RT-PCR analysis of IL-12, TNF-α, IL-6, IL-1β, IL-10, TGF-β, IFN-α, and IFN-β were carried out as outlined in Materials and Methods. Each bars represents the mean (se) of five biological replicates at each time-point (*, P<0.05; **, P<0.01).

Antigen uptake and T cell proliferation

The Nb-induced changes in the morphology, surface phenotype, and gene expression profile of lung macrophages suggested that there could also be alterations in cell function. To test the functional capabilities of these cells, we isolatedCD11c⁺ cells from the lungs at days 0, 4, and 13–15 PI to determine their ability to take-up antigen and inhibit T cell proliferation. The CD11c⁺ cells isolated from days 13–15 post-Nb lungs have a significantly enhanced ability to take up antigen when compared with the CD11c⁺ cells isolated from uninfected and day 4 post-Nb lungs (Fig. 6A). Interestingly, the macrophages isolated from days 13–15 PI also showed an increased capacity to inhibit CD4 T cell proliferation in response to plate-bound CD3 and soluble CD28 when compared with macrophages isolated from WT and day 4 post-Nb lungs (Fig. 6B). We tested the possibility that this enhanced ability of days 13–15 macrophages to sequester antigen along with their elevated levels of MHCII expression (Fig. 2) correlated with an increased capacity to induce antigen-specific CD4⁺ T cell proliferation. Lung macrophages isolated from day 0-uninfected and days 13–15 post-Nb lungs were pulsed with OVA peptide (323–339) and incubated with CFSE-labeled CD4⁺ DO.11.10-transgenic T cells. In contrast to the robust proliferation observed when DC presented the peptide, the T cells incubated with macrophages isolated from days 13–15 PI lungs showed limited proliferation (Fig. 6C). However, this limited ability to promote T cell proliferation was not significantly different from that measured from the lung macrophages isolated from uninfected lungs (Fig. 6C). These data illustrate that days 13–15 PI macrophages show a lack of ability to mediate antigen-specific T cell proliferation, despite increased antigen incorporation and elevated MHCII levels.

Fig. 6.

Nb-induced CD11c⁺ lung macrophages have an increased phagocytic capacity but are poor mediators of T cell activation. Macrophages were isolated from whole lung tissue using CD11c magnetic beads. (A) Changes in the phagocytic capacity of CD11c lung macrophages isolated at days 0, 4, and 15 PI. (B) PI CD11c⁺ lung macrophages modulate T cell proliferation. CD4⁺ T cells (1×10⁵) were loaded with CFSE and stimulated for 4 days with plate-bound anti-CD3 plus soluble anti-CD28 in the presence of 1 × 10⁵ CD11c⁺ lung macrophages isolated from lungs at day 0, day 4, or day 13 PI. The level of proliferation was evaluated by flow cytometry. A dotted gray line represents the results of unstimulated T cells. Results are representative of three to five independent experiments. (C) CFSE-stained DO11.10-transgenic CD4⁺ T cells (1×10⁵) were incubated with spleen-derived DC of CD11c⁺ lung macrophages isolated from lungs at day 0 or day 13 PI. The DC and lung macrophages pulsed with 10 μg/ml OVA peptide (323–339) for 60 min prior to adding the transgenic T cells (**, P<0.01).

DISCUSSION

Infection with Nb results in a number of immediate and long-term changes to the immunological status of the lungs, including quantitative and phenotypic changes in the macrophage population [10, 13, 17, 42]. In the current study, we describe two distinct morphological forms that develop within the lung macrophage population after a hookworm infection—a vacuolated, foamy phenotype that was supplanted within a few days PI by a persistent population of enlarged macrophages that contained dense granules in the cytoplasm (Fig. 1). It is not clear if the origin of these cells is connected. Although it is possible that foamy macrophages transform into the cells with the granular morphology, it is equally likely that these distinct cell types represent at least two different populations whose phenotypes are influenced by the dynamics of the cellular and cytokine changes that persist in the PI lung environment [42].

Macrophages with the granular phenotype, which first appeared between days 4 and 8 PI in the parenchyma and between days 8 and 13 in the BAL, are readily detectible through day 100 PI (Fig. 1) [43], suggesting they are an enduring element of the post-Nb lung. It is not clear if the granular phenotype is maintained as a consequence of long-lived cells, or it represents scavenging of granules released after cell death. Lung macrophages are distinguished by their longevity, with estimates of their half-life ranging between 30 days and >2 years [44,45,46]; thus, it is conceivable that the granular cells observed at 100 days PI are the same cells formed within a few days after Nb traversed the lungs.

The composition of the granules appears to be complex. Although it is apparent that SPs make up a portion of macrophage granules, some macrophages also stained positive for Prussian blue (data not shown), suggesting the sequestration of iron, possibly through the uptake of RBCs. During the first 2–3 days of infection, hemorrhage is evident as a consequence of the mechanical damage caused by the Nb larvae as they enter the lungs [13] and could be the source of the iron. In addition, coincident with the development of the granular phenotype, a large number of cells infiltrating the lungs undergo apoptosis, a significant percentage of which appears to be eosinophils (data not shown). Therefore, it is likely that macrophage granularity is a result of the incorporation of SPs and cellular debris released as a result of the structural damage and cellular infiltration caused by the larvae’s migration through the airways. Histological staining for Fizz1 and YM1 revealed that Fizz1 is distributed uniformly throughout the cytoplasm of the granular macrophages and does not appear to be concentrated in granules (data not shown). In contrast, YM1 staining is punctate in nature and is concentrated in specific areas of the cytoplasm, suggesting that YM1 could be a component of the granules (data not shown). Additional work will be required to confirm the distribution of YM1 within the cells.

Measuring the cellular composition of the BAL fluid has been used traditionally as a rapid method to monitor cellular lung infiltrates, typically during acute inflammation. The results presented here suggest that BAL-based analysis of pulmonary macrophage populations does not accurately reflect the quantitative and qualitative changes in the macrophage populations during acute inflammation or after inflammation was resolved. Although BAL-based analysis may be useful as a rapid, low-resolution measure of the cellular content of the lungs during inflammation, a more extensive protocol, such as the bead-based isolation protocol used here, may be required to gain an accurate evaluation of cellular changes in inflamed and postinflammatory lungs.

One confounding issue associated with the CD11c bead-separation protocol is the coisolation of lung DCs (estimated at ∼5% of total cells) that could have influenced the outcome of the expression and functional analyses. In C57/BL6 mice, a combination of autofluorescence and MHCII levels has been proposed to differentiate lung DC and macrophage populations [47], but these criteria do not appear to be suitable for the separation of these cell types from the lungs of BALB/c mice (data not shown). Although F4/80 has been used traditionally as a macrophage-specific surface marker, its use in the lung is compromised by low levels of F4/80 on lung macrophages and expression levels that vary in the context of inflammation (Fig. 2) [48, 49]. To increase the purity of lung macrophage preparations for procedures such as adoptive transfers, it may be necessary to add a step to specifically deplete the lung DCs [50].

Gene expression analysis of lung-derived macrophages confirmed earlier observations that there is a rapid and persistent induction of AAMs in the pulmonary environment as a result of a helminth infection [10, 13]. Expression profiles of day 2 PI cells isolated from whole lung showed induction of the genes encoding ARG1, FIZZ1, YM1, and AMCase and the lectin MSR1, all hallmark indicators of alternative activation [5]. AAMs are induced by a combination of the rapid induction of Th2 cytokines [13] and tissue injury [51] caused by the Nb larvae as they enter the lungs and then sustained by the long-term changes to the immunological environment of the lung [42].

Mature macrophages have long been considered end-stage cells with no or limited capacity to proliferate. However, recent evidence indicates that under certain conditions, lung macrophages do have the capability to divide [52], and the expression and cellular data from this study support this idea. Day 2 PI macrophages expressed elevated levels of a number cell cycle-associated genes (Fig. 3) and incorporated BrdU, suggesting that at least a portion of the increases in macrophages observed on day 4 PI is a result of resident cell proliferation. However, this capacity to proliferate appears to be short-lived, as the cell cycle genes are no longer elevated by day 8 PI. Lung macrophages are derived from blood monocytes and take on their unique characteristics only after a period of development in the pulmonary environment [52]. Therefore, it is possible that the capacity to proliferate provides an immediate source of differentiated lung macrophages that can play a role in limiting the extent of inflammation and promoting repair of cellular and structural damage.

The AAM phenotype is induced by engaging the IL-4R or IL-13R and is accompanied by the dramatic up-regulation of genes encoding ARGI, FIZZI, YMI, and AMCase [5]. In addition to the traditional AAM expression profile, the macrophages isolated on days 4 and 8 PI expressed the gene encoding ArgII (Fig. 3), a molecule not typically associated with alternative activation. Although it is possible that ARGII production represents a novel marker of alternative activation in the lung, it is also likely that argII expression is a reflection of an environment, where the macrophages are receiving signals that modify the IL-4/IL-13 AAM phenotype. Clearance of apoptotic cells may be one of those signals. Although argII expression is largely independent of Th2 cytokine signaling [53], it is strongly induced in macrophages during the incorporation of apoptotic bodies, where it plays a role in inhibiting classical macrophage activation [31]. In addition, post-Nb AAMs strongly up-regulated complement components such as C1q and C3a, which are known to increase the ability of lung macrophages to phagocytize and incorporate SPs and apoptotic granulocytes [38].

Furthermore, the up-regulation of genes that encode proteins involved in the metabolism of lipids such as CD68, Oldr1, PU.1, peroxisome proliferator-activated receptor (PPAR)γ, and Pex11γ (a PPARγ target gene) also supports the notion that Th2-independent stimuli are important cosignals in the post-Nb lung. The importance of lipid metabolism and the activation of PPARγ and other PPAR-associated factors have often been cited as a major factor responsible for initiating anti-inflammatory qualities in macrophages [54, 55]. Recent studies have also shown that PPARγ activation greatly enhances alternative macrophage development and along with IL-4, is a dominant signal responsible for this phenotype [56, 57]. Furthermore, it has been demonstrated that activating PPARγ in the absence of IL-4 and IL-13 is sufficient to drive M2 qualities in macrophages [58], suggesting it may play an independent role in alternative activation. In particular, the activation of these factors appears to be essential for the regulation of pulmonary inflammation by lung macrophages and may represent a crucial regulatory component capable of shifting lung macrophages to a protective phenotype [59, 60].

Over time, there were changes in the surface and cytokine profiles of lung AAMs that suggested a transition to a more anti-inflammatory state. The peak in surface expression of CD40, CD80, and CD86 on day 4 PI was followed at day 8 PI by a marked reduction in these conventional costimulatory molecules and an increase in PDL-1 and PDL-2 expression. These changes at the surface were accompanied by a significant increase in IL-10 transcription. Interestingly, PDL-1 has been shown to be up-regulated by AAMs during helminth infections and was demonstrated as an AAM mechanism of T cell suppression [61, 62]

Lung AAMs also appear to exert control on the pulmonary environment by increasing the transcription of genes encoding chemokines such as CCL11, CCL17, and CCL24, which are known to be involved in cellular recruitment during Th2 immune responses [63,64,65]. CCL17, which was up-regulated more than ten-fold on each day of infection, has been shown to be induced by AAMs as a result of IL-4 signaling [66]. CCL17 is also known to be involved in the recruitment of Th2 cells during allergic responses and asthma, further illustrating its importance in driving Th2 responses in the lung [67,68,69,70].

Interestingly, day 13 post-Nb AAMs also possess a significantly enhanced ability to incorporate antigen when compared with the lung macrophages isolated from day 4 post-Nb and day 0 uninfected lungs. The biological significance of this enhanced ability to phagocytize on a per-cell basis becomes even more striking when one considers that there are approximately six times as many macrophages present in the lungs on day 13 post-Nb than are found in uninfected controls. As expected, similar to control lung macrophages, post-Nb macrophages showed a limited ability to mediate antigen-specific CD4 T cell proliferation when compared with DC. These data suggest that post-Nb AAMs may act as an antigen sink capable of sequestering large amounts of antigen and preventing its uptake by DC, thus preventing the initiation of a strong, adaptive immune response. Furthermore, days 13–15 AAMs also showed a modestly enhanced capability to check the proliferation of CD4 T cells activated with plate-bound anti-CD3 and soluble anti-CD28, illustrating another mechanism by which AAMs are capable of dampening an adaptive immune response (Fig. 6).

In summary, alternative activation has often been viewed as an “off-on” phenotype downstream of IL-4 and IL-13 signaling. The results presented here illustrate that the development of lung AAMs is a dynamic process that appears to progress in a stepwise manner. Lung macrophages are exposed to many dominant signals as a result of the structural damage and cellular infiltration initiated during and after a Nb infection. This is evident by the distinct sets of genes that are up- and down-regulated during the various days of infection and well after initial Th2 cytokine signals are received. Included among these signals are the metabolism of oxidated lipids, the incorporation of SPs, and the exposure to apoptotic bodies found in the post-Nb lung. The notion of multiple signals contributing to the progression of lung macrophages into the alternative phenotype is illustrated further by acute transitions in gross phenotype, surface marker characteristics, cytokine production, chemokine production, and general function between days 0 and 15 PI. These data expand significantly upon past observations and more fully define the dynamics of AAMs that develop in the lung as a result of an infectious insult. Th2-independent signals are an important component of AAM in vivo that are often overlooked and should be studied further in natural systems of infection. It is likely that the lack of Th2-independent signals in other systems of study is the reason why the in vitro AAMs by IL-4 and IL-13 is considerably less intense in nature than the AAMs during a natural infection [8].