Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The lung, one of the most immunologically challenged organs of the body, is continually exposed to both innocuous and pathogenic environmental antigens. Thus, local pulmonary immune responses against invading pathogens are critical for host defense. To initiate such immune responses requires the presence of effective antigen-presenting cells (APCs) in the lung. Dendritic cells (DCs), the most efficient professional APCs, have the unique ability to induce primary immune responses in T cells (1, 2). In the lung, they are distributed within the airway epithelium, the lung parenchyma, the connective tissue surrounding major airways and vessels, the pleura, and in the pulmonary vascular bed (3). As in other nonlymphoid organs, pulmonary DCs are believed to be functionally immature and engaged primarily in immune surveillance and the uptake of antigen (7). Following antigen acquisition, and in response to locally produced chemokines/cytokines, DCs begin to mature as they migrate via afferent lymphatics into the T cell areas of regional lymph nodes (LNs) (8). DCs migrating from the periphery may not necessarily be the cells that present antigens, as they may transfer the captured antigen to other DCs for presentation in the LNs (11). Although it is clear that DCs play a pivotal role in the induction of cell-mediated immunity in general, their role in initiating a granulomatous immune response in the lung has not been examined in detail.

Immune granulomatous inflammation is a unique response to certain pathogens, including a variety of mycobacteria (12, 13) and parasites (14). Classically, granulomatous inflammation has been defined as a dense accumulation of macrophages, lymphocytes, epithelioid cells, and fibroblasts around indigestible particles or antigens (15). Although it is well known that T cell mediated immunity is involved, the role of DC in granulomatous inflammation has tended to be overlooked. Unlike macrophages that rapidly digest endocytosed antigens, DCs retain on their surface antigenic peptides bound to major histocompatibility complex (MHC) class II antigen, as well as important costimulatory molecules, for prolonged periods of time (16). However, DCs are capable of considerable plasticity in their response to antigens. The cytokine milieu in which they find themselves and the type of antigen to which they are exposed determines the type of T cell response that is elicited (17).

A number of studies indicate that DC-derived interleukin (IL)-12 is involved in the induction of mycobacterial and microbial immunity (18). Following exposure to Toxoplasma gondii (20) or Leishmania donovani (22), DCs and not macrophages are the first cells to secrete IL-12. These data suggest that DCs may act as the priming APCs for T-cell responses in granulomatous inflammation of the lungs. We hypothesized that the first step in immune-mediated granulomatous inflammation is the endocytosis by immature DCs of an immobilized antigen, which is then followed by their migration to local LNs, where they prime T cells. We examined the role of DCs in pulmonary granulomas by using a well-defined mouse model in which purified protein derivative (PPD) of Mycobacterium bovis is covalently bound to Sepharose 4B beads and embolized to the lungs (12). Here we show by in situ hybridization that IL-12-expressing DCs are detected in granulomas and that a number of inflammatory cytokine mRNAs are elevated in the lung, as determined by RNase protection assay. In addition, DCs isolated from granuloma-bearing lungs as well as from thoracic LNs have antigenic PPD peptides on their surface and are capable of stimulating PPD-specific T cell proliferation in the absence of exogenously added antigens. Our results support the premise that DCs play a key role in the initiation of cell-mediated immune-granuloma formation.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals

Female C57BL/6 mice, 6-8 wk old, were obtained from Charles River Laboratories (Kingston, NY). Mice were housed in restricted-access animal facilities at the Massachusetts General Hospital; cages, food and bedding were autoclaved. The mice were handled in a hood by gloved and masked personnel at all times. All experiments were conducted with prior approval from the Subcommittee on Research Animal Care of the Massachusetts General Hospital.

Reagents and Antibodies

PPD of M. bovis strain AN-5 was obtained from the National Veterinary Services Laboratories, Ames, IA. Hybridoma TIB120 that produces monoclonal antibody (mAb) M5/114 (rat anti-mouse MHC class II) was from American Type Culture Collection (ATCC), Manassas, VA. MOMA2 mAb (rat anti-mouse monocyte/macrophage) was from Biosource International, Camarillo, CA. Rat anti-mouse CD3, CD4, CD8 mAbs, hamster anti-mouse CD11c mAb, phycoerythrin (PE)-conjugated hamster anti-mouse CD11c mAb, fluorescein isothiocyanate (FITC)-conjugated rat anti-mouse B220 mAb (CD45R), and rat anti-mouse Fc II/III receptor (CD16/CD32) mAb were from BD PharMingen, San Diego, CA. Polyclonal rabbit anti-PPD was purchased from Fitzgerald Industries International, Concord, MA. Secondary antibodies included biotinylated rabbit anti-rat immunoglobulin G (IgG), goat anti-rabbit IgG (Vector Laboratories, Burlingame, CA), and biotin-conjugated mouse anti-hamster IgG (BD PharMingen). Recombinant mouse granulocyte macrophage-colony-stimulating factor (GM-CSF) was a gift of Genetics Institute, Cambridge, MA. Recombinant human IL-2 was obtained from the National Institutes of Health AIDS Research and Reference Reagent Program, Rockville, MD. [³H]-thymidine was from Perkin-Elmer Life Sciences, Inc., Boston, MA. All other reagents were from Sigma Chemical Co., St. Louis, MO.

Immunization and Granuloma Induction

Mice were anesthetized by intraperitoneal injection of 4% chloral hydrate (0.4 mg/g of body weight). They were immunized by subcutaneous injection at the base of the tail with an emulsion of PPD (0.2 mg/mouse) in complete Freund's adjuvant (CFA) (Difco Laboratories, Detroit, MI). Normal nonimmunized mice served as controls. Two weeks later, the mice were challenged by intravenous injection of 1 × 10⁴ PPD-coupled cyanogen bromide (CNBr)-activated Sepharose 4B beads (Amersham Pharmacia Biotech Inc., Piscataway, NJ) in 0.5 ml phosphate-buffered saline (PBS) or control beads. PPD was coupled to beads according to the manufacturer's protocol. Briefly, 75 mg of beads were swollen and coupled with 2.5 mg of PPD in 0.626 ml of coupling buffer (0.1 M NaHCO₃, 0.5 M NaCl, pH 8.3) on a rotary mixer for 2 h at room temperature. After washing with coupling buffer, the remaining active sites were blocked with 0.1 M Tris-HCl buffer (pH 8.0) for 2 h at room temperature. Control beads were processed as above without PPD. The beads were washed using three alternate cycles of 0.5 M NaCl in 0.1 M acetate buffer (pH 4.0) followed by 0.5 M NaCl in 0.1 M Tris-HCl buffer (pH 8.0). They were resuspended in sterile PBS at 2 × 10⁴ beads/ml. At 6 and 12 h, and at 1, 3, and 7 d, the lungs of anesthetized mice were inflated with 0.6 ml of 1:1 OCT in PBS. Five blocks of lung per mouse were embedded in Tissue Tek OCT (Sakura Finetek, USA Inc., Torrance, CA) and stored at 80°C. Cryostat sections, 3-µm thick, were air-dried overnight and immunolabeled.

Immunostaining and Morphometric Analysis of Granulomas

For cell immunophenotyping and detecting PPD antigen, cryostat sections were stained using the avidin-biotin immunoperoxidase technique as described (23). Briefly, after fixing in acetone for 10 min at room temperature, sections were blocked with normal serum of the same species from which the biotinylated 2° antibody was obtained. Endogenous peroxidase activity was inhibited with 0.3% H₂O₂ in PBS, and endogenous biotin and avidin activity was blocked by sequential incubation with 100 µg/ml avidin and 20 µg/ml biotin in PBS for 20 min each. Sections were incubated for 1 h with 1° antibody or PBS as control, followed by incubation with biotinylated 2° antibody for 45 min. To detect M5/114, MOMA2, and anti-PPD antibodies, sections were incubated with avidin-biotin peroxidase complex (ABC Elite kit, Vector Laboratories) for 30 min and the reaction product was generated by incubation with 0.03% H₂O₂, 0.03% 3-amino-9-ethylcarbazole (Aldrich Chemical Co., Milwaukee, WI), 5% N-N-dimethylformamide in 0.1 M acetate buffer, pH 5.0. To detect CD11c, CD4, CD8 mAbs the sections were incubated with biotinylated 2° antibodies followed by horseradish peroxidase-conjugated streptavidin for 30 min, and the reaction product was generated using a diaminobenzidine tetrahydrochloride substrate kit (BD PharMingen). Sections were counterstained with Gill's hematoxylin No. 2 (Fisher Scientific, Fairlawn, NJ) and coverslipped using glycergel (Dako Corp., Carpinteria, CA).

The size of each granuloma was measured on coded slides. Maximum and minimum diameters of the granuloma and central bead were measured separately using a 1 cm² eye piece graticule, divided into 10 × 10 squares. The radius of the granuloma (R) and bead (r) was calculated from the mean of the maximum and minimum diameters (×1/20 mm). The area of the granulomas was calculated as (R²  r²) × 1/400 mm². A minimum of 36 granulomas per category was measured. The number of immunolabeled cells in each granuloma was counted. and their number divided by the area of the granuloma.

DC Isolation from Lungs

DCs were isolated from the lungs of nonimmunized and PPD-immunized mice 1 and 3 d after the injection of noncoated or PPD-coated beads and from lungs of untreated mice. To avoid contamination with peripheral blood leukocytes and alveolar macrophages, the lungs were perfused via the pulmonary artery with 12 ml and then lavaged with six 0.8-ml aliquots of 0.6 mM ethylene diaminetetraacetic acid/PBS, respectively. To facilitate dissection of the lung parenchyma, 1 ml of SeeKem agarose solution (FMC Bioproducts, Rockland, ME) solution (1% agarose, 5% fetal calf serum [FCS] in PBS) was instilled intratracheally (4). After tying off the trachea, the lungs were removed and immersed in ice-cold PBS for 30 min to solidify the agarose. The lung parenchyma was dissected from the airways, minced and digested with 150 U/ml collagenase, type I (Worthington Biochemical Corp., Freehold, NJ) and 50 U/ml DNase I in complete medium (CM) (RPMI 1640, 1% P/S, 50 µM 2-mercaptoethanol [2-ME], 5 mM Hepes), 10% FCS for 90 min at 37°C. Low-density cells were retrieved by bovine serum albumin density gradient sedimentation (1.080 g/ml, refractive index 1.3845) (4). Briefly, 2.5 ml of 2 × 10⁷ cells/ml were suspended in the above-described albumin solution and overlaid with 1 ml of two parts dense BSA to one part PBS. After centrifugation at 10,000 × g for 30 min at 4°C, low-density cells were harvested at the interface, and incubated in culture dishes with CM, 5% FCS for 2 h at 37°C. Nonadherent cells were removed by gentle rinsing with PBS at 37°C, and the adherent fraction was cultured overnight at 37°C in CM, 5% FCS, 250 U/ml GM-CSF. The next day floating cells were collected. Dead cells were removed by centrifugation with Histopaque 1,077. DCs were enriched by positive, immunomagnetic cell separation on a Vario MACS device (Miltenyi Biotech, Auburn, CA). Briefly, after blocking with 10% normal mouse serum in MACS buffer (0.5% BSA, 2 mM EDTA, 50 U/ml DNase I in PBS, pH 7.3) for 5 min at 6°C, the cells were incubated with 20 µg/ml M5/114 mAb in MACS buffer for 15 min at 6°C. They were washed and incubated with goat anti-rat IgG conjugated to magnetic beads (Miltenyi Biotech) in MACS buffer for 15 min at 6°C. After washing, the cells were applied to a MACS RS⁺ column, placed in the magnetic field, and the column was washed with 7× 500 µl of MACS buffer. After removing the column from the magnetic field, the positive cell fraction was eluted. The preparations were enriched for MHC class II⁺ cells (~ 90% purity) and yielded ~ 3-5 × 10⁴ DC/mouse.

DC Isolation from Thoracic LNs

Thoracic LNs were harvested from 12-14 granuloma-bearing or control mice, and DCs were isolated (24). Briefly, the LNs were teased in 100 U/ml collagenase, 50 U/ml DNase I in CM, 10% FCS. The released cells were filtered through a cell strainer (70-µm pore size; Becton Dickinson, Sunnyvale, CA), and pooled in CM, 10% FCS on ice. The remaining tissue fragments were digested with 400 U/ml collagenase, 50 U/ml DNase I in CM, 10% FCS with gentle agitation for 30 min at 37°C. The digested fragments were passed through an 80-mesh metal screen, filtered through a cell strainer, and pooled in CM, 10% FCS on ice. Low-density cells were retrieved by BSA density gradient sedimentation and incubated in CM, 5% FCS, 250 U/ml GM-CSF overnight at 37°C. The floating cells, enriched for DC, were collected and CD11c⁺ B220 DCs were sorted from CD11c B220⁺ B cells on a FACS Vantage SE (Becton Dickinson) using PE-CD11c and FITC-B220 mAbs. The sorted fractions were > 92% pure.

Generation of PPD and Hen Egg Lysozyme-Specific T Cells

Mice were immunized with an emulsion of 150 µg of PPD or 100 µg of hen egg lysozyme (HEL) in CFA injected subcutaneously in the base of the tail. Two weeks later inguinal LNs were harvested, teased, and the released cells passed through a cell strainer. They were plated in 24-well culture plates, 2 × 10⁶ cells/ml, in CM, 5% FCS with 150 µg/ml of PPD or 100 µg/ml of HEL, respectively. IL-2 (10 U/ml) was added to the medium on Day 7. Medium was refreshed twice a week by removing 0.5 ml of spent medium and adding 0.5 ml of fresh CM, 5% FCS with 10 U/ml IL-2. Cells were maintained between 5 × 10⁵ and 9 × 10⁵ cells/ml at all times. Every 4 wk the cultures were restimulated with PPD or HEL, respectively, using irradiated (3,000 rad) syngeneic spleen cells at a ratio of 10 spleen cells to 1 T cell. T cells that were 4 wk from the last restimulation were used in antigen presentation assays. Antigen specificity of the T cells was tested periodically using irradiated spleen cells as APC.

Antigen Presentation Assay

Isolated lung or LN DCs were irradiated (1,000 rad), suspended in CM, 5% FCS and plated in 96-well flat-bottom culture plates at 10⁴, 10³, or 10² cells/well. PPD- or HEL-specific T cells (5 × 10⁴ cells/well) were added with or without exogenous PPD or HEL (30 µg/well). After incubating for 72 h at 37°C, the cells were pulsed with [³H]-thymidine (1 µCi/well) for 18 h. They were harvested in a cell harvester (Skatron Inc., Sterling, VA) and radioactivity was counted in a Tri-Carb liquid -scintillation spectrometer (Packard Instrument Co., Downers Grove, IL). In each assay, irradiated spleen cells, 4 × 10⁵ cells/well, with PPD- or HEL-specific T cells and added PPD or HEL, respectively, served as positive controls. Negative controls included: (i) the addition of HEL-specific T cells to lung DC harvested from PPD granuloma bearing lungs, and (ii) the addition of HEL (30 µg/well) to isolated lung DCs and PPD-specific T cells. T cell proliferative response was calculated as DC + PPD-T cpm / PPD-T cpm. Data are expressed as mean ± SEM.

Synthesis of IL-12 p40 Riboprobe for In Situ Hybridization

Murine IL-12 p40 cDNA in pBluescript II SK+ (ATCC #87595) was used as template to generate a 252 bp IL-12 fragment by PCR. The primers consisted of a forward primer containing an EcoRI site, 5'-TGGCTTATGAATTCCCTCACCTGTGACACG CCTGA-3' and a reverse primer containing a HindIII site 5'-TCA CAGCAAAGCTTGGTGCTTCACACTTCAGGAAAG-3'. The amplified cDNA was digested with EcoRI and HindIII, cloned into pBluescript II SK+ and confirmed by dideoxy sequencing. The 252 bp IL-12 fragment in pBluescript SK+ was linearized by digestion either with EcoRI or HindIII. Digoxigenin-labeled antisense and sense RNA probes (252 nt, corresponding to bases 175-426 of Genbank Acc. No. M86671) were prepared by reverse transcription using either T7 or T3 polymerase, respectively, using a DIG RNA labeling kit (Roche Molecular Biochemicals, Indianapolis, IN).

Detection of IL-12 p40 mRNA by In Situ Hybridization

For in situ hybridization granulomatous lungs were prepared as follows. Anesthetized mice were bled and the lungs sequentially perfused with 24 ml of 0.6 mM EDTA in PBS followed by 24 ml of 4% paraformaldehyde in PBS via the pulmonary artery. The lungs were then lavaged with five 0.7-ml aliquots of 2% paraformaldehyde, 30% sucrose in PBS. After intratracheal instillation of the last 0.7 ml, the trachea was tied off and the lungs were removed and stored overnight in the latter solution at 4°C. Blocks of lung were embedded in OCT and stored at 80°C.

Frozen sections (10 µm thick) were cut at -35°C and air-dried on poly-L-lysine-coated slides. They were treated with 1 µg/ml proteinase K in 0.1 M Tris-HCl, 0.05 M EDTA (pH 8.0) for 15 min at 37°C. After washing with PBS, sections were acetylated with 0.25% acetic anhydride in 0.1 M triethanolamine-HCl, pH 8.0 for 10 min at room temperature. The sections were washed twice with 2× saline sodium citrate (SSC), and prehybridized with hybridization solution (50% formamide, 4× SSC, 250 µg/ml yeast RNA, 1× Denhardt's, 10% dextran sulfate) at 37°C for 1 h. Digoxigenin-labeled antisense or sense RNA probe (1 µg/ml) in hybridization solution was applied and the sections were incubated at 42°C overnight. They were washed in 2× SSC and 1× SSC, each for 1 h at RT, 0.5× SSC for 30 min at 37°C, and 0.5× SSC for 30 min at room temperature. To detect bound digoxigenin-labeled probes, the sections were blocked in Buffer I (100 mM Tris-HCl, 150 mM NaCl, pH 7.5), 2% sheep serum, 0.1% TX-100. They were then incubated for 3 h with alkaline phosphatase- coupled anti-digoxigenin antibody diluted 1:250 in Buffer I with 1% sheep serum, 0.1% TX-100. The sections were washed in Buffer I and equilibrated in Buffer II (100 mM Tris-HCl, 100 mM NaCl, 50 mM MgCl₂, pH 9.5). Color was developed by incubating in 4-nitro-blue tetrazolium and 5-bromo-4-chloro-3-indolyl-phosphate in Buffer II (Roche Molecular Biochemicals) according to the manufacturer's protocol.

RNAse Protection Assay

Experimental and control lungs were placed in RNALater (Ambion Inc., Austin, TX) and stored at 4°C. They were homogenized with a polytron homogenizer using the Ultorspec-II RNA Isolation System (Biotecx, Houston, TX). Total RNA was isolated according to the manufacturer's instructions.

The MCK-2b Multi-Probe Template Set (BD Pharmingen) was used to generate [-³³P]UTP-labeled antisense probes with the Maxiscript Kit (Ambion) following the manufacturer's instructions. Probes were hybridized to 35 µg of total RNA at 56°C overnight and RNAse digestion was performed using the RPA III kit (Ambion) according to the manufacturer's protocol. The protected fragments were resolved on a 5% polyacrylamide sequencing gel. The dried gel was analyzed using a Model GS-363 Molecular Imager System (BioRad Laboratories, Hercules, CA) and bands were quantified by densitometry using BioRad Quantity One software. Each cytokine band was normalized to the housekeeping glyceraldehyde-3-phosphate dehydrogenase RNA.

Statistics

Values from each group were compared by an analysis of variance (Sigma Plot; SPSS Inc., Chicago, IL). The data are presented as the mean ± standard error of the mean. Differences were considered statistically significant when P < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Morphometric Analysis of Granulomas

The purpose of the present study was to identify in situ the cells that are involved in initiating a granulomatous immune response and to determine the role of DCs in this process. Our efforts, therefore, focused primarily on Days 1 and 3 after the injection of antigen-coated beads. At earlier time points (6 and 12 h), no bead-associated cells were detected, and by Day 7 the size of the granulomas had diminished significantly (Figure 1). By Day 1 a mixture of inflammatory cells were aggregated around the beads; their number and the size of the resulting granulomas were 3-fold (P < 0.001) larger and more cellular on Day 1 than in age-matched, nonimmunized animals (Figure 1). By Day 3, however, the difference had decreased to 2-fold (P < 0.001).

View larger version (15K): [in this window] [in a new window]   Figure 1.   Morphometry of granulomas. PPD-coated beads were injected IV into nonimmunized (open bars) or PPD-immunized (solid bars) mice, and the lungs were harvested 1, 3, or 7 d later. Frozen sections were immunolabeled with M5/114 mAb. The size of a minimum of 36 granulomas in each category was measured on coded slides as described in MATERIALS AND METHODS. The asterisks indicate a statistically significant difference of P < 0.001.

To identify and enumerate the cells in the granulomas, DCs, macrophages, and T cells were immunolabeled, counted in situ, and their number related to the size of the granuloma. DCs were identified as relatively large cells with lobulated nuclei that were MHC class II⁺ or CD11c⁺. In immunized mice, significant numbers of DCs, macrophages, and T cells were recruited to the beads by Day 1 (Figure 2A). Their number per unit area increased, albeit at a slower rate, by Day 3 (Figure 2B). By contrast, in nonimmunized mice the initial recruitment was delayed and did not approach levels seen in immunized mice until Day 3 (Figures 2A and 2B). The number of large MHC class II⁺ cells was consistently greater than CD11c⁺ cells (1,915 versus 261/mm² on Day 1 and 2,160 versus 705/mm² on Day 3 in immunized mice) (Figure 2), suggesting that MHC class II⁺ immature DCs and some macrophages were included in these counts. The number of CD11c⁺ cells, therefore, represents the more highly differentiated DCs in these granulomas.

View larger version (27K): [in this window] [in a new window]   Figure 2.   Quantification of cells in granulomas. The mean number of M5/114+, MOMA2+, CD11c+, CD4+, and CD8+ cells per granuloma area in nonimmunized (open bars) and PPD-immunized (solid bars) mice was counted on Days 1 and 3 after the injection of PPD-coated beads. Significantly larger numbers of these cells were present in immunized than in nonimmunized mice at Day 1 (A) (P < 0.001). The difference in the number of M5/114+ and MOMA2+ cells was less by Day 3 (P < 0.01) and was no different for CD11c+ DCs and CD4+, CD8+ T cells (B). The predominant cells were MOMA2+ macrophages. The number of CD11c+ DC/mm2 was ~ 3.6-fold (Day 1) and 1.1-fold (Day 3) higher in PPD-immunized than in nonimmunized mice. A minimum of 36 granulomas per category was counted on coded slides. Asterisks indicate a statistically significant difference (P < 0.01).

Immunohistochemical Phenotype of Granuloma-Associated Cells

To determine the phenotype of the cells and their location in the granulomas, we next analyzed immunostained granulomas in situ. Regardless of whether the animals were immunized, MHC II⁺, CD11c⁺ DCs were among the cells that were found in close proximity to the beads (Figures 3a-3h). Their number increased by Day 3 (Figures 3b, 3d, 3f, and 3h), at which time they were observed both adjacent to the beads as well as dispersed throughout the granuloma (Figures 3b and 3d). Both CD4⁺ (Figures 3m-3p) and to a lesser degree CD8⁺ (not shown) T cells were noted in the granulomas. Their number was significantly greater in immunized as compared with nonimmunized mice on Day 1, but this difference was diminished by Day 3. Both types of T cells were located primarily at the periphery of the granuloma. The distribution of CD11c⁺ DCs was more difficult to determine, as there appeared to be artifactual background staining (Figures 3e- 3h). However, when intensely stained cells with lobulated nuclei were counted, fewer CD11c⁺ cells were observed in the granulomas of nonimmunized mice as compared with those in immunized mice. Furthermore, the distribution of the intensely labeled CD11c⁺ cells (Figures 3f and 3h) coincided approximately with those of the MHC class II⁺ cells (Figures 3b and 3d). Cells with the morphologic features of DCs were not labeled with anti-CD8 and few were labeled with anti-CD4. Immunostaining of granulomas with anti-PPD antibody clearly revealed the presence of PPD antigen on the surface of PPD-coated beads on Day 1 and somewhat reduced staining was observed by Day 3 (data not shown).

View larger version (109K): [in this window] [in a new window]   Figure 3.   Immuno-phenotype of granuloma-associated cells. Frozen sections of lungs of nonimmunized or PPD-immunized mice on Days 1 or 3 after the intravenous injection of PPD-coated beads were immunolabeled for MHC class II (a-d), CD11c (e-h), macrophage antigen (i-l), and CD4 (m-p). In nonimmunized mice on Day 1, the beads were surrounded by a limited number of inflammatory cells (a, e, i, and m). By contrast, on Day 1 in immunized mice, PPD-coated beads were surrounded by a mixture of inflammatory cells (c, g, k, and o) that included MHC II+ cells with dendritic profiles (c), CD11c+ DCs (g), MOMA2+ macrophages (k), CD4+ (o), and CD8+ T cells (not shown). By Day 3, granulomas in both nonimmunized and immunized mice contained increased numbers of MHC II+ cells with dendritic shapes (b and d) and MOMA2+ macrophages (j and l). Intensely stained CD11c+ DCs (f and h), CD4+ (n and p), and CD8+ (not shown) T cells, however, were more numerous in the granulomas of immunized than in nonimmunized mice. Scale bar: 100 µm.

RNAse Protection Assay Reveals Elevated Levels of Several Inflammatory Cytokine mRNAs

To examine a sufficient number of mice in each group, RNAse protection assays were conducted on whole-lung mRNA preparations. Using this approach, four to six mice per experimental group were examined. In both nonimmunized and immunized mice on Days 1 and 3, IL-1, IL-1Ra, IL-6, interferon-, and macrophage inhibitory factor were elevated above normal controls (Figure 4). There was no change in IL-1 levels, and IL-18 mRNA was below control levels on Day 1 and returned to control values by Day 3. The density of the radioactive bands for IL-12 p35 and IL-12 p40 mRNA was weak in the total RNA harvested from the granulomatous lungs and could not be reliably measured. We therefore elected to examine granulomas for the presence of IL-12 p40 mRNA by in situ hybridization.

View larger version (55K): [in this window] [in a new window]   Figure 4.   RNAse protection assay. Total RNA from whole lung was harvested from nonimmunized () and immunized (+) mice at Days 1 and 3 after intravenous injection of PPD-coated beads (open bars, 1 d; shaded bars, 1 d+; striped bars, 3 d; striped and shaded bars, 3 d+). Normal, untreated mice served as controls (solid bars). A representative gel and the mean of four separate determinations are shown graphically. Except for IL-18, the levels of IL-1, IL-1Ra, IL-6, and IFN- are elevated above controls. The faint band density of IL-12 p35 and IL-12 p40 could not be reliably measured.

IL-12 p40 mRNA Expression in Granuloma Associated DCs

We initially attempted to use a full-length, 1-kb, IL-12 p40 riboprobe on frozen sections with little success, and opted, therefore, to use a shorter 252-nt probe on formaldehyde-fixed tissue (25). A Blast search using this probe sequence did not detect any homology with other cRNA sequences. Sections from immunized mice obtained at Days 1 and 3 after the injection of antigen-coated beads, were hybridized with a 252-nt, digoxigenin-labeled antisense IL-12 p40 riboprobe. Relatively large, elongated cells, similar in appearance to MHC class II⁺, CD11c⁺ DCs, were detected using the antisense, but not the sense, riboprobe (Figures 5A and 5B). Similar labeling was observed on Day 3 (not shown).

View larger version (89K): [in this window] [in a new window]   Figure 5.   In situ hybridization for IL-12 p40 mRNA. Frozen sections of fixed lungs of immunized mice, at Day 1 after injection of PPD-coated beads, were hybridized with digoxigenin-labeled IL-12 p40 antisense (a) and sense (b) riboprobes. IL-12 p40 mRNA is present in large, elongated cells that are in a similar distribution as those seen in M5/114 and CD11c immunolabeled granulomas. Scale bar: 50 µm.

PPD Antigen Is Present on the Surface of DCs Harvested from Granulomatous Lungs

We initially attempted to specifically isolate the granulomas with their included DCs from the lungs. However, because there was no connective tissue associated with these recently formed granulomas, the retrieved beads lacked any adherent cells. We therefore harvested DCs from the whole lung parenchyma 1 and 3 d after IV injection of PPD-coupled beads into nonimmunized and immunized mice. The presence of PPD antigenic peptides on the surface of these cells was then assessed by adding PPD-sensitized T cells to graded numbers of lung DCs in the absence of exogenously added antigen. DCs from the granulomatous lungs 1 d after the injection of beads induced significant antigen-mediated T cell proliferation. The T cell proliferative response was higher using DCs derived from immunized (43.5 ± 2.8) than from nonimmunized mice (24.1 ± 1.9) (P < 0.001) (Figure 6A). The proliferation index elicited by DC from nonimmunized and PPD-immunized mice was 8.7- and 15.7-fold higher, respectively, than that of nonimmunized mice injected with noncoupled beads (P < 0.001). By contrast, DCs harvested from the lungs of immunized mice 3 d (Figure 6B) after the induction of granulomas elicited a significantly lower T cell proliferative response (7.0 ± 1.0) than at Day 1. The response in nonimmunized mice at Day 3 was similar to controls despite the fact that granulomas were clearly present in these animals (Figure 3).

View larger version (14K): [in this window] [in a new window]   Figure 6.   DCs were isolated from granulomatous lungs 1 (A) and 3 (B) d after intravenous injection of either PPD-coated beads (Pb) or noncoated beads (Ncb) into nonimmunized (Ni) or PPD-immunized (Pi) mice. DCs were also isolated from thoracic lymph nodes of immunized mice 1 and 3 d after injection of PPD-coated beads (C). DCs isolated from untreated mice served as normal controls (Control). Antigen presentation assays were conducted with isolated lung or lymph node DCs and PPD-sensitized T cells without exogenously added antigen. Negative controls included DCs and HEL-sensitized T cells (HEL-Control) and DC from nonimmunized mice injected with noncoated beads. Positive controls (not shown) using lung DCs, PPD-sensitized T cells, and exogenous PPD-antigen elicited a [3H]-thymidine incorporation of 120,755 ± 5,950 cpm and a proliferative index of 556 ± 38. The latter data were derived from 12 separate experiments. The remaining data are expressed as mean ± standard error of the mean from a representative experiment of a minimum of three experiments each.

Antigenic Peptides Are Detected on Thoracic LN DCs

DCs from thoracic LNs of immunized mice injected with PPD beads induced significant antigen-specific T cell proliferation. In contrast to those isolated from the granulomatous lungs, LN DCs elicited a T cell proliferative response on both Days 1 and 3. However, as with lung DCs, the T cell proliferative response elicited by LN DCs was highest on Day 1 (84.1 ± 4.4) and declined by Day 3 (51.0 ± 3.5) (P < 0.001) (Figure 6C). No T cell proliferative response was observed when HEL-immune T cells were added to LN DCs. Lymph node DCs from normal mice did not elicit a proliferative response from syngeneic PPD-sensitized T cells.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Mobilization of DCs is pivotal to immune surveillance and to eliciting an immune response (2, 7). The sentinal and migratory functions of DCs in the lung has been examined extensively using intratracheal instillation of both soluble (9, 26, 27) and particulate (10) antigens. The role of DCs in a pulmonary granulomatous immune response, however, has been less well studied. To explore the contribution of DCs in cell-mediated immune granulomatous response in the lung we used a well-defined experimental model in which PPD-coated Sepharose beads are embolized into the lung (12, 28). Although vascular embolization does not mimic the inhalational route of mycobacterial infection, it provides an experimental model in which synchronously produced foci of granulomatous inflammation can be examined. As originally reported, this self-limiting model of granulomatous inflammation in CBA/J mice results in regression of the granulomas by the eighth day after injection (12). Using C57BL/J mice, we found that the granulomas were already substantially reduced in size by Day 7. We therefore focused our studies on granulomas at Days 1 and 3 after embolization, a time when they were still increasing in size. We compared the number and phenotype of the granuloma-associated cells as well as the antigen-presenting function of lung and thoracic lymph node DCs in animals that were either not immunized or immunized 2 wk before embolization.

Not surprisingly, the formation of granulomas was accelerated in immunized mice, resulting on Day 1 in substantially larger granulomas with greater numbers of DCs, macrophages, and T cells per unit granuloma area than in nonimmunized mice. The cells adjacent to the beads were primarily DCs and macrophages, whereas T cells were arrayed in the peripheral zones of the granuloma. This was best appreciated in the smaller granulomas formed in nonimmunized mice, in which MHC class II⁺, CD11c⁺ DC were among the first cells to localize close to the beads. By Day 3 DCs were detected throughout the granuloma, suggesting that these cells were actively trafficking into and out of the granuloma.

Results of antigen presentation assays with DCs isolated from both the lung and thoracic lymph nodes clearly showed that DC had in fact endocytosed and transported antigen to thoracic lymph nodes. These DCs induced the proliferation of PPD-sensitized T cells in vitro without the addition of exogenous antigen. The T cell response was specific because HEL immune T cells failed to respond. Surprisingly, however, DCs isolated from the lungs of immunized mice on Day 3 produced a markedly reduced T cell proliferative response and those from nonimmunized mice failed to do so. To determine whether this was due to a loss of antigen from the beads, the lungs were immunolabeled with an anti-PPD antibody. The degree of positive immunostaining on the beads was strongest on Day 1; however, reaction product was still detectable on the beads on Day 3. Although immunocytochemical staining neither quantifies nor assesses the condition of the antigen remaining on the bead, antigen presentation assays clearly detected antigen-bearing DCs in thoracic lymph nodes on both Days 1 and 3. These observations suggest that factors present in the granuloma on Day 3 markedly downregulated the antigen-presenting function of DCs.

Our original goal was to specifically isolate those DCs that were associated with granulomas. However, because of the lack of collagen deposition, the granulomas did not withstand the isolation procedure. For this reason we had to resort to harvesting DCs from the entire lung parenchyma. Although yielding sufficient numbers of cells, the dilutional effect of non-granuloma-associated lung DCs significantly reduced the radioactive counts due to antigen specific T cell proliferation. We estimated the fraction of DCs associated with the granulomas based on the measured size and number of the granulomas, their DC content, and the size and thickness of the lung sections and the total lung volume. In immunized mice the granuloma-associated M5/114⁺ DCs comprised ~ 11.5 and 35.0% of total lung DCs on Days 1 and 3, respectively, whereas in nonimmunized animals the granuloma-associated DCs constituted 1.2 and 12.3% of total lung DCs during the same period of time. Based on these estimates, the level of antigen-induced T cell proliferation underestimated the antigen-presenting capacity of antigen-primed DCs. These considerations notwithstanding, the ability of the isolated lung DCs to present antigen was significantly depressed on Day 3 relative to Day 1. To account for these observations we considered the possibility that IL-10 released from granuloma and/or alveolar macrophages (29, 30) or even from DCs themselves (31), might have inhibited DC function by downregulating IL-12 expression (32). However, no IL-10 mRNA was detected by RNAse protection assays on either Days 1 or 3, confirming previous observations in this model (28). Further studies will be required to determine whether other inhibitory factors are involved, including macrophage-derived NO, TGF-, or prostaglandin E₂, factors that are known to inhibit lung DC function (29).

The induced expression of IL-12 in DCs is usually relatively short-lived, and occurs during the first 24 h following an inflammatory stimulus (32). Recently it has been shown that infection of murine DCs with M. tuberculosis stimulates the secretion of IL-12 by these cells (33). However, as DCs mature in vitro they become refractory to bacterial IL-12 inducers (34). In the present study IL-12 mRNA was detected in granuloma-associated cells, some of which had the elongated morphology of DCs, by in situ hybridization and RNAse protection assays on both Days 1 and 3. This suggested either a continuous turnover of granuloma-associated DCs and/or a contribution by macrophages. It is, therefore, likely that the cells expressing IL-12 mRNA in granulomas on Day 3 represents a subset of newly arrived immature DCs and also some macrophages. To account for the significantly reduced APC function of lung DCs on Day 3, but the vigorous antigen presentation by thoracic lymph node DCs on both Days 1 and 3, the following possibilities are considered: (i) the inhibitory effect of factors prevailing in the granuloma milieu may be reversed as DCs transit via lymphatics to local lymph nodes; (ii) Antigenic peptides are known to be retained on the surface of DCs for up to 1 wk (35) and may be transferred to bystander DC in lymph nodes (11). Therefore, antigen presented by lymph node DCs harvested on Day 3 may, in part, represent peptides that were endocytosed and processed during the first 24 h of the granulomatous response. The reduced accessory cell function of granuloma associated DCs at Day 3 could explain, in part, the self-limiting response of the granulomatous reaction observed in this model. However, to further examine the cell-mediated granulomatous response to mycobacteria per se will require the use of the organism itself.