Effect of tumour necrosis factor-α receptor 1 genetic deletion on carrageenan-induced acute inflammation: a comparison with etanercept

Authors


S. Cuzzocrea, Institute of Pharmacology, School of Medicine, University of Messina, via C. Valeria, Torre Biologica, Policlinico Universitario, Messina 98123, Italy.
E-mail: salvator@unime.it

Summary

In the present study, we used tumour necrosis factor-α receptor 1 knock-out mice (TNF-αR1KO) to evaluate an in vivo role of TNF-αR1 on the pathogenesis of inflammatory diseases. We used a murine model of carrageenan-induced acute inflammation (pleurisy), a preclinical model of airway inflammation. The data proved that TNF-αR1KO were resistant to carrageenan-induced acute inflammation compared with TNF-α wild-type mice. TNF-αR1KO showed a significant reduction in accumulation of pleural exudate and in the number of inflammatory cells, in lung infiltration of polymorphonuclear leucocytes and lipid peroxidation and showed a decreased production of nitrite/nitrate in pleural exudates. Furthermore, the intensity and degree of the adhesion molecule intercellular adhesion molecule-1 and P-selectin, Fas ligand (FasL), inducible nitric oxide sythase and nitrotyrosine determined by immunohistochemical analysis were reduced markedly in lung tissues from TNF-αR1KO at 4 h and 24 h after carrageenan injection. Moreover, TNF-α and interleukin-1β concentrations were reduced in inflamed areas and in pleural exudates from TNF-αR1KO. To support the results generated using pleural inflammation, carrageenan-induced paw oedema models were also performed. In order to elucidate whether the observed anti-inflammatory effects were related to the inhibition of TNF-α, we also investigated the effect of etanercept, a TNF-α soluble receptor construct, on carrageenan-induced pleurisy. The treatment with etanercept (5 mg/kg subcutaneously 2 h before the carrageenan injection) reduces markedly both laboratory and histological signs of carrageenan-induced pleurisy. Our results showed that administration of etanercept resulted in the same outcome as that of deletion of the TNF-αR1 receptor, adding a new insight to TNF-α as an excellent target by therapeutic applications.

Introduction

The inflammatory process is characterized invariably by a production of prostaglandins, leukotrienes, histamine, bradykinin, platelet-activating factor and by the release of chemicals from tissues and migrating cells [1]. Carrageenan (CAR)-induced local inflammation is used commonly to evaluate anti-inflammatory effects of non-steroidal anti-inflammatory drugs (NSAIDs). Therefore, CAR-induced local inflammation (paw oedema or pleurisy) is a useful model to assess the contribution of mediators involved in cellular alterations during the inflammatory process.

In particular, the initial phase of acute inflammation (0–1 h), which is not inhibited by NSAIDs such as indomethacin or aspirin, has been attributed to the release of histamine, 5-hydroxytryptamine and bradykinin, followed by a late phase (1–6 h) sustained mainly by prostaglandin release and attributed to the induction of inducible cyclooxygenase-2 in the tissue [2]. It appears that the onset of CAR acute inflammation has been linked to neutrophil infiltration and the production of neutrophil-derived free radicals, such as hydrogen peroxide, superoxide and hydroxyl radicals, as well as to the release of other neutrophil-derived mediators [3].

Tumour necrosis factor (TNF)-α is the prototypic member of a cytokine family which regulates essential biological functions (e.g. cell differentiation, proliferation, survival, apoptosis) and a broad spectrum of responses to stress and injury [4]. It is produced primarily by immune cells such as monocytes and macrophages, but it can also be released by many other cell types, including acinar cells. Membrane-bound or soluble TNF-α interacts with two different surface receptors, TNF-α receptor 1 (TNF-αR1, also known as p55) and TNF-αR2 (also known as p75) [5]. Although the extracellular domains of TNF-αR1 and TNF-αR2 are homologous and manifest similar affinity for TNF-α, the cytoplasmic regions of the two receptors are distinct and mediate different downstream events. Although most cell lines and primary tissues express both isoforms [6], the structural and functional characterization of TNF−αR2 is less well understood, because it is not activated efficiently in vitro[7,8]. After exposure to TNF-α, target cells may down-regulate their responsiveness to the cytokine by shedding the receptors into the circulation. A natural mechanism which has been hypothesized to counteract excessive concentrations of circulating TNF-α (and the subsequent enhanced surface receptor activation) is the release of soluble receptors. The two soluble receptor forms (sTNF-αR1 and sTNF-αR2) have longer half-lives than TNF-α, and their concentration may reflect TNF-α activity [9].

A primary role for TNF-α in inflammation (e.g. sepsis, endotoxaemic shock and acute pancreatitis) is suggested by several studies conducted using cell lines, animal models and humans [10–12]. In inflammation, over-production of TNF-α is pivotal in the induction of inflammatory genes, cell death, endothelial up-regulation and in the recruitment and activation of immune cells [13,14]. TNF-α has also been regarded as one of the major mediators of systemic progression and tissue damage in severe disease [15].

However, the biological significance of TNF-αR shedding is unclear. It could represent a neutralizing mechanism to counteract excessive TNF-α activity, or it has been suggested that in relatively low concentrations sTNF-αR may serve as carriers of TNF-α to distant organs. Furthermore, sTNF-αR stabilize the structure of TNF-α trimeric, thereby prolonging its half-life and augmenting its biological effects [9].

Etanercept is a dimeric fusion protein consisting of the extracellular ligand-binding portion of the human 75 kilodalton (p75) TNF-αR linked to the Fc portion of human immunoglobulin G1 (IgG1), and it can bind to two TNF-α molecules blocking their interaction with cell surface TNF-αRs and rendering TNF-α biologically inactive. TNF-α inactivation is 1000 times stronger than inactivation by p75 monomeric TNF-αR [16,17]. Etanercept inhibits the activity of TNF-αin vitro and has been examined in vivo for its effects in different animal model systems of inflammatory and autoimmune diseases [18]. In clinical settings, etanercept has been tested in numerous trials and approved for the treatment of rheumatoid arthritis, juvenile rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis and plaque psoriasis [19–23].

In our study, we have investigated the role of TNF-α in conditions associated with experimental acute inflammation using TNF-αR1 knock-out (TNF-αR1KO) mice. Moreover, the role of blockage through dimeric fusion protein etanercept has not been investigated in rodent models of paw oedema and lung injury. Injection of CAR in mice may constitute a useful model for evaluation of the effects of potential inhibitors of TNF-α-related pathways in vivo, as also suggested by Rocha et al. [24]. Thus, we have also investigated whether etanercept treatment attenuates acute inflammation.

The aims of the present study are: (i) to assess the effects observed in TNF-αR1KO mice on the inflammatory response in a murine model of acute inflammation; and (ii) to compare the results with those observed with pharmacological inhibition of TNF-α by means of etanercept.

Materials and methods

Animals

Male mice (4–5 weeks old, 20–22 g) with a targeted disruption of the TNF-αR1KO and corresponding eterozygote wild-type controls (TNF-αWT) were purchased from Jackson Laboratories (Charles River, Milan, Italy). Mice were housed in stainless steel cages in a room kept at 22 ± 1°C with a 12-h light/12-h dark cycle (lights on at 6 a.m.). The animals were acclimated to their environment for 1 week and had ad libitum access to tap water and standard rodent diet. The study was approved by the University of Messina Review Board for the care of animals. All animal experiments complied with regulations in Italy (D.M. 116192), Europe (O.J. of E.C. L 358/1 12/18/1986) and the United States (Animal Welfare Assurance no. A5594-01; Department of Health and Human Services, Washington, DC, USA).

Carrageenan-induced pleurisy

Carrageenan-induced pleurisy was induced as described previously [25]. We anaesthetized the mice with isoflurane and made a skin incision at the level of the left sixth intercostal space. The underlying muscle was dissected and saline (0·2 ml) or saline containing 1% (w/v) λ-CAR (0·2 ml; Sigma-Aldrich, Milan, Italy) was injected into the pleural cavity. The skin incision was closed with a suture and the animals were allowed to recover. At 4 h and 24 h after the injection of CAR, the animals were killed. The chest was opened carefully and the pleural cavity rinsed with 2 ml of saline solution containing heparin (5 U/ml) and indomethacin (10 μg/ml). The exudate and washing solution were removed by aspiration and the total volume measured. Any exudate which was contaminated with blood was discarded. The amount of exudate was calculated by subtracting the volume injected (2 ml) from the total volume recovered. The leucocytes in the exudate were suspended in phosphate-buffered saline (PBS; 0·01 M, pH 7·4) and counted with an optical microscope in a Burker's chamber after vital Trypan Blue staining. The following groups of animals were used. (i) CAR-WT group: WT mice were subjected to injection of 1% λ-CAR in the pleural cavity (n = 20); (ii) CAR KO group: TNF-αR1KO mice were subjected to injection of 1% λ-CAR in the pleural cavity (n = 20); (iii) sham WT group: WT mice were subjected to pleural injection of sterile saline (n = 20); (iv) sham KO group: TNF-α-R1KO mice were subjected to pleural injection of sterile saline (n = 20); (v) CAR-WT + etanercept group: the aame as CAR-WT group, except for the pre-administration of etanercept (5 mg/kg dissolved subcutaneously in saline solution) which was given 2 h before the CAR injection (n = 20); and (vi) sham WT +  etanercept group: the same as the sham WT group except for the pre-administration of etanercept (5 mg/kg subcutaneously dissolved in saline solution), which was given 2 h before the saline injection (n = 20). Pleural exudates and lung tissues were collected at 4 (n = 10) and 24 h (n = 10) after CAR injection. At each time-point after pleurisy, lung tissues were collected for histological analysis, for immunohistochemical studies, for myeloperoxidase (MPO) activity and thiobarbituric acid-reactant substances measurement; pleural exudate were collected for inflammatory cell count, measurement of cytokines and nitrite–nitrate concentration.

Carrageenan-induced paw oedema

Paw oedema was induced as described previously [26] by a subplantar injection of 50 μl of sterile saline containing 1% (wt/vol) λ-CAR or sterile saline alone into the right hind paw. Foot volumes were measured using a water plethysmometer (Ugo Basile, Milan, Italy) immediately before injection and every hour for 5 h. The increase in paw volume was evaluated as the difference between the paw volume at each time-point and the basal paw volume (time 0). The increase in paw volume was taken as oedema volume. Mice were allocated randomly into the following groups. (i) CAR-WT group: WT mice were subjected to subplantar injection of sterile saline containing 1% λ-CAR (n = 10); (ii) CAR KO group: TNF-αR1KO mice were subjected to subplantar injection of sterile saline containing 1% λ-CAR (n = 10); (iii) sham WT group: WT mice were subjected to subplantar injection of sterile saline (n = 10); (iv) sham WT + etanercept group: the same as the WT sham group except for the administration of etanercept (5 mg/kg subcutaneously dissolved in saline solution), which was given 2 h before saline injection (n = 10); (v) sham KO group: TNF-αR1KO mice were subjected to subplantar injection of sterile saline (n = 10); and (vi) CAR-WT + etanercept group: the same as the CAR-WT group except for the administration of etanercept (5 mg/kg subcutaneously dissolved in saline solution), which was given 2 h before the CAR injection (n = 10). After 4 h following CAR or saline injection, the subplantar oedematous area from paw was excised and prepared for further determinations. The dose of etanercept 5 mg/kg used here was based on previous in vivo studies [27].

Histological examination

Lungs were taken 4 h and 24 h after CAR or vehicle injection. Tissues were fixed for 1 week in 10% (w/v) PBS-buffered formaldehyde solution at room temperature, dehydrated using graded ethanol and embedded in Paraplast (Sherwood Medical, Mahwah, NJ, USA). Lung sections were then deparaffinized with xylene and stained with haematoxylin and eosin. All sections were studied using light microscopy (Dialux 22 Leitz).

Myeloperoxidase activity

Myeloperoxidase activity, an indicator of polymorphonuclear leucocyte (PMN) accumulation, was determined in the paw and lung tissues after CAR-injection as described previously [28]. Subplantar and lung tissues obtained from 10 animals per group were homogenized in a solution containing 0·5% (w/v) hexadecyltrimethyl-ammonium bromide dissolved in 10 mM potassium phosphate buffer (pH 7) using a Polytron homogenizer (three cycles of 10 s at maximum speed) and centrifuged for 30 min at 20 000 g at 4°C. An aliquot of the supernatant was then allowed to react with a solution of tetramethylbenzidine (1·6 mM) and 0·1 mM hydrogen peroxide. The rate of change in absorbance was measured spectrophotometrically at 650 nm. MPO activity was defined as the quantity of enzyme degrading 1 μmol of peroxide/min at 37°C and was expressed in units/g of wet tissue.

Measurement of cytokines

Tumour necrosis factor-α and interleukin (IL)-1β concentrations in pleural exudates and lung tissue were evaluated using a colorimetric commercial enztme-linked immunosorbent assay (ELISA) kit (Calbiochem-Novabiochem Corporation, Milan, Italy) at 4 h and 24 h after the induction of pleurisy by CAR injection, as described previously [29]. The lower detection limit of the assay was of 10 pg/ml.

Immunohistochemical localization of TNF-α, IL-1β, P-selectin, intercellular adhesion molecule-1, inducible nitric oxide sythase, nitrotyrosine and Fas ligand

Left lobe of lungs were fixed in 10% (w/v) PBS-buffered formaldehyde and 8 μm sections were prepared from paraffin-embedded tissues. After deparaffinization, endogenous peroxidase was quenched with 0·3% (v/v) hydrogen peroxide in 60% (v/v) methanol for 30 min. The sections were permeablized with 0·1% (w/v) Triton X-100 in PBS for 20 min. Non-specific adsorption was minimized by incubating the section in 2% (v/v) normal goat serum in PBS for 20 min. Endogenous biotin or avidin binding sites were blocked by sequential incubation for 15 min with biotin and avidin (Vector, D.B.A., Milan, Italy) respectively. Sections were incubated overnight with (i) anti-IL-1β antibody (Santa Cruz Biotechnology, D.B.A., Milan, Italy; 1 : 500 in PBS w/v), (ii) anti-TNF-α antibody (Santa Cruz Biotechnology; 1 : 500 in PBS w/v), (iii) anti-Fas ligand (FasL) antibody (Santa Cruz Biotechnology; 1 : 500 in PBS v/v), (iv) anti-inducible nitric oxide sythase (iNOS) antibody (Transduction Laboratories, D.B.A., Milan, Italy; 1 : 500 in PBS v/v), (v) anti-nitrotyrosine antibody (Cayman, D.B.A., Milan, Italy; 1 : 1000 in PBS v/v), (vi) anti-intercellular adhesion molecule (ICAM) antibody (Pharmigen; 1 : 1000 in PBS v/v) or (vii) anti-P-selectin antibody (Santa Cruz Biotechnology; 1 : 500 in PBS v/v). Sections were washed with PBS and incubated with secondary antibody. Specific labelling was detected with a biotin-conjugated goat anti-rabbit IgG and avidin–biotin peroxidase complex (Vector). The counterstain was developed with 3′3′-diaminobenzidene (brown colour) and nuclear fast red (red background). To verify the binding specificity for IL-1β, TNF-α, P-selectin, ICAM-1, iNOS or FasL some sections were also incubated with only the primary antibody (no secondary) or only the secondary antibody (no primary). In order to confirm that the immunoreactions for the nitrotyrosine were specific, some sections were also incubated with the primary antibody (anti-nitrotyrosine) in the presence of excess nitrotyrosine (10mM) to verify the binding specificity. In these situations no positive staining was found in the sections indicating that the immunoreaction was positive in all the experiments carried out.

Immunohistochemical photographs (five photographs from each sample collected from all mice in each experimental group) were assessed as described previously [28] by using Imaging Densitometer (AxioVision, Zeiss, Milan, Italy) and a computer program (AxioVision). In particular, the densitometry analysis was carried out in sections in which the lung was orientated in order to observe all the histological portions. In this type of section it is possible to evaluate the presence/absence of positive staining. Therefore, the densitometry data obtained represent all these differences.

Measurement of nitrite–nitrate concentration

Total nitrite in pleural exudates, an indicator of nitric oxide (NO) synthesis, was measured as described previously [28]. Briefly, the nitrate in the sample was reduced first to nitrite by incubation with nitrate reductase (670 mU/ml) and β-nicotinamide adenine dinucleotide 3′-phosphate (160 μM) at room temperature for 3 h. The total nitrite concentration in the samples was then measured using the Griess reaction by adding 100 μl of Griess reagent (0·1% (w/v) naphthylethylendiamide dihydrochloride in H2O and 1% (w/v) sulphanilamide in 5% (v/v) concentrated H3PO4; vol. 1 : 1) to the 100 μl sample. The optical density at 550 nm (OD550) was measured using an ELISA microplate reader (SLT-Laboratory Instruments, Salzburg, Austria). Nitrite concentrations were calculated by comparison with OD550 of standard solutions of sodium nitrite prepared in H2O.

Thiobarbituric acid-reactant substances measurement

Thiobarbituric acid-reactant substances measurement, which is considered a good indicator of lipid peroxidation, was determined in lung tissues, as described previously [30]. Thiobarbituric acid-reactant substances were calculated by comparison with OD650 of standard solutions of 1,1,3,3-tetramethoxypropan 99% malondialdehyde bis (dimethyl acetal) 99% (MDA) (Sigma-Aldrich, Milan, Italy). The absorbance of the supernatant was measured by spectrophotometry at 650 nm.

Materials

All compounds were obtained from Sigma-Aldrich. All other chemicals were of the highest commercial grade available. All stock solutions were prepared in non-pyrogenic saline (0·9% NaCl; Baxter, Milan, Italy). Etanercept was from Wyeth Pharmaceuticals (Philadephia, PA, USA).

Statistical evaluation

All values in the figures and text are expressed as mean ± standard error of the mean from 10 mice for each group. For the in vivo studies n represents the number of animals studied. In the experiments involving histology or immunohistochemistry, the figures shown are representative of at least three experiments (histological or immunohistochemistry colouration) performed on different experimental days on the tissue sections collected from all the animals in each group. The results were analysed by one-way analysis of variance (anova) followed by Bonferroni's post hoc test for multiple comparisons and assessed with two-way anova for repeated measures and followed by Student's t-test. A P-value of less than 0·05 was considered statistically significant.

Results

Effects of TNF-αR1 gene deletion and etanercept administration on CAR-induced pleurisy, lung MPO activity, histological examination

To analyse the possible influence of TNF-α receptors during acute inflammation in the lung, we examined the effect of TNF-αR1 gene deletion on CAR-induced pleurisy, as evaluated at 4 h and 24 h after injection. All WT mice which had received CAR developed acute pleurisy, producing turbid exudate (Fig. 1a). When compared with the number of cells collected from the pleural space from sham WT mice and sham TNF-αR1KO mice, injection of CAR induced a significant increase in the number of PMNs at 4 h and 24 h (Fig. 1b). The presence of pleural exudate (Fig. 1a) and the number of inflammatory cells (Fig. 1b) in the pleural cavity at 4 h and 24 h after CAR administration was reduced significantly in the absence of a functional TNF-αR1 gene (P < 0·01 respectively). Similarly, treatment of WT mice with etanercept reduced significantly the pleural exudates formation and the inflammatory cell infiltration in the pleural cavity (Fig. 1a and b).

Figure 1.

Effects of tumour necrosis factor (TNF)-α gene deletion and etanercept administration on CAR-induced pleural exudate production (a) and accumulation of polymorphonuclear cells in the pleural cavity (b). At 4 h (polymorphonuclear leucocytes are the predominant cell type) and 24 h (peak of total inflammatory cell number and exudate volume), the total volume was recovered by aspiration. A significant production in pleural exudate (a) and polymorphonuclear cell infiltration (b) was observed in the pleural cavity from wild-type (WT) mice at 4 h and 24 h after CAR administration. The absence of TNF-α receptor 1 gene in mice as well as the treatment of WT mice with etanercept reduced significantly the presence of pleural exudate (a) and the number of inflammatory cells in the pleural cavity (b) at 4 h and 24 h after CAR injection. Data are means ± standard error of the mean of 10 mice for each group. *P < 0·01 versus sham; °P < 0·01 versus CAR-WT group. KO, knock-out; CAR, carrageenan.

The important presence of inflammatory cells in the pleural cavity appeared to be related to the influx of leucocytes into the lung tissue. MPO activity was elevated significantly at 4 h and 24 h after CAR injection in WT mice (Fig. 2, P < 0·01). In lung from TNF-αR1KO mice, MPO activity was reduced significantly at 4 h and 24 h in comparison with those of WT animals (Fig. 2, P < 0·01). Similarly, the treatment of WT mice with etanercept reduced significantly the neutrophils infiltration in the lung tissues at 4 h and at 24 h after CAR injection (Fig. 2, P < 0·01).

Figure 2.

Effects of tumour necrosis factor (TNF)-αR1 gene deletion and etanercept administration on CAR-induced neutrophils infiltration in lung tissues. Myeloperoxidase (MPO) activity was elevated significantly at 4 h and 24 h after CAR administration in wild-type (WT) mice. In TNF-α receptor 1 KO mice lung MPO activity was reduced significantly at 4 h and 24 h. Similarly, the treatment of WT mice with etanercept (5 mg/kg administered subcutaneously 2 h prior to CAR) reduced significantly neutrophil infiltration in the lung tissues at 4 h and at 24 h after CAR administration. Data are means ± standard error of the mean of 10 mice for each group. *P < 0·01 versus sham; °P < 0·01 versus CAR-WT group. KO, knock-out; CAR, carrageenan.

No histological alterations were observed in the lung tissues collected from sham WT mice (Fig. 3a) and from sham TNF-αR1KO mice (Fig. 3b) at 4 h. In contrast, histological examination of lung sections collected at 4 h from all CAR-WT mice showed tissue injury as well as inflammatory cells infiltration (Fig. 3c). At 24 h after CAR injection a significant augmentation in lung injury as well as a significant presence of inflammatory cells (Fig. 3d) was observed in the lung tissues collected from WT mice. The absence of TNF-αR1 in mice led to a significant reduction in lung injury and inflammatory cell infiltration at 4 h (Fig. 3e) as well as at 24 h after CAR injection (Fig. 3g). Similarly, the treatment of WT mice with etanercept reduced the lung injury significantly at 4 h (Fig. 3f) and 24 h (Fig. 3h) after CAR injection.

Figure 3.

Effects of tumour necrosis factor (TNF)-α gene deletion and etanercept administration on lung injury. Lung tissues collected at 4 h and 24 h were stained with haematoxylin and eosin. Histological examination of lung sections collected at 4 h from all wild-type (WT) carrageenan-injected mice showed tissue injury as well as inflammatory cells infiltration (c). At 24 h after carrageenan administration a significant augmentation of lung injury as well as a significant presence in inflammatory cells (d) was observed in the lung tissues collected from WT mice. The absence of TNF-α receptor 1 (TNF-αR1) in mice in TNF-αR1 knock-out mice led to a significant reduction in lung injury and inflammatory cells infiltration at 4 h (e) and 24 h after carrageenan administration (g). Similarly, the treatment of WT mice with etanercept (5 mg/kg administered subcutaneously 2 h prior carrageenan) reduced significantly at 4 h (f) and 24 h (h) after carrageenan injection of the lung injury. No injury was observed in the lung tissues collected from sham WT mice (a) and from sham TNF-αR1KO mice (b). Figure is representative of at least three experiments performed on different experimental days.

Effects of TNF-αR1 gene deletion and etanercept administration on cytokine concentrations and immunohistochemical localization in the lung

A substantial increase in TNF-α and IL-1β production was found in pleural exudates (Fig. 4a and b, P < 0·01) and in the lung tissues (Fig. 4c and d, P < 0·01) collected from mice at 4 h and 24 h after CAR injection (Fig. 4 respectively). Pleural exudate and lung tissue production of TNF-α and IL-1β were reduced significantly in CAR-injected TNF-αR1KO mice as well as in WT mice treated with etanercept (Fig. 4, P < 0·01). Therefore, we also evaluated the TNF-α and IL-1β expression in the lung tissues by immunohistochemical detection. Tissue sections obtained from WT animals at 4 h (Fig. 5c; see densitometry Fig. 7) and 24 h (data not shown) after CAR injection demonstrate positive staining for TNF-α localized mainly in the infiltrated inflammatory cells and pneumocytes as well as in vascular wall. In CAR-injected TNF-αR1KO mice, no positive staining for TNF-α was observed in the lung tissues collected at 4 h (Fig. 5d; see densitometry Fig. 7) and 24 h (data not shown). Similarly, the treatment of WT mice with etanercept reduced positive staining visibly and significantly for TNF-α in infiltrated inflammatory cells and pneumocytes as well as in vascular wall in the lung tissues collected at 4 h (Fig. 5e; see densitometry Fig. 7) and at 24 h (data not shown). At 4 h (Fig. 6c; see densitometry Fig. 7) and 24 h (data not shown) after CAR injection, positive staining for IL-1β localized mainly in the infiltrated inflammatory cells was observed in lung tissue sections obtained from WT animals. In CAR-injected TNF-αR1KO mice, the staining for IL-1β in the infiltrated inflammatory cells and pneumocytes as well as in vascular wall was reduced significantly in the lung tissues collected at 4 h (Fig. 6d; see densitometry Fig. 7) and 24 h (data not shown). Similarly, the treatment of WT mice with etanercept reduced significantly positive staining for IL-1β in the lung tissues collected at 4 h (Fig. 6e; see densitometry Fig. 7) and at 24 h (data not shown).

Figure 4.

Effects of tumour necrosis factor-α receptor 1 (TNF-αR1) gene deletion and etanercept administration on lung TNF-α and interleukin (IL)-1β levels. TNF-α and IL-1β production was evaluated in the pleural exudate and lung tissues collected at 4 h and 24 h after CAR injection using an enzyme-linked immunosorbent assay kit. A significant production TNF-α (a) and IL-1β (b) was observed in pleural exudate collected from wild-type (WT) mice. The absence of TNF-αR1 gene in mice as well as the treatment of WT mice with etanercept reduced significantly the pleural exudate production of TNF-α (a) and IL-1β (b). Similarly, a significant increase in TNF-α (c) and IL-1β (d) was observed in the lung tissues from CAR-injected WT mice at 4 h and 24 h after CAR. In the lung tissues from CAR-injected TNF-αR1KO mice as well as of WT mice which received etanercept TNF-α (c) and IL-1β (d) levels were reduced significantly in comparison with those of WT animals. Data are means ± standard error of the mean of 10 mice for each group. *P < 0·01 versus SHAM; °P < 0·01 versus CAR-WT group. KO, knock-out; CAR, carrageenan.

Figure 5.

Immunohistochemical localization of tumour necrosis factor (TNF)-α in the lung. Lung sections were taken 4 h after injection of carrageenan. No positive staining for TNF-α was observed in the lung tissues collected from sham wild-type (WT) mice (a) and from sham TNF-α receptor 1 knock-out (TNF-αR1KO) mice (b). In contrast, positive staining for TNF-α (c), localized mainly in the infiltrated inflammatory cells and pneumocytes as well as in vascular wall (see arrows), was observed in lung sections from WT animals. In carrageenan-injected TNF-αR1KO mice, no positive staining for TNF-α was observed in the lung tissues (d). Similarly, the treatment of WT mice with etanercept (5 mg/kg administered subcutaneously 2 h prior to carrageenan) reduced positive staining visibly and significantly for TNF-α in the lung tissues (e). Figure is representative of at least three experiments performed on different experimental days.

Figure 7.

Typical densitometry evaluation. Densitometry analysis of immunocytochemistry photographs (n = 5) for intercellular adhesion molecule 1 (ICAM-1), tumour necrosis factor (TNF)-α, interleukin-1 (IL-1), P-selectin, inducible nitric oxide sythase (NOS), nitrotyrosine and Fas ligand (FasL) from lung was assessed. The assay was carried out by using Optilab Graftek software on a Macintosh personal computer (CPU G3-266). Data are expressed as percentage of total tissue area; n.d., not detectable. *P < 0·01 versus sham. °P < 0·01 versus carrageenan (CAR). WT, wild-type; KO, knock-out.

Figure 6.

Immunohistochemical localization of interleukin (IL)-1β in the lung. Lung sections shown were taken 4 h after injection of carrageenan. No positive staining for IL-1β was observed in the lung tissues collected from sham wild-type (WT) mice (a) and from sham tumour necrosis factor-α receptor 1 knock-out (TNF-αR1KO) mice (b). At 4 h after carrageenan administration, positive staining for IL-1β localized in the infiltrated inflammatory cells and pneumocytes as well as in vascular wall (c, see arrows) was observed in lung tissue sections obtained from WT animals. The absence of a functional TNF-αR1 gene resulted in a significant reduction of the positive staining for IL-1β (d) in the infiltrated inflammatory cells and pneumocytes as well as in vascular wall (e). Similarly, treatment of WT mice with etanercept (5 mg/kg administered subcutaneously 2 h prior to carrageenan) reduced positive staining visibly and significantly for IL-1β in the lung tissues (e). Figure is representative of at least three experiments performed on different experimental days.

Effects of TNF-αR1 gene deletion and etanercept administration on CAR-induced immunohistochemical localization of adhesion molecules (ICAM-1 and P-selectin) in the lung tissues after pleurisy

Enhanced staining for ICAM-1 and P-selectin was evident in lung sections obtained from WT mice at 4 h (Figs 8c and 9c respectively; see densitometry Fig. 7) and at 24 h (data not shown). In inflamed TNF-αR1KO mice, the positive immunostaining for ICAM-1 and P-selectin was reduced significantly in tissues collected at 4 h (Figs 8d and 9d respectively; see densitometry Fig. 7) and at 24 h (data not shown). Similarly, treatment of TNF-αR1WT mice with etanercept reduced positive immunostaining for ICAM-1 and P-selectin in the lung tissues collected at 4 h (Figs 8e and 9e respectively; see densitometry Fig. 7) and 24 h (data not shown).

Figure 8.

Immunohistochemical localization of intercellular adhesion molecule 1 (ICAM-1) in the lung. Lung tissues shown were taken 4 h after injection of carrageenan. No positive staining for ICAM-1 was found in lung tissue section from sham wild-type (WT) mice (a) and sham tumour necrosis factor-α receptor 1 knock-out (TNF-αR1KO) mice (b). In contrast, an increase in imununohistochemical staining for ICAM-1 was observed in lung tissue sections obtained from WT mice (c) localized mainly around the vessels (see arrows). In carrageenan-treated TNF-αR1KO mice, the positive immunostaining in the lung for ICAM-1 was reduced visibly and significantly (d). Similarly, the treatment of TNF-α-R1 WT mice with etanercept (5 mg/kg administered subcutaneously 2 h prior to carrageenan) reduced the positive immunostaining for ICAM-1 in the lung tissues (e). Figure is representative of at least three experiments performed on different experimental days.

Figure 9.

Immunohistochemical localization of P-selectin in the lung. Lung tissues shown were taken 4 h after injection of carrageenan. No positive staining for P-selectin was found in lung tissue sections from sham wild-type (WT) mice (a) and from sham tumour necrosis factor-α receptor 1 knock-out (TNF-αR1KO) mice (b). In contrast, an increase in imununohistochemical staining for P-selectin was observed in lung tissue sections obtained from WT mice (c) localized mainly around the vessels (see arrows). In carrageenan-treated TNF-αR1KO mice, the positive immunostaining for P-selectin was reduced visibly and significantly (d). Similarly, the treatment of TNF-α-R1 WT mice with etanercept (5 mg/kg administered subcutaneously 2 h prior to carrageeenan) reduced the positive immunostaining for P-selectin in the lung tissues (e). Figure is representative of at least three experiments performed on different experimental days.

Effects of TNF-αR1 gene deletion and etanercept administration on iNOS expression and nitrite–nitrate concentration

No positive staining for iNOS was observed in the lung tissues obtained from sham WT mice (Fig. 10a) and sham TNF-αR1KO mice (Fig. 10b). In contrast, tissue sections obtained from WT animals at 4 h (Fig. 10c; see densitometry Fig. 7) and 24 h (data not shown) after CAR injection demonstrate positive staining for iNOS localized mainly in the infiltrated inflammatory cells. In CAR-injected TNF-αR1KO mice, no positive staining for iNOS was observed in the lung tissues collected at 4 h (Fig. 10d) and 24 h (data not shown). Similarly, the treatment of WT mice with etanercept reduced positive staining visibly and significantly for iNOS in the infiltrated inflammatory cells in the lung tissues collected at 4 h (Fig. 10e) and 24 h (data not shown). NO levels were also increased significantly in the pleural exudate obtained from WT mice at 4 h and 24 h after CAR injection (Fig. 10f, P < 0·01). NO levels were reduced significantly in CAR-injected TNF-αR1KO mice as well as in WT mice treated with etanercept (Fig. 10f).

Figure 10.

Effects of tumour necrosis factor-α receptor 1 (TNF-αR1) gene deletion and etanercept administration on inducible nitric oxide synthase (iNOS) expression and nitrite/nitrate formation. iNOS expression was studied by immunohistochemical localization in the lung collected at 4 h after injection of carrageenan. No positive staining for iNOS was observed in the lung tissues obtained from sham wild-type (WT) mice (a) and sham TNF-αR1 knock-out (KO) mice (b). In contrast, tissue sections obtained from WT animals at 4 h after carrageenan injection demonstrate positive staining for iNOS localized mainly in the infiltrated inflammatory cells (c, see arrows). In carrageenan-injected TNF-αR1KO mice, no positive staining for iNOS were observed in the lung tissues (d). Similarly, the treatment of WT mice with etanercept (5 mg/kg administered subcutaneously 2 h prior to carrageenan) reduced positive staining visibly and significantly for iNOS in the infiltrated inflammatory cells in the lung tissues (e). Nitrite/nitrate levels were also increased significantly in the exudate obtained from WT mice at 4 h and 24 h after carrageenan injection (f). Nitric oxide exudate levels were reduced significantly in carrageenan-injected TNF-αR1KO mice as well as in WT mice treated with etanercept (5 mg/kg administered subcutaneously 2 h prior to carrageenan) (f). Figure is representative of at least three experiments performed on different experimental days. Data are means ± standard error of the mean of 10 mice for each group. *P < 0·01 versus sham; °P < 0·01 versus carrageenan-WT group.

Effects of TNF-αR1 gene deletion and etanercept administration on nitrotyrosine formation and thiobarbituric acid-reactant substances measurement

Tissue sections obtained from WT animals at 4 h (Fig. 11c, see densitometry Fig. 7) and 24 h (data not shown) after CAR injection demonstrate positive staining for nitrotyrosine localized mainly in the infiltrated inflammatory cells. In CAR-injected TNF-αR1KO mice, no positive staining for nitrotyrosine was observed in the lung tissues collected at 4 h (Fig. 11d) and 24 h (data not shown). Similarly, the treatment of WT mice with etanercept reduced positive staining visibly and significantly for nitrotyrosine in infiltrated inflammatory cells in tissues collected at 4 h (Fig. 11e) and 24 h (data not shown). In addition, at 4 h and 24 h after CAR-induced pleurisy, MDA levels were also measured in the lungs as an indicator of lipid peroxidation. As shown in Fig. 11f, MDA levels were increased significantly in the lungs from CAR-injected mice. Lipid peroxidation was attenuated significantly in CAR-injected TNF-αR1KO mice as well as in WT mice treated with etanercept (Fig. 11f).

Figure 11.

Effects of tumour necrosis factor-α receptor 1 (TNF-αR1) gene deletion and etanercept administration on nitrotyrosine formation and lipid peroxidation. Nitrotyrosine formation was evaluated by immunohistochemical localization in the lungs collected at 4 h after injection of carrageenan. No positive staining for nitrotyrosine was observed in the lung tissues obtained from sham wild-type (WT) mice (a) and sham TNF-αR1 knock-out (KO) mice (b). In contrast, tissue sections obtained from WT animals at 4 h after carrageenan administration demonstrate positive staining for nitrotyrosine localized mainly in the infiltrated inflammatory cells (c, see arrows). In carrageenan-treated TNF-αR1KO mice, no positive staining for nitrotyrosine was observed in the lung tissues (d). Similarly, the treatment of WT mice with etanercept (5 mg/kg administered intraperitoneally 2 h prior to carrageenan) reduced positive staining visibly and significantly for nitrotyrosine in the infiltrated inflammatory cells in the lung tissues (e). In addition, at 4 h and 24 h after carrageenan-induced pleurisy, malondialdehyde bis (dimethyl acetal) 99% levels were increased significantly in the lungs from carrageenan-treated mice at 4 h and 24 h after carrageenan administration (f). Lipid peroxidation was attenuated significantly in carrageenan- treated TNF-αR1KO mice as well as in WT mice treated with etanercept (5 mg/kg administered subcutaneously 2 h prior to carrageenan) (f). Figure is representative of at least three experiments performed on different experimental days. Data are means ± standard error of the mean of 10 mice for each group. *P < 0·01 versus sham; °P < 0·01 versus carrageenan-WT group.

Immunohistochemical localization of FasL in the lung after CAR-induced pleurisy

The potential effect of TNF-αR1 gene deletion on apoptosis in acute lung inflammation was evaluated by immunohistochemical detection of FasL. At 4 h (Fig. 12c; see densitometry Fig. 7) and 24 h (data not shown) after CAR injection, positive staining for FasL is detected readily in the lung tissues from WT mice localized mainly in vascular wall and pneumocytes. The presence of positive staining for FasL were reduced significantly in the absence of a functional TNF-αR1 gene at 4 h (Fig. 12d) and at 24 h (data not shown) after CAR. Similarly, the treatment of WT mice with etanercept reduced positive staining for FasL visibly and significantly in the lung tissues collected at 4 h (Fig. 12e) and 24 h (data not shown).

Figure 12.

Immunohistochemical localization of Fas ligand (FasL) in the lung. Lung tissues were taken 4 h after injection of carrageenan. No positive staining for FasL was observed in the lung tissues from sham wild-type (WT) mice (a) and sham tumour necrosis factor-α receptor 1 knock-out (TNF-αR1KO) mice (b). Positive staining for FasL (c) localized mainly in vascular wall and in pneumocytes (see arrows) were detected in the lung tissues from WT mice. The absence of a functional TNF-αR1 gene in TNF-αR1KO mice resulted in a significant reduction of the positive staining for FasL (d) in the infiltrated inflammatory cells and pneumocytes as well as in vascular wall (e). Similarly, the treatment of WT mice with etanercept (5 mg/kg administered subcutaneously 2 h prior to carrageenan) reduced positive staining for FasL visibly and significantly in the lung tissues (e). Figure is representative of at least three experiments performed on different experimental days.

Effects of TNF-αR1 gene deletion and etanercept administration on CAR-induced paw oedema and MPO activity

Utilizing a well-established model of acute inflammatory response (CAR-induced paw inflammation) [31], the putative role of TNF-α in acute inflammation was investigated. No paw oedema formation was observed in the sham WT mice and sham TNF-αR1KO mice (data not shown). Subplantar injection of CAR in WT mice leads to a time-dependent development of inflammation, which peaks within 4–5 h (Fig. 13a). The absence of TNF-αR1 in mice led to a significant reduction of paw oedema formation at all time-points (Fig. 13a) as well as a reduction in infiltration of neutrophils (Fig. 13b), as indicated by MPO activity. MPO activity levels were increased significantly (Fig. 13b, P < 0·01) in the paw from WT mice at 4 h after CAR injection when compared with sham WT mice and sham TNF-αR1KO mice (Fig. 13b). Similarly, the treatment of WT mice CAR-injected and treated with etanercept reduced significantly paw oedema formation and MPO activity at all time-points (Fig. 13a and b, P < 0·01).

Figure 13.

Effects of tumour necrosis factor-α receptor 1 (TNF-αR1) gene deletion and etanercept administration on paw oedema development (a) and neuthophil infiltration (b) induced by carrageenan subplantar injection in the mice. The paw volume was measured before the subplantar injection and at 1-h intervals up to 5 h. The oedema volume is the difference in the paw volume at each time-point and the basal paw volume. Myeloperoxidase (MPO) activity was increased significantly in the paw from wild-type (WT) mice collected at 4 h after carrageenan (b). The absence of TNF-αR1 gene in knock-out mice (TNF-αR1KO) as well as the treatment of WT mice with etanercept reduced significantly the carrageenan-induced paw oedema development at all time-points (a). Moreover, MPO activity was reduced in significantly carrageenan-treated TNF-αR1KO mice as well as in WT mice pretreated with etanercept (b). Data are means ± standard error of the mean of 10 mice for each group. #P < 0·01 versus carrageenan-WT group at the indicated time-points. *P < 0·01 versus SHAM; °P < 0·01 versus carrageenan-WT group.

Discussion

Our study provides evidence that genetic (mice with a targeted deletion of the TNF-αR1 gene, TNF-αR1KO) or pharmacological (TNF-αR1WT mice treated with etanercept) inhibition of TNF-α exerts a protective effect against the pathological changes in response to acute inflammation. In particular we have demonstrated clearly that TNF-α genetic or pharmacological inhibition reduce: (i) the development of CAR-induced pleurisy; (ii) the development of CAR-induced paw oedema; (iii) the infiltration of PMNs; (iv) the degree of proinflammatory cytokine production in the pleural exudates and lung tissue; (v) the expression of FasL; and (vi) the degree of tissues (lung) injury caused by injection of CAR. All these findings support the view that TNF-α receptor modulates the degree of acute inflammation in mice.

There is good evidence that TNF-α and IL-1β help to propagate the extension of local or systemic inflammatory process [32,33]. We confirm here that CAR injection causes a significant increase in TNF-α and IL-1β in the pleural exudates and lung tissues which probably contribute, in different capacities, to the evolution of acute inflammation. As expected, the TNF-α levels were abolished in TNF-αR1KO mice and reduced significantly in WT mice treated with etanercept. Interestingly, the levels of IL-1β are also reduced significantly in TNF-αR1KO mice as well as in the WT mice treated with etanercept, suggesting that TNF-α modulates the activation and subsequent expression of proinflammatory genes.

Various studies have demonstrated clearly that endothelial cells are major regulators of the neutrophil traffic mastering the processes of chemoattraction, adhesion and migration from the vasculature to the tissues [34–36]. Injured endothelial cells produce proinflammatory cytokines, including TNF-α, which can up-regulate endothelial expression of P-selectin and ICAM-1 in an autocrine fashion [37,38]. We confirm in the present study that CAR injection induced the expression of P-selectin and ICAM-1 on endothelial cells, and this resulted in a significant neutrophil infiltration. The pharmacological TNF-α inhibition with etanercept, as well as the TNF-αR1 genetic deletion, attenuated up-regulation of P-selectin and ICAM-1 in the lung tissue. The absence of an increased expression of adhesion molecules in the lung from etanercept-treated mice and TNF-αR1KO mice correlated with the reduction of leucocyte infiltration, as assessed by the specific granulocyte enzyme MPO, and with moderation of tissue damage, as evaluated by histological examination. Neutrophils play a crucial role in the development and full manifestation of acute inflammation [39]. Their infiltration into inflamed tissue contributes to the inactivation of foreign antigens and to the remodelling of injured tissue, but an exaggerated recruitment accounts for tissue destruction via the production of reactive oxygen metabolites [40], granule enzymes and cytokines that amplify further the inflammatory response. Therefore, the reduced neutrophil recruitment seen in TNF-αR1KO mice and in WT mice treated with etanercept represents an important additional mechanism subtended to its protective anti-inflammatory effects.

There is a large amount of evidence that the production of reactive oxygen species (ROS) such as hydrogen peroxide, superoxide and hydroxyl radicals at the site of inflammation contribute to tissue damage [41–43]. Inhibitors of NOS activity reduce the development of CAR-induced inflammation and support a role for NO in the pathophysiology associated with this model of inflammation [44]. We demonstrate here that the formation of nitrite and nitrate (metabolites of NO in water), as well as the induction of iNOS protein in lung tissues caused by injection of CAR into the pleural cavity, is reduced in lungs from TNF-αR1KO mice as well as in WT mice pretreated with etanercept. This finding confirms that TNF-α amplifies the induction of iNOS caused by CAR in the lung as well as induction of iNOS caused, for example, by injection of endotoxin in rodents in vivo, was mediated by endogenous TNF-α[15]; polyclonal antibodies against this cytokine in fact abolish the induction of iNOS caused by endotoxin [45]. Also endogenous IL-1β plays an important role in the induction of iNOS, such as endogenous IL-1 receptor antagonists, reducing the degree of iNOS induction caused by injection of lipopolysaccharide in rodents [46]. As levels of TNF-α and IL-1 are significantly lower in the exudate and in lung tissues obtained from TNF-αR1KO mice as well as in WT mice pretreated with etanercept, we propose that the reduction of iNOS protein and activity observed in mice are secondary to a reduced formation of TNF-α but also to the subsequent reduction of IL-1β. This reduction in the expression of iNOS in the lung tissues obtained from TNF-αR1KO mice as well as WT mice pretreated with etanercept may also contribute to the attenuation of nitrotyrosine formation and lipid peroxidation in the lung in CAR-injected animals. Nitrotyrosine formation, along with its detection by immunostaining, was proposed initially as a relatively specific marker for the detection of the endogenous formation ‘footprint’ of peroxynitrite [47]. There is, however, recent evidence that other reactions can also induce tyrosine nitration; e.g. reaction of nitrite with hypoclorous acid and the reaction of MPO with hydrogen peroxide can lead to the formation of nitrotyrosine [48]. Increased nitrotyrosine staining is considered therefore as an indicator of ‘increased nitrosative stress’ rather than a specific marker of the generation of peroxynitrite.

Generation of ROS has been implicated in induction of cell death and inflammation in the paw and lung tissues after CAR injection [3,41]. Furthermore, cell death induced by ROS depends on FasL expression mediated by redox-sensitive activation of nuclear factor-κB [49]. FasL plays a central role in apoptosis induced by a variety of chemical and physical insults [50]. Recently it has been pointed out that Fas–FasL signalling plays a central role in acute inflammation (e.g. acute lung injury) [51,52]. We confirm here that the inflammatory process (CAR-induced pleurisy) leads to a substantial activation of FasL in the lung tissues which probably contributes in different capacities to the evolution of acute inflammation. In the present study, we found that FasL activation was reduced significantly in lungs from TNF-αR1KO mice as well as in WT mice pretreated with etanercept. Moreover, FasL activation also induced a proinflammatory response characterized by a release of IL-1β and chemokines macrophage inflammatory protein (MIP)-1α, MIP-1β and MIP-2 [50]. Reduction in the FasL activation in the lungs obtained from TNF-αR1KO mice as well as in WT mice pretreated with etanercept contributes to the attenuation of IL-1β production found in the pleural exudates and in the lung.

The overall view that arises from our data is that administration of etanercept has beneficial effects on acute inflammation and on tissue injury associated to CAR-induced acute inflammation in mice. In particular, the main findings of our study are the following: etanercept (i) ameliorates the course of CAR-induced acute inflammation; (ii) reduces lung injury and neutrophil infiltration; (iii) reduces the formation of proinflammatory cytokines; (iv) reduces iNOS expression and nitrite/nitrate exudate levels and nitrotyrosine formation and; (v) reduces the degree of apoptosis. It is known that blocking TNF-α impairs innate anti-microbial defences (especially against intracellular pathogens), may cause opportunistic infections, and could even be harmful in those conditions in which microbial growth contributes to the disease pathogenesis, as observed in clinical trials directed at antagonizing TNF-α in human sepsis. These studies showed only a modest impact on survival, albeit in heterogeneous patient populations. The present report suggests that TNF-α neutralization could be potentially useful in acute inflammation (e.g. lung injury) of non-septic origin where microbial proliferation does not occur, and this is in accordance with the results obtained neutralizing TNF-α in experimental models of endotoxemia and acute pancreatitis. The effects of etanercept are similar to those observed in TNF-αR1KO mice. However, it is important to point out that in both the KO and etanercept groups that received CAR injections, most of the markers were not abolished but attenuated significantly. This confirms clearly that TNF-α is an important, but not the only, mediator of the inflammatory process. Moreover, further experiments with multiple approaches can be expected to unlock the full potential of the TNF-α pathway for therapeutic purposes. In conclusion, we have demonstrated in vivo that the pharmacological inhibition of TNF-α by etanercept attenuates the development of acute inflammation in mice and the administration of this dimeric fusion protein results in the same outcome as that of deletion of the receptor. Knocking out a TNF-α receptor p55 gene reproduced the characteristics observed when etanercept rendered TNF-α biologically inactive. Our results show direct evidence that TNF-α plays a pivotal role in acute inflammation and that it is a excellent target by therapeutic applications.

Acknowledgements

This study was supported by a grant from the Caminiti SRL foundation. The authors would like to thank Giovanni Pergolizzi and Carmelo La Spada for their excellent technical assistance during this study, Mrs Caterina Cutrona for secretarial assistance and Miss Valentina Malvagni for editorial assistance with the manuscript.

Ancillary