Expression of Toll-like receptor 9 in mouse and human lungs

Abstract

Toll-like receptors (TLR) recognize conserved molecular motifs of microorganisms, and constitute an important part of the innate immune system. Numerous studies have shown the importance of these receptors, including TLR9, in establishing effective immune responses to a broad range of infections, and in disorders such as COPD. TLR9 detects unmethylated DNA and is expressed in a wide range of immune cells in mice and humans, as well as other species. Most TLR9 expression studies have been done on cultured or isolated cells, but none that we know of on intact lung. Because cell-specific expression of TLR9 is important to understand its precise role in lung physiology, we tested mouse and human lung tissues for expression of TLR9 mRNA and protein with in situ hybridization and immunohistochemistry, respectively. We found TLR9 mRNA and protein expression in bronchial epithelium, vascular endothelium, alveolar septal cells and alveolar macrophages in both species. Immuno-electron microscopy delineated TLR9 expression in plasma membrane, cytoplasm and the nucleus of various lung cells. Lungs from human cases of COPD had significantly increased numbers of TLR9-positive cells. These are the first data showing TLR9 mRNA and protein expression in intact human and mouse lungs. The data may be useful for clarifying the role of TLR9 in the contributions of specific cells to lung physiology.

Introduction

The innate immune system plays important role in protecting mucosal surfaces. The toll-like receptors (TLRs) are one of the most studied members of the innate immune system. The TLRs expressed on endothelium, epithelium and immune cells regulate the induction of inflammation and protection against pathogens. In organs such as the lung these receptors have been implicated in many physiological alterations. Recently, a role for TLR4 has been shown in reduced forced expiratory volume in workers in swine confinement buildings (Senthilselvan et al., 2009) and in the clearance of numerous pathogens (Rutz et al., 2004; Kalis et al., 2005; Bhan et al., 2008). Although TLR4 has been studied the most, other receptors such as TLR2 and TLR9 have also been gaining in interest as potentially important receptors. Interestingly, the effects of TLRs on lung inflammation have not always been clear. For example, TLR9 has been shown in various studies to induce (Schwartz et al., 1997; Knuefermann et al., 2007), prevent (Parilla et al., 2006) or modify (Schwartz et al., 1999) lung inflammatory responses.

TLR9 is a membrane-bound receptor found primarily associated with endosomes (Rutz et al., 2004). It binds non-methylated CpG sequences of DNA of a given structure (Krieg, 2002). The effects of non-methylated DNA were first observed in mice (Kuramoto et al., 1992) and TLR9 as the receptor responsible for much of the immune response to this ligand was first elucidated in mice as well (Hemmi et al., 2000). Since then, much work has been done to determine the effects of non-methylated CpG oligonucleotides on TLR9 as well as its expression and localization.

Experiments on localization of the TLR9 receptor have focused primarily on cultured isolated immune cells. These studies established TLR9 expression in B cells (Krieg et al., 1995; Hornung et al., 2002), neutrophils (Schwartz et al., 1997; Jozsef et al., 2004), and eosinophils (Wong et al., 2007). The information on the expression of TLR9 even in immune cells is far from complete, as indicated by the complexity of its expression in monocytes and macrophages. Whereas mice show expression of TLR9 in macrophages (An et al., 2002), the expression in humans was observed only in professional antigen-presenting cells such as dendritic cells (Kadowaki et al., 2001; O'Mahony et al., 2008) in the lung. Interestingly, TLR9 was absent in lung but not splenic dendritic cells of mice (Chen et al., 2006). Alveolar macrophages are vital in the clearance of a variety of pathogens as well as debris from the alveolar space (Bowden, 1984). These cells typically express a larger panel of innate immune receptors (Schneberger et al., 2011a). There is controversy over the expression of TLR9 in mouse and human alveolar macrophages (Fernandez et al., 2004; Suzuki et al., 2005). It is possible that the controversy on the expression of TLR9 in various cells has arisen because studies were done in isolated cultured cells.

The story of TLR9 expression is further complicated by availability of very little in situ data from intact organs such as the lung. Intriguingly, to our knowledge, with the exception of one study on neoplastic lungs (Droemann et al., 2005) there are no data on TLR9 expression in intact lungs of mouse or human. The study of spatial receptor expression in intact lungs is important because of the role of intercellular communication and cell-specific contributions in inflammatory responses. Therefore, we used immunohistochemistry, immuno-electron microscopy, and in situ hybridization to study the expression of TLR9 mRNA and protein in intact lungs of mice and humans and to test the hypothesis that lung expression of TLR9 in mouse and human would follow a pattern similar to that seen in other species.

Materials and methods

Lung tissues

Six- to 8-week-old C3HeB/FeJ mice (n = 5) were purchased (Jackson Laboratories, Bar Harbor, ME) and housed at the Animal Care Unit of the Western College of Veterinary Medicine for 1 week following approval by the institution in accordance with the Canadian Council on Animal Care guidelines. Lungs from euthanized animals were processed and embedded in paraffin.

Paraffin-embedded human lung samples from patients with (n = 4) and without chronic obstructive pulmonary diseases (COPD; n = 4) were obtained following consent from the archives of the Department of Pathology in the College of Medicine at the University of Saskatchewan.

Immunohistochemistry

All lung sections were stained with mouse anti-human TLR9 antibody (1 : 50 dilution; IMG-305a, Imgenex, San Diego, CA) as described previously (Schneberger et al., 2009), and incubated at 4 °C for 16 h followed by incubation with horseradish peroxidase-conjugated goat anti-mouse antibody (1 : 75) for 1 h at room temperature (P0447,Dako, Ontario, Canada). Color was developed using a color development kit and counterstained with methyl green (Vector Laboratories, Ontario, Canada). Immunohistochemical controls included omission of the primary antibody or the secondary antibody as negative controls or staining with von Willebrand factor (vWF) antibody, which stains endothelial cells as a positive control.

Immuno-electron microscopy

Immuno-electron microscopy was carried out as described previously (Schneberger et al., 2009) with TLR9 antibody (1 : 50) and 20 nm gold-conjugated anti-mouse secondary antibody (1 : 100; Fitzgerald Industries International, Concord, MA, USA) for 1 h. The controls included protocols without the primary antibody and staining of sections with vWF antibody.

In situ hybridization

Human TLR9 sequence obtained from Invivogen (pUNO1-hTLR9a) was cut with HindIII and SmaI and ligated into pSPT18 (Roche, Laval, QC, Canada). This plasmid was then transfected and expanded in HIT-DH5α competent cells (Real Biotech Corp., Taipei, Taiwan) as per manufacturer's specifications. Mouse TLR9 in vector pCR II-TOPO (25) was kindly provided by Dr. Serge Rivest (Laval University, Quebec City). Plasmids were purified using GelElute DNA purification kit (Sigma, Oakville, ON, Canada). Purified human plasmid insert was confirmed by cutting with HindIII and SmaI and looking for insert fragment on ethidium bromide acrylamide gel. DIG-labeled probes for TLR9 were generated using the Roche DIG RNA Labeling Kit (Roche) according to manufacturer's instructions. Probe concentration was determined using the Roche DIG Luminescent Detection Kit (Roche) protocol.

Fresh sections were prepared and allowed to adhere to slides for 2 h at 60 °C followed by de-paraffinization and re-hydration (Parbhakar et al., 2004). Sections were treated as per manufacturer's protocol (Biochain, Hayward, CA). The sections were incubated with 2 ng mL⁻¹ DIG-labeled probe for 16 h at 45 °C followed by blocking for 1 h with 1× block buffer. AP-conjugated anti-DIG antibody (1 : 500) was applied for 4 h at room temperature and washed before incubation for 4 h with BCIP-NBT. Control sections were fixed and treated without addition of probe.

Numerical cell count

Cells stained with TLR9 were counted in randomly selected five fields of sections from COPD lungs (n = 4) and those from patients with no identified lung problems (n = 4). The cells were counted at 400× magnification.

Statistical analysis

All values are given as the mean ± standard deviation. A one-way Mann–Whitney test was used to determine significance between normal and COPD lungs.

Results

TLR9 mRNA in mouse lung

TLR9 mRNA was detected in mouse lungs with in situ hybridization. Staining without probe showed no background staining (Fig. 1B). Whereas negative control lung sections lacked any staining, TLR9 mRNA was localized in vascular endothelium (Fig. 1C), airway epithelium (Fig. 1A,E), alveolar macrophages and septal cells (Fig. 1A,D).

Fig. 1.

TLR9 in situ hybridization on mouse lungs. Lung sections stained using TLR9 DIG-labeled RNA show staining (arrows) in several cell types (A). Lung sections from mice stained without a specific probe (B) lacks staining in any tissue. High magnification shows staining in vascular endothelium (arrowhead) (C), septal (arrow) as well as alveolar macrophages (chevron) (D), and bronchial epithelium (double arrow) (E). Scale bar: 100 μm.

TLR9 protein in mouse lung

We determined TLR9 protein expression and distribution in mouse lungs with immunohistochemistry. Lung sections stained without primary (Fig. 2A) or secondary antibody (data not shown) lacked staining, whereas vWF antibody stained vascular endothelium but not the airway epithelium (Fig. 2B). Lungs showed TLR9 protein in bronchiolar epithelium (Fig. 2C,2D), vascular endothelium (Fig. 2C,2E) and alveolar macrophages and the septal cells (Fig. 2C,2F).

Fig. 2.

TLR9 immunohistochemistry on mouse lungs. Lung sections from mice stained with only a secondary antibody (A) lack staining in alveolar septa, whereas those stained with vWF antibody (B) shows staining in endothelium (arrowhead) alone. Lung sections stained using TLR9 antibody show staining in several tissues (arrows) (C), including bronchial epithelium (double arrow) (D), vascular endothelium (arrowhead) (E), and septal (arrow) as well as alveolar macrophages (chevron) (F). Scale bar: 100 μm.

We confirmed the identity of the cells and the subcellular localization of TLR9 with immuno-gold electron microscopy. TLR9 staining was observed in neutrophils (Fig. 3), Type-II alveolar epithelial cells (Fig. 4) and the alveolar macrophages (Fig. 5). Whereas nuclear and cytoplasmic staining for TLR9 was noticed in Type-II alveolar epithelial cells and alveolar macrophages, only cytoplasmic staining was present in neutrophils.

Fig. 3.

TLR9 staining in a mouse neutrophil. TLR9 staining (arrows) observed in mouse lung neutrophil, epithelium (arrowhead), and endothelium (chevron). N, nucleus; AS, alveolar space. Scale bar: 4 μm.

Fig. 4.

TLR9 staining in a mouse type-II cell. TLR9 staining (arrows) observed in a mouse lung type-II cell. N, nucleus; AS, alveolar space; LB, lamellar bodies. Scale bar: 4 μm.

Fig. 5.

TLR9 staining in a mouse alveolar macrophage cell. TLR9 staining (arrows) observed in a mouse alveolar macrophage. N, nucleus; AS, alveolar space. Scale bar: 4 μm.

TLR9 mRNA expression in human lung

In situ hybridization showed TLR9 mRNA in lungs from healthy (Fig. 6A) and COPD (Fig. 6B) humans. Lungs from human cases of COPD appeared to show more intense staining for TLR9 mRNA in the resident as well as many recruited cells (Fig. 6A,B). The negative control lacked positive reaction (Fig. 6C). Large and small lung vessels showed TLR9 mRNA expression in their endothelium (Fig. 6D,E). Large septal cells, most likely macrophages, and alveolar macrophages were also positive for TLR9 mRNA (Fig. 6F).

Fig. 6.

Human lung TLR9 in situ RNA hybridization. Lung sections from normal (A) or COPD (B) patients stained using TLR9 DIG-labeled RNA probe show staining (arrows) in several cell types with a clear influx of TLR9 positive cells in the bronchus. Staining without a RNA probe shows lack of staining in any tissue (C). High magnification shows staining in bronchial epithelium (double arrow) (E), vascular endothelium (arrowhead) (F), and septal (arrow) as well as alveolar (chevron) macrophages (G). Scale bar: 100 μm.

TLR9 protein expression in human lung

Immunohistochemistry of lungs without primary (Fig. 7A) or secondary antibody (data not shown) lacked staining, whereas vascular endothelium reacted with vWF-stained vascular endothelium (Fig. 7B). TLR9 antibody reacted with alveolar septal cells and alveolar macrophages in normal (Fig. 7C) and COPD (Fig. 7D) lungs. Lungs also showed TLR9 protein expression in vascular endothelium (Fig. 7E), bronchiolar epithelium (Fig. 7F), alveolar septal cells (Fig. 7G) and alveolar macrophages (Fig. 7H). The number of TLR9-positive septal cells was increased in the COPD lungs compared with normal human lungs (Fig. 7I).

Fig. 7.

Human lung TLR9 immunohistochemistry. Lung sections from human patients stained with only a secondary antibody (A) lack staining in alveolar septa, whereas those stained with vWF antibody (B) show staining in endothelium (arrowhead) alone. Lung sections stained using TLR9 antibody show staining (arrows) in several cell types in both normal (C) as well as COPD patients (D). High magnification shows staining in bronchial epithelium (double arrow) (E), vascular endothelium (arrowhead) (F), and septal (G) as well as alveolar macrophages (chevron) (H). Scale bar: 100 μm. Field counts of control and COPD patients showed a greater density of TLR9 positive cells in those with COPD (P < 0.01).

Immuno-electron microscopy used for fine detailing showed TLR9 expression in cytoplasm, nucleus and the plasma membrane of alveolar macrophages (Fig. 8). Type-II alveolar epithelial cells contained TLR9 in their cytoplasm (Fig. 9).

Fig. 8.

TLR9 staining in human alveolar macrophage. TLR9 staining (arrows) observed in a human alveolar macrophage. N, nucleus; AS, alveolar space. Scale bar: 4 μm.

Fig. 9.

TLR9 staining in a human type-II cell. TLR9 staining (arrows) observed in a mouse lung type-II cell. N, nucleus; AS, alveolar space. Scale bar: 4 μm.

Discussion

We provide a description of in situ expression of TLR9 protein and mRNA in intact lungs of mice and humans through the use of multiple morphological methods. Although preliminary, these data are important because of the very limited information available on TLR9 expression in intact lungs. Furthermore, these data are needed for mechanistic studies on the role of TLR9 in lung inflammation. The lung comprises many types of resident and transiting cells. The highly coordinated and integrated communication of various cells with each other leads to physiological responses in the lung. Cell responses are largely governed by the type and amount of expression of immune receptors such as TLR9, and inter-cellular communications. Therefore, it is important to understand precise cell-specific expression of receptors such as TLR9 in intact organs. The data reported here show TLR9 mRNA and protein expression in vascular endothelium, bronchiolar epithelium, alveolar septal cells, and alveolar macrophages in lungs of both mice and humans.

We used a combination of methods to detect TLR9 mRNA and protein expression in mouse and human lungs. The reason for studying both human and mouse lungs is that although mice are the most commonly used animal model to study human lung physiology, the basal expression of proteins and mRNA may not be similar between humans and mice. We studied TLR9 protein and mRNA expression in situ in intact lungs of both species because the mRNA may not always be translated into its protein. Lastly, we employed immuno-gold electron microscopy to clarify the sub-cellular expression of TLR9 protein and resolve TLR9 expression in cells of the alveolar septa.

Innate immune receptors such as TLR9 recognize microbial motifs and their expression on epithelia and alveolar macrophage forms the first line of defense in the lung. Interestingly, there is controversy regarding the expression of TLR9 in monocytes and macrophages. The data from isolated cells show that expression of TLR9 appears to be primarily restricted to certain human dendritic cells, particularly plasmacytoid dendritic cells and alveolar macrophages (Kadowaki et al., 2001; Hornung et al., 2002; O'Mahony et al., 2008), whereas in mouse, monocytes and alveolar macrophages express the receptor, but dendritic cells do not. This pattern is more restricted in mouse lung than in other tissues, as these dendritic cells do express TLR9 in the mouse spleen (Chen et al., 2006). Further, some species such as cattle and horses have a highly phagocytic lung macrophage population, known as pulmonary intravascular macrophages, which bind to and reside in lung capillaries (Staub, 1994). These cells also strongly express TLR9, and, in fact depletion of these cells can greatly reduce overall TLR9 mRNA in the lungs of these animals (Schneberger et al., 2009, 2011b).

There has been some dispute over the expression of TLR9 in alveolar macrophages within both human and mouse lungs. Fernandez et al. (2004) found functional inhibition of interleukin (IL)-10R in alveolar macrophages stimulated with CpG-DNA, but Suzuki et al. (2005) found no mRNA expression in lavaged mouse alveolar macrophages. However, Kiemer et al. (2008) showed that expression of TLR9 mRNA and protein does in fact occur in these cells, something we further confirm in this study, but in whole lung tissue.

This information is of particular use as the location of a receptor within the lung can greatly alter its ability to detect its ligand. For example, particulate size has a clear role in determining where inhaled debris is likely to be deposited (Bowden, 1984). Thus, expression of TLR9 in the bronchial epithelium allows for unmethylated DNA to be sensed on much larger particles than would be encountered in the alveolar space. Additionally, as the half-life of naked DNA in circulation is only a matter of minutes (Kawabata et al., 1995), local tissue receptor interaction may be more important than DNA reaching circulating cells.

Alveolar septum comprises epithelial and endothelial cells that may share a common basement membrane, resulting in a thin blood-air barrier to facilitate gas exchange. While in situ hybridization and immunohistochemical staining showed TLR9 expression in alveolar septum of mouse and human lungs, we demonstrated distinct ultrastructural evidence of TLR9 expression in Type-I and Type-II alveolar epithelial cells and microvascular endothelial cells. Previously, Platz et al. (2004) showed expression in lung epithelial cell line, and induction of cytokines in response to CpG stimulation. Similarly, Li et al. (2004) detected TLR9 in cultured pulmonary endothelial cells and showed that CpG DNA induced expression of proinflammatory proteins such as IL-8 and ICAM-1. The presence of TLR9 in vascular and alveolar cells of the septum in intact human and mouse lungs would enable these cells to sense inhaled and blood-borne CpG.

We also examined autopsied lung tissues from COPD patients. The COPD lungs contained large numbers of recruited cells in the airways and highly thickened alveolar septa. Many of the recruited inflammatory cells expressed TLR9 mRNA and protein. Although we did not seek to identify these specific cells, work by others has suggested that these cells could include myeloid dendritic cells (Schaumann et al., 2008), eosinophils (Ilmarinen et al., 2009) and neutrophils (Jozsef et al., 2004), all of which express TLR9 and may account for the positive staining of cells in the vascular space. This increased TLR9 mRNA expression has been shown to occur, such as in idiopathic pulmonary fibrosis and non-small cell lung cancer (Samara et al., 2012).

It should be noted that, as found in tumor cell lines, the expression of TLR9 is no guarantee of receptor signaling (Assaf et al., 2009). Obviously, an influx of many inflammatory cells expressing TLR9 would increase the receptor burden in the lung and result in skewed or exaggerated subsequent inflammatory responses. The therapeutic intervention that will balance the number of cells expressing a particular receptor such as TLR9 may result in better outcomes.

Author contribution

D. Schneberger played a role in concept/design, data collection and analysis, and writing of manuscript. S. Caldwell aided in data collection. R. Kanthan helped in data/sample acquisition and approval of article. B. Singh helped in concept/design, analysis, writing/editing manuscript and approval of article.