Pyrin-only protein 2 limits inflammation but improves protection against bacteria

Abstract

Pyrin domain-only proteins (POPs) are recently evolved, primate-specific proteins demonstrated in vitro as negative regulators of inflammatory responses. However, their in vivo function is not understood. Of the four known POPs, only POP2 is reported to regulate NF-κB-dependent transcription and multiple inflammasomes. Here we use a transgenic mouse-expressing POP2 controlled by its endogenous human promotor to study the immunological functions of POP2. Despite having significantly reduced inflammatory cytokine responses to LPS and bacterial infection, POP2 transgenic mice are more resistant to bacterial infection than wild-type mice. In a pulmonary tularaemia model, POP2 enhances IFN-γ production, modulates neutrophil numbers, improves macrophage functions, increases bacterial control and diminishes lung pathology. Thus, unlike other POPs thought to diminish innate protection, POP2 reduces detrimental inflammation while preserving and enhancing protective immunity. Our findings suggest that POP2 acts as a high-order regulator balancing cellular function and inflammation with broad implications for inflammation-associated diseases and therapeutic intervention.

Figure 1. Expression of POP2 in transgenic mice.

(a) Detection of POP2 Tg expression by RT–PCR in perfused organs (Supplementary Fig. 3). (b,c) Relative POP2 Tg basal mRNA levels (normalized to GAPDH) in various tissue and cellular compartments by qPCR. The data are shown as mean fold change relative to LMC±s.d. (n=4). (d) The frequency (%) of major myeloid-lineage cells (CD11c⁺ dendritic cells, CD11b⁺ myeloid cells, F4/80⁺ macrophages (CD11b⁺CD11c⁻Gr-1⁻F4/80⁺) and Gr-1⁺ neutrophils (CD11b⁺CD11c⁻Gr-1⁺F4/80⁻)) in spleen. The data are shown as mean±s.d. (n=4). To determine significance, Student’s t-test was performed. *P<0.05. LN, lymph nodes; mLN, mesenteric LN; iLN, inguinal LN; bLN, brachial LN; BM, bone marrow.

Figure 2. POP2 inhibits cytokine production in mouse macrophages.

(a–c) Mean level±s.d. of TNF and IL-6 measured from culture supernatants of (a) BMDM, (b) peritoneal macrophages or (c) splenic macrophages from POP2Tg (n=4) and LMC (n=4) mice stimulated with LPS (100 ng ml⁻¹) for 24 h. (d–f) Mean level±s.d. of IL-1β and IL-18 measured from culture supernatants of (d) BMDM, (e) peritoneal macrophages or (f) splenic macrophages from POP2Tg (n=4) and LMC (n=4) mice stimulated with LPS (100 ng ml⁻¹) for 24 h plus ATP (5 mM) for the last 30 min. (g,h) Mean level±s.d. of IL-1β, IL-18 and TNF measured from culture supernatants of BMDM from POP2Tg (n=4) and LMC (n=4) mice infected for 24 h with (g) F. tularensis LVS (MOI=100) or (h) F. novicida (MOI=100). To determine significance, Student’s t-test was used. *P<0.05.

Figure 3. POP2 regulates the AIM2 inflammasome.

(a) Mean level±s.d. of mature IL-1β measured from culture supernatants of pcDNA3 or POP2Tg J774A.1 transfectants treated with poly (dA:dT) (1 μg ml⁻¹) for 18 h or (b) infected with F. novicida (MOI=100 for 24 h), L. monocytogenes (MOI=5 for 6 h) or S. Typhimurium (MOI=10 for 6 h) (n=3). (c) Immunohistochemical detection of Asc-speck formation (indicated by arrows) and per cent Asc specks in J774A.1 transfectants infected with F. novicida (MOI=100). (n>100 cells) (d) Mean level±s.d. of IL-1β and IL-18 measured from culture supernatants of BMDM from POP2Tg (n=4) and LMC (n=4) mice treated with poly (dA:dT) (1 μg ml⁻¹) for 18 h. (e) Mean level±s.d. of mature IL-1β measured from culture supernatants of HEK293T cells transiently transfected with human ASC, AIM2 and/or POP2 constructs for 20 h (n=3). To determine significance, Student’s t-test was used. *P<0.05, ND, none detected.

Figure 4. POP2 tempers the in vivo inflammatory cytokine response.

(a) Mean serum levels±s.d. of IL-1β, IL-18, TNF, IL-6, IL-12p40, IL-12p70, MCP-1 and IL-10 in POP2 (n=4) and LMC (n=4) mice receiving a sublethal dose (15 mg kg⁻¹) of LPS for 24 h by i.p. injection. To determine significance, Student’s t-test was used. (b) Survival of POP2 Tg (n=14) and LMC (n=14) mice after i.p. injection with a lethal dose (50 mg kg⁻¹) of LPS. Data represent per cent of surviving mice from at least two independent experiments. To determine significance, Log-rank test was used. *P<0.05, **P<0.01, ***P<0.001.

Figure 5. POP2 improves resistance to bacterial infections.

(a–c) Per cent survival and mean body weight loss±s.d. in (a) POP2 (n=12) and LMC (n=14) mice after subcutanesous infection with 1.5 × 10⁵ c.f.u. of F. novicida, (b) POP2 (n=10) and LMC (n=10) mice after intranasal (i.n.) infection with 500 c.f.u. (LD50) of F. tularensis (left panel) and in POP2 (n=12) and LMC (n=12) mice after i.n. infection with 1,000 c.f.u. (lethal) of F. tularensis (right panel) and (c) POP2 (n=12) and LMC (n=12) mice after i.n. infection with 1 × 10⁵ c.f.u. of S. pneumoniae. Data represent per cent of surviving mice and accompanying weight loss from at least two independent experiments. To determine significance for survival data, the Log-rank test was used, and for weight data, Student’s t-test was used. P values are shown where significance was detected.

Figure 6. POP2 limits acute inflammation and increases IFN-γ in pulmonary tularaemia.

(a) Mean total numbers±s.d. of Ly6G⁺ neutrophils and F4/80⁺ macrophages in lungs of POP2 (n=4) and LMC (n=4) mice infected with F. tularensis (Ft) LVS (MOI=100). (b) Mean total numbers±s.d. of polymorphonucleated-MDSC (pMDSC) and mono-nucleated-MDSC (mMDSC) in lungs of POP2 (n=4) and LMC (n=4) mice infected with Ft LVS (MOI=100). (c) Representative histologic lung sections from mice infected with Ft LVS showing inflammatory foci and necrosis (‘necrotizing inflammation’) at 6 dpi in LMC mice (top panel) and POP2 mice (bottom panel) (haematoxylin and eosin; Scale bar, 100 μm). Lung pathology scores for LMC (n=6) and POP2 (n=6) mice infected with Ft LVS (1,000 c.f.u.; LD100) (right graph) were calculated by analysis of lung sections for location, type and extent of inflammation and necrosis (see ‘Methods’ section). (d–g) POP2 (n=6) and LMC (n=6) mice were infected with Ft LVS (1,000 c.f.u.; LD100) and shown are (d) mean levels±s.d. of IL-1β, IL-18, IL-6 and TNF from lung homogenates, (e) bacterial burden in lungs, (f) mean levels±s.d. of IFN-γ measured from lung homogenates and (g) mean total numbers±s.d. of IFN-γ⁺ immune cells in lungs. (h) Mean bacterial count (c.f.u.)±s.d. from POP2 (n=4) and LMC (n=4) BMDM infected with Ft LVS (MOI=100) with and without administration of recombinant IFN-γ (100 IU). (i) Per cent survival of POP2 (n=12) and LMC (n=12) mice treated with 300 μg anti-IFN-γ antibody (catalogue number BE0055, BioXCell) on days −1, 2 and 4 or isotype control (catalogue number BE0088, BioXCell) and infected with Ft LVS (500 c.f.u., LD50). Survival of infected POP2 mice receiving isotype control differs significantly from that of POP2 mice receiving anti-IFN-γ (P<0.001) To determine significance, for a,b,d and h, Student’s t-test was used (*P<0.05); for c, scores were evaluated using the Mann–Whitney test (*P<0.05); for e–g, Student’s t-test was used (*P<0.05, **P<0.01), and for i, the Log-rank test, was used.