Host response of platelet-activating factor receptor-deficient mice during pulmonary tuberculosis

Abstract

Platelet-activating factor (PAF) is a phospholipid with potent, diverse actions, which has been implicated as an important mediator in host defence against several intracellular pathogens. To determine the role of PAF in host defence in pulmonary tuberculosis, PAF receptor-deficient (PAFR−/−) and wild-type (PAFR+/+) mice were infected intranasally with a virulent strain of Mycobacterium tuberculosis. Mycobacterial outgrowth in lungs and liver did not differ significantly between PAFR−/− and PAFR+/+ mice at 2 or 6 weeks postinfection. After 28 weeks, 86% of PAFR−/− mice and 79% of PAFR+/+ mice had died (non-significant). In addition, both mouse strains were indistinguishable with respect to histopathology, the recruitment and activation of lymphocytes, and cytokine concentrations in the lung. These data suggest that PAF is not involved in the protective immune response to tuberculosis.

Introduction

Tuberculosis (TB) is a re-emerging disease, affecting patients in both developing and industrialized countries.¹ The increasing incidence of antibiotic resistance, together with synergism between human immunodeficiency virus (HIV) and TB, has increased our interest in this important infectious disease and in mechanisms contributing to antimicrobial host defence. Resistance to mycobacterial infections is mediated mainly by macrophages and T cells and requires the formation of granulomas, characterized by lymphocytes, macrophages and granulocytes.² Their interaction is dependent on the interplay of cytokines and chemokines produced by different inflammatory cells.²

Platelet-activating factor (PAF) is a potent phospholipid mediator that plays an important role in inflammatory and immune responses.³ PAF is produced by a large number of cells, including platelets, endothelial cells, stromal cells, lymphoid tissue and neutrophils.⁴ The biological activity of PAF is mediated through a specific G-protein-coupled receptor (PAFR) on the membrane of responsive cells, which has been identified on many haemopoietic cells, including neutrophils, dendritic cells, macrophages and monocytes.^4,5 Recent studies have suggested that endogenous PAF may play an important role in stimulating an adequate immune response to intracellular microorganisms, such as Leishmania amazonensis and Trypanosoma cruzi. Indeed, treatment with PAF antagonists was found to increase the outgrowth of microorganisms and mortality in murine models of these infections.^6,7 In accordance, PAF reduced the intracellular growth of Leishmania and Trypanosoma in macrophages.^6,7 Notably, M. tuberculosis is an intracellular microorganism that uses macrophages as its natural environment in the host, and many of the host defence mechanisms known to be important for the protection against M. tuberculosis are also involved in the protective immune response to other intracellular pathogens, including Leishmania and Trypanosoma.^2,8–10 These findings led us to hypothesize that PAF may also be important for host defence against M. tuberculosis. Therefore, in the present study we sought to determine the role of PAF in the immunopathology of TB.

Materials and methods

Mice

PAFR gene-deficient (PAFR−/−) mice were generated as described previously.¹¹ For the experiments described here, female PAFR−/− mice, backcrossed seven times to a C57BL/6 background, and female wild-type C57BL/6 (PAFR+/+) mice (Harlan Sprague Dawley Inc., Horst, the Netherlands), were used at 6–8 weeks of age. The Animal Care and Use Committee of the University of Amsterdam (Amsterdam, the Netherlands) approved all experiments.

Experimental infection

Pulmonary TB was induced exactly as described previously.^12–14 Briefly, a virulent laboratory strain of M. tuberculosis H37Rv was grown for 4 days in liquid Dubois medium containing 0·01% Tween-80. A replicate culture was incubated at 37°, harvested at mid-log phase, and stored in aliquots at −70°. For each experiment, a vial was thawed and washed twice with sterile 0·9% NaCl. Mice were anaesthetized by inhalation of isoflurane (Abbott Laboratories, Kent, UK) and infected with 1 × 10⁵ live bacilli in 50 µl of saline, as determined by viable counts on 7H11 Middlebrook agar plates. Bacterial administration was performed intranasally, as described previously.^12–14 Survival was monitored for 200 days in 14 PAFR−/− and 14 PAFR+/+ mice. In addition, groups of eight mice per time-point were killed 2 or 6 weeks postinfection, and the lungs and one lobus of the liver were removed aseptically. Organs were homogenized using a tissue homogeniser (Biospec Products, Bartlesville, OK), in 5 volumes of sterile 0·9% NaCl, and 10-fold serial dilutions were plated on Middlebrook 7H11 agar plates to determine bacterial loads. Colonies were counted after 21 days of incubation at 37°. Numbers of colony-forming units (CFU) are provided as total in the lungs or as total/g of liver tissue. For cytokine measurements, lung homogenates were diluted 1 : 1 in lysis buffer (150 mm NaCl, 15 mm Tris, 1 mm MgCl.H₂O, 1 mm CaCl_2, 1% Triton-X-100, 100 µg/ml pepstatin A, leupeptin, and aprotinin), and incubated on ice for 30 min. Supernatants were sterilized using a 0·22-µm filter (Corning, Corning, NY) and frozen at −20° until required.

Histological analysis

The right lungs of six PAFR−/− and six wild-type PAFR+/+ mice were removed 2 or 6 weeks after intranasal inoculation with M. tuberculosis and then fixed for 24 hr in 4% paraformaldehyde in phosphate-buffered saline (PBS). After embedding in paraffin wax, 4-µm-thick sections were stained with haematoxylin & eosin or the Ziehl–Neelsen (ZN) stain for acid-fast bacilli. All slides were coded and semiquantitatively scored for the total area of inflammation (percentage of surface of the slide) and granuloma format by a pathologist. In separate experiments, organs of six uninfected PAFR−/− and wild-type PAFR+/+ were harvested and examined as described above.

Fluorescence-activated cell sorter (FACS) analysis

For FACS analysis, pulmonary cell suspensions were obtained using an automated disaggregation device (Medimachine System; Dako, Glostrup, Denmark) and processed as described previously.¹³ Cells from two mice per group (n = 10) were pooled for each time-point (yielding five samples per group for FACS analysis) and then adjusted to a concentration of 4 × 10⁶ cells/ml of FACS buffer (PBS supplemented with 0·5% bovine serum albumin, 0·01% NaN₃, and 100 mm EDTA). Immunostaining for cell-surface molecules was performed for 30 min at 4° using antibodies (Abs) directly labelled against CD3 [anti-CD3 phycoerythrin (PE)], CD4 (anti-CD4 CyChrome), CD8 [anti-CD8 fluorescein isothiocyanate (FITC); anti-CD8 peridinin chlorophyll protein (PerCP)], CD25 (anti-CD25 FITC) and CD69 (anti-CD69 FITC). All Abs were used at concentrations recommended by the manufacturer (PharMingen, San Diego, CA). To correct for non-specific staining, an appropriate control Ab (rat IgG2; PharMingen) was used. The number of positive cells was obtained by setting a quadrant marker for non-specific staining.

Cytokine measurements

Interferon-γ (IFN-γ) and interleukin (IL)-4 concentrations were measured using commercially available enzyme-linked immunosorbent assay (ELISA) reagents, according to the instructions of the manufacturer (R & D Systems, Abingdon, UK).

Statistical analysis

All values are expressed as mean ± standard error of the mean (SEM). Comparisons were performed using Mann–Whitney U-tests. For comparison of survival curves, Kaplan-Meier analysis with a log rank test was used. P-Values of ≤ 0·05 were considered statistically significant.

Results

Survival

PAFR−/− and PAFR+/+ mice were inoculated intranasally with 10⁵ live M. tuberculosis bacilli and their survival was monitored for a time-period of 200 days (Fig. 1). Although PAFR−/− mice tended to succumb to TB earlier than PAFR+/+ mice, the difference between the two strains was not significant. Overall, the survival rate was 14% for PAFR−/− mice and 21% for PAFR+/+ mice (not significant).

Figure 1.

Platelet-activating factor (PAF) receptor deficiency does not influence survival during murine lung tuberculosis. Survival of PAF receptor-deficient (PAFR−/−) and wild-type (PAFR+/+) mice infected intranasally with 10⁵ Mycobacterium tuberculosis colony-forming units (CFU) (n = 14 per group). No significant difference was found in lethality between the two strains of mice.

Mycobacterial outgrowth

Next, the numbers of M. tuberculosis CFU were determined in lungs and livers of PAFR−/− and PAFR+/+ mice at 2 and 6 weeks after intranasal infection. Both organs contained a similar number of M. tuberculosis CFU in PAFR−/− and PAFR+/+ mice at each time-point (Fig. 2).

Figure 2.

Platelet-activating factor (PAF) receptor deficiency does not influence mycobacterial outgrowth in lungs or liver during murine tuberculosis. Bacterial outgrowth is represented, in colony-forming units (CFU)/ml of organ, in PAF receptor-deficient (PAFR−/−) and wild-type (PAFR+/+) mice in lungs (a) and livers (b) at 2 and 6 weeks after intranasal infection with 10⁵Mycobacterium tuberculosis CFU. Data represent mean values ± standard error (SE) (n = 8/group). SE values in Fig. 2(a) were too small to show on the figure. No significant differences were found in mycobacterial outgrowth.

Cellular recruitment to lungs

The histology of parenchymatous organs of 8–10-week-old PAFR−/− and PAFR+/+ mice, without M. tuberculosis infection, was similar and displayed no signs of abnormalities (data not shown). Histopathological examination of lungs from PAFR−/− and PAFR+/+ mice at 2 and 6 weeks after intranasal infection with M. tuberculosis revealed no differences between the two mouse strains. Figure 3 shows representative slides of lungs from mice killed 6 weeks after intranasal infection. At this time-point, dense and diffuse infiltrates were found in the lungs of both mouse strains; the percentage of inflamed parenchyma was similar in both groups (data not shown). To obtain further insight into the cellular composition of the pulmonary infiltrates, we analysed whole lung cell suspension by FACS analysis. The percentages of CD4⁺ and CD8⁺ lymphocytes did not differ significantly between PAFR−/− and PAFR+/+ mice; furthermore, the surface expression of CD25 and CD69 on T cells was similar in both mouse strains (shown for the 6-week postinfection time-point in Table 1).

Figure 3.

No differences in histopathology were observed between platelet-activating factor (PAF) receptor-deficient (PAFR−/−) and wild-type (PAFR+/+) mice. (a) Representative slides of lung tissue of PAFR+/+ mice, 6 weeks after intranasal infection with 10⁵ Mycobacterium tuberculosis colony-forming units (CFU), showed a diffuse inflammatory infiltrate which was almost confluent. Macrophages were the most predominant cell type observed, together with small number of lymphocytes (haematoxylin & eosin staining; original magnification × 25). A comparable picture was observed in PAFR−/− mice (b) 6 weeks postinfection (haematoxylin & eosin staining, original magnification × 25). Slides are representative for six mice per strain.

Table 1.

Cellular composition and cytokine concentrations in lungs

Cells (104/ml)	PAFR+/+	PAFR−/−
Total cells	290·0 ± 31·1	278·0 ± 45
Cell subsets (percentage of total)
″CD4+	67·2 ± 0·8	71·3 ± 1·2
″CD8+	27·1 ± 1·1	22·0 ± 0·9
″CD4+/CD69+	9·9 ± 0·9	11·5 ± 2·4
″CD4+/CD25+	8·6 ± 0·7	11·1 ± 1·6
″CD8+/CD69+	16·7 ± 2·2	15·2 ± 0·6
″CD8+/CD25+	1·7 ± 0·2	2·0 ± 0·4
Cytokines (ng/ml)
″IFN-γ	7·75 ± 0·64	7·68 ± 0·64
″IL-4	5·96 ± 0·75	6·67 ± 0·64

Total cell counts and lymphocyte typing were performed on pulmonary cell suspensions 6 weeks postinfection, as described in the Materials and methods. Fluorescence-activated cell sorter (FACS) analysis was performed on pooled cells from two mice for each analysis from a total of 10 mice per group (i.e. yielding five samples per mouse strain). FACS results are expressed as the percentage of CD4⁺, CD8⁺, CD25⁺ and CD69⁺ within the CD3⁺ population [i.e. for each of the five samples per mouse strain the percentage of positive cells relative to the total number of CD3⁺ cells was determined, and from these data means ± standard error (SE) were calculated]. Cytokine data were obtained from eight mice per group and data are expressed as mean ± SE.

PAFR−/−, platelet-activating factor receptor gene-deficient mice; PAFR+/+, wild-type platelet-activating factor mice.

Lung IFN-γ and IL-4 concentrations

Cytokine concentrations in lung homogenates of 8–10-week-old PAFR−/− and PAFR+/+ mice, without M. tuberculosis infection, were either low or undetectable, with no differences between groups. IFN-γ and IL-4 concentrations in lung homogenates obtained at 2 and 6 weeks postinfection were similar in PAFR−/− and PAFR+/+ mice (shown for the 6-week postinfection time-point in Table 1).

Discussion

PAF has been implicated as a protective mediator in the host response to several intracellular pathogens. The data presented here argue against such a protective role of PAF in pulmonary TB. Indeed, intranasal infection with live M. tuberculosis was associated with similar mortality rates in PAFR−/− and PAFR+/+ mice, and the mycobacterial loads in lungs and liver, determined during the early phase of the infection when all animals were still alive, did not differ significantly between the two mouse strains.

Host defence against TB, at least in part, relies on CD4⁺ and CD8⁺ T cells.² We therefore determined the number of T cells in whole-lung cell suspensions and, in addition, obtained insight into their activation state by measuring the surface expression of CD25 and CD69. Theoretically, PAF can inhibit certain lymphocyte functions. Indeed, PAF has been found to reduce proliferation of CD4⁺ T cells induced by phytohaemagglutinin, which is associated with a reduced expression of CD25.¹⁵ PAF also suppresses the mitogen-stimulated production of IL-2 by human lymphocytes.¹⁶ However, to our knowledge little, if anything, is known about the effects of PAF on lymphocyte activation in vivo. We here demonstrate that deficiency of the PAFR does not influence the recruitment or activation of CD4⁺ and CD8⁺ lymphocytes during pulmonary TB.

The clinical outcome of pulmonary TB is considered to be dependent on a type 1-mediated host response.² We therefore determined whether PAFR deficiency influences the type 1/type 2 balance by measuring the concentration of the type 1 cytokine IFN-γ and the type 2 cytokine, IL-4, in lung homogenates of infected PAFR−/− and PAFR+/+ mice. However, no differences in the pulmonary concentrations of these cytokines were found between these two strains.

Our assumption, that PAF could be involved in the protective immune response to TB, was primarily based on its reported protective role in experimental infections of mice with L. amazonensis, T. cruzi and Candida albicans.^6,7,17 From the present study it remains unclear why PAF does not contribute to protective immunity in TB. PAFR−/− mice are capable of producing PAF, yet PAF cannot exert any biological effect owing to the absence of its receptor.⁴ Knowledge of the production of PAF in TB, either experimentally induced or in patients, is, to the best of our knowledge, not available. In this respect it is important to realize that PAF measurements do not necessarily provide insight into the production of this lipid mediator, as PAF that is synthesized remains predominantly in cell-associated form.¹⁸ Clearly, further research is warranted to dissect the distinct molecular mechanisms that contribute to an adequate immune response to different intracellular pathogens.