Comments
Abstract
Chronic perivascular inflammation is a prominent feature in the lungs of idiopathic pulmonary arterial hypertension. Although the proportions of conventional dendritic cells (cDCs) and plasmacytoid DCs are increased in idiopathic pulmonary arterial hypertension lungs, it remains unknown whether activated cDCs play a pathogenic role. The Tnfaip3 gene encodes the ubiquitin-binding protein A20, which is a negative regulator of NF-κB, critically involved in DC activation. Targeting of Tnfaip3/A20 in cDCs was achieved by Clec9a (DNGR1)-Cre–mediated excision of the Tnfaip3 gene in Tnfaip3DNGR1-KO mice. Mice were evaluated for signs of pulmonary hypertension (PH) using right heart catheterization, echocardiography, and measurement of the Fulton index. Inflammation was assessed by immunohistochemistry and flow cytometry. Pulmonary cDCs and monocyte-derived DCs from 31-week-old Tnfaip3DNGR1-KO mice showed modulated expression of cell surface activation markers compared with Tnfaip3DNGR1-WT mice. Tnfaip3DNGR1-KO mice developed elevated right ventricular systolic pressure and right ventricular hypertrophy. The lungs of these mice displayed increased vascular remodeling and perivascular and peribronchial immune cell infiltration resembling tertiary lymphoid organs. Proportions of activated T cells and expression of IL-1β, IL-6, and IL-10 were enhanced in the lungs of Tnfaip3DNGR1-KO mice. Autoreactive IgA and IgG1 was detected in BAL and autoreactive IgA recognizing pulmonary endothelial antigens was present in the serum of Tnfaip3DNGR1-KO mice. All signs of PH were ameliorated in Tnfaip3DNGR1-KO mice by anti–IL-6 antibody treatment. These results indicate that activation of the NF-κB pathway in DCs, through deletion of A20/Tnfaip3, leads to experimental PH with accompanied pulmonary inflammation in an IL-6–dependent fashion.
In our manuscript, a novel mouse model is presented. The Tnfaip3DNGR1-KO mice show enhanced activation of the NF-κB pathway in dendritic cells by selective knockout of the Tnfaip3 gene, which encodes for ubiquitin-binding protein A20. To our knowledge, this is the first study that shows that enhanced dendritic cell activation is sufficient for the development of pulmonary hypertension in an IL-6–dependent fashion. We believe this study will help to further elaborate and understand the possible role for dendritic cells in the pathophysiology of pulmonary hypertension.
Pulmonary arterial hypertension (PAH) is a debilitating disease in which increased pulmonary vascular pressure leads to right ventricular (RV) hypertrophy, heart failure, and eventually death (1). An increasing body of evidence supports dysregulated immunity in the pathobiology of PAH (2–4). Patients with idiopathic PAH (IPAH) showed enhanced expression of circulating inflammatory cytokines, such as IL-6, IL-8, IL-10, and IL-12p70, which was found to correlate with survival (5, 6). In patients with IPAH, CD11c+ dendritic cell (DC) numbers are enhanced in the lung, accumulating around remodeled pulmonary arteries (7). In the parenchyma, DCs present with an immature phenotype, indicated by an increased number of immature DC-SIGN+ DCs (7). The numbers of both conventional DCs (cDCs) and plasmacytoid DCs (pDCs) are augmented in total lung cell suspensions of patients with IPAH compared with controls (8). In contrast, in peripheral blood, cDC numbers are decreased in patients with IPAH (9), suggesting migration to the lungs. Finally, lungs of patients with IPAH often display perivascular lymphocytic infiltration and formation of tertiary lymphoid organs (TLOs) in comparison with healthy controls (10). These TLOs consist of T cells, B cells, mast cells, and DCs (10, 11).
DCs are crucial for the induction and maintenance of lung TLO formation during pulmonary infection in mice (12). DCs in TLOs often show an activated phenotype, suggesting that they exert one of their main functions at this site, that is, antigen presentation and activation of T cells (13, 14). DCs can be divided into cDCs, monocyte-derived DCs (mo-DCs), and pDCs. cDCs can be subdivided into type 1 cDCs (cDC1s), which are efficient at presentation of exogenous self-antigens in major histocompatibility complex class I (MHC-I) and can induce CD8+ T-cell responses, and type 2 cDCs (cDC2s), which are capable of eliciting T-helper cell type 2 (Th2) and Th17-cell responses. Upon inflammation, mo-DCs arise and are able to produce large amounts of chemokines, leading to attraction of T cells to the site of inflammation (15). Finally, pDCs are best known for producing large amounts of type I IFNs during viral infections.
By exerting these functions, DCs play a pivotal role in the balance between immunity and tolerance. DC triggering by microbial products via Toll-like receptors or cytokines leads to activation of the NF-κB pathway. Upon translocation to the nucleus, NF-κB is responsible for the transcription of numerous proinflammatory and cell-survival genes. Following stimulation, DCs upregulate costimulatory molecules, produce various inflammatory cytokines such as IL-6, and drive T-cell priming and effector differentiation (16–18). A breach of tolerance through inappropriate DC activation can lead to several autoimmune diseases (17). Especially cDCs are involved in the control of tolerance, which is less evident for pDCs and moDCs.
Because DC activation is a critical step in immune modulation, their activation must be tightly controlled. One of several mechanisms for this control involves the key regulatory ubiquitin-binding protein A20, encoded by the TNFAIP3 gene. Several studies have shown a crucial role for TNFAIP3/A20 in controlling inflammatory disease in vivo (19–22). Polymorphisms in the TNFAIP3 locus are significantly associated with several autoimmune conditions, including pulmonary hypertension (PH) (23). Strikingly, a specific SNP in the TNFAIP3 locus has been associated with PAH development in patients with systemic sclerosis (24).
Currently, it has not been elucidated whether defects in DC subsets contribute to the pathogenesis of PAH. Considering that cDCs are increased in the lungs of IPAH, our aim was to investigate whether constitutive activation of cDCs in mice would result in the development of PH. To this end, we used the Tnfaip3DNGR1-KO mouse model with a Clec9a/DNGR1-Cre–mediated deletion of the Tnfaip3 gene, which specifically targets cDC1s, and to a lesser extent cDC2s and mo-DCs, but not pDCs (25). Previous work from our group indicated that in these mice, the Tnfaip3 gene was deleted in ∼95% of the cDC1s, ∼35% of the cDC2s, and ∼45% of the mo-DCs present in the liver, based on yellow fluorescent protein (YFP) expression as a tracer of DNGR1-Cre–driven gene deletion (22). At the age of ∼30 weeks, these mice generally showed chronic liver inflammatory infiltrates surrounding the portal triads that was often associated with development of IgA autoantibodies recognizing liver periportal antigens.
In the current study, we found that Tnfaip3DNGR1-KO mice developed experimental PH over time, characterized by increased RV systolic pressure, RV hypertrophy, extensive lymphocytic infiltration, vascular remodeling, and increased cytokine (IL-6) production in the lung. Blockade of IL-6 in Tnfaip3DNGR1-KO mice ameliorated the experimental PH phenotype to wild-type (WT) level, indicating a crucial role for IL-6 in DC-driven PH pathophysiology.
Male and female C57Bl/6 mice harboring a conditional Tnfaip3 allele flanked by LoxP sites (26) were crossed to a transgenic line expressing the Cre recombinase under the control of the promotor region of the C-type lectin domain family 9a (Clec9a) gene, which encodes DNGR1 (the DC NK lectin group receptor 1) (25), generating Tnfaip3fl/flxClec9a+/Cre mice (Tnfaip3DNGR1-KO mice), as previously described (22). Tnfaip3fl/flxClec9a+/+ littermates (Tnfaip3DNGR1-WT mice) served as controls. All mice were killed between 24 and 31 weeks of age.
To identify cells that have undergone DNGR1-Cre–mediated recombination, Tnfaip3+/+xClec9a+/Cre mice and Tnfaip3fl/flxClec9a+/Cre mice were crossed to Rosa26-Stopfl/fl-YFP mice (27), yielding Tnfaip3DNGR1-ROSA-WT and Tnfaip3DNGR1-ROSA-KO mice, respectively. Rag1−/− mice (28) were on the C57Bl/6 background. Mice were housed under specific pathogen-free conditions and had ad libitum access to food and water. All experiments were approved by the animal ethical committee of the Erasmus MC.
Mice were anesthetized using urethane, which produced anesthesia with spontaneous breathing. Subsequently, thorax hairs were removed by hair removal cream for optimal ultrasound imaging. Using a microscan transducer (VEVO2100; Visual Sonics), RV wall thickness, pulmonary arterial acceleration time (PAAT), stroke volume, and cardiac output were measured as described (29).
For evaluation of airway resistance (Rn), tissue damping (G), and elastance coefficient (H), mice underwent measurements on a small-animal ventilator (flexiVent; EMKA, SCIREQ Inc.) after placement of a tracheal canule after sedation through urethane. To prevent spontaneous breathing, mice received curare intraperitoneally (30).
In anesthetized mice, a tracheal cannula was placed for ventilation (miniVent type 845; Hugo Sachs Elektronik). Then a midline sternal incision was made, and the RV of the heart was punctured at the apex using a small-gauge needle. A pressure catheter (Miller Inc.) was positioned in the RV of the heart for RV systolic pressure (RVSP) and RV end-diastolic pressure (RVeDP) measurement (31). Pressures were recorded and analyzed using WinDaq (DataQ instruments) and Matlab (the Mathworks).
Following excision of the heart, the RV free wall was separated from the left ventricle (LV) and the septum (s). The RV and the LV+s were weighed separately (29).
For RT-PCR analysis, the postcaval lung lobe and the heart (after separation of RV and LV/S) were stored in −80°C until processing of the material. Lung and heart tissue were homogenized using lysis buffer with 2-mercaptoethanol and 1/4’’ ceramic spheres (6540-034; MP Biomedicals) by shaking for 40 seconds (Fastprep-24 5G; MP Biomedicals). Subsequently, RNA was extracted using TRI reagent (T9424; Sigma) and cDNA synthesis was performed using a RevertAid H minus First Strand cDNA synthesis Kit (K1632; Thermoscientific). The cDNA was used for measuring IL-1β, IL-6, IL-10, TGFβ (transforming growth factor β), FGF2, VEGF and HGF expression by RT-PCR in a 7300 Real-time PCR system (Applied Biosystems). For control, a housekeeping gene was also measured as an internal control, to which the relative mRNA expression was determined. Primers used for RT-PCR are given in Table E1 in the data supplement.
The left lung was either frozen in Tissue-TEK O.C.T. (VWR International) solution and kept at −80°C until further processing into cryosections or fixed with 4% paraformaldehyde (Carl Roth) before paraffin embedding. For immunohistochemical stainings, 6-μm-thick cryosections were fixed in acetone. Antigen retrieval on paraffin sections were established using citrate buffer (Sigma Aldrich). Five-micrometer-thick paraffin-embedded lung sections were stained with hematoxylin and eosin or Elastica von Giessen.
To assess pulmonary vascular remodeling, cryosections were stained for ACTA2 and CD31. Using Adobe Photoshop, the ratio between αSMA-positive and total pulmonary artery diameter was calculated. This was performed for a minimal of five arteries per mouse for small-sized pulmonary arteries (20–50 μm).
To visualize immune cells, cryosections were stained for CD138, IgD, IgM, GL7, Ki67, IgA, CD3 and B220, using standard procedures. The primary antibodies and their dilutions, used for immunohistochemistry, are listed in Table E2. Sections were incubated for 1 hour with the primary antibodies. After washing, slides were incubated for 30 minutes with secondary antibodies (Table E3). On paraffin sections, which were stained for CD3, the anti-Rabbit ABC Peroxidase Kit was used (Vector Labs). Diaminobenzene and Fast Blue Alkaline phosphatase substrates were used to retrieve specific staining.
BAL, mediastinal lymph nodes, and lungs were obtained and used for flow cytometry. BAL was centrifuged at 400 × g for 3 minutes, after which the supernatant was stored at −80°C for further analysis and the single cells were used for flow cytometry. Mediastinal lymph nodes were homogenized through a 100-μm cell strainer. Lung single-cell suspensions were obtained by digesting with Liberase (Roche) for 30 minutes at 37°C. After digestion, the lungs were homogenized using a 100-μm cell strainer (Fischer Scientific). Finally, erythroid cells were lysed using osmotic lysis buffer.
Flow cytometry surface and intracellular staining procedures have been described previously (32). Monoclonal antibodies used for flow cytometric analyses are listed in Table E3. For all experiments, dead cells were excluded using fixable AmCyan viability dye (eBioscience). To measure cytokine production, cells were stimulated with 10 ng/ml PMA (Sigma-Aldrich) and 250 ng/ml ionomycin (Sigma-Aldrich) in the presence of GolgiStop (BD Biosciences) for 4 hours at 37°C. Data were acquired using an LSR II flow cytometer (BD Biosciences) with FACS Diva software with a minimal of 100 cells per gate of interest and analyzed by FlowJo version 9 (Tree Star Inc software).
For quantification of total Ig levels in BAL fluid and serum, Nunc Microwell plates (Life Technologies) were coated with 1 μg/ml goat-anti-mouse IgM, IgA, or an IgG isotype (IgG1, IgG2a, IgG2b, IgG3) (Southern Biotech) overnight at 4°C. Wells were blocked with 10% FCS (Capricorn Scientific) in PBS (Thermo Scientific) for 1 hour. Standards, BAL, and serum were diluted in PBS and incubated for 3 hours at room temperature. Depending on the isotype to be measured, anti-mouse biotin–labeled IgM, IgA, or IgG isotype (Southern Biotech) was incubated for 1 hour. Streptavidin-HRP (eBioscience) and Tetramethylbenzidine substrate (eBioscience) was used to develop the ELISA, and optical density was measured at 450 nm on a Microplate Reader (Bio-Rad) (22).
For human epithelial type 2 cell (HEp-2) autoreactivity assessment, BAL fluid and serum samples (1/100 diluted in PBS) were incubated on Kallestad HEp-2 slides (Bio-Rad Laboratories), followed by AF488-labeled donkey anti-mouse IgM antibodies (Invitrogen) or anti-mouse IgG (Invitrogen). Analysis was performed on a Meta311 confocal microscope (Zeiss).
For detection of anticardiolipin antibodies, Nunc Microwell plates were coated with 10 μg/ml cardiolipin from bovine heart (Sigma) in ethanol and left to dry overnight. For detection of anti-dsDNA antibodies, 20 μg/ml dsDNA from calf thymus was coated overnight on precoated poly-l-lysine microwells. Wells were blocked with 2% BSA/PBS for 2 hours, after which serum (diluted in multiple series) was incubated for 2 hours. Depending on the isotype, anti-mouse IgG biotin/streptavidin-HRP or anti-mouse IgA biotin/streptavidin-HRP were used to develop the ELISA. Optical density was measured at 450 nm on a Microplate Reader (22).
For serum autoreactive IgA binding to tissues, cryosectioned Rag1−/− mouse lungs were used. After 10 minutes of acetone fixation (Sigma) and 10 minutes of block with 10% normal goat serum, serum was differently diluted for WT (1:33) and knockout (KO) (1:100) mice to compensate for differences in total serum IgA levels. Anti-mouse IgA biotin/streptavidin (BD) was used to stain bound IgA. Goat anti-Rat-AP (Sigma) was the secondary antibody used for 30 minutes, after which New Fuchsine (Sigma) was used to detect specific binding (22).
To evaluate the effects of neutralization of IL-6 in vivo, mice were treated with anti–IL-6 antibodies. Starting at the age of 24 weeks, mice received 0.3 mg of anti–IL-6 mAb (clone 20F3) or isotype-matched control mAb, anti–β-Gal (clone GL113) intraperitoneally twice weekly for a total period of 8 weeks.
Statistical significance of data was calculated using a nonparametric Mann-Whitney U test or Bonferroni’s multiple comparison test ANOVA. P values <0.05 were considered significant. All analyses were performed using Prism (GraphPad Software). Data are presented as mean values, together with symbols for individual measurements.
To evaluate the effects of DNGR1-Cre–mediated Tnfaip3/A20 deletion in the lungs, Tnfaip3DNGR1-KO mice were investigated for pulmonary DC subset distribution and conditional deletion efficiency. In 31-week-old Tnfaip3DNGR1-KO mice, cDC1, cDC2, and mo-DC numbers were significantly increased compared with Tnfaip3DNGR1-WT mice (Figures 1A and 1B). To investigate the efficiency of DNGR1-Cre–mediated deletion, YFP expression was evaluated in crosses with Rosa26-Stopfl/fl-YFP mice (25, 27). In line with our previous findings in liver DC populations (22, 25), ∼90% of pulmonary cDC1s in Tnfaip3DNGR1-ROSA-WT mice were targeted by DNGR1-Cre, which unexpectedly declined to ∼50% in Tnfaip3DNGR1-ROSA-KO mice (Figure 1C). In cDC2s and mo-DCs, DNGR1-Cre targeting efficiency was lower and not significantly different between Tnfaip3DNGR1-ROSA-WT and Tnfaip3DNGR1-ROSA-KO mice (∼25% and ∼20%, respectively) (Figure 1C). In pDCs, YFP expression was ∼10% in Tnfaip3DNGR1-ROSA-WT and was even lower in Tnfaip3DNGR1-ROSA-KO mice. Using fluorescence microscopy, we observed that YFP+ cells specifically accumulated perivascularly and within pulmonary lymphocytic infiltrations in the lungs of Tnfaip3DNGR1-ROSA-KO mice. By contrast, only low numbers of YFP+ cells were found in Tnfaip3DNGR1-ROSA-WT mice (Figure 1D).
Figure 1. DNGR1-Cre–mediated Tnfaip3/A20 deletion induces an increase of cDC1s, cDC2s, and mo-DCs in the lung. (A) Gating strategy for pulmonary dendritic cell (1-DC) subset evaluation; conventional type 1 (cDC1, CD11c+MHCII+CD103+) and type 2 (cDC2, CD11c+MHCII+CD11b+, CD64−) DCs, monocyte-derived DCs (mo-DCs, CD11c+MHCII+CD11b+CD64+), and plasmacytoid DCs (pDCs, CD11c+Ly6C+CD11b−CD64−). (B) Absolute cell numbers for pulmonary cDC1s, cDC2s, pDCs, and mo-DCs in the indicated mouse strains: Tnfaip3+/+xClec9a+/crexRosa26-Stopfl/fl-YFP
[More] [Minimize]In summary, DNGR1-Cre–mediated targeting efficiency and subset-specific tracing in control mice was similar to earlier reports (22, 25). Strikingly, compared with Tnfaip3DNGR1-ROSA-WT mice, YFP expression in cDC1s Tnfaip3DNGR1-KO mice dropped from ∼90% to ∼50%, suggesting a selective disadvantage of A20-deficient cDC1s. Furthermore, in Tnfaip3DNGR1-ROSA-KO mice, YFP+ DCs were located near pulmonary vessels.
Next, we investigated whether deletion of the Tnfaip3 gene affected surface expression of typical DC activation markers. No significant differences between Tnfaip3DNGR1-KO mice and WT mice were observed for MHCI expression on any of the three DC subsets (Figure 2A). However, MHCI expression was enhanced in YFP+ cDC1s and YFP+ mo-DCs in Tnfaip3DNGR1-KO mice in comparison with their WT counterparts (Figure 2B). Because MHCI expression was also significantly increased in YFP− pulmonary cDC1s in Tnfaip3DNGR1-KO mice, compared with WT mice, we conclude that MHCI expression was regulated in cDC1s by Tnfaip3/A20 in a cell-nonintrinsic fashion. For mo-DCs, upregulation of MHCI expression appeared to be cell-intrinsically regulated, because YFP− pulmonary mo-DCs and WT DCs showed a similar MHCI surface expression level.
Figure 2. DNGR1-Cre–mediated Tnfaip3/A20 deletion induces both cell-intrinsic and cell-extrinsic effects. Activation status of pulmonary DC subsets was analyzed by flow cytometry. (A) Expression of MHCI, CD86, CD40, and programmed death-ligand 1 (PDL1) was quantified in gated cDC1s, cDC2s, pDCs, and mo-DCs in lungs of the indicated mice and is shown as histogram overlays. (B) Cell-intrinsic versus cell-extrinsic effects of Tnfaip3/A20 deletion in DCs on the expression levels of MHCI, CD86, CD40, and PDL1, as revealed by a separate analysis of YFP− and YFP+ DC subsets. Results are presented as mean values of 3–4 mice per group and representative of two or more independent experiments. *P < 0.05, **P < 0.01, and ***P < 0.001. MFI = median fluorescence intensity.
[More] [Minimize]In the YFP− and YFP+ fractions for cDC1s, cDC2s, and mo-DCs, MHCII expression appeared to be slightly decreased when compared with WT (data not shown). However, this analysis is complicated by the use of the MHCII expression level to gate the individual DC subsets, as shown in Figure 1A. In accordance with previous reports (22, 33), CD86 expression was decreased cDC1s in Tnfaip3DNGR1-KO mice compared with WT, in both YFP+ and YPF− cDC1s. CD40 expression was generally low on all three DC subsets, although it was increased for YFP+ cDC1s, cDC2s, and mo-DCs in Tnfaip3DNGR1-KO mice compared with WT cDC1s. Considering CD40 expression was not increased in YFP− cDC1s, cDC2s, and mo-DCs, we conclude that the increase of CD40 expression in the three subsets of Tnfaip3DNGR1-KO mice is regulated in a cell-intrinsic manner. Compared with WT mice, in Tnfaip3DNGR1-KO mice, programmed death-ligand 1 (PDL1) expression was significantly increased in pulmonary cDC1s, cDC2s, and mo-DCs (Figure 2A), in both the YFP− and the YFP+ fractions (Figure 2B). Such a cell-nonintrinsic regulation of PDL1 expression on DCs would parallel our previous finding in the context of Tnfaip3/A20-deficient DCs (33), that PDL1 expression on DCs can be induced by high levels of IFN-γ.
In summary, upon DNGR1-Cre–mediated Tnfaip3 deletion, cell surface expression of activation markers on cDC1, cDC2, and mo-DC populations was regulated by both cell-intrinsic and cell-extrinsic mechanisms.
To investigate whether pulmonary homeostasis is changed by Tnfaip3/A20-deficient DCs, we evaluated lungs of Tnfaip3DNGR1-KO mice by histological analysis at 31 weeks of age. In the small pulmonary arteries of Tnfaip3DNGR1-KO mice, we observed increased perivascular inflammatory infiltration and the presence of CD11c-positive cells (Figure 3A). Furthermore, RT-PCR experiments and flow cytometric analysis was performed in heart tissue of Tnfaip3DNGR1-KO mice, showing increased Itgax (encoding CD11c) expression and increased proportions of DCs, in both the left and right ventricle of the heart (Figure E1).
Figure 3. Lymphocytic infiltrates in the lungs of Tnfaip3DNGR1-KO mice consist of activated T and B cells. (A) Hematoxylin and eosin staining (arrowheads indicating pulmonary arteries [PAs]) and immunohistochemistry staining for CD11c combined with αSMA (arrowheads indicating CD11c+ cells) was performed on tissue sections of lungs of 31-week-old Tnfaip3DNGR1 mice. Scale bars, 50 μm. (B) T-cell (CD3, red) and B-cell (B220, blue) staining on pulmonary cryosections. IgD (red) and GL7 (blue) staining for germinal center (GC) B cells. CD138 (red) and IgA (blue) staining for plasma cells. IgD (red) and Ki67 (blue). Scale bars, 200 μm. (C–E) Flow cytometric quantification of the indicated subsets of CD4+ T cells (C), Tregs (D), and CD8+ T cells (E). (F–H) Flow cytometry analysis for follicular T-helper cells (PD-1+CXCR5+CD4+ T cells) (F), total B cells and GC B cells (GL7+CD95+CD19+) (G), and plasma cells (CD138+B220+) (H). *P < 0.05. Results are presented as mean values with symbols for individual mice (5–8 per group), representative of two or more independent experiments. Tfh = follicular T-helper cell; Tregs = regulatory T cells.
[More] [Minimize]Because lymphocytic infiltrations in the lungs of patients with PAH contain increased numbers of T and B cells (10), we explored the inflammatory landscape in the lungs of Tnfaip3DNGR1-KO mice. The peribronchial infiltrates consisted of distinct T-cell (CD3+) and B-cell (B220+) zones (Figure 3B). A staining for the plasma cell marker CD138 and IgA indicated the presence of IgA-producing plasma cells in both peribronchially and perivascularly infiltrated areas in the lungs of Tnfaip3DNGR1-KO mice, while being absent in Tnfaip3DNGR1-WT mice. Of note, CD138 staining was also observed in bronchial epithelial cells (Figure 3B), paralleling reported findings (34). Evaluation of lymphocytic infiltrates also showed specific IgD+ zones and proliferating cells (Ki67+) (Figure 3B).
Next, we used flow cytometry to quantify T and B cells and to investigate their phenotype in the lungs. The numbers of CD4+ Th cells were increased in Tnfaip3DNGR1-KO mice compared with Tnfaip3DNGR1-WT mice, although this increase did not reach significance. The proportions of activated/memory (CD44high) and proliferating CD44+Ki67+ Th cells were significantly higher in Tnfaip3DNGR1-KO mice than in Tnfaip3DNGR1-WT mice (Figure 3C). Regulatory T cells were unaltered in cell number, but an augmented proportion of CD44high activated/memory regulatory T cells was observed in Tnfaip3DNGR1-KO mice, compared with Tnfaip3DNGR1-WT mice (Figure 3D). The total numbers of CD8+ T cells were not significantly increased in the lungs of Tnfaip3DNGR1-KO mice compared with Tnfaip3DNGR1-WT mice, but the proportions of CD44high activated/memory and proliferating CD44+/Ki67+ CD8+ T cells were significantly higher in Tnfaip3DNGR1-KO mice (Figure 3E).
The presence of specific T- and B-cell zones in the lung, as detected by histology, was indicative for the formation of TLO-like structures. Therefore, lungs were investigated for follicular Th cells (Tfh) and B-cell subsets. The frequency of Tfh cells, defined as CXCR5+/PD1+, within the fraction of total CD4+ T cells was increased in Tnfaip3DNGR1-KO mice compared with Tnfaip3DNGR1-WT mice (Figure 3F). Although the number of total B cells did not differ between the two genotypes, the proportions of germinal center (GC) (CD95+/GL7+) B cells and CD138+ plasma cells were increased in lungs of Tnfaip3DNGR1-KO mice compared with Tnfaip3DNGR1-WT mice (Figures 3G and 3H).
Because recruitment of macrophages also plays an important role in pulmonary vascular remodeling (35, 36), an F4/80 staining was performed, showing the perivascular presence of macrophages in Tnfaip3DNGR1-KO mice (Figure E2).
From these findings we conclude that the pulmonary inflammatory infiltrates in Tnfaip3DNGR1-KO mice contained dendritic cells and macrophages as well as lymphocytes present in TLO-like structures. CD4+ and CD8+ T cells showed enhanced activation and proliferation, and GC B cells and plasma cells were present. In summary, these data indicate that T and B cells were most likely chronically activated in inflammatory lesions in the lungs of Tnfaip3DNGR1-KO mice.
Considering the pulmonary perivascular infiltration, we investigated possible vascular remodeling in Tnfaip3DNGR1-KO mice. Pulmonary vascular remodeling with increased lymphocytic infiltration and vascular muscular wall thickening was observed in the small (20–50 μm) arteries of the lungs of Tnfaip3DNGR1-KO mice compared with Tnfaip3DNGR1-WT mice at 24 and 31 weeks of age (Figures 4A and 4B). Quantification of the muscularization of the small pulmonary arteries by staining for αSMA showed that this was significantly enhanced in 31-week-old Tnfaip3DNGR1-KO mice (Figures 4B and 4C).
Figure 4. DNGR1-Cre–specific Tnfaip3/A20 deletion leads to a pulmonary hypertension phenotype in Tnfaip3DNGR1-KO mice. (A) Elastin von Giessen–stained lung histology of Tnfaip3DNGR1-WT and Tnfaip3DNGR1-KO mice for representative sections for small PAs (20–50 μm). Scale bars, 50 μm. Arrowheads indicate PAs. (B) Using immunohistochemistry staining for CD31 and αSMA in the lung, vascular muscular wall thickness was evaluated and (C) quantified. Scale bars, 100 μm. (D) Right ventricular systolic pressure (RVSP) and RVeDP, as determined using right heart catheterization, in the indicated mice. (E) Echocardiographic measurements in the indicated mice for pulmonary arterial acceleration time (PAAT) and RVWT. (F) Right ventricular hypertrophy assessed using the Fulton index (right ventricle [RV]/left ventricle+septum [LV+s]), as well as separate LV+s and RV measurements for the indicated mice. (G) Correlation of RVSP and PAAT in both Tnfaip3DNGR1-WT and Tnfaip3DNGR1-KO mice at 31 weeks of age. (H) Relative mRNA expression of vascular remodeling factors in the indicated tissues of 31-week-old Tnfaip3DNGR1-KO mice. Expression in 24-week-old Tnfaip3DNGR1-WT mice was set to 100%. Results are presented as symbols for individual mice, together with mean values of 7–15 mice per group and representative of three or more independent experiments. *P < 0.05, **P < 0.01, and ***P < 0.001. HGF = hepatocyte growth factor; RVeDP = right ventricular end-diastolic pressure; RVWT = right ventricular wall thickness; TGFβ = transforming growth factor β; VEGFα = vascular endothelial growth factor α.
[More] [Minimize]Because arteries in Tnfaip3DNGR1-KO mice showed vascular remodeling, we evaluated these mice for signs of PH at the age of 24 and 31 weeks. No significant differences in RVSP or RVeDP were found between 24-week-old Tnfaip3DNGR1-KO and Tnfaip3DNGR1-WT mice. Strikingly, at the age of 31 weeks, both RVSP and RVeDP were significantly elevated in Tnfaip3DNGR1-KO mice (Figure 4D). Echocardiographic evaluation showed a decreased PAAT and increased RV wall thickness in 31-week-old Tnfaip3DNGR1-KO mice compared with Tnfaip3DNGR1-WT mice (Figure 4E). To investigate cardiac remodeling due to elevated RVSP, right ventricular hypertrophy (RVH) was explored by determining the Fulton index. In both 24- and 31-week-old mice, an augmentation in the Fulton index was observed in Tnfaip3DNGR1-KO mice compared with Tnfaip3DNGR1-WT mice. The elevated Fulton index indicative for RVH in Tnfaip3DNGR1-KO mice was due to increased RV weight and not to a decreased LV+s weight (Figure 4F). The PAAT determined by echocardiography showed a significant inverse correlation with RVSP (Figure 4G). No correlations were found between a higher grade of vascular remodeling and more pronounced PH in our mice (data not shown). The PH phenotype was likely not induced by obstructive airway pathology or hypoxia, as no differences were seen in lung function or pulmonary hypoxia factor expression (Hif1a, Bnip3, and Slc2a1 mRNA) between Tnfaip3DNGR1-KO mice and Tnfaip3DNGR1-WT mice (Figure E3 and data not shown).
Known mediators involved in vascular remodeling are TGFβ, VEGFA, and HGF. TGFβ mRNA expression was significantly increased in lungs (at 31 wk), RV (at 31 wk), and LV+s (both at 24 and at 31 wk) from Tnfaip3DNGR1-KO mice compared with control mice (Figure 4H). Although the lungs of 31-week-old Tnfaip3DNGR1-KO mice did not show increased Vegf expression, it was increased in RV and LV+s. Hgf transcription was specifically increased in the RV of Tnfaip3DNGR1-KO mice compared with Tnfaip3DNGR1-WT at 31 weeks (Figure 4H). An αSMA staining on heart tissue of 31-week-old Tnfaip3DNGR1-WT/KO showed no differences between vascular endothelial cells of Tnfaip3DNGR1-KO and WT mice (data not shown).
Taken together, these data show that 31-week-old Tnfaip3DNGR1-KO mice developed experimental PH with RVH and increased pulmonary and RV/LV expression of vascular remodeling/growth factors.
We previously reported increased levels of total IgG1 and IgA, but no other isotypes, in serum of Tnfaip3DNGR1-KO mice compared with control mice (22). When we determined Ig levels in BAL fluid, we observed increased total IgG1 and IgA in Tnfaip3DNGR1-KO mice compared with Tnfaip3DNGR1-WT mice (Figure 5A). To evaluate the presence of autoantibodies, we screened HEp-2 cells and found staining with IgG present in the serum—but not in BAL fluid—in three out of six Tnfaip3DNGR1-KO mice (Figure 5B). These HEp-2 IgG staining patterns were nuclear speckled and antimitochondrial (37). In contrast, none of the four WT control sera were positive.
Figure 5. Tnfaip3DNGR1-KO mice show IgG and IgA autoantibodies in BAL fluid and serum. (A) Quantification of BAL fluid immunoglobulins (IgG1 and IgA) in Tnfaip3DNGR1WT/KO mice. Results are presented as symbols for individual mice, together with mean values of 30–32 mice per group (B) Immunofluorescence assessment of autoreactive IgG immunoglobulins from 31-week-old mice of the indicated genotype using human epithelial type 2 cells. (C) Quantification of autoreactive IgG1 and IgA for dsDNA and cardiolipin in serum from 31-week-old mice using ELISA. Results are presented as symbols for individual mice, together with mean values of 9–10 mice per group. (D) Immunohistochemical evaluation for the presence of autoreactive IgA in Tnfaip3DNGR1-WT and Tnfaip3DNGR1-KO mouse serum, using lung tissue from Rag1−/− mice (arrows indicating IgA-positive staining). Scale bars, 100 μm. *P < 0.05, **P < 0.01, and ***P < 0.001.
[More] [Minimize]Previously, we have shown that both anti-dsDNA IgG1 and anticardiolipin IgG1 concentrations were significantly enhanced in the serum of Tnfaip3DNGR1-KO mice compared with WT controls, as measured at an age of 31 weeks (22). Because anti-dsDNA and anticardiolipin Igs can be found in autoimmune diseases associated with PAH (38), we also investigated whether Tnfaip3DNGR1-KO mice showed antibodies with these specificities in BAL fluid. The levels of dsDNA- and cardiolipin-specific autoantibodies—both IgG1 and IgA—were significantly increased in Tnfaip3DNGR1-KO mice compared with WT controls (Figure 5C).
We next investigated whether autoantibodies would recognize pulmonary antigens and stained lung sections of Rag1−/− mice with serum of Tnfaip3DNGR1-KO and Tnfaip3DNGR1-WT mice. In Rag1−/− mice, endogenous Igs are absent, facilitating the detection of binding of serum autoantibodies in tissues. IgA, but not IgG, in the serum of Tnfaip3DNGR1-KO mice specifically recognized antigens present in pulmonary endothelial cells (Figure 5D).
Summarizing, in Tnfaip3DNGR1-KO mice levels of total IgA and IgG1 and auto-antibodies recognizing dsDNA and cardiolipin levels were elevated in serum and BAL fluid. In these mice, circulating IgA contained specific reactivity toward the pulmonary vasculature.
In patients with IPAH, the increased presence of inflammatory cytokines in serum negatively correlated with patient survival (5). Especially IL-1β, IL-6, and IL-10 serum levels were increased in patients with IPAH versus healthy controls. Therefore, we investigated cytokine expression in the lungs and hearts of Tnfaip3DNGR1-KO mice. Increased pulmonary Il1b, Il6, and Il10 mRNA expression was observed in 31-week-old Tnfaip3DNGR1-KO mice compared with Tnfaip3DNGR1-WT mice (Figure 6A). In contrast, only Il1b—but not Il6 or Il10—mRNA expression was significantly increased in the RV of Tnfaip3DNGR1-KO mice; no significant differences were observed for the three cytokines examined in LV+s (Figure 6B).
Figure 6. Tnfaip3DNGR1-KO mice show increased IL1β, IL6, and IL10 mRNA expression in the lungs and heart. (A and B) Expression of IL-1β (Il1b), IL-6 (Il6), and IL-10 (Il10) mRNA in lung (A) and in the indicated parts of the heart (B). (C and D) Intracellular flow cytometry analysis of the indicated cytokines. Data are shown as proportions of cytokine-positive CD4+ T cells (C) or CD8+ T cells (D). Results are presented as symbols for individual mice, together with mean values of 15 versus 16 (A and B) and 7 versus 6 (C and D) mice per group. *P < 0.05 and **P < 0.01.
[More] [Minimize]We next evaluated whether increased proportions of inflammatory cytokine–producing CD4+ and CD8+ T cells were present in the lungs of Tnfaip3DNGR1-KO mice. Strikingly, IL-10 and IFN-γ–producing CD4+ and CD8+ T cells were significantly increased in Tnfaip3DNGR1-KO mice compared with Tnfaip3DNGR1-WT mice (Figures 6C and 6D). Furthermore, the CD4+ T-cell population in the lungs of Tnfaip3DNGR1-KO mice contained increased proportions of IL-17A–producing cells (Figure 6C). No differences in the frequencies of TNF-α–producing CD4+ or CD8+ T cells were found between Tnfaip3DNGR1-KO mice and WT control mice.
Taken together, lungs of Tnfaip3DNGR1-KO mice had increased mRNA expression of IL-6 and IL-10, accompanied by increased proportions of IL-10+ and IFN-γ+ CD4+ and CD8+ T cells and increased proportions of IL-17A+ CD4+ T cells.
Because IL-6 is thought to have a prominent role in PAH pathology (5, 39) and given the increased expression of Il6 in the lungs of Tnfaip3DNGR1-KO mice, we investigated whether neutralization of IL-6 could ameliorate the PH phenotype. To this end, Tnfaip3DNGR1-WT and Tnfaip3DNGR1-KO mice received either IL-6–neutralizing antibodies or an isotype control antibody intraperitoneally twice per week starting at the age of 24 weeks. Mice were investigated for signs of development of PH at week 31 (Figure 7A).
Figure 7. Anti–IL-6 treatment ameliorates the pulmonary hypertension phenotype in Tnfaip3DNGR1-KO mice. (A) Experimental setup for anti–IL-6 antibody treatment and isotype treatment in 31-week-old Tnfaip3DNGR1-KO mice and WT mice. (B) Right heart catheterization measurements for RVSP and RVeDP values in the indicated mouse groups. (C) Echocardiographic measurements for PAAT values in the indicated mouse groups. (D) Right ventricular hypertrophy assessment using the Fulton Index (RV/LV+s) in the indicated mouse groups. (E) αSMA-stained lung histology of Tnfaip3DNGR1-WT and Tnfaip3DNGR1-KO mice for representative sections of large and small PAs (arrowheads indicating PAs). (F) SMC per total artery area, as determined by quantification of vascular muscular wall thickness using CD31 and αSMA immunohistochemistry. (G) Proportions of T-helper cell type 17 (Th17) cells as a percentage of CD4+ T cells. *P < 0.05 and ***P < 0.001. Results are presented as symbols for individual mice, together with mean values of 9–10 (B and D) and 4–7 (F and G) mice per group and are representative of two independent experiments. SMC = smooth muscle cells per total artery area, as determined by quantification of vascular muscular wall thickness using CD31 and αSMA immunohistochemistry.
[More] [Minimize]As expected, an increased RVSP/RVedP, a trend toward a decreased PAAT, and a significantly increased Fulton index were found in isotype-treated Tnfaip3DNGR1-KO compared with Tnfaip3DNGR1-WT mice. No differences between isotype and anti–IL-6–treated Tnfaip3DNGR1-WT mice were detected. However, in Tnfaip3DNGR1-KO mice, anti–IL-6 treatment prevented the increase in RVSP/RVeDP compared with isotype-treated Tnfaip3DNGR1-KO mice (Figure 7B). Echocardiographic analysis showed, albeit not significant, normalization of the PAAT following anti–IL-6 treatment in Tnfaip3DNGR1-KO mice, when compared with isotype-treated Tnfaip3DNGR1-KO mice (Figure 7C). Furthermore, the enhanced Fulton index observed in control-treated Tnfaip3DNGR1-KO mice was diminished after anti–IL-6 treatment in Tnfaip3DNGR1-KO mice, reaching values similar to Tnfaip3DNGR1-WT mice (Figure 7D).
Pulmonary lymphocytic infiltration was unaffected by IL-6 blockade in Tnfaip3DNGR1-KO mice (Figure 7E). The small- to medium-sized pulmonary arteries of isotype-control Tnfaip3DNGR1-KO mice exhibited significantly increased muscularization compared with Tnfaip3DNGR1-WT mice (Figure 7E, quantified in Figure 7F). After anti–IL-6 treatment, the increase in muscularization was only moderate (Figures 7E and 7F).
Anti–IL-6 treatment did not affect the numbers of total CD4+ or CD8+ T cells or the proportions of actively cycling CD44+Ki67+ CD4+ or CD8+ T cells (Figure E4). Only the proportions of IL-17+CD4+ T cells, which were increased in isotype-treated Tnfaip3DNGR1-KO mice, were significantly decreased after anti–IL-6 treatment (Figure 7G). Anti–IL-6 therapy did not influence the amount of IgA and IgG1 in BAL fluid in Tnfaip3DNGR1-KO mice (data not shown).
In conclusion, these data showed amelioration of the experimental PH phenotype in Tnfaip3DNGR1-KO mice upon IL-6 neutralization, indicating a crucial role for an active IL-6 pathway in the pathophysiology of PH in Tnfaip3DNGR1-KO mice.
Increasing evidence suggests a prominent role for an activated immune system in the pathophysiology of PAH and experimental PH models (5, 40–43). Although it is known that in IPAH lungs, TLO formation and increased DC numbers are often present (10, 11), our study investigated the role of DCs in the pathophysiology of PAH in more detail in a mouse model.
In this study, we investigated Tnfaip3DNGR1-KO mice, in which the Tnfaip3/A20 gene—a negative regulator of the NF-κB pathway—was deleted mainly in cDC1s using the Clec9a(DNGR1)-Cre transgene. These mice developed a PH phenotype, as indicated by increased RVSP, RVH, perivascular lymphocytic infiltration, and vascular remodeling. In comparison with established PH models, such as the monocrotaline rat model, our model seems less pronounced in vascular remodeling. Nevertheless, striking parallels exist, as in monocrotaline-exposed rats the mean number of arterial DCs was found to be increased during development of vasculopathy (7). In this model, arterial DC accumulation precedes pulmonary arterial thickening and hemodynamic alteration and is constantly present in remodeled vessels, indicating that DC influx is not merely a consequence of increased pulmonary arterial pressure.
Upon evaluation of known PH-inducing factors in Tnfaip3DNGR1-KO mice, we concluded that the experimental PH phenotype was not likely to be induced by obstructive airway pathology or hypoxia. In particular, no differences were seen in lung function or in pulmonary expression of hypoxia factors including Hif1a, Bnip3, and Slc2a1, between Tnfaip3DNGR1-KO mice and Tnfaip3DNGR1-WT mice (Figure E3 and data not shown). Furthermore, in previous work performed by our group (22), inflammatory autoimmune hepatic injury was found in Tnfaip3DNGR1-KO mice. Considering that no signs of portopulmonary hypertension were observed, portopulmonary hypertension seems unlikely to be the causal factor for the experimental PH phenotype in Tnfaip3DNGR1-KO mice. Nor did we find any signs of LV involvement or signs of LV failure resembling World Health Organization PH group 2.
Most likely, pulmonary inflammation occurs concomitantly with vascular remodeling, both of which appear to be present at 24 weeks of age. Analysis at 24 weeks showed a trend toward increased RVSP. Also, a trend of arterial remodeling was noticed in 24-week-old Tnfaip3DNGR1-KO mice (Figure 4C). These findings suggest that 24-week-old mice are in transition to develop PH, with a near-significant RVSP increase, a significantly increased Fulton Index, and a trend of vascular remodeling. Additionally, Elastica von Giessen stainings of the hearts of 24-week-old Tnfaip3DNGR1-KO and WT mice showed right ventricular hypertrophy and no differences in collages deposition or cardiomyocyte shape and size (T. Koudstaal, unpublished results).
However, the severity of the PH phenotype is variable in Tnfaip3DNGR1-KO mice, with RVSP values ranging from mild or no PH to severe PH. At 31 weeks, ∼50% of the mice have developed PH with increased RVSP, increased Fulton index, and vascular remodeling. Variability of the Tnfaip3/A20 gene deletion efficiency across different DC subsets and cell-extrinsic effects of the targeted deletion might partly be an explanation as to why not all mice develop experimental PH.
DNGR1-Cre–specific deletion of Tnfaip3 led to increased numbers of pulmonary cDC1s, cDC2s, and mo-DCs. Specifically, cDCs harbored an altered activation status, for example, enhanced MHCI, CD40, and PDL1 expression, which was regulated by both cell-intrinsic and cell-extrinsic effects. The increase in PDL1 is most likely caused by cell-extrinsic factors, for which IFN-γ is a likely candidate as the proportion of IFN-γ–producing T cells is augmented in Tnfaip3DNGR1-KO mice (Figure 6). Possibly, this increased expression of PDL1 might be a feedback mechanism, induced by the increased production of IFN-γ or IL-12 (33).
Currently, the exact mechanism by which experimental PH develops in Tnfaip3DNGR1-KO mice is unclear. Although cDC1s were mainly affected by DNGR1-mediated ablation, cDC1s, cDC2s, and mo-DCs showed an increase in cell number. This might partially be due to targeting of a minor fraction of cDC2s and mo-DCs (25). These finding make it difficult to determine which DC subset is exactly contributing to the PH phenotype. Moreover, the direct and indirect effects of activated DCs on other immune cells, such as T cells and B cells, and pulmonary vasculature remain to be characterized in detail. Importantly, however, our results do show that this experimental PH model is likely to be IL-6 dependent and that activation of the NF-κB pathway in DCs is sufficient to induce experimental PH in mice. Previous research has shown that Tnfaip3-deficient DCs are able to produce a variety of inflammatory cytokines such as IL-6 (44). Although we did not study whether vascular remodeling is a direct effect of Tnfaip3-deficient cDCs, the presence of YFP-expressing DCs around blood vessels suggests a direct interaction. Possibly, Tnfaip3-deficient DCs may contribute to vascular remodeling by production of cytokines and growth factors and/or by attraction of other immune cells.
The translatability of our findings in the Tnfaip3DNGR1-KO mice to human PAH remains to be investigated. Nevertheless, the findings of 1) increased numbers of DCs in remodeled pulmonary vessels of patients with IPAH (7) and 2) a specific SNP in the TNFAIP3 locus that is associated with PAH development in patients with systemic sclerosis may support a pathological role of activated DCs in human PH. It is currently unknown whether the reported increased presence of DCs in the lungs of patients with IPAH is causal or a circumstantial effect.
DCs excel at antigen presentation to T cells, leading to T-cell activation and differentiation (16–18, 20). The level of Tnfaip3 expression in DCs has been shown to control T-cell differentiation, as Tnfaip3-deficient DCs promote Th17-cell differentiation through increased expression of IL-1β, IL-6, and IL-23 (20, 44). In Tnfaip3DNGR1-KO mice, a modest increase in Th17 cells was observed, which was lower after anti–IL-6 treatment. Also, increased proportions of IFN-γ–producing Th1 cells and IL-10–producing Th cells were found in Tnfaip3DNGR1-KO mice. These findings may support involvement of both Th17 cells and Th1 cells in IPAH and PAH pathology (45, 46).
Conventional DC1 s are especially equipped to promote CD8+ T-cell activation (47). In support of cDC1 activation in Tnfaip3DNGR1-KO mice, the number of CD8+ T cells, their proliferative capacity, and the proportion of IFN-γ and IL-10–producing cells was higher in lungs of Tnfaip3DNGR1-KO mice compared with control mice. Strikingly, also in patients with IPAH, a dramatic increase in pulmonary CD8+ T cells has been observed, specifically located in the adventitia of pulmonary vessels (11, 48). These results are suggestive of a role for both Th1 and Th17 cells as well as activated CD8+ T cells in the pathogenesis of PAH, although direct causality has not been shown yet.
The presence of TLOs is a hallmark characteristic for chronic immune activation with a specific role for GC formation where B cells are activated and can become antibody-producing plasma cells. In IPAH lungs, TLOs are present near the vasculature and harbor Tfh cells, activated B cells, and antibody-producing plasma cells (10).
In Tnfaip3DNGR1-KO mice, inflammatory lesions with separated T- and B-cell zones were observed in the lung as well. The observed increase in GC B cells, Tfh cells, and plasma cells suggests that these inflammatory lesions are active TLOs in Tnfaip3DNGR1-KO mice. It is conceivable that activation of these cells led to the enhanced total IgG1 and IgA levels found in serum (22) and BAL fluid in Tnfaip3DNGR1-KO mice. Specifically, autoreactive IgA recognizing lung vasculature was found in serum—but not in BAL fluid—of Tnfaip3DNGR1-KO mice. Autoreactive IgA might be secreted by lung plasma cells, and rapidly bound to the lung vasculature, whereby residual IgA concentrations were too small to be detected. Moreover, IgA staining was less pronounced in the smaller vasculature. Approximately 40% of patients with IPAH are known to have autoantibodies (31, 49) that may specifically target endothelial cell surface antigens (50).
In the Tnfaip3DNGR1-KO mouse model, it is unclear whether T and B cells are mobilized from bone marrow/lymphatic tissue to the lungs or whether they are already stationary pulmonary resident cells. Future experiments are required to provide further insight into these possible mechanisms.
In patients with PAH, several cytokines are enhanced in the peripheral blood and correlate with survival (5). Several lines of evidence support a prominent role for IL-6, as 1) serum IL-6 concentrations correlate with survival (5), 2) IL-6–overexpressing transgenic mice spontaneously develop a PH phenotype (43), and 3) IL-6 receptor expression and signaling is crucial for PAH development and progression (39). In Tnfaip3DNGR1-KO mice, we observed a higher mRNA expression of the pro- and antiinflammatory cytokines IL-1β, IL-6, and IL-10, in both lungs and hearts (Figure 6). However, DCs, B cells, and macrophages are known producers of these inflammatory cytokines. In previous work by our group (44), Tnfaip3-deficient DCs have been shown to promote Th17-cell differentiation through increased expression of IL-1β, IL-6, and IL-23. Possibly, DCs are one of the main contributors to the increased mRNA expression of IL-1β, IL-6, and IL-10. Strikingly, membrane-bound IL-6 expression on DCs is required for the differentiation and priming of pathogenic Th17 cells (51). Our data also showed enhanced proportions of Th17 cells in Tnfaip3DNGR1-KO mice, which were reduced upon blockade of IL-6. Neutralization of IL-6 in Tnfaip3DNGR1-KO mice ameliorated the experimental PH phenotype, providing evidence that IL-6 is a major contributor to PH development in these mice. Possibly, anti–IL-6 treatment of KO mice may attenuate the pulmonary vascular remodeling through a different mode of action, including through reduction of PA-smooth muscle cell survival (39).
Tnfaip3DNGR1-KO mice develop an experimental PH phenotype characterized by increased RVSP, RVH, perivascular lymphocytic infiltration, and vascular remodeling. To our knowledge, this is the first study that shows that that activation of the NF-κB pathway in DCs is sufficient for the development of experimental PH in an IL-6–dependent fashion.
The authors thank Prof. Caetano Reis e Sousa for providing critical mouse strains. They also thank the animal caretakers in their animal experimental facility. Also, they thank Odilia Corneth and Jasper Rip for their help in staining and analyzing the B cells and plasma cells in Tnfaip3DNGR1-KO mice.
| 1. | Galiè N, Humbert M, Vachiery JL, Gibbs S, Lang I, Torbicki A, et al. 2015 ESC/ERS guidelines for the diagnosis and treatment of pulmonary hypertension: the Joint Task Force for the diagnosis and treatment of pulmonary hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS). Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC), International Society for Heart and Lung Transplantation (ISHLT). Eur Respir J 2015;46:903–975. |
| 2. | Rabinovitch M, Guignabert C, Humbert M, Nicolls MR. Inflammation and immunity in the pathogenesis of pulmonary arterial hypertension. Circ Res 2014;115:165–175. |
| 3. | Huertas A, Perros F, Tu L, Cohen-Kaminsky S, Montani D, Dorfmüller P, et al. Immune dysregulation and endothelial dysfunction in pulmonary arterial hypertension: a complex interplay. Circulation 2014;129:1332–1340. |
| 4. | Humbert M, Guignabert C, Bonnet S, Dorfmüller P, Klinger JR, Nicolls MR, et al. Pathology and pathobiology of pulmonary hypertension: state of the art and research perspectives. Eur Respir J 2019;53:1801887. |
| 5. | Soon E, Holmes AM, Treacy CM, Doughty NJ, Southgate L, Machado RD, et al. Elevated levels of inflammatory cytokines predict survival in idiopathic and familial pulmonary arterial hypertension. Circulation 2010;122:920–927. |
| 6. | Cracowski JL, Chabot F, Labarère J, Faure P, Degano B, Schwebel C, et al. Proinflammatory cytokine levels are linked to death in pulmonary arterial hypertension. Eur Respir J 2014;43:915–917. |
| 7. | Perros F, Dorfmüller P, Souza R, Durand-Gasselin I, Mussot S, Mazmanian M, et al. Dendritic cell recruitment in lesions of human and experimental pulmonary hypertension. Eur Respir J 2007;29:462–468. |
| 8. | Marsh LM, Jandl K, Grünig G, Foris V, Bashir M, Ghanim B, et al. The inflammatory cell landscape in the lungs of patients with idiopathic pulmonary arterial hypertension. Eur Respir J 2018;51:1701214. |
| 9. | Wang W, Yan H, Zhu W, Cui Y, Chen J, Wang X, et al. Impairment of monocyte-derived dendritic cells in idiopathic pulmonary arterial hypertension. J Clin Immunol 2009;29:705–713. |
| 10. | Perros F, Dorfmüller P, Montani D, Hammad H, Waelput W, Girerd B, et al. Pulmonary lymphoid neogenesis in idiopathic pulmonary arterial hypertension. Am J Respir Crit Care Med 2012;185:311–321. |
| 11. | Savai R, Pullamsetti SS, Kolbe J, Bieniek E, Voswinckel R, Fink L, et al. Immune and inflammatory cell involvement in the pathology of idiopathic pulmonary arterial hypertension. Am J Respir Crit Care Med 2012;186:897–908. |
| 12. | GeurtsvanKessel CH, Willart MAM, Bergen IM, van Rijt LS, Muskens F, Elewaut D, et al. Dendritic cells are crucial for maintenance of tertiary lymphoid structures in the lung of influenza virus-infected mice. J Exp Med 2009;206:2339–2349. |
| 13. | Nacionales DC, Kelly KM, Lee PY, Zhuang H, Li Y, Weinstein JS, et al. Type I interferon production by tertiary lymphoid tissue developing in response to 2,6,10,14-tetramethyl-pentadecane (pristane). Am J Pathol 2006;168:1227–1240. |
| 14. | Neyt K, Perros F, GeurtsvanKessel CH, Hammad H, Lambrecht BN. Tertiary lymphoid organs in infection and autoimmunity. Trends Immunol 2012;33:297–305. |
| 15. | Guilliams M, Ginhoux F, Jakubzick C, Naik SH, Onai N, Schraml BU, et al. Dendritic cells, monocytes and macrophages: a unified nomenclature based on ontogeny. Nat Rev Immunol 2014;14:571–578. |
| 16. | Lewis KL, Reizis B. Dendritic cells: arbiters of immunity and immunological tolerance. Cold Spring Harb Perspect Biol 2012;4:a007401. |
| 17. | Ganguly D, Haak S, Sisirak V, Reizis B. The role of dendritic cells in autoimmunity. Nat Rev Immunol 2013;13:566–577. |
| 18. | Blanco P, Palucka AK, Pascual V, Banchereau J. Dendritic cells and cytokines in human inflammatory and autoimmune diseases. Cytokine Growth Factor Rev 2008;19:41–52. |
| 19. | Hammer GE, Turer EE, Taylor KE, Fang CJ, Advincula R, Oshima S, et al. Expression of A20 by dendritic cells preserves immune homeostasis and prevents colitis and spondyloarthritis. Nat Immunol 2011;12:1184–1193. |
| 20. | Kool M, van Loo G, Waelput W, De Prijck S, Muskens F, Sze M, et al. The ubiquitin-editing protein A20 prevents dendritic cell activation, recognition of apoptotic cells, and systemic autoimmunity. Immunity 2011;35:82–96. |
| 21. | Matmati M, Jacques P, Maelfait J, Verheugen E, Kool M, Sze M, et al. A20 (TNFAIP3) deficiency in myeloid cells triggers erosive polyarthritis resembling rheumatoid arthritis. Nat Genet 2011;43:908–912. |
| 22. | Das T, Bergen IM, Koudstaal T, van Hulst JAC, van Loo G, Boonstra A, et al. DNGR1-mediated deletion of A20/Tnfaip3 in dendritic cells alters T and B-cell homeostasis and promotes autoimmune liver pathology. J Autoimmun 2019;102:167–178. |
| 23. | Ma A, Malynn BA. A20: linking a complex regulator of ubiquitylation to immunity and human disease. Nat Rev Immunol 2012;12:774–785. |
| 24. | Dieudé P, Guedj M, Wipff J, Ruiz B, Riemekasten G, Matucci-Cerinic M, et al. Association of the TNFAIP3 rs5029939 variant with systemic sclerosis in the European Caucasian population. Ann Rheum Dis 2010;69:1958–1964. |
| 25. | Schraml BU, van Blijswijk J, Zelenay S, Whitney PG, Filby A, Acton SE, et al. Genetic tracing via DNGR-1 expression history defines dendritic cells as a hematopoietic lineage. Cell 2013;154:843–858. |
| 26. | Vereecke L, Sze M, Mc Guire C, Rogiers B, Chu Y, Schmidt-Supprian M, et al. Enterocyte-specific A20 deficiency sensitizes to tumor necrosis factor-induced toxicity and experimental colitis. J Exp Med 2010;207:1513–1523. |
| 27. | Srinivas S, Watanabe T, Lin CS, William CM, Tanabe Y, Jessell TM, et al. Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus. BMC Dev Biol 2001;1:4. |
| 28. | Mombaerts P, Iacomini J, Johnson RS, Herrup K, Tonegawa S, Papaioannou VE. RAG-1-deficient mice have no mature B and T lymphocytes. Cell 1992;68:869–877. |
| 29. | de Raaf MA, Herrmann FE, Schalij I, de Man FS, Vonk-Noordegraaf A, Guignabert C, et al. Tyrosine kinase inhibitor BIBF1000 does not hamper right ventricular pressure adaptation in rats. Am J Physiol Heart Circ Physiol 2016;311:H604–H612. |
| 30. | Hammad H, Kool M, Soullié T, Narumiya S, Trottein F, Hoogsteden HC, et al. Activation of the D prostanoid 1 receptor suppresses asthma by modulation of lung dendritic cell function and induction of regulatory T cells. J Exp Med 2007;204:357–367. |
| 31. | Arends SJ, Damoiseaux J, Duijvestijn A, Debrus-Palmans L, Boomars K, Broers B, et al. Prevalence of anti-endothelial cell antibodies in idiopathic pulmonary arterial hypertension. Eur Respir J 2010;35:923–925. |
| 32. | Pal Singh S, de Bruijn MJW, de Almeida MP, Meijers RWJ, Nitschke L, Langerak AW, et al. Identification of distinct unmutated chronic lymphocytic leukemia subsets in mice based on their T cell dependency. Front Immunol 2018;9:1996. |
| 33. | Vroman H, van Uden D, Bergen IM, van Hulst JAC, Lukkes M, van Loo G, et al. Tnfaip3 expression in pulmonary conventional type 1 Langerin-expressing dendritic cells regulates T helper 2-mediated airway inflammation in mice. Allergy [online ahead of print] 23 Apr 2020; DOI: 10.1111/all.14334. |
| 34. | Hayashida K, Johnston DR, Goldberger O, Park PW. Syndecan-1 expression in epithelial cells is induced by transforming growth factor beta through a PKA-dependent pathway. J Biol Chem 2006;281:24365–24374. |
| 35. | Redinus K, Baek JH, Yalamanoglu A, Shin HKH, Moldova R, Harral JW, et al. An Hb-mediated circulating macrophage contributing to pulmonary vascular remodeling in sickle cell disease. JCI Insight 2019;4:e127860. |
| 36. | Tian W, Jiang X, Tamosiuniene R, Sung YK, Qian J, Dhillon G, et al. Blocking macrophage leukotriene b4 prevents endothelial injury and reverses pulmonary hypertension. Sci Transl Med 2013;5:200ra117. |
| 37. | Chan EK, Damoiseaux J, Carballo OG, Conrad K, de Melo Cruvinel W, Francescantonio PL, et al. Report of the first international consensus on standardized nomenclature of antinuclear antibody HEp-2 cell patterns 2014-2015. Front Immunol 2015;6:412. |
| 38. | Parthvi R, Sikachi RR, Agrawal A, Adial A, Vulisha A, Khanijo S, et al. Pulmonary hypertension associated with antiphospholipid antibody: call for a screening tool? Intractable Rare Dis Res 2017;6:163–171. |
| 39. | Tamura Y, Phan C, Tu L, Le Hiress M, Thuillet R, Jutant EM, et al. Ectopic upregulation of membrane-bound IL6R drives vascular remodeling in pulmonary arterial hypertension. J Clin Invest 2018;128:1956–1970. |
| 40. | Taraseviciene-Stewart L, Kasahara Y, Alger L, Hirth P, Mc Mahon G, Waltenberger J, et al. Inhibition of the VEGF receptor 2 combined with chronic hypoxia causes cell death-dependent pulmonary endothelial cell proliferation and severe pulmonary hypertension. FASEB J 2001;15:427–438. |
| 41. | Wang GS, Qian GS, Mao BL, Cai WQ, Chen WZ, Chen Y. [Changes of interleukin-6 and Janus kinases in rats with hypoxic pulmonary hypertension] [in Chinese]. Zhonghua Jie He He Hu Xi Za Zhi 2003;26:664–667. |
| 42. | Kasahara Y, Kimura H, Kurosu K, Sugito K, Mukaida N, Matsushima K, et al. MCAF/MCP-1 protein expression in a rat model for pulmonary hypertension induced by monocrotaline. Chest 1998;114(1 Suppl):67S. |
| 43. | Steiner MK, Syrkina OL, Kolliputi N, Mark EJ, Hales CA, Waxman AB. Interleukin-6 overexpression induces pulmonary hypertension. Circ Res 2009;104:236–244, 28p following 244. |
| 44. | Vroman H, Bergen IM, van Hulst JAC, van Nimwegen M, van Uden D, Schuijs MJ, et al. TNF-α-induced protein 3 levels in lung dendritic cells instruct TH2 or TH17 cell differentiation in eosinophilic or neutrophilic asthma. J Allergy Clin Immunol 2018;141:1620–1633, e12. |
| 45. | Hautefort A, Girerd B, Montani D, Cohen-Kaminsky S, Price L, Lambrecht BN, et al. T-helper 17 cell polarization in pulmonary arterial hypertension. Chest 2015;147:1610–1620. |
| 46. | Ross DJ, Strieter RM, Fishbein MC, Ardehali A, Belperio JA. Type I immune response cytokine-chemokine cascade is associated with pulmonary arterial hypertension. J Heart Lung Transplant 2012;31:865–873. |
| 47. | Theisen D, Murphy K. The role of cDC1s in vivo: CD8 T cell priming through cross-presentation. F1000 Res 2017;6:98. |
| 48. | Austin ED, Rock MT, Mosse CA, Vnencak-Jones CL, Yoder SM, Robbins IM, et al. T lymphocyte subset abnormalities in the blood and lung in pulmonary arterial hypertension. Respir Med 2010;104:454–462. |
| 49. | Rich S, Kieras K, Hart K, Groves BM, Stobo JD, Brundage BH. Antinuclear antibodies in primary pulmonary hypertension. J Am Coll Cardiol 1986;8:1307–1311. |
| 50. | Dib H, Tamby MC, Bussone G, Regent A, Berezné A, Lafine C, et al. Targets of anti-endothelial cell antibodies in pulmonary hypertension and scleroderma. Eur Respir J 2012;39:1405–1414. |
| 51. | Heink S, Yogev N, Garbers C, Herwerth M, Aly L, Gasperi C, et al. Trans-presentation of IL-6 by dendritic cells is required for the priming of pathogenic TH17 cells. Nat Immunol 2017;18:74–85. |
* Co–senior authors.
Supported by the Dutch Heart Foundation (2016T052) (M.K.). T.K. was supported by an unrestricted grant by Actelion Pharmaceuticals and the Pulmonary Hypertension Patient Association. D.M. was supported by the Netherlands CardioVascular Research Initiative, the Dutch Heart Foundation, the Dutch Federation of University Medical Centers, the Netherlands Organization for Health Research and Development, and the Royal Netherlands Academy of Science (CVON-2012-08 PHAEDRA). H.J.B. was supported by the Netherlands CardioVascular Research Initiative, the Dutch Heart Foundation, Dutch Federation of University Medical Centres, the Netherlands Organisation for Health Research and Development, and the Royal Netherlands Academy of Sciences (CVON-2012-08 PHAEDRA, CVON-2018-29 PHAEDRA-IMPACT, and CVON-2017-10 Dolphin-Genesis).
Author Contributions: T.K., J.A.C.v.H., K.A.B., M.K., and R.W.H. designed the experiments. T.K., J.A.C.v.H., T.D., S.F.H.N., D.M., I.M.B., M.A.d.R., H.J.B., L.B., and J.G.J.V.A. performed experiments and analyzed/interpreted data. G.v.L. provided critical mouse strains. T.K., K.A.B., M.K., and R.W.H. wrote the manuscript. All authors read and approved the final manuscript.
This article has a data supplement, which is accessible from this issue’s table of contents at www.atsjournals.org.
Originally Published in Press as DOI: 10.1165/rcmb.2019-0443OC on August 5, 2020
Author disclosures are available with the text of this article at www.atsjournals.org.
