Peptidylarginine Deiminase 2 in Murine Antiviral and Autoimmune Antibody Responses

Figures

• 

  Figure 1. In collagen-induced arthritis (CIA), PAD2 is required for normal anti-collagen IgG levels, but not numbers of anti-collagen antibody secreting cells. CIA was induced in PAD2+/+ and PAD2−/− mice on the DBA/1J background with the first injection at week zero. (a) Sera were collected at indicated time points post-CIA induction and anti-collagen IgG levels quantified by ELISA and expressed as a value relative to a CIA serum standard (mean ± SEM, n = 23 mice, 6 experiments). Gray arrows indicate CIA-inducing injections. Bone marrow-derived anti-collagen IgG antibody secreting cells (ASCs) were quantified by (b) ELISpot and (c) anti-collagen ELISA limiting dilution assay (LDA) 17 weeks post-CIA induction (symbols indicate ASC numbers in individual mice, horizontal bar indicates mean ± SEM, n = 10 mice, 3 experiments). For all panels, PAD2+/+ versus PAD2−/− data were compared by paired t test and ∗p < 0.05.

• 

  Figure 2. T cells in collagen-induced arthritis (CIA) are unaffected by loss of PAD2. Seventeen weeks after the first injection to induce CIA, splenocytes were harvested and stained for flow cytometry. Debris, clustered cells, and dead cells were excluded using forward scatter, side scatter, and a viability dye. (a) Percent of CD3+ splenocytes that are CD4+ and CD8+ (n = 10 mice, 3 experiments). Activation status of these CD3+ cells was evaluated with (b) representative plots as well as mean percent ± SEM of (c) CD4+ and (d) CD8+ populations that are effector (CD62L-CD44+), memory (CD62L+CD44+), and naïve (CD62L+CD44-) shown. The percent of CD4+ cells that produce IL-17A, IL-4, and IFNγ was also quantified with (e) representative plots and (f) mean percent ± SEM graphed (n = 12 mice, 3 experiments). For all panels, PAD2+/+ versus PAD2−/− data were compared by paired t test, and no comparisons were significant.

• 

  Figure 3. Arthritis severity in collagen-induced arthritis (CIA) is unaltered in the absence of PAD2. Six weeks after the first injection to induce CIA, (a) clinical scores and (b) paw thickness were assessed weekly (n = 23 mice, 6 experiments). Controls (n = 9), which received no injections and were cohoused with experimental mice, were also evaluated. At 17 weeks after the first injection, the front paws were fixed, embedded, sectioned, and stained with hematoxylin and eosin. (c) Representative images at 100x. Bar indicates 200 microns. (d) The extent of pannus development, cartilage destruction, and ankylosis were scored in a blinded manner (n = 23 mice, 6 experiments). All graphs depict mean ± SEM. PAD2+/+ versus PAD2−/− data were compared by paired t test, and no comparisons were significant.

• 

  Figure 4. DBA/1J mice are highly susceptible to influenza infection. Mice were infected with different doses of PR8 or PR8-OVA followed by daily health monitoring. Kaplan–Meier curves depict survival of DBA/1J mice after (a) PR8 infection (100 PFU n = 8, 33.3 PFU n = 2, 10 PFU n = 2, 3 PFU n = 2) or (b) PR8-OVA infection (50 PFU n = 3, 10 PFU n = 2).

• 

  Figure 5. PAD2 is required for hemagglutination inhibiting antibody titers in DBA/1J mice after influenza infection. PAD2+/+ and PAD2−/− mice on the DBA/1J background were infected with PR8-OVA. (a) Sera were collected at the indicated time points and used in a hemagglutination inhibition (HI) assay (n = 18, 5 experiments). (b) Sera at 12 weeks postinfection were used to detect anti-hemagglutinin (HA) IgG, IgA, and IgM by ELISA (n = 17‐18). (c) Body weights were obtained daily for 14 days postinfection and reported as percent of weight on the day of infection (n = 18, 5 experiments). For all panels, PAD2+/+ versus PAD2−/− data were compared by paired t test, graphs depict mean ± SEM, and ∗p < 0.05.

• 

  Figure 6. PAD2 is required for normal protection from influenza reinfection. PAD2+/+ and PAD2−/− mice on the DBA/1J background were infected with PR8-OVA, followed by lethal rechallenge 12 weeks later with PR8. (a) Sera were collected at the indicated time points and used in a hemagglutination inhibition (HI) assay (n = 11, 3 experiments). (b) Sera from two weeks postchallenge were used to detect anti-hemagglutinin (HA) IgG, IgA, and IgM by ELISA (n = 9‐11). (c) Body weights were obtained daily for 14 days postchallenge infection and reported as percent of weight on the day of challenge infection (n = 11, 3 experiments). For all panels, PAD2+/+ versus PAD2−/− data were compared by paired t test, graphs depict mean ± SEM, and ∗p < 0.05.