There is growing evidence that neutrophils influence host resistance during influenza virus infection; however, factors that regulate neutrophil migration to the lung during viral infection are unclear. Activation of the aryl hydrocarbon receptor (AhR) by the pollutant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD or dioxin) results in an increased number of neutrophils in the lung after influenza virus infection. The mechanism of AhR-mediated neutrophilia does not involve elevated levels of soluble neutrophil chemoattractants, upregulated adhesion molecules on pulmonary neutrophils, delayed neutrophil apoptosis, or increased vascular damage. In this study, we determined whether AhR activation increases neutrophil numbers systemically or only in the infected lung, and whether AhR-regulated events within the hematopoietic system underlie the dioxin-induced increase in pulmonary neutrophils observed during influenza virus infection. We report here that AhR activation does not increase neutrophil numbers systemically or increase neutrophil production in hematopoietic tissue, suggesting that the elevated number of neutrophils is restricted to the site of antigen challenge. The generation of CD45.2AhR−/− → CD45.1AhR+/+ bone marrow chimeric mice demonstrates that even when hematopoietic cells lack the AhR, TCDD treatment still results in twice as many pulmonary neutrophils compared with control-treated, infected CD45.2AhR−/− → CD45.1AhR+/+ chimeric mice. This finding reveals that AhR-mediated events extrinsic to bone marrow–derived cells affect the directional migration of neutrophils to the infected lung. These results suggest that the lung contains important and heretofore overlooked targets of AhR regulation, unveiling a novel mechanism for controlling neutrophil recruitment to the infected lung.
Although vaccination and improved drug therapy have dramatically reduced mortality from infectious diseases, respiratory viral infections remain among the leading causes of mortality worldwide (World Health Organization, 2005). With the continued emergence of new viral strains, influenza viruses in particular pose a threat to human health and the global economy. The immune response to influenza virus relies on the activation of cells from the innate and adaptive arms of the immune system, leading ultimately to the creation of virus-specific CD8+ cytotoxic T lymphocytes (CTL), and antibodies (Gerhard, 2001; Woodland et al., 2001). During a primary infection, the generation of CTL and virus-specific antibodies takes 7–10 days, during which time cells of the innate immune system emigrate to the lung and presumably keep the infection at bay. In contrast to the well-characterized role of lymphocytes in antiviral immune responses, less is known about the precise role that cells of the innate immune system, and neutrophils in particular, play during respiratory viral infections.
It is becoming increasingly clear that the recruitment of neutrophils to the lung requires exquisite control. Evidence for this includes studies demonstrating that the accumulation of excess neutrophils is associated with host tissue damage and increased mortality (Fernandez et al., 2001; Patel et al., 1999; Shimizu et al., 1999; Teske et al., 2005). On the other hand, depletion of neutrophils can diminish survival following infection (Bliss et al., 2001; Sayles and Johnson, 1996; Stephens-Romero et al., 2005; Tumpey et al., 2005). Thus, there is growing evidence that the magnitude of neutrophil influx to the lung probably plays a very important role in the host's ability to survive viral infection. Therefore, it is important to understand what factors influence the differential recruitment of neutrophils to the lung upon infection with different subtypes of influenza A virus or among different individuals.
We have recently reported that activation of the aryl hydrocarbon receptor (AhR) markedly increases the number of neutrophils in lungs of mice infected with influenza A virus (Teske et al., 2005). This AhR-mediated increase in the number of neutrophils peaks on the seventh day of infection, was observed in both the airways and lung interstium, but was not observed in the absence of infection. Furthermore, by depleting neutrophils in vivo, we were able to improve the survival of infected mice, suggesting that AhR-mediated recruitment of excess neutrophils to the lung contributes to the decreased survival from influenza virus (Teske et al., 2005). The AhR is a member of the Per-Arnt-Sim family of transcriptional regulators and plays a role in xenobiotic metabolism and development (Gu et al., 2000). The AhR is activated by a diverse spectrum of ligands, including plant-derived natural compounds, tryptophan metabolites, and environmental contaminants (Denison and Nagy, 2003). Of the AhR agonists characterized thus far, the pollutant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD or “dioxin”) binds to it with the highest affinity, and is often used as a prototypical AhR agonist. In addition to TCDD, other pollutants that bind and activate the AhR include coplanar polychlorinated biphenyls and polyaromatic hydrocarbons (PAH), such as benzo[a]pyrene and 7,12-dimethylbenzanthracene, which are found in cigarette smoke and diesel exhaust (Behnisch et al., 2003; Denison and Nagy, 2003). In short, humans are exposed to AhR ligands daily through ingestion and inhalation (Charnley and Doull, 2005; Schecter et al., 2001). Moreover, diminished host resistance and altered immune function following exposure to PAH-containing pollutants correlates with an increased incidence of influenza and other respiratory infections (Burchiel and Luster, 2001; Sopori and Kozak, 1998). In rodents, AhR activation impairs survival following infection with influenza virus (Burleson et al., 1996; Luebke et al., 2002; Teske et al., 2005; Warren et al., 2000), further illustrating the relationship between exposure to AhR ligands and altered host resistance to infection.
The AhR is broadly expressed in mammalian tissues, and cells of the immune system, including neutrophils, have been reported to express it (Ackermann et al., 1989; Lang et al., 1998; Lawrence et al., 1996; Williams et al., 1996; Yamamoto et al., 2004,). AhR is also found in both the human and rodent lung (Dolwick et al., 1993; Lang et al., 1998; Thatcher et al., 2007; Yamamoto et al., 2004). Furthermore, in humans and rodents exposure to AhR agonists has been linked to enhanced pulmonary inflammation, including increased neutrophil influx to the lung (Diaz-Sanchez et al., 2000; Harrod et al., 2003; Luebke et al., 2002; Teske et al., 2005; Warren et al., 2000). However, the mechanism by which AhR activation enhances the directional migration of neutrophils has proved difficult to determine. The immune system is a very well known and sensitive target organ for the toxicity of dioxins and related compounds, and studies using AhR-deficient mice demonstrate that their toxicity is AhR dependent (Kerkvliet et al., 2002; Neff-LaFord et al., 2007; Teske et al., 2005; Vorderstrasse et al., 2001; and our unpublished observations). In particular, we have previously reported that increased neutrophilic inflammation in lungs of TCDD-treated mice infected with influenza virus is AhR dependent (Teske et al., 2005). Therefore, much effort to delineate the mechanism by which AhR activation enhances neutrophil recruitment has focused on an immune-mediated mechanism. However, neither the infection-induced increase in soluble neutrophil chemoattractants nor the upregulation of adhesion molecules on neutrophils is perturbed when the AhR is activated by TCDD (Teske et al., 2005). Likewise, the functional activity of neutrophils from infected, TCDD-treated mice was equivalent to neutrophils from vehicle control-treated mice (Teske et al., 2005).
In addition to affecting known soluble neutrophil chemoattractants, it is possible that the increase in the number of neutrophils results from nonspecific leakage of leukocytes from the blood into the lung. However, the following pieces of evidence do not support this idea: (1) the number of macrophages in lungs of infected mice is the same, regardless of TCDD treatment (Warren et al., 2000); (2) AhR activation by TCDD reduces the number of lymphocytes in lungs of infected mice (Mitchell and Lawrence, 2003; Warren et al., 2000); and (3) compared with infected controls, there is no change in the levels of protein or LDH in lavage fluid from infected mice treated with TCDD (Bohn et al., 2005). Collectively, these observations suggest that the directional migration of excess neutrophils to the lung during influenza virus infection is likely the result of AhR-dependent deregulation of a neutrophil-specific recruitment processes. To characterize how activation of the AhR deregulates neutrophil-specific processes, we first determined whether AhR activation enhances the number of neutrophils systemically or only at the site of antigen challenge (i.e., the lung). We then determined whether AhR-mediated events within or extrinsic to the immune system drive the recruitment of excess neutrophils to the lungs of TCDD-treated, infected mice. The findings from these studies provide novel information regarding how AhR activation alters neutrophil recruitment to the lung, and emphasizes that environmental exposure to AhR ligands may have a profound effect on disease outcome during infection with a common respiratory virus.
MATERIALS AND METHODS
Animals, TCDD treatment, and infection.
Female C57BL/6, B6-Ly5.2/Cr congenic, and B6.129S7-Ifngtm1Ts/J (GKO) mice were purchased from The Jackson Laboratory (Bar Harbor, ME) or NCI-Frederick (Frederick, MD) at 4–6 weeks of age. A breeding colony of C57BL/6-AhRtm1Bra (AhR−/−) mice was maintained as previously described (Teske et al., 2005). All mice were provided with water and food at libitum and housed under pathogen-free conditions in microisolator cages (two to five mice per cage). All animal treatment was performed in accordance with protocols approved by the Institutional Animal Care and Use Committee.
TCDD (≥ 99% pure, Cambridge Isotope Laboratories, Inc., Woburn, MA) was dissolved in anisole and diluted in peanut oil to a concentration of 1 μg/ml. Mice were gavaged with a single dose of TCDD (10 μg/kg body weight). Vehicle–control mice were gavaged with a single dose of peanut oil-anisole vehicle. Mice were infected intranasally (i.n.) with 120 hemagglutinating units (HAU) of murine-adapted human influenza virus, A/HKX31 (X31; H3N2), or 107 plaque forming units (PFU) of influenza virus, A/Memphis/102/72 (Mem-72; H3N2; 107 PFU of Mem-72 are equivalent to 125 HAU), in 25 μl of endotoxin-tested phosphate buffered saline (PBS). Vehicle-treated C57BL/6, B6-Ly5.2/Cr, and AhR−/− mice typically survive infection with 120 HAU X31 or 107 PFU Mem-72 (Lawrence et al., 2006; Mitchell and Lawrence, 2003; Warren et al., 2000). Vehicle- and TCDD-treated, mock-infected mice were used as controls for the infection, and received 25 μl of endotoxin-tested PBS (i.n.). Data from mock-infected mice are defined as the zero time point (i.e., day 0). Infection was performed under general anesthesia (2,2,2-tribromoethanol, Sigma-Aldrich, St Louis, MO).
Collection of immune cells.
To obtain airway-derived immune cells, lungs were lavaged with cold serum-free RPMI 1640 media or PBS (Teske et al., 2005; Warren et al., 2000). Airway-derived immune cells were separated from bronchoalveolar lavage (BAL) fluid by centrifugation and enumerated using a Coulter counter (Beckman Coulter Corp., Miami, FL). Lung-derived immune cells were obtained by digesting lungs with collagenase as previously described (Teske et al., 2005). Immune cells were separated from erthrocytes and parenchymal lung cells using Lympholyte-M media (Cedarlane Laboratories, Hornby, Ontario, Canada). Bone marrow cells were collected from femurs of both hind legs as previously described (Vorderstrasse et al., 2004). Single-cell suspensions of spleen and mediastinal lymph node (MLN) cells were prepared by pressing each organ between the frosted ends of two microscope slides (Warren et al., 2000). Spleen and MLN cells from individual animals were resuspended in cold RPMI 1640 containing 2.5% fetal bovine serum and 10mM 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid or N-(2-hydroxyethyl)piperazine-N′-(2-ethanesulfonic acid) (HEPES). Erythrocytes were removed from marrow, spleen and MLN cells by hypotonic lysis and cells were counted using a Coulter counter. In some experiments, cellular viability was assessed by trypan blue dye exclusion, and was generally ≥ 90%.
Neutrophil chemotaxis assay.
Migration of peritoneal neutrophils was assessed using a fluorimetric 96-well trans-well migration kit (3 μm pore size; Chemicon Intl., Temecula, CA). The 3 μm pore size allows for migration of neutrophils but not macrophages (Cataisson et al., 2005). To obtain neutrophils, untreated and uninfected mice were injected (ip) with 0.5 ml of Brewer's medium (VWR Scientific, Arlington Heights, IL), and 3 h after injection peritoneal exudates were collected by lavage. At this point in time, about 50% of the cells in the peritoneal cavity are neutrophils, as defined using hematoxylin and eosin staining (data not shown). Peritoneal lavage cells (2 × 105) containing approximately 50% neutrophils (determined by differential cell counts) were resuspended in 100 μl serum-free RPMI 1640 media, containing 1% bovine serum albumin and 10mM HEPES and placed in the upper compartment of the transwell assay. BAL fluid (150 μl) derived from vehicle- and TCDD-treated, infected (X31) mice or from vehicle- and TCDD-treated mock-infected mice was placed into the lower compartment. After 45 min at 37°C, 5% CO2, the migration of peritoneal neutrophils from the upper to the lower compartments was determined using a fluorescence plate reader. To determine the number of neutrophils, a standard curve was generated by incubating known numbers (0–140,000) of peritoneal neutrophils with the fluorescent CyQuant GR dye provided with the kit. Results were validated using duplicate wells in which extravasated cells were counted with a hemacytometer, cytospun and identified using hematoxylin and eosin staining (data not shown).
Livers from vehicle- and TCDD-exposed mice were excised, fixed in 10% formalin, and embedded into paraffin, and 5-μm tissue slices were prepared. The tissue slices were mounted and stained with hematoxylin and eosin at the Washington Animal Disease Diagnostic Laboratory (Washington State University). Slides were blinded with regard to treatment and evaluated by a veterinary pathologist (A.A.B.).
Peripheral white blood cell analyses.
Mice were sacrificed by overdose with Avertin, and blood was collected by cardiac puncture into ethylenediaminetetracaetic acid–coated syringes. Complete blood counts were performed by the Veterinary Medicine Clinical Pathology Laboratory (College of Veterinary Medicine, Washington State University). Total white blood cell counts were determined using a Sereno-Baker, System 9010+CP, hematology analyzer (ABX Diagnostics, Irvine, CA). Blood smears were stained with Wright-Geimsa stain, and differential cell counts were performed manually by counting a total of 100 cells per slide without knowledge of treatment group.
Differential analysis of BAL cells.
Using a cytological centrifuge, lavage cells (1 × 104) were spun onto a microscope slide, fixed, and stained with hematoxylin and eosin (LeukoStat; Fisher Scientific, Pittsburgh, PA). The number of macrophages, neutrophils, and lymphocytes were enumerated manually by differential counts of 200 cells per slide. Differential cell counts were performed on blinded slides.
Cells were stained with different combinations of the following fluorescent conjugated monoclonal antibodies: fluorescent isothiocyanate (FITC)–labeled anti-Gr-1 or PE-labeled anti-Gr-1; FITC-labeled anti-CD45.2; PE-labeled anti-CD8 or APC-labeled anti-CD8; TC-labeled anti-CD44; APC-labeled anti-CD62L; PE-labeled anti-CD45.1 or TC-labeled anti-CD45.1. To validate that Gr-1+ cells were neutrophils, we enriched Gr-1+ cells from total lung-derived immune cells using immunomagnetic beads (Miltenyl Biotec, Auburn, CA), stained the cells with hematoxylin and eosin, and examined the cells by light microscopy. The cells were morphologically neutrophils, as evidenced by their multilobular nuclei and hematoxylin/eosin staining pattern (data not shown). To identify influenza virus-specific CD8+ T cells in C57BL/6 mice, MLN cells were incubated with PE-labeled DbNP366-374 tetrameric complexes (Beckman Coulter Inc., San Diego, CA). To assess nonspecific fluorescence, appropriately labeled, isotype-matched antibodies were used as negative controls. Antibodies and isotypic controls were purchased from eBioscience (San Diego, CA) and BD Biosciences (San Jose, CA). Data from 50,000 to 100,000 cells were collected by listmode acquisition using a FACSort flow cytometer (Becton Dickenson, San Jose, CA). WinList software (Verity Software, Topsham, ME) was used for data analyses.
Generation of bone marrow chimeric mice.
Four-week-old, female C57BL/6 mice (CD45.2+ phenotype) and B6-Ly5.2/Cr congenic mice (CD45.1+ phenotype) were purchased from NCI-Frederick (MD). The mice received sterile-filtered, acidified water (pH 3.0) supplemented with 1 mg/ml oxytetracycline HCl (terramycin) and were fed irradiated food (Pico-Vac mouse diet 20, Purina Mills, St Louis, MO) for the duration of the study. To ablate the endogenous immune system, anesthetized (Suprane; deflurane, Baxter Healthcare Corp., Deerfield, IL) CD45.1+AhR+/+ mice were irradiated with a cumulative dose of 1200 rad, given as two separate doses of 600 rad 3.5 h apart (Philips SL-15 linear accelerator, Radiology Section, College of Veterinary Medicine, Washington State University). One hour after the second dose of irradiation, 1.5 × 106 bone marrow cells from either CD45.2+AhR+/+ or CD45.2+AhR−/− donor C57BL/6 mice were injected into the tail vein of the irradiated CD45.1+AhR+/+ recipient mice. Lethally irradiated CD45.1+AhR+/+ mice that did not receive bone marrow cells from donor mice served as controls for the irradiation. These mice died within 10 days of irradiation (data not shown). Five weeks after irradiation, CD45.2AhR+/+ → CD45.1AhR+/+ and CD45.2AhR−/− → CD45.1AhR+/+ chimeras were treated with peanut oil vehicle or TCDD (10 μg/kg body weight) one day prior to infection with 120 HAU X31.
All statistical analyses were conducted using StatView statistical software (SAS, Cary, NC). Using one-way analysis of variance, followed by a Fisher's protected least significant difference post hoc test, mean values from each treatment group were compared at a specific point in time and over time. Where applicable, an unpaired two-sided t-test was used to compare mean values from each treatment group at a specific point in time. Values of p ≤ 0.05 serve as the basis for designation of statistical significance for all studies.
An Equivalent Number of Neutrophils Migrate toward Lung Lavage Fluid from TCDD-Treated, Infected Mice and Vehicle-Treated, Infected Mice
We have previously examined the role of numerous neutrophil chemoattractants, including tumor necrosis factor-alpha, interleukin-1 (IL-1), IL-6, keratinocyte chemoattractant, macrophage inflammatory protein-(MIP)-1α, MIP-2, lipopolysaccharide-induced CXC chemokine, granulocyte-macrophage colony stimulating factor, and complement split product C5a in the lungs of infected mice (Neff-LaFord et al., 2003; Teske et al., 2005). Although levels of these chemoattractants increase with influenza virus infection, AhR activation does not further enhance the infection-associated increase. However, a cytokine that is enhanced in lungs of infected, TCDD-treated mice is interferon-γ (IFN-γ) (Neff-LaFord et al., 2007; Warren et al., 2000). IFN-γ is a proinflammatory cytokine that has been reported to be chemoattractive for neutrophils in a murine models of hyperoxia-mediated lung damage and infection with the parasite Litomosoides sigmodontis (Saeftel et al., 2001; Yamada et al., 2004). Therefore, using IFN-γ–deficient (GKO) mice, we determined whether elevated pulmonary levels of IFN-γ enhance recruitment of neutrophils to the lungs of TCDD-treated, infected mice. As shown in Figure 1, the number of neutrophils in the lung was elevated in TCDD-exposed wild-type and GKO mice infected with influenza virus compared with vehicle-treated infected controls. Although GKO mice may produce a different profile of cytokines during infection with influenza virus, this finding indicates AhR-mediated increases in pulmonary IFN-γ levels do not drive the recruitment of excess number of neutrophils to the lungs. In fact, the neutrophils recruited to the lung are probably the major source of the increased pulmonary IFN-γ (Neff-LaFord et al., 2007).
We next sought to determine whether an AhR-mediated deregulation of a novel neutrophil chemoattractant underlies the recruitment of excess neutrophils to the lung. To test this, we compared the migration of peritoneal neutrophils from untreated mice toward BAL fluid derived from vehicle- and TCDD-exposed, infected mice. Compared with migration of untreated neutrophils toward BAL fluid from vehicle- and TCDD-treated, mock-infected mice, fourfold more neutrophils migrated toward BAL fluid from vehicle- and TCDD-treated, infected mice (Fig. 2). However, treatment with TCDD did not further enhance this strong infection-associated increase in neutrophil chemotaxis. The data from this migration study in combination with our other neutrophil chemoattractant data strongly suggest that elevated levels of soluble, stable neutrophil chemoattractants in the lungs of TCDD-treated, infected mice are not a likely mechanistic explanation for the excessive number of neutrophils in mice treated with TCDD.
AhR-Driven Recruitment of Excess Neutrophils during Influenza Infection is Limited to the Lungs
We have previously shown that activation of AhR does not upregulate expression of adhesion molecules on pulmonary neutrophils, delay pulmonary neutrophil apoptosis, or enhance vascular damage in the lungs of infected mice (Bohn et al., 2005; Teske et al., 2005). In the absence of an effect of AhR regulation on these mechanisms, an AhR-mediated increase in the production or overall number of circulating neutrophils provides a potential alternative mechanistic explanation for the enhanced number of neutrophils in the lungs of TCDD-treated, infected mice. To examine whether activation of the AhR increases neutrophil numbers systemically, we determined the percentage and number of neutrophils in the bone marrow and peripheral blood over the course of influenza infection. Although infection with influenza virus generally elevated the percentage and number of neutrophils in the bone marrow (Figs. 3A and 3B) and blood (Figs. 3C and 3D), AhR activation did not further increase the percentage or number of neutrophils relative to the vehicle control group. Another indicator of enhanced neutrophil production is an increase in band cells (immature neutrophils). Exposure to TCDD did not alter the percentage or number of band cells in the blood of infected mice (data not shown). The absence of an enhanced number of neutrophils in the blood or bone marrow suggests that activation of AhR does not directly increase neutrophil production.
To further investigate possible systemic effects on neutrophil numbers, we assessed whether treatment with TCDD increases the number of neutrophils at other anatomical sites during respiratory viral infection. To do so, we examined the number of neutrophils in the spleen and liver, because influenza virus does not infect or replicate in these organs. As shown in Figures 4A and 4B, exposure to TCDD did not elevate the percentage or number of neutrophils in the spleen of infected mice. With regard to the liver, there was no evidence of generalized hepatitis in any of the treatment groups. Livers from all treatment groups contained only a low number of hematopoietic foci as well as a few small, discrete foci containing hepatocellular necrosis and inflammation (Figs. 4C and 4F). The latter foci consisted predominantly of lymphocytic inflammation and were seen in slightly higher numbers in the TCDD-treated, mock-infected treatment group. Based on these observations it does not appear that AhR activation significantly elevates the number of neutrophils in the liver. In summary, these results show, that in the context of influenza virus infection, AhR activation does not enhance the number of neutrophils in bone marrow, blood, spleen, or liver, suggesting that the excess number of pulmonary neutrophils in TCDD-treated, infected mice is restricted to the site of antigen challenge (i.e., the lung).
Establishment and Validation of Bone Marrow Chimeric Mice
Given that the lung expresses high levels of the AhR compared with immune cells (Dolwick et al., 1993; Lang et al., 1998; Yamamoto et al., 2004), alterations in AhR-regulated events within the lung are a logical mechanism for the recruitment of excess neutrophils to the lungs of TCDD-treated, infected mice. To determine whether AhR-driven events external to the immune system underlie the elevated number of neutrophils in the lungs of TCDD-treated mice, we generated bone marrow chimeric mice in which immune cells lack the AhR, but all other tissues express a functional AhR. To accomplish this, we reconstituted lethally irradiated AhR+/+ recipient mice (CD45.1+ phenotype) with hematopoietic cells (CD45.2+ phenotype) from either wild-type (AhR+/+) or AhR-deficient (AhR−/−) congenic donor mice, generating CD45.2AhR+/+ → CD45.1AhR+/+ and CD45.2AhR−/− → CD45.1AhR+/+ chimeric mice. Five weeks after irradiation, we validated the success of reconstitution of the hematopoietic system in these chimeric mice. As shown in Figure 5A, greater than 90% of the hematpoietic cells were of donor origin in both CD45.2AhR+/+ → CD45.1AhR+/+ and CD45.2AhR−/− → CD45.1AhR+/+ chimeras, indicating that chimerism was successfully established regardless of AhR status of the donor immune system. In addition to establishing chimerism, we determined that the bone marrow chimeric mice were immunocompetent. To do so, we monitored survival of CD45.2AhR+/+ → CD45.1AhR+/+ and CD45.2AhR−/− → CD45.1AhR+/+ chimeric mice for 14 days following infection with influenza virus X31. As illustrated in Figure 5B, regardless of the AhR status of immune cells, all chimeric mice survived this infection, demonstrating that we successfully generated immunocompetent bone marrow chimeric mice.
Increased Number of Neutrophils in TCDD-Treated, Infected Mice is Dependent on AhR-Driven Events at the Site of Antigen Challenge, but not Directly within Bone Marrow–Derived Cells
Using chimeric mice, we determined whether AhR activation increases the number of neutrophils in the lungs when immune cells lack the AhR. As expected, compared with vehicle-treated, infected chimeric mice, there was a twofold increase in the number of neutrophils in the lungs of TCDD-exposed, infected mice in which immune cells express a functional AhR (Fig. 6A). Interestingly, we also detected two times more CD45.2AhR−/− (donor-derived) neutrophils in the lungs of TCDD-exposed, infected CD45.2AhR−/− → CD45.1AhR+/+ chimeras (Fig. 6B). That is, even when the immune cells lack the AhR, exposure to TCDD still results in excess recruitment of neutrophils to the infected lung. This novel finding strongly suggests that the enhanced number of neutrophils in lungs of TCDD-treated mice is mediated by alterations in AhR-regulated events that are external to the immune system.
It has been previously shown that AhR activation suppresses T cell–dependent immune responses to a variety of antigens (see Lawrence and Kerkvliet, 2006 for review), including the proliferation and differentiation of CD8+ T cells in influenza virus–infected mice (Lawrence et al., 2006; Mitchell and Lawrence, 2003). Moreover, the suppressive effect of TCDD on T cell responses is AhR dependent, and appears to require the AhR within T cells (Kerkvliet et al., 2002; Lawrence et al., 2006; Vorderstrasse et al., 2001). Therefore, our observation that the recruitment of excess neutrophils to the lung does not require the AhR in the hematopoietic system suggests that AhR-mediated deregulation of neutrophil responses may differ from suppressed T cell responses, which we predict would not be sensitive to TCDD-induced suppression when the hematopoietic system lacks the AhR. To formally test this, we determined the number of CD8+ T cells in the lung and the number of virus-specific (DbNP366-374+) CD8+ T cells in the MLN of the same group of infected chimeric mice. When bone marrow–derived cells are AhR+/+, we observed a twofold suppression in the number of CD8+ T cells in the lungs and a 10-fold decrease in virus-specific CD8+ T cells in the MLN of TCDD-treated mice (Fig. 7). However, when hematopoietic cells lack the AhR, exposure to TCDD failed to suppress the expansion of CD8+ T cells in the draining lymph node and their migration to the lungs of infected mice. Therefore, although AhR-dependent events within bone marrow–derived cells are required for the suppression of the CD8+ T cell response, AhR-mediated events extrinsic to hematopoietic cells underlie the recruitment of excess neutrophils to the lung.
We present here for the first time that the mechanism of AhR-mediated alterations in immune function differs between innate and adaptive immunity. AhR-driven events directly within hematopoietic cells underlie suppression of the CD8+ T cell response to infection with influenza virus (Lawrence et al., 2006). In contrast, AhR-driven events external to bone marrow–derived cells mediate the recruitment of excess neutrophils to the lungs of TCDD-treated, infected mice. Furthermore, in the context of respiratory viral infection, we did not detect an increased number of neutrophils in the bone marrow or blood of mice exposed to TCDD. Likewise, AhR activation did not enhance the number of neutrophils in the spleen or liver, two anatomical sites in which influenza virus does not infect or replicate. Taken together, these findings suggest that AhR-driven events at the site of antigen challenge (i.e., the lung) mediate the excessive number of neutrophils in the lungs of mice infected with influenza virus.
Possible AhR-driven events within the lung that could affect the directional migration of neutrophils include alterations in either soluble or cell-associated mediators. Collectively, the past findings from our laboratory in conjunction with the results from the migration assay presented here indicate that the recruitment of excess neutrophils to the lungs of TCDD-exposed mice is mediated by either non-soluble or extremely labile soluble factors. With regard to possible labile, soluble factors, arachidonic acid metabolites, such as leukotriene-(LTB)4 and prostaglandin-(PG)E2, are chemoattractive for neutrophils in models of lung inflammation (Alba-Loureiro et al., 2004; Cuzzocrea et al., 2003). In the context of influenza virus infection, LTB4 has been shown to be elevated in the lungs of influenza virus–infected mice (Hennet et al., 1992). Likewise, LTC4 was increased in lungs of humans infected with influenza virus (Gentile et al., 2003). However, activation of the AhR did not alter levels of LTB4, LTC4, and PGE2 in several different experimental systems (Lawrence and Kerkvliet, 1998; Lee et al., 1998), suggesting that production of these soluble mediators is not affected by AhR activation either. Instead, the accumulated data suggest that deregulation of cell-associated molecules provides a more plausible mechanistic explanation for the enhanced number of neutrophils in the lungs of TCDD-exposed mice. Within the pulmonary vasculature and structural cells of the lung, there are numerous cell-associated factors important for neutrophil recruitment. These include matrix metalloproteases and their inhibitors, toll-like receptors, and cellular adhesion molecules (Burns et al., 2003; Wagner and Roth, 2000). Potential modulation of these factors within the lung by AhR has not been well studied.
In summary, the findings from this study illustrate the complexity of the mechanism by which AhR activation influences the pulmonary immune response to viral infection. In contrast to the suppression of CD8+ T cell responses, which appear to result from AhR activation within the immune system, AhR-driven enhancement of neutrophil recruitment does not. Furthermore, the influx of neutrophils to the infected lung is not simply due to elevated neutrophil numbers systemically, or to enhanced pulmonary vascular permeability (Bohn et al., 2005), nor is it governed by an increase in soluble neutrophil chemoattractants in the lung (Teske et al., 2005). Instead, we show for the first time that the excess neutrophilia is caused by AhR-driven events extrinsic to the immune system, suggesting that AhR-mediated events within the lung influence neutrophil recruitment, and thereby alter the outcome from respiratory viral infection. Importantly, this consequence of AhR activation is not unique to a single subtype of influenza A virus or a single mouse strain. We have observed enhanced neutrophilia in C57BL/6, C57BL/10, C3H/HeNCr, and C3H/HeJCr mice infected with two different strains of influenza A virus (i.e., A/HKX31 and A/Memphis/102/72; data not shown).
In addition to understanding how AhR activation affects the immune response to influenza virus, this discovery is significant in a broader sense because it suggests that the lung is a very important and overlooked target for AhR ligands. There is mounting evidence that environmental exposure to AhR agonists adversely affects human health. For example, poorer health in disadvantaged communities, increased asthma in urban areas, cancer, and distinct clinical outcomes from individuals with the same type of infection all point to links between environmental exposures and the etiology or differential severity of numerous disease states. Therefore, the ability of AhR to modulate the directional migration of neutrophils to the lung suggests that exposure to ubiquitous AhR ligands may contribute to the etiology of these diseases. It also suggests that as we improve our understanding of the molecular mechanism, modulation of AhR function and AhR-dependent processes serve as potential therapeutic targets to regulate neutrophil recruitment and the pathology associated with disease.
National Institutes of Health grants (R01-ES10619, K02-ES012409 to B.P.L.) and grant T32-GM008336). J.P.H is the recipient of a Seattle Chapter ARCS Scholarship.
We would like to thank and gratefully acknowledge Ms Jennifer Cundiff and Mr Kevin Kipp for their excellent technical assistance with these studies. We thank Dr Allen Silverstone (SUNY Upstate Medical University, Syracuse, NY) and Dr David Shepherd (University of Montana, Missoula, MT) for helpful advice with the bone marrow chimeras, Drs Douglas Call and James Krueger (Washington State University) for their assistance with the chemotaxis assay, and Dr Janean Fidel and Mr. Robert Houston (Washington State University) for irradiating the mice. Finally, we thank Drs Beth Vorderstrasse (Washington State University) and Haley Neff-LaFord (University of Washington) for their helpful discussion of our data and critical review of this manuscript.