Influenza A (H1N1) virus infection triggers severe pulmonary inflammation in lupus-prone mice following viral clearance

Abstract

Each year, up to one fifth of the United States population is infected with influenza virus. Although mortality rates are low, hundreds of thousands are hospitalized each year in the United States. Specific high risk groups, such as those with suppressed or dysregulated immune systems, are at greater danger for influenza complications. Respiratory infections are a common cause of hospitalizations and early mortality in patients with systemic lupus erythematosus (SLE); however, whether this increased infection risk is a consequence of the underlying dysregulated immune background and/or immunosuppressing drugs is unknown. To evaluate the influenza immune response in the context of lupus, as well as assess the effect of infection on autoimmune disease in a controlled setting, we infected lupus-prone MRL/MpJ-Fas(lpr) mice with influenza virus A PR/8/34 H1N1. Interestingly, we found that Fas(lpr) mice generated more influenza A virus specific T cells with less neutrophil accumulation in the lung during acute infection. Moreover, Fas(lpr) mice produced fewer flu-specific IgG and IgM antibodies, but effectively cleared the virus. Further, increased extrinsic apoptosis during influenza infection led to a delay in autoimmune disease pathology with decreased severity of splenomegaly and kidney disease. Following primary influenza A infection, Fas(lpr) mice had severe complications during the contraction and resolution phase with widespread severe pulmonary inflammation. Our findings suggest that influenza infection may not exacerbate autoimmune pathology in mice during acute infection as a direct result of virus induced apoptosis. Additionally, autoimmunity drives an enhanced antigen-specific T cell response to clear the virus, but persisting pulmonary inflammation following viral clearance may cause complications in this lupus animal model.

Keywords: Influenza; Lupus; MRL-Fas(lpr); Pulmonary inflammation; SLE.

Figure 1. MRL-Fas^lpr mice clear IAV infection with reduced morbidity and less pulmonary pathology

Ten to 12 week old B6, MRL, and MRL-Fas^lpr mice (females only) were infected i.n. with 200 EID⁵⁰ of PR8 IAV or left uninfected. All PBS mice in Figure 1 are MRL-Fas^lpr mice. Weights were measured daily and the percent starting weight was calculated and plotted over time for 20 days (a). In separate experiments, on days 0, 7, 10, or 20 post infection, the middle right lobe was collected to access viral burden by expression of the influenza M1 gene by RTPCR (b). Formalin-fixed paraffin embedded (FFPE) serial sections were stained with hematoxylin and eosin (H&E) at day 7 post infection to examine pulmonary pathology. Representative images were captured with .5X and 10X objective with original magnification at x5 and x100, respectively. Black boxes in .5X image represent the area of 10X image (c). The severity of pulmonary inflammation was blindly scored on a scale of 0-4 on day 7 post infection, with a score of 4 being the most severe (d). The number of specific cell populations in the lung was determined at day 10 post infection by flow cytometry (e). CD4+ and CD8+ T cells were gated using forward by side scatter, then by gating on CD3+ cells following by gating on the respective CD4 or CD8 marker. B cells were gated on forward by side scatter and CD19+ and B220+ cells. Total myeloid cells were gated by size using forward by side scatter and CD11c+ and/or CD11b+ expression and neutrophils were further calculated based on CD11b+ and GR1+ expression. Total frequencies were used to back calculate total numbers per lung. The data points represent the mean (±SD) of values from 4-6 mice (a-e). ns=not significant, *p≤0.05, **p≤0.005, ***p≤0.0005.

Figure 2. Increased IAV-specific T cells and higher TNFα levels are found in the IAV infected lung of MRL-Fas^lpr mice

Mice were infected as in Figure 1. All PBS mice in Figure 2 a-d are MRL-Fas^lpr mice. The number of antigen (Ag)-specific cells were determined by flow cytometry on day 7 post infection. The total number CD8+ T cells producing TNFα (a) and granzyme B (c), and CD4+ T cells producing TNFα (b) and IFNγ (d) following stimulation with inactivated IAV are shown. Total cytokine levels of TNFα (e), IFNγ (f) and IL-10 (g) in the lung were determined from lung homogenates by ELISA. The data points represent the mean (±SD) of values from 4-6 mice (a-g). ns= not significant, *p≤0.05, **p≤0.005, ***p≤0.0005.

Figure 3. Virus-specific antibodies are severely reduced while autoantibody production remains unaltered in IAV-infected MRL-Fas^lpr mice

All mice were infected as in Figure 1. At days 10 (black) and 20 (white) post infection blood was collected and processed for serum. Serum was either treated with RDE for influenza antibody analysis or used untreated for autoantibody analysis. Total IgM (a) and IgG (b) IAV-specific antibodies were determined by ELISA. Neutralization capacity was determined by the hemagglutination inhibition assay and HAI titer was calculated (c). Dilutions were made and total IgG (d) and autoantibody concentrations for total anti-nuclear antibodies (e) and anti-dsDNA antibodies (f) were analyzed in uninfected and infected mouse sera by ELISA. The data points represent the mean (±SD) of values from 4-6 mice (a-f). ns= not significant, *p≤0.05.

Figure 4. IAV-infected MRL-Fas^lpr mice have temporary reduction in autoimmune organ damage

Splenomegaly and kidney function were assessed at day 10 and day 20 post infection in IAV-infected mice as described in Figure 1. Spleens were taken and weighed before further analysis (a). Serum creatinine was measured in IAV infected and PBS treated mice at day 20 post infection (b). Representative kidney sections stained with periodic acid-Schiff (PAS) from B6, MRL, and MRL-Fas^lpr mice that were infected or not infected with IAV at 10-12 weeks of age (c). Representative glomeruli are shown in each section. Kidney sections were scored in a blinded analysis in PAS stained kidney sections based on both acute (mesangial matrix expansion, focal necrosis and glomerular hypercellularity) and chronic (interstitial fibrosis, medullary inflammation, glomerulosclerosis, and tubular atrophy) glomerulonephritis parameters on a scale of 0-10. Scores were combined for each kidney section to give a cumulative acute (d) and chronic (e) kidney disease score. The data points represent the mean (±SD) of values from 4-6 mice (a-e). ns= not significant, *p≤0.05, **p≤0.005, ***p≤0.0005.

Figure 5. Extrinsic influenza-induced apoptosis alleviates splenomegaly in MRL-Fas^lpr mice

Mice were infected as previously described (Figure 1). Spleens from infected and uninfected mice at day 10 post infection were formalin-fixed paraffin embedded and terminal deoxynucleotidyltransferase-mediated dUDP nick-end labeling (TUNEL, green) stained to determine cells undergoing apoptosis (red arrows point to TUNEL expressing cells). Representative images are shown (20x objective) (a). Original magnification x200. Apoptotic cells were quantitated by counting the number of TUNEL+ cells per 20x field (10 images taken of individual mouse lobes) (b). IAV infected MRL-Fas^lpr mice were treated with caspase-8 inhibitor (Z-IETD-FMK, 0.1 μM) or a negative control (Z-FA-FMK) daily for 6 days starting at day 4 post infection. Mice were sacrificed at day 10 post infection and spleens were weighed (c). PBS treated mice are MRL-Fas^lpr mice (c). The data points represent the mean (±SD) of values from 4-6 mice (a-c). ns= not significant, *p≤0.05, **p≤0.005.

Figure 6. Lupus prone MRL-FAS^lpr mice are unable to resolve pulmonary inflammation following IAV infection

MRL and MRL-Fas^lpr mice were infected with IAV PR8 with 200 EID⁵⁰. Mice were sacrificed at day 40 post infection and lung tissue was formalin fixed and paraffin embedded for pathology analysis. Slides were H&E stained (a) and scored for disease severity on a scale of 0-4 (b). Representative pictures were taken using a Zeiss Axioplan Microscope at 0.5x and 10x objective, original magnification of x50 and x100, respectively. Black boxes in .5X image represent the area of 10X image. Spleen weight was measured in grams for each individual mouse spleen (c). Kidneys were formalin fixed and paraffin embedded before staining with PAS. Both acute and chronic disease parameters (as mentioned in Figure 5) were blindly scored for each mouse kidney and scores were combined for a cumulative score of acute (d) and chronic (e) glomerulonephritis. The data points represent the mean (±SD) of values from 4-6 mice (a-e). ns= not significant, *p≤0.05, **p≤0.005, ***p≤0.0005.