Background
The activation of Toll-Like Receptors (TLRs), a family of innate immune receptors, is believed to be an important step in the initiation of the inflammatory response raised against numerous pathogens. TLR3 is a mammalian pattern recognition receptor that recognizes double-stranded (ds) RNA as well as the synthetic ds RNA analog poly-riboinosinic-ribocytidylic acid (poly(I:C)) [
1]. Activation of TLR3 by poly(I:C) or by endogenous mRNA ligands, such as those released from necrotic cells [
2], induces secretion of pro-inflammatory cytokines and chemokines, a finding that suggests that TLR3 agonists modulate disease outcome during infection-associated inflammation [
3]. Thus, long-term activation of TLR3
in vivo is thought to occur in the context of viral infection [
4] or necrosis associated with inflammation [
2].
In vitro studies have demonstrated that stimulation of lung epithelial cells with poly(I:C) elicited the secretion of multiple cytokines, chemokines, the induction of transcription factors and increased expression of TLRs [
3]. It has also been demonstrated that poly(I:C) enhanced bradykinin- and [des-Arg
9]-bradykinin-induced contractions of tracheal explants
in vitro, an effect mediated by C-jun-amino-terminal kinase (JNK) and nuclear factor kappa B (NF-kB) signaling pathways [
5]. Taken together, these data suggest that TLR3 activation may have a physiological consequence in the lung. Further, these data demonstrate that ligation of TLR3 initiates cascades of phosphorylation and transcriptional activation events that result in the production of numerous inflammatory cytokines that are thought to contribute to innate immunity [
5]. Overall, these data suggest that sustained TLR3 activation can be a critical component in the modulation of infection-associated inflammatory diseases.
Exacerbations in respiratory diseases such as asthma and chronic obstructive pulmonary disease (COPD) are characterized by the worsening of symptoms and a decline in lung function. Viral infections are associated with respiratory disease exacerbations [
6] and may be associated with progression of disease. Secretion of pro-inflammatory cytokines in the lungs following viral infection represents a crucial step in promoting the inflammatory response in various lung diseases [
7,
8]. A better understanding of the effects of TLR3 activation may provide insight into the mechanisms underlying virally-induced respiratory disease exacerbations.
In the current study we examined the effects of TLR3 activation in vivo. We sought to induce long term activation of TLR3 to mimic the physiologic disease state associated with virally-induced disease exacerbations. Administration of poly(I:C) to the lungs of mice induced a marked impairment of lung function that was accompanied by the production of pro-inflammatory mediators and inflammatory cell recruitment into the airways. TLR3 appears to play a role in the effects of poly(I:C) since TLR3 KO mice were partially protected. Taken together, our data suggest an important role for TLR3 activation in impairment of lung function.
Methods
Poly(I:C) induced cytokine secretion in BEAS-2B cells
The SV-40-transformed normal human bronchial epithelial cell line, BEAS-2B (ATCC, VA) was cultured in LHC-9 media without additional supplements. (Biosource, CA). 1 × 106 cells were seeded in collagen type I-coated T75 flasks (BD, NJ) and split every 2–3 days using 0.25% trypsin/ethylenediaminetetraacetic acid (EDTA) (Gibco, CA). Poly(I:C) (Amersham, NJ) was dissolved in phosphate-buffered saline (10 mM phosphate, 150 mM NaCl, pH 7.4; phosphate buffered saline (PBS)) at a concentration of 2 mg/ml and aliquots were stored at -20°C. For poly(I:C) stimulation, cells were incubated at 37°C with different concentrations of poly(I:C). Supernatants were collected after 24 hours and stored at -20°C or assayed immediately for cytokine secretion using a multi-plex bead assay (Biosource, CA) for detection of interferon-alpha (IFNα), interferon-gamma (IFNγ), interleukin-1-beta (IL-1β), interleukin-10 (IL10), interleukin-12p70 (IL12p70), tumor necrosis factor-alpha (TNFα), Chemokine (C-C motif) ligand 3 (CCL3), interleukin-6 (IL-6), interleukin-8 (IL-8), Chemokine (C-C motif) ligand 2 (CCL2), Chemokine (C-C motif) ligand 5 (CCL5), and Chemokine (C-X-C motif) ligand 3 (CXCL10). Limits of detection for the analytes range from 3 – 20 pg/ml. Sample acquisition and analysis was performed using the Luminex 100S with StarStation software (Applied Cytometry Systems).
Administration of Poly(I:C) to the lungs of mice
Female C57BL/6 mice wild-type (WT) (12 weeks old) or female TLR3 knock-out (KO) mice (C57BL/6; 12 weeks old, ACE animals, PA) were anesthetized with isoflurane and different doses (10–100 μg) of poly(I:C) in 50 μl sterile PBS, or PBS alone, were administered intranasally (I.N.) Mice received three administrations of poly(I:C) (or PBS) with a 24 hour rest period between each administration. KO mice were fully backcrossed to C57BL/6 background to at least N10.
All animal care was performed according to the Guide for the Care and Use of Laboratory animals and the Institutional Animal Care and Use Committee approved all studies.
Whole Body Plethysmography
Twenty-four hours following the last poly(I:C) (or PBS) administration, lung function without provocation (baseline) and airway hyperresponsiveness (AHR) to methacholine were measured using whole body plethysmography (BUXCO system). The mice were placed into the whole body plethysmograph chamber and allowed to acclimate for at least 5 minutes. Following baseline readings, mice were exposed to increasing doses of nebulized methacholine (Sigma, MO). The nebulized methacholine was administered for 2 minutes, followed by a 5-minute data collection period, followed by a 10-minute rest period before subsequent increasing-dose methacholine challenges. The increased airflow resistance was measured as Enhanced Pause (Penh) and is represented as the average penh value over the 5-minute recording period.
Invasive measures of lung function
Twenty-four hours following the last poly(I:C) (or PBS) administration, lung function and increased lung resistance in response to methacholine were measured using invasive measures of lung function (BUXCO system). Mice were anesthetized with 50 mg/kg sodium pentobarbital (Nembutal, Abbot Labs, IL). The trachea was cannulated with a 19 gauge cannula and the mouse was connected to a mechanical ventilator, with breath frequency of 120 and stroke volume of 0.3 mL. The mouse was connected to the plethysmograph for lung function measurements. After establishing a stable baseline of lung resistance, methacholine was administered I.V. through the tail vein (240 μg/kg). The peak resistance measured over 3 minutes was recorded.
Measurement of lung inflammation
Following lung function measurements, mice were sacrificed by CO2 asphyxiation and the lungs were cannulated. Bronchoalveolar lavages (BAL) were performed by injecting 1 mL of PBS into the lungs and retrieving the effluent. The lung tissues were removed and frozen. The BALs were centrifuged (1200 rpm, 10 minutes) and the cell-free supernatants were collected and stored at -80°C until analysis. The cell pellet was resuspended in 200 μl PBS for total and differential cell counts using a hemacytometer (on Wright's – Giemsa-stained cytospin preparations).
Measurement of proteins in bronchoalveolar lavage samples
The cell-free supernatants were collected and stored at -80°C until used for analyses. The multiplex assay was performed following the manufacturer's protocol and the LINCOplex Multiplex Immunoassay Kit (LINCO Research, St. Charles, MO). Analytes included in the analysis were MIP1α, Granulocyte Macrophage Colony Stimulating Factor (GMCSF), JE, KC, RANTES, IFNγ, IL-1α, IL-1β, Granulocyte Colony Stimulating Factor (GCSF), CXCL10, IL-2, IL-4, IL-5, IL-6, IL-7, IL-9, IL-10, IL-12(p70), IL-13, IL-15, IL-17 and TNFα. Limits of detection for the analytes range from 3 – 20 pg/ml.
Measurement of lung mRNA expression
Following collection of BAL samples, the right lobes of the lung were removed and placed in Trizol total RNA isolation reagent (Life Technologies, Gaithersburg, MD). RNA was isolated using manufacturer's instructions of the Qiagen Rneasy Mini kit (Qiagen, Valencia, CA). Total RNA (2 μg) from pooled groups was then reverse transcribed using the OmniScript RT kit (Qiagen, Valencia, CA) according to the manufacturer's protocol. One hundred nanograms of cDNA was then amplified using both the TaqMan® Low Density Immune Profiling Array cards (Applied Biosystems, Foster City, CA), or microfluidic cards, and custom Low Density Array cards. Primer-probes with genes of interest were plated in a 384 well format following the manufacturer's protocol for Real-Time PCR. Data are normalized to 18s rRNA and represent fold change over PBS treated mice.
Histological Analysis
Following BAL collection, the left lobes were inflated with 10% neutral buffered formalin under constant pressure then immersed in additional fixative, the right lobes were clamped with hemostats and ligated. Tissue was processed by routine methods, oriented so as to provide coronal sections and 5 micron mid-coronal sections cut and stained with hematoxylin and eosin.
Morphometric analysis
A Nikon Eclipse E800 (Nikon Corporation, Tokyo, Japan) microscope was equipped with an Evolution™ MP 5.0 RTV color camera (Media Cybernetics, Inc. Silver Spring, MD). Images were captured and analyzed using Image-Pro Plus software version 5.1 (Media Cybernetics, Inc. Silver Spring, MD). GraphPad Prism version 4.03 (GraphPad Software, Inc. San Diego, CA) was used to interpret, analyze and graph the raw data. SigmaStat Statistical Software version 2.03 (SPSS, Inc. Chicago, IL) was used to perform statistical analysis on the collected data. Using the Auto-Pro tool within the Image-Pro Plus software, custom written macros were used to perform the analysis. Six TLR3 KO mice treated with poly(I:C), six WT mice treated with poly(I:C), four TLR3 KO mice treated with PBS and six WT mice treated with PBS were imaged and analyzed. No imaging or analysis was performed on areas of the lung that were torn, damaged, or folded.
Tissue Density
From each lung, five fields were randomly selected and imaged using a 20× objective lens. The total area of the tissue was measured and the ratio of total area of tissue to total area of field calculated.
Tissue Cellularity
From each lung, five fields were randomly selected and imaged using a 20× objective lens. The total area of the nuclei was measured and the ratio of total area of nuclei to total area of field calculated.
Airway Cellularity
From each lung, five airways were chosen and imaged using a 40× objective lens. A line of 100 μm in length was superimposed on the airway at a random location. The number of nuclei within the fixed distance were counted and recorded.
Airway Mucosal Height
From each lung, five airways were chosen and imaged using a 40× objective lens. The image was segmented so as to include only the airway mucosa and the average thickness of the airway mucosa was measured using the curve thickness algorithm built into ImagePro. This algorithm parses the mucosa into 30,000 arc segments, measures the thickness of the mucosa at each arc segment and calculated the average thickness for the mucosa.
Statistical analysis
Specific statistical methods are described in the figure legends. Graphs and summary statistics were also used to assess the results. All statistical tests were 2-sided. Except for where noted, all p-values presented are unadjusted for multiple comparisons.
Discussion
Exacerbations of respiratory diseases such as asthma and COPD are often associated with concomitant respiratory viral infections. Since TLR3 is activated by viral dsRNA, the purpose of the current study was to better understand the functional consequences of TLR3 activation in vivo. Administration of poly(I:C), a synthetic TLR3 ligand, to the lungs of mice induced marked inflammation accompanied by impaired lung function. TLR3 KO mice were partially protected from the effects of poly(I:C) demonstrating the involvement of TLR3. These data provide further support for a role of TLR3 in respiratory diseases and suggest a potential mechanisitic pathway for viral exacerbations.
Upon activation, TLR3 recruits a Toll-IL-1 receptor (TIR) – related adaptor protein inducing interferon (TRIF), which activates both IFN-regulatory factor 3 (IRF3) and NF-kB [
12] and [
13]. In our model, following poly(I:C) administration to the lungs, there was an up regulation of TLR3, -2, -7, and 9 gene expression and their associated signaling molecules. Previous
in vitro studies have demonstrated that activation of TLR3 with poly(I:C) induces up regulation of its own expression as well as the expression of other TLRs. For example, poly(I:C) up regulates mRNA for TLR2, 3 and 4 in airway smooth muscle cells [
14] and TLR2, 3, 6 and 10 in lung epithelial cells [
3].
In vivo, the up regulation of TLR mRNA expression may have occurred as a result of expression of TLRs on infiltrating cells or through up regulation on resident lung cells. Indeed, monocytes express all of the known TLRs [
15]. In contrast, neutrophils have been shown to express all the TLRs except TLR3 [
16]. Within the lung, all of the known TLRs have been found to be expressed by human primary bronchial epithelial [
3] and smooth muscle cells [
14]. The up regulation of multiple members of the TLR family, as a consequence of activation of one TLR, may indicate the creation of an environment of hyper-responsiveness to pathogen insult whereby, an exacerbation event could be triggered in the event that the lung is exposed to other toll-ligands. In support of this hypothesis, it has been shown that infection of airway epithelial cells with
Hemophilus influenza induced the secretion of CXCL-8, up regulated TLR3 expression and increased the responsiveness to a secondary challenge of
Rhinovirus. Interestingly, inhibition of TLR3 with small interfering RNA, inhibited the
Rhinovirus-induced CXCL-8 production [
17]. In addition this same group demonstrated that pretreatment with
Rhinovirus resulted in delayed bacterial clearance when a secondary infection was induced using nontypeable
Hemophilus influenza. Sajjan et al. showed that this may be the result of decreases in transepithelial resistance or compromised tight junctions and loss of zona occludins-1 and junctional adhesion molecule-1 [
18]. Taken together these studies suggest that activation of TLRs, such as TLR3 can result in a perturbation of the local environment, specifically dysregulation of the airway epithelium thereby supporting an environment primed for an exacerbation. We are currently focusing efforts in our laboratory toward identifying the composition of the mononuclear cell populations in this model including the activation state of various cell types including dendritic cells. In a review by Fe
et. al. it is summarized that TLR3 can induce a variety of cytokines in human dendritic cells including IFNβ, and CXCL10 [
19].
In vivo TLR3 agonism by synthetic dsRNA also resulted in a profound up regulation of the expression and secretion of multiple pro-inflammatory cytokines, chemokines, and growth factors.
In vitro studies have demonstrated that activation of TLR3 by dsRNA on different cell types including natural killer cells [
20], epithelial cells [
3,
21,
22], and smooth muscle cells [
14] results in increased expression and/or secretion of pro-inflammatory cytokines including IL-6, CXCL-8, CCL-2, CCL-5, CXCL-10, GM-CSF, TNFα and IFNγ. A likely source of cytokines following poly(I:C) administration may be the airway epithelium since activation of BEAS-2B cells
in vitro induced a profile of pro-inflammatory cytokines similar to that observed following
in vivo poly(I:C) challenge. TLR3 has been identified and functionally characterized in mouse tracheal muscle [
23] and in primary human small airway epithelial cells [
21,
3,
22]. Previous
in vitro studies have also demonstrated the secretion of inflammatory mediators following TLR3 activation of epithelial cells[
3,
3,
21]. The up regulation of pro-inflammatory cytokines and chemokines provides an inflammatory milieu supporting the infiltration of inflammatory cells into the airways and lung interstitium. Accompanying the inflammation-rich pathology was the presence of bronchial epithelial cell hypertrophy. The hypertrophic cells extended into the secondary and tertiary airways. Epithelial cell hypertrophy is normally associated with increased mucus production [
23]. However, in the current study, there was no evidence for increased mucus production by PAS staining. Given the distribution of goblet cells in normal mouse airways, which is restricted to the main bronchi and primary bronchioles, the data suggest that the hypertrophic epithelial cells are not mucus-producing goblet cells.
Along with the demonstration that poly(I:C), acting as a TLR3 ligand, results in an inflammatory response
in vivo, the study presents a novel finding that stimulation of TLR3 results in a measurable impairment of lung function both without provocation and characterized by increased AHR to methacholine. Similar changes in baseline lung function have also been described in mice exposed to Respiratory Syncytial virus (RSV) [
24]. Recent studies have demonstrated that pre-exposure of mouse tracheas to poly(I:C) in vitro increases the expression of bradykinin B1 and B2 receptors on the smooth muscle and confers AHR to bradykinin [
25]. Notably, inhibition of the bradykinin B1 receptor confers protection from acetylcholine-induced AHR following allergen sensitization and challenge [
26]. In contrast, AHR to histamine following
parainfluenza-3 infection in guinea pigs was inhibited by a bradykinin B2 receptor antagonist [
27]. Taken together these data suggest a role for bradykinin in TLR3-induced airway dysfunction. In the current study some, but not all, functional responses were protected in TLR3 KO mice following multiple administrations of poly(I:C). Specifically, they were protected from baseline lung function changes in response to poly(I:C), however protection from AHR in response to provocation with methacholine did not result in significant protection. Further, the pro-inflammatory mediators produced following poly(I:C) administration were not modulated in TLR3 KO mice. Unpublished data from our laboratory has shown that TLR3 KO mice were significantly protected from a single administration of poly(I:C) with respect to pro-inflammatory mediators in the bronchoalveolar lavage (data not shown), indicating that mediators released in response to acute activation with poly(I:C) may be more TLR3 dependent. This data suggests that another receptor for poly(I:C) may be available. Indeed, since a percentage of TLR3 KO mice succumb to poly(I:C)-induced shock, it suggests that poly(I:C) may still signal in the absence of TLR3 [
1]. Indeed, dsRNA can also signal through dsRNA-dependent protein kinase (PKR) [
28], RIGI [
29] and MDA-5 [
30]. The potential redundancy in the dsRNA downstream pathways may be an explanation for the incomplete protection observed in TLR3 KO mice.
Understanding the different signaling pathways involved in recognition of dsRNA by the host has been a major area of focus by many researchers. Le Goffic
et al. demonstrated that sensing of
influenza A virus by TLR3 and RIG-I regulates a pro-inflammatory response. In contrast, RIG-I but not MDA-5 also mediates type I IFN-dependent antiviral signaling response[
31]. Use of non-poly(I:C) TLR3 ligands is necessary to further define the impact of TLR3-specific signaling on pulmonary pathophysiology. Interestingly, TLR3 KO mice demonstrate protection from
influenza A virus-induced lung function impairment accompanied by reduced inflammation and improved survival [
32].
These data taken along with the inflammatory consequences of TLR3 activation suggest that sustained TLR3 activation may also contribute to severe exacerbations of chronic pulmonary diseases. In summary, the data presented in this study suggest that sustained TLR3 activation may play an important role in respiratory disease pathogenesis. A better understanding of the effects of TLR3 activation will provide additional insight into the mechanisms underlying virus-induced exacerbations associated with respiratory diseases. Additionally, these studies provide an opportunity to identify suitable targets for therapeutic intervention for respiratory disease exacerbations.
Competing interests
NCS, JS, HAR, KAS, RJL, DDE, PJB, LAM, PA M, RAB, LRS, DEG, RTS, MLM, and AMD are current or former employees of Centocor Research & Development, Inc. RAF and LA declare that they have no competing interests.
Authors' contributions
NCS conceived of the study and participated in its design and coordination as well as all analysis. JS, LAM, LRS, DEG, RTS, MLM, and AMD participated in the design and coordination of the studies. HAR, and KAS executed the in-life portion of the studies. RJL carried out the BEAS2B studies. DDE and PJB carried out the histopath analysis of the lungs. PAM carried out the statistical analysis of all data sets. RAB carried out the analysis of cellular infiltrates in the lung. RAF and LA made the TLR3 KO mice and gave input on the design of the studies and the manuscript. All authors read and approved the final manuscript.