Skip to main content
NCBI home page
Immunology logoLink to Immunology
. 2002 Jun;106(2):257–266. doi: 10.1046/j.1365-2567.2002.01403.x

The role of prostaglandin E2 in the immunopathogenesis of experimental pulmonary tuberculosis

Javier Rangel Moreno †, Iris Estrada García †, María de la Luz García Hernández *, Diana Aguilar Leon *, Ricardo Marquez *, Rogelio Hernández Pando *
PMCID: PMC1782721  PMID: 12047755

Abstract

Prostaglandins (PG) are potent mediators of intercellular communication, and PGE2 at high concentration is immunosuppressive for T-cell-mediated immunity. We studied the kinetics of PGE2 production and the expression of the enzymes related to its synthesis during the course of experimental pulmonary tuberculosis. Secondly, we analysed the pathological and immunological changes produced by the pharmacological suppression of PG production. In BALB/c mice infected via the trachea with Mycobacterium tuberculosis H37Rv there is an initial phase of partial resistance, dominated by type 1 cytokines plus tumour necrosis factor-α (TNF-α) and expression of the inducible form of nitric oxide synthase (iNOS), followed by a phase of progressive disease. During the early phase of the infection some activated macrophages located in the alveolar-capillary interstitium and in granulomas showed strong PGE2 immunostaining. However, PGE2 concentrations were relatively low and stable. Animals in this early phase of infection were treated with niflumic acid, a potent and specific blocker of cyclo-oxygenase 2, the rate-limiting enzyme of PG production. In comparison with control animals, the suppression of PG synthesis produced higher inflammation and expression of TNF-α, interleukin-1α and interferon-γ (IFN-γ), but almost complete disappearance of iNOS expression, which coexisted with a significant increment of bacterial load. The late progressive phase in this experimental model is characterized by progressive pneumonia, small granulomas and diminished expression of IFN-γ, TNF-α and iNOS in coexistence with high expression of IL-4. Strong PGE2 immunostaining was seen in foamy macrophages localized in the pneumonic areas, and the PGE2 concentration was four-fold higher in this late phase of infection than during the early phase. When PG production was suppressed in animals suffering advanced phase infection, a significant reduction of pneumonia and bacillus load with striking increment of granuloma size was seen, and the expression of IFN-γ, TNF-α and iNOS was also improved. These findings demonstrate a significant participation of PGE2 in the pathogenesis of pulmonary tuberculosis, showing that during the early phase of the infection there are low PGE2 concentrations which contribute to iNOS expression permitting the temporal control of bacillus growth, while the high PGE2 concentrations during the late phase of the disease contribute to down-regulate cell-mediated immunity, permitting disease progression.

Introduction

Cell signalling is accompanied by rapid remodelling of membrane lipids by activated lipases which generate bioactive lipids that can serve as cellular mediators.1 One of these bioactive lipids is arachidonic acid, which is the principal substrate for the production of the so-called eicosanoids.1,2 Eicosanoids affect many physiological and pathological events,1,2 they are synthesized by two major classes of enzymes: cyclo-oxygenases (COX) and lipoxygenases which produce prostaglandins (PG) and leukotrienes respectively.1

The COX pathway is mediated by two different enzymes: COX-1, which is a constitutively expressed enzyme, and COX-2 a highly inducible enzyme which is expressed in inflamed tissue after exposure to diverse factors, including cytokines.1–3 The most intensively studied PG are those of the E series.2 Monocytes and alveolar macrophages are very efficient cells producing PGE after COX-2 stimulation by interleukin-1 (IL-1), tumour necrosis factor-α (TNF-α), and mycobacterial antigens.4–7

High concentrations of PGE2 have significant immunosuppressive effects,2 including decrement of lymphocyte proliferation, natural killer activity and major histocompatibility complex class II expression.8,9 PGE2 also inhibits the production of T helper type 1 (Th1) cytokines,10–13 shuts off macrophage activation14 and suppresses the production of IL-1 and TNF-α.15 These PGE2 activities could be important in tuberculosis, since it has been shown that in mice and humans mycobacterial infection is mainly controlled by macrophage activation induced through Th1 type cytokines.16 Interferon-γ (IFN-γ) and TNF-α are central to this process by inducing macrophage activation and inducible nitric oxide synthase (iNOS) expression. The NO thus produced is essential, at least for mice, to kill intracellular mycobacteria.17

This protective activity fails if there is a marked release of Th2-type cytokines.18–20 This interplay of cytokines is clearly depicted in a BALB/c model of pulmonary tuberculosis following intratracheal inoculation.20,21 In this model, an initial phase is dominated by high production of Th1 cytokines which, in coexistence with high levels of TNF-α and iNOS, temporarily control the infection. Incipient and mature granulomas are found in this phase. After the 4th week of infection, there is a drop in cells expressing IL-2, TNF-α and iNOS. Gradually, pneumonic areas prevail over granulomas. Pneumonia, in coexistence with a high bacterial burden, causes death. Thus, one important issue in the study of the immunopathology of tuberculosis is the characterization of the diverse factors which contribute to the progressive deterioration of the type 1 cytokine pattern, which is maximal during the advanced phase of the infection and probably contributes to the cell-mediated immune deficiency permitting mycobacterial multiplication, tissue damage and death. One potential factor that might contribute to this phenomenon is PGE2 overproduction.

In this study, we first analysed the kinetics of COX expression and PGE2 production during the course of pulmonary tuberculosis in the above mentioned model in BALB/c mice.

Next we examined the immunological and pathological changes after pharmacological suppression of PG production in tuberculous animals during the early and late phases of infection.

Materials and Methods

Experimental model of progressive pulmonary tuberculosis

Male BALB/c mice at 6–8 weeks of age were infected by direct intratracheal injection with 106 viable Mycobacteria tuberculosis H37Rv in 100 µl of phosphate-buffered saline (PBS), as described elsewhere.20,21 Infected mice were maintained in cages fitted with microisolators connected to negative pressure. Two experiments were performed, and the data were pooled. All procedures were performed in a laminar flow cabinet in a biosafety level III facility. Mice were killed by exsanguination at 1, 3, 7, 14, 21, 28, 60 and 90 days. The protocol was approved by the Ethical Committee for Experimentation in Animals of the National Institute of Medical Sciences and Nutrition in Mexico.

Preparation of tissue samples for histopathology and immunohistochemistry

For histological study, lungs were prepared as described.20–22 For immunohistochemistry, lung sections were mounted on silane-covered slides, deparaffinized, and the endogenous peroxidase was quenched with 0·03% H2O2 in absolute methanol. Lung sections were incubated overnight at room temperature with rabbit-specific polyclonal antibodies against mouse iNOS, COX-1, COX-2 and PGE synthase (Cayman Lab, Billerica, MD, USA) diluted to 1/400 in PBS. Bound antibodies were detected with goat anti-rabbit immunoglobulin G (IgG) labelled with peroxidase (Dako, Carpinteria, CA) and diaminobenzidine. The slides were counterstained with haematoxylin. For immunohistochemical detection of PGE2, the lungs were perfused with OCT compound (Miles Laboratory, Elkhart, IN), frozen in liquid nitrogen, and sectioned with a cryostat. Frozen sections were incubated for 2 hr at room temperature with polyclonal rabbit antibodies against PGE2 (Sigma, St. Louis, MO, USA) diluted 1/50 in PBS. Bound antibodies were revealed with polyclonal goat anti-rabbit antibodies and streptoavidin peroxidase and counterstained with haematoxylin. The percentage of positive cells located in the alveolar-capillary interstitium, perivascular and peribronchial inflammation, granulomas and pneumonia were determined with a computerized image analyser (Qwin Leica, Leica Imaging Systems, Cambridge, UK).22 The negative control consisted of performing the whole procedure with noninfected mouse lung. A further control was the use of normal rabbit sera or isotype non-relevant antibody instead of the primary antibody.

RT-PCR analysis of COX-1, COX-2, PGE synthase and IL-17 in lung homogenates

After the mice were killed, the lungs were removed, the hilar lymph nodes and thymus were eliminated, and the tissue was immediately frozen by immersion in liquid nitrogen. RNA was isolated using the reagent trizol (Gibco BRL, MD) as described previously.23

The cDNA was synthesized using Moloney murine leukaemia virus reverse transcriptase (Gibco BRL, Rockville, MD, USA) and oligo-dT-priming. The expression of COX-1, COX-2, PGE synthase and IL-17 was determined by reverse transcription–polymerase chain reaction (RT-PCR) using 2 µg of cDNA from each point pool (four lungs). After initial denaturation at 94° for 45 seconds, the primers for COX-1, IL-17, COX-2 and PGE synthase were annealed at 57°, 58°, 60° and 61°, respectively, for 45 seconds, a first extension at 72° for 120 seconds, and another extension cycle for 7 min after the completion of 30 cycles. These temperatures and numbers of cycles were defined after the corroboration that in these conditions the PCR was in exponential phase. The PCR products were electrophoresed on 6% polyacrylamide gels, stained with a silver kit (Promega, Madison, WI, USA) following the recommendations of the manufacturer, and analysed with an image densitometer (Bio7-Rad, Hercules, CA) coupled to a computer program (Molecular Analyst 1·4).

The primer sequences were: COX-1: sense: 5′CTGGCCGGATTGGTGGAGGTAG3, antisense: 5′ATAGGGGCAGGTCTTGGTGTTGAG3′; COX-2: sense: 5′CCGGGTTGCTGGGGGAAGA3′, antisense: 5′GTGGCTGTTTTGGTAGGCTGTGGA3′; PGE synthase: sense: 5′TTTGCCAACCCCGAGGATGC3′, antisense: 5′TACCCAGGACCCCAGGACCAGAC3′; IL-17: sense: 5′CCAAACACTGAGGCCAAGGACT3′, antisense: 5′GCCCACACCCACCAGCATCT3′.

The glyceraldehyde-3-phosphate dehydrogenase expression was used as control for RNA content and integrity.

PGE2 quantification in lung homogenates by enzyme-linked immunosorbent assay (ELISA)

After the mice were killed, the lungs were removed and immediately frozen by immersion in liquid nitrogen. The lungs from four infected mice in each time interval were homogenized in 3 ml of ethanol at 70%, using a polytron (Tissue homogenizer, Kinematica, Luzern, Switzerland). Then, the lung homogenate was centrifuged at 960 g for 10 min to eliminate cellular debris. One millilitre of supernatant was collected in sterile tubes, dried by vaccum centrifugation, solubilized with 50 µl of acetone and spotted into thin-layer chromatography plates. A standard of PGE2 (Sigma) (1 µg) dissolved in acetone was added. The plate was developed with a mixture of chloroform, methanol, acetic acid and water (80 : 18 : 1 : 0·8 vol. proportions). The PG standard and lungs samples were revealed by 3·5% of phosphomolybdic acid dissolved in methanol. After calculating the retention factor of the standard, the areas on the plate where PGE2 from the lung samples were localized, were scraped off and deposited in clean tubes for elution in 2 ml of absolute ethanol. Then, the extract was concentrated at dryness by vaccum centrifugation and reconstituted in 450 µl of enzymatic immunoassay (EIA) buffer.

The PGE2 concentrations in pg/ml/mg of the lipidic pulmonary extract were quantified using the EIA provided by Cayman Chemical Co (Billerica, MD, USA), following the recommendation of the manufacturer.

Determining the best COX-2 blocker drug and its appropriate dose

The specific COX-2 blocker drugs: niflumic acid (NA), mefenamic acid and nimesulide were obtained from Cayman Chemical Co. Drugs were dissolved in PBS and administered twice per day at different doses (0·5, 1 and 2 mg) by intragastric cannula in a volume of 2 ml for 28 days into five infected mice. Mice were then killed, their lungs were removed and homogenized to quantify PGE2 by ELISA as described above.

Blocking of PG production by administration of NA during the early and advanced phases of the infection

The experiment designated to compare the efficiency of the different COX-2 blocking drugs demonstrated that NA was the best. Thus, each mouse received 500 µg/ml of NA by intragastric cannula, twice per day. Two experiments were performed and the data were pooled. In each experiment the infected animals were divided into two groups. Group one received NA daily for the first 45 days of infection, starting on day one. Group two received NA from 2 to 3 months after initiation of the infection. Control groups for both conditions received the diluent only by the intragastric route on the same days. Groups of eight mice were killed by exsanguination after 1, 3, 7, 14, 21, 28 and 45 days after NA administration in the acute infection protocol, and after 15 and 30 days of NA treatment in the chronic infection protocol. One lung, right or left, was used for histological analysis, and the other lung was frozen in liquid nitrogen for other determinations. Using an automated image analyser and lung sections stained with haematoxylin & eosin, we determined the area in μm2 occupied by the inflammatory infiltrate in the alveolar-capillary interstitium and blood vessel wall, as well as the granuloma size and the percentage of lung surface affected by pneumonia.22

Determination of colony-forming units (CFU) in infected lungs

The right or left lung from four mice, NA-treated or untreated controls, in two different experiments per time interval were homogenized with a Polytron in sterile isotonic saline and plated for CFU counting as described elsewhere.22

RT-PCR analysis of cytokines and iNOS in lung homogenates

Four lungs, right or left, were used to isolate RNA from NA-treated mice and untreated control animals at each time-point using Trizol as described previously23,24. The cDNA was synthesized using Moloney murine leukaemia virus reverse transcriptase (Gibco BRL) and oligo-dT priming. Two micrograms of cDNA from lung pools for each time-point were used to determine by RT-PCR the expression of IL-1α, TNF-α, iNOS, IFN-γ, IL-4 and IL-10 as described previously.20–22 The PCR products were electrophoresed on 6% polyacrylamide gels, stained with a silver kit and analysed with an image densitometer coupled to a computer program. Expression of glyceraldehyde-3-phosphate dehydrogenase was used to control for RNA content and integrity per separate in each time-point interval.

Measurment of cutaneous delayed type hypersensitivity (DTH)

The antigen and procedure are described in detail elsewhere.20–22

Statistics

A one-way analysis of variance (anova) and Student's t-test were used to compare PGE2 lung concentrations during the early and late phases of infection, as well as the morphometry, DTH responses and CFU determinations in infected mice treated with NA and in untreated control animals. A difference of P<0·05 was considered as significant.

Results

Kinetics of the expression of the COX pathway and PGE2 production during the course of experimental pulmonary tuberculosis

The PGE2 quantification in lung homogenates by ELISA showed stable low concentrations during the first month of infection, followed by a significant increment of two-fold at day 60 and four-fold at day 90 post-infection when compared to day 28 of infection (Fig. 1). These results correlated with the kinetics of the gene expression of COX-1, COX-2 and PGE synthase determined by RT-PCR which showed a progressive increment that reached a peak at day 90 of infection (Fig. 1). Moreover, the expression of IL-17, which has been demonstrated to be a potent inducer of PG synthesis,25 was only detected during the advanced phase of infection (Fig. 1). The immunohistochemistry and automated morphometry showed that the most common COX-2, PG synthase and PGE2 immunostained cells were macrophages (Fig. 2a). At the beginning of the infection, days 1 and 3, these immunostained cells showed the morphological features of activated macrophages (large cells with abundant and compact cytoplasm), and were located in the alveolar-capillary interstitium, where they constituted 20 ± 7% of the total inflammatory cells. Granulomas started their formation after 2 weeks of infection. These granulomas also had activated macrophages with immunoreactivity to COX-2 and PGE, but during the whole course of the infection they were maintained at a stable percentage from 13 to 16% ± 7%. In contrast, macrophages embedded in the pneumonic areas which showed positive immunostaining to PGE2 and COX-2 showed vacuolated cytoplasm, and corresponded to the 17 ± 5% and 25 ± 6% of the total inflammatory cells at days 60 and 90 respectively (Fig. 2b). Considering that these pneumonic areas increased progressively during infection, it was evident that these vacuolated cells were the most important source of PGE2 during the advanced phase of the infection. Interestingly, these pneumonic areas also showed COX-1 immunostained cells, but they were monocytes located in the alveolar-capillary interstitium (Fig. 2c).

Figure 1.

Figure 1

Open in a new tab

Top panel shows the kinetics of COX-1, COX-2, PGE synthase and IL-17 mRNA expression during experimental pulmonary tuberculosis. The glyceraldehyde-3-phosphate dehydrogenase expression (G3PDH) was used as control for RNA content and integrity. Lower panel shows the PGE2 concentrations (pg/ml) in lung lipidic extracts during the course of pulmonary tuberculosis in BALB/c mice infected by the intratracheal route. Asterisks indicate statistical significance (P<0·05) when compared to day 28.

Figure 2.

Figure 2

Open in a new tab

Representative histopathology and immunohistochemistry of PGE2 and the enzymes related to its synthesis during experimental pulmonary tuberculosis, as well as the effect of the pharmacological suppression of the prostaglandin production by niflumic acid (NA) treatment, during the early and late phases of infection. (a) Interstitial activated macrophages show strong PGE2 immunostaining after 1 day of infection. (b) Pneumonic areas in animals with 3 months of infection have numerous vacuolated macrophages with strong COX-2 immunostaining. (c) In the same pneumonic areas, there are monocytoid cells with COX-1 immunoreactivity. (d) After 3 days of infection, there is mild mononuclear inflammatory infiltrate around venules. (e) The perivenular inflammation is wider in infected mice treated with NA during the early phase of the infection. (f) Infected mouse with 3 months of infection shows extensive areas affected by pneumonia. (g) In contrast, mice treated with NA during the late phase of the infection showed focal pneumonic areas. (h) Pneumonic areas after 90 days of infection show few macrophages with iNOS immunostaining. (i) In comparison, chronic infected animals treated for 1 month with NA show less pneumonia, with more iNOS immunostained cells. Controls to demonstrate the specificity of staining consisted of lung tissue incubated with normal rabbit sera (j) or isotype irrelevant antibody instead of the primary antibody (k). A further control was performing the whole procedure on uninfected mouse lung (l).

The effects of blocking COX-2 activity during the early phase of infection

The experiment conducted to select the best drug and dose to block PGE production showed that NA was the most efficient drug, since the daily intragastric administration of 1 mg to tuberculous mice produced a 96% lower PGE2 concentration in comparison with the control untreated group, while mefenamic acid and nimesulide produced 81% and 85% lower concentrations than the controls respectively. Thus, we chose NA to suppress PGE2 production, starting its administration on the first day of infection until 45 days post-infection.

In comparison with the control untreated mice, animals treated with NA showed a significant increment of interstitial and perivascular inflammation during the 1st and 2nd weeks of infection (Figs 2 and 3). Pneumonic areas started their formation after 1 month of infection affecting small areas of the lung surface, while granulomas, which started their formation after 2 weeks of infection, did not show any size change with NA treatment (data not shown).

Figure 3.

Figure 3

Open in a new tab

Top panel shows the area occupied by the inflammatory infiltrate in the alveolar-capillary interstitium and around venules determined by automated morphometry, comparing control untreated mice (black bars) and animals treated with niflumic acid (NA) (white bars) during the early phase of the infection. Lower panel shows the granuloma size and the percentage of lung surface affected by pneumonia in mice treated with NA during the late phase of infection (white bars) in comparison with control animals (black bars). Data are means from eight mice in two different experiments and three random fields from each pulmonary lobe. Asterisk indicates statistical significance (P<0·05).

The pharmacological suppression of PG production during the early phase of the infection evoked higher expression of TNF-α and IFN-γ, and to a lesser extent of IL-1α (Fig. 4), in association with higher DTH, although this was not statistically significant (Fig. 5).

Figure 4.

Figure 4

Open in a new tab

Kinetics of cytokines and iNOS mRNA expression comparing infected control mice (C/black bars) with tuberculous animals treated with niflumic acid (NA) during the early phase of the infection (T/white bars). The glyceraldehyde-3-phosphate dehydrogenase expression (G3PDH) was used as control for RNA content and integrity. Mice treated with NA showed higher IL-1α, TNF-α and IFN-γ expression, with very low mRNA concentrations of iNOS than control mice.

Figure 5.

Figure 5

Open in a new tab

Delayed hypersensitivity responses to soluble antigens of M. tuberculosis in tuberculous animals. Swelling was measured 24 hr after challenge. Animals treated with niflumic acid (white bars) during the early (top panel) and late (lower panel) phases of infection show higher responses than control untreated mice (black bars). Asterisk indicates statistical significance (P<0·05).

Interestingly, iNOS expression in these NA-treated animals was completely suppressed (Fig. 4), in coexistence with significant increments of CFU (Fig. 6).

Figure 6.

Figure 6

Open in a new tab

Colony-forming units in the lungs of BALB/c mice killed at intervals after intratracheal infection with M. tuberculosis, comparing animals treated with niflumic acid (NA) (white bars) and control untreated mice (black bars). Supplementation of NA during the early phase of infection produced higher bacterial load (top panel). In contrast, blocking prostaglandin production during the late phase of the infection produced lower bacterial load after 1 month of treatment (lower panel). Before 14 days of infection, CFU were too few to be plotted. Asterisk indicates statistical significance (P<0·05).

The effects of blocking COX-2 activity during the advanced phase of infection

We started to suppress PGE2 production in mice at day 60 post-infection, since PGE2 concentrations doubled at this time (Fig. 1), killing animals after 2 and 4 weeks of NA treatment. The morphometric analysis showed a striking increment of granuloma area, with significant decrement in the percentage of the lung affected by pneumonia after 1 month of NA administration (Figs 2 and 3).

The pharmacological suppression of PG production during the late phase of the infection induced an increment of the gene expression of IL-1α, TNF-α and IFN-γ. In contrast to the infected mice treated during the early phase, those animals treated during the chronic phase showed a striking increment of iNOS expression (Fig. 7). We corroborated this latter result by immunohistochemistry and automated image analysis, observing that after 15 and 30 days of NA administration the percentage of iNOS immunostained macrophages in pneumonic areas increased 30% and 40% respectively (Fig. 2). The gene expression of IL-4 and IL-10 was lower in chronically infected animals after 30 days of NA administration (Fig. 7), in coexistence with a significant decrement of bacillary load (Fig. 6).

Figure 7.

Figure 7

Open in a new tab

Kinetics of cytokines and nitric oxide synthase (iNOS) mRNA expression comparing infected control mice (C/black bars) with tuberculous animals treated with niflumic acid (NA) during the chronic phase of the infection (T/white bars). The glyceraldehyde-3-phosphate dehydrogenase expression (G3PDH) was used as control for RNA content and integrity. Mice treated with NA showed higher IL-1α, TNF-α, IFN-γ and iNOS expression than control mice.

Treated mice during the advanced phase of the infection showed a significant increment in DTH responses when compared with infected untreated animals (Fig. 5).

Discussion

We report here on the kinetics of PGE2 production and the expression of the enzymes involved in PG synthesis during the course of experimental pulmonary tuberculosis, as well as the pathological and immunological changes produced by the pharmacological suppression of PG production.

Two main findings are reported: (1) the production of PGE2 and COX expression are progressively increased during the course of the infection; and (2) there is a different effect on the pathology and immune response when PG production is suppressed during the early or during the late phase of the disease.

Our immunohistochemistry results showed that macrophages were the most important source of PGE2. During the early phase of the infection many of the PGE2 and COX-2 immunostained cells were activated macrophages located in the alveolar-capillary interstitium and granulomas. We have previously shown in this experimental model that activated macrophages are an important source of IL-1α and TNF-α,21 and it has been demonstrated that both cytokines are potent inducers of COX-2 expression and PG synthesis.4,5 Thus, in agreement with previous reports of in vitro experiments,7 our results suggest that in vivo these pro-inflammatory cytokines, besides mycobacterial antigens, might be good inducers of PGE2 production. However, the PGE2 concentrations in the lung during the early phase of the infection were stable and lower in comparison with those determined during the advanced phase of the disease. Interestingly, more inflammation with higher expression of TNF-α, IL-1α and IFN-γ was seen when PG production was inhibited by the administration of NA in infected mice during the first 45 days of infection, but paradoxically a significant increment of CFU was observed in these animals.

The explanation of this contradictory observation is that PGE2 regulates iNOS depending on its concentration. Indeed, in vitro studies clearly demonstrated that low concentrations of PGE2 are able to stimulate the expression of iNOS and the release of NO, while high concentrations are inhibitory.26 Our results of iNOS gene expression are in agreement with these in vitro observations, showing that in comparison with control animals, iNOS is not expressed during the early phase of the infection in mice treated with NA. Thus, it seems that the low and stable production of PGE2 during the early phase of the infection contributes to the expression of iNOS and NO production, which is essential for intracellular mycobacterial killing.17

In this experimental model, the progressive phase of the infection started 1 month after intratracheal instillation of M. tuberculosis, and is characterized by progressive pneumonia, reduction of granuloma size, decrement of IL-1α and TNF-α production, low iNOS expression, and progressive increment of IL-4 production.20,21,24 Interestingly, the pneumonic areas show a progressive increment of macrophages that exhibit numerous cytoplasmic vacuoles which contain mycobacterial lipids.21 These vacuolated macrophages usually contain numerous bacilli and their cytoplasm shows faint TNF-α and iNOS immunostaining, but strong immunoreactivity to transforming growth factor-β.21,24 Although, this chronic phase started at 1 month of infection, the amount of PGE2 doubled until 2 months of infection. The combination of low expression of pro-inflammatory cytokines with high concentration of anti-inflammatory cytokines like IL-4 should inhibit PG production.27,28 However, the concentration of PGE2 during the advanced phase of the infection, particularly at day 90, was four-fold higher than in the early phase, in combination with the highest COX-2 and PGE synthase gene expression, and strong PGE2 and COX-2 immunostaining in foamy macrophages located in pneumonic areas. Thus, it seems that some factors are contributing to stimulate PGE production in this phase. Our results showed that at least the high expression of IL-17 and COX-1 are some of these contributing factors.

IL-17 is a potent inducer of PG production.25 Our RT-PCR study clearly showed that IL-17 was exclusively expressed during the progressive phase of the disease, when the highest expression of COX-1 was also seen, and numerous monocytoid cells showed strong COX-1 immunostaining, suggesting that during the progressive phase of the disease there is also PG production by the activity of COX-1, its cellular source being the recently emigrated monocytes that are continuously recruited in the pneumonic areas.

High concentration of PGE2 is a potent inhibitor of TNF-α and Th1 cytokine production, as well as of NO production due to iNOS down-regulation.10,13,15,26 These abnormalities were observed during the late phase of our experimental model.21,22,24 Our results suggested that PGE2 participates in this decrement of the protective cellular immunity, since infected mice treated with NA after 2 months of infection showed improvement of the expression of TNF-α, IFN-γ and iNOS, in combination with bigger granulomas, less pneumonia, higher DTH and lower bacillus load. Moreover, the IL-4 and IL-10 expression was also decreased after 1 month of NA administration. It is important to consider that COX inhibitors increase leukotriene synthesis, because blocking of COX shifts arachidonic acid to the lipoxygenase system. Leukotriene B4 in particular stimulates the production of IL-1, IL-2 and IFN-γ,29 which is the opposite to the effect of PGE2. Thus, part of the improvement of protective cell-mediated immunity in tuberculous mice treated with NA during late infection could be secondary to high production of leukotriene B4.

In conclusion, our results suggest a significant participation of PGE2 in the protection and progression of pulmonary tuberculosis depending on its concentration. During early infection there is mild inflammation with low and stable concentrations of PGE2 which contribute to an efficient iNOS expression permitting temporal control of the infection. Then, during the late phase of the infection, there is a high concentration of PGE2 which participates in the down-regulation of the cell-mediated immunity permitting disease progression.

Acknowledgments

This work was supported by CONACyT, Grant no. 28700M and by the International Centre for Genetic Engineering and Biotechnology. Javier Rangel was a recipient of a PhD scholarship also provided by CONACyT. The authors thank Dr Oscar Rojas for his comments and technical assistance.

References

  • 1.Serhan CN, Haeggström JZ, Leslie CC. Lipid mediator networks in cell signaling: update and impact of cytokines. FASEB J. 1996;10:1147–58. doi: 10.1096/fasebj.10.10.8751717. [DOI] [PubMed] [Google Scholar]
  • 2.Phipps RP, Stein SH, Roper RL. A new view of prostaglandin E regulation of the immune response. Immunol Today. 1991;12:349–52. doi: 10.1016/0167-5699(91)90064-Z. [DOI] [PubMed] [Google Scholar]
  • 3.Smith WL, Dewitt DL. Biochemistry of prostaglandin endoperoxidase H synthases (cyclooxygenases 1 and 2) Semin Nephrol. 1995;15:178–89. [Google Scholar]
  • 4.Ristimaki A, Garfinkel S, Wessendorf J, Maciag T, Hla T. Induction of cyclooxygenase by Interleukin 1 alpha. J Biol Chem. 1994;269:1769–75. [PubMed] [Google Scholar]
  • 5.Last Barney K, Homon CA, Faanes RB, Merluzzi VJ. Synergistic and overlapping activities of tumor necrosis factor-α and IL-1. J Immunol. 1988;141:527–30. [PubMed] [Google Scholar]
  • 6.Hempel SL, Monick MM, Hunninghake GW. Lipopolysacharide induces prostaglandin H synthase-2 protein and mRNA in human alveolar macrophages and blood monocytes. J Clin Invest. 1994;93:391–6. doi: 10.1172/JCI116971. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Edwards CK, Hedegaard HB, Zlotnik A, Gangadharam PR, Johnston RB, Pabst MJ. Chronic infection due to Mycobacterium intracellulare in mice association with macrophage release of prostaglandin E2 and reversal by injection of indomethacin, muramyl dipeptide, or interferon-γ. J Immunol. 1986;136:1820–7. [PubMed] [Google Scholar]
  • 8.Goto TR, Herberman RB, Maluish A, Strong DM. Cyclic AMP as a mediator of prostaglandin E induced suppression of human natural killer cell activity. J Immunol. 1983;130:1350–5. [PubMed] [Google Scholar]
  • 9.Snyder DS, Beller DI, Unanue ER. Prostaglandins modulate macrophage Ia expression. Nature. 1982;299:163–5. doi: 10.1038/299163a0. [DOI] [PubMed] [Google Scholar]
  • 10.Betz M, Fox BS. Prostaglandin E2 inhibits production of Th-1 lymphokines but not of Th-2 lymphokines. J Immunol. 1991;146:108–13. [PubMed] [Google Scholar]
  • 11.Wu CY, Wang K, McDyer JF, Seder RA. Prostaglandin E2 and dexamethasone inhibit IL-12 receptor expression and IL-12 responsiveness. J Immunol. 1998;161:2723–30. [PubMed] [Google Scholar]
  • 12.Tineke CTM, van der Pouw K, Boeije LCM, Smeenk RJT, Wijdenes J, Aarden LA. Prostaglandin-E2 is a potent inhibitor of human interleukin 12 production. J Exp Med. 1995;181:775–9. doi: 10.1084/jem.181.2.775. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Kuroda E, Suguira T, Zeki K, Yoshida Y, Yamashita U. Sensitivity difference to the suppressive effect of prostaglandin E2 among mouse strains: a possible mechanism to polarize Th2 type response in BALB/c mice. J Immunol. 2000;164:2386–95. doi: 10.4049/jimmunol.164.5.2386. [DOI] [PubMed] [Google Scholar]
  • 14.Schultz RM, Pavlidis NA, Stylos WA, Chirigos MA. Regulation of macrophage tumoricidal function: a role of prostaglandins of the E series. Science. 1978;202:320–1. doi: 10.1126/science.694537. [DOI] [PubMed] [Google Scholar]
  • 15.Renz H, Gong JH, Schmidt A, Nain M, Gemsa D. Release of tumor necrosis factor-α from macrophages. Enhancement and suppression are dose-dependently regulated by prostaglandin E2 and cyclic nucleotides. J Immunol. 1988;141:2388–93. [PubMed] [Google Scholar]
  • 16.Rook GAW, Hernández Pando R. The pathogenesis of tuberculosis. Annu Rev Microbiol. 1996;50:259–84. doi: 10.1146/annurev.micro.50.1.259. [DOI] [PubMed] [Google Scholar]
  • 17.Chan JYX, Magliozzo RS, Bloom BR. Killing of virulent Mycobacterium tuberculosis by reactive nitrogen intermediates produced by activated macrophages. J Exp Med. 1992;175:1111–22. doi: 10.1084/jem.175.4.1111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Seah GT, Scott M, Rook GAW. Type 2 cytokine gene activation and its relationship to extent of disease in patients with tuberculosis. J Infect Dis. 2000;181:385–9. doi: 10.1086/315200. [DOI] [PubMed] [Google Scholar]
  • 19.Wangoo A, Sparer T, Brown I, et al. Contribution of Th1 and Th2 cells to protection and pathology in experimental models of granulomatous disease. J Immunol. 2001;166:3432–9. doi: 10.4049/jimmunol.166.5.3432. [DOI] [PubMed] [Google Scholar]
  • 20.Hernández Pando R, Orozco EH, Sampieri A, Pavón L, Velasquillo C, Larriva Sahd J, Madrid MV. Correlation between kinetics of Th1/Th2 cells and pathology in a murine model of experimental pulmonary tuberculosis. Immunology. 1996;89:26–33. [PMC free article] [PubMed] [Google Scholar]
  • 21.Hernández Pando R, Orozco EH, Arriaga AK, Sampieri A, Larriva Sahd J, Madrid Marina V. Analysis of the local kinetics and localization of interleukin 1α, tumor necrosis factor α and transforming growth factor β during the course of experimental pulmonary tuberculosis. Immunology. 1997;90:607–17. doi: 10.1046/j.1365-2567.1997.00193.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Hernández Pando R, Pavón L, Orozco EH, Rangel J, Rook GAW. Interactions between hormone-mediated and vaccine mediated immunotherapy for pulmonary tuberculosis in BALB/c mice. Immunology. 2000;100:391–8. doi: 10.1046/j.1365-2567.2000.00054.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Chomzynski PA. Reagent for the single step simultaneous isolation of RNA, DNA and proteins from cells and tissue samples. Biotechniques. 1994;15:532–9. [PubMed] [Google Scholar]
  • 24.Hernandez Pando R, Schön T, Orozco EH, Serafin M, Estrada GI. Expression of nitric oxide synthase and nitrotyrosine during the evolution of experimental pulmonary tuberculosis. Exp Toxic Pathol. 2001;53:257–65. doi: 10.1078/0940-2993-00182. [DOI] [PubMed] [Google Scholar]
  • 25.Fossiez F, Djossou Chomarat P, et al. T cell Interleukin 17 induces stromal cells to produce proinflammatory and hematopoietic cytokines. J Exp Med. 1996;183:2593–603. doi: 10.1084/jem.183.6.2593. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Milano S, Arcoleo F, Dieli M, D'Agostino R, D'Agostino P, De Nucci G, Cillari E. Prostaglandin E2 regulates inducible nitric oxide synthase in the murine macrophage cell line J774. Prostaglandins. 1995;49:105–15. doi: 10.1016/0090-6980(94)00004-g. [DOI] [PubMed] [Google Scholar]
  • 27.Niiro H, Otsuka T, Ogami E, et al. MAP kinase pathways as a route for regulatory mechanisms of IL-10 and IL-4 which inhibit COX-2 expression in human monocytes. Biochem Biophys Res Commun. 1998;250:200–5. doi: 10.1006/bbrc.1998.9287. [DOI] [PubMed] [Google Scholar]
  • 28.Dworsky R, Sheller JR. Differential sensitivities of human blood monocytes and alveolar macrophages to the inhibition of prostaglandin endoperoxide synthase 2 by interleukin 4. Prostaglandins. 1997;53:237–51. doi: 10.1016/s0090-6980(97)89598-6. [DOI] [PubMed] [Google Scholar]
  • 29.Hwang O. Essential fatty acids and immune response. FASEB J. 1989;3:2052–61. doi: 10.1096/fasebj.3.9.2501132. [DOI] [PubMed] [Google Scholar]

Articles from Immunology are provided here courtesy of British Society for Immunology

RESOURCES