Galectin-3 functions as an alarmin: pathogenic role for sepsis development in murine respiratory tularemia

Abstract

Sepsis is a complex immune disorder with a mortality rate of 20-50% and currently has no therapeutic interventions. It is thus critical to identify and characterize molecules/factors responsible for its development. We have recently shown that pulmonary infection with Francisella results in sepsis development. As extensive cell death is a prominent feature of sepsis, we hypothesized that host endogenous molecules called alarmins released from dead or dying host cells cause a hyperinflammatory response culminating in sepsis development. In the current study we investigated the role of galectin-3, a mammalian β-galactoside binding lectin, as an alarmin in sepsis development during F. novicida infection. We observed an upregulated expression and extracellular release of galectin-3 in the lungs of mice undergoing lethal pulmonary infection with virulent strain of F. novicida but not in those infected with a non-lethal, attenuated strain of the bacteria. In comparison with their wild-type C57Bl/6 counterparts, F. novicida infected galectin-3 deficient (galectin-3(-/-)) mice demonstrated significantly reduced leukocyte infiltration, particularly neutrophils in their lungs. They also exhibited a marked decrease in inflammatory cytokines, vascular injury markers, and neutrophil-associated inflammatory mediators. Concomitantly, in-vitro pre-treatment of primary neutrophils and macrophages with recombinant galectin-3 augmented F. novicida-induced activation of these cells. Correlating with the reduced inflammatory response, F. novicida infected galectin-3(-/-) mice exhibited improved lung architecture with reduced cell death and improved survival over wild-type mice, despite similar bacterial burden. Collectively, these findings suggest that galectin-3 functions as an alarmin by augmenting the inflammatory response in sepsis development during pulmonary F. novicida infection.

Figure 1. Upregulated expression and extracellular release of Galectin-3 in lungs during respiratory F. novicida infection.

(A) Total RNA was extracted by Trizol method from lungs harvested at the indicated times after infection with the Wild-type bacteria (WT) or from mice vaccinated with an attenuated mutant strain followed by challenge with WT bacteria (Mut/WT mice). The mRNA levels of Galectin-3 were analyzed by real-time PCR as described in Materials and Methods and are expressed as fold changes over the levels in mock control mice. Data shown are the averages of 3–4 mice per group. Statistically significant differences are denoted by asterisks (**p<0.005). (B) In-situ IF staining of frozen lung sections from mock infected and WT U112 infected or Mut/WT mice harvested at 3 d. p.i Lung harvested 3 weeks after vaccination with the mutant alone (Mut-3 wk) served as controls for Mut/WT mice. The sections were stained for galectin-3 (red) using a purified rat anti-mouse galectin-3 antibody followed by Alexa-546 conjugated chicken anti-rat antibody. Nuclei (blue) were stained with 4′6′ diamidino-2-phenylindol-dilactate (DAPI). Magnification×200. Insets depict extracellular galectin-3 in WT F. novicida infected mouse lungs (B2’) and cytosolic galectin-3 in Mut/WT (B4’) mouse lungs.

Figure 2. Galectin-3^−/− mice display reduced levels of inflammatory mediators in lungs after pulmonary infection with F.n.

The lungs from WT mock infected (WT-M), galectin-3^−/− mock infected (Gal-3^−/−M), WT F. novicida infected (WT-Inf) or galectin-3^−/− F. novicida infected mice (Gal3^−/−Inf) were harvested at 3 d.p.i., homogenized with protease inhibitors in PBS and analyzed commercially for rodent multi-analyte profiles (Rules-Based Medicine, Austin, TX). (A), levels of vascular injury markers; (B), levels of inflammatory cytokines; and (C), levels of neutrophil attractant chemokines and activation markers in lung homogenates. Results shown are from 3–4 mice per group from 3 different experiments. CRP; C-reactive protein, MMP-9; matrix metalloproteinase 9, MPO; myeloperoxidase. *p<0.05; **p<0.005.

Figure 3. Galectin-3^−/− mice display reduced accumulation of neutrophils in lungs during F.n. infection.

Frozen sections of lungs harvested at 3 d. p.i. from mock infected and F. novicida infected WT or galectin-3^−/− mice were co-stained with antibodies against myeloid cell markers CD11b (red) and Gr1 (green). A high co-expression of both markers is depicted by yellow color in infected WT lungs while cells infiltrating lungs of galectin-3^−/− mice exhibited expression of only CD11b. Nuclei (blue) were stained with 4′6′ diamidino-2-phenylindol-dilactate (DAPI). Magnification×200. Asterisks depict lesions in the lungs.

Figure 4. Galectin-3 regulates F.n. infection induced inflammatory response in-vitro.

(A). Bone marrow derived macrophages (BMDMs) were isolated from wild-type and galectin-3^−/− mice as described in Methods. The cells were infected with wild-type F.n. Strain U112 at MOI of 50 and the culture supernatents were collected 24 h after infection. The amount of TNF-α and IL-6 were measured in the supernatents by Sandwich ELISA. (B). BMDMs from C57Bl/6 wild-type mice were infected with wild-type F.n. Strain U112 at an MOI of 50 with or without pretreatment with 10 µg/ml of purified recombinant galectin-3. Culture supernatants were collected 24 h after infection and the amount of TNF-α and IL-6 were measured by ELISA. The experiment was repeated three times with similar results. (C). Peritoneal neutrophils were isolated from mice 12–14 h after injection with 4% thioglycollate and were stimulated with F. novicida at an MOI 50 with or without pre-treatment with purified recombinant galectin-3 (10 µg/ml). Stimulation with galectin-3 alone or phorbol myristate acetate (PMA, 10 ng/ml) was used as a control. Production of reactive oxygen species was measured one hour post-stimulation by flow-cytometry using Fc-OxyBURST dye following the manufacturer’s instructions. Numbers in black on the plots depict percent of ROS positive cells and the numbers in green represent median fluorescence intensity (MFI) of individual cells. Dot plots from a representative of 3 independent experiments are shown. Statistical analysis between the data sets was performed by Student’s t test where **p<0.005; ***p<0.001.

Figure 5. Galectin-3 deficiency leads to improved lung pathology, reduced leukocyte accumulation and reduced cell death upon pulmonary F.n. infection.

(A) Lungs from mock infected and F.n. infected wild-type (WT) or galectin-3^−/− mice were harvested at the septic phase (3 d. p.i.), embedded in optimal-cutting-temperature (OCT) compound, and sectioned as described in Materials and Methods. The frozen sections were stained with Hematoxylin and Eosin. The images obtained are representatives of three experiments performed, and in each experiment each group contained three mice. Magnification, ×200. (B). Lungs from mock infected and F.n. infected WT or galectin-3^−/− mice were harvested 3 days after intranasal infection. Total immune cells infiltrating the lungs were isolated by collagenase treatment of lungs as described in Materials and Methods. Total numbers of viable immune cells were counted by trypan blue exclusion staining (n = 5–6). Statistical analysis between the data sets was performed by Student’s t test where **p<0.005. (C). Frozen lung sections from mock infected and Francisella infected WT or galectin-3^−/− mice were processed for in-situ TUNEL staining for detection of DNA fragmentation (red) in nuclei. Nuclei (blue) were stained with 4′,6′-diamidino-2-phenylindole dilactate. Magnification, ×100.

Figure 6. Galectin-3^−/− mice show improved survival during pulmonary F.n. infection.

(A). Fifteen C57Bl/6 WT and 17 galectin-3^−/− mice in 3 separate experiments were inoculated intranasally with F. novicida and were monitored for survival daily for 2 weeks. The improved survival of galectin-3^−/− mice compared to WT mice was statistically significant, as determined by Kaplan-Meier log-rank analysis (P value*** = 0.0003). (B) Bacterial burdens in lungs harvested from F. novicida infected WT and galectin-3^−/− mice at 3 d.p.i. Lungs and liver were harvested from infected mice, homogenized as described in Materials and Methods. Tissue homogenates and blood harvested at the same time, were serially diluted and plated on TSA plates to enumerate bacterial burdens. In this representative of three independent experiments, each group contained three to five mice.