Granzyme A inhibition reduces inflammation and increases survival during abdominal sepsis

Abstract

Aims: Peritonitis is one of the most common causes of sepsis, a serious syndrome characterized by a dysregulated systemic inflammatory response. Recent evidence suggests that Granzyme A (GzmA), a serine protease mainly expressed by NK and T cells, could act as a proinflammatory mediator and could play an important role in the pathogenesis of sepsis. This work aims to analyze the role and the therapeutic potential of GzmA in the pathogenesis of peritoneal sepsis. Methods: The level of extracellular GzmA as well as GzmA activity were analyzed in serum from healthy volunteers and patients with confirmed peritonitis and were correlated with the Sequential Organ Failure Assessment (SOFA) score. Peritonitis was induced in C57Bl/6 (WT) and GzmA^-/- mice by cecal ligation and puncture (CLP). Mice were treated intraperitoneally with antibiotics alone or in combination serpinb6b, a specific GzmA inhibitor, for 5 days. Mouse survival was monitored during 14 days, levels of some proinflammatory cytokines were measured in serum and bacterial load and diversity was analyzed in blood and spleen at different times. Results: Clinically, elevated GzmA was observed in serum from patients with abdominal sepsis suggesting that GzmA plays an important role in this pathology. In the CLP model GzmA deficient mice, or WT mice treated with an extracellular GzmA inhibitor, showed increased survival, which correlated with a reduction in proinflammatory markers in both serum and peritoneal lavage fluid. GzmA deficiency did not influence bacterial load in blood and spleen and GzmA did not affect bacterial replication in macrophages in vitro, indicating that GzmA has no role in bacterial control. Analysis of GzmA in lymphoid cells following CLP showed that it was mainly expressed by NK cells. Mechanistically, we found that extracellular active GzmA acts as a proinflammatory mediator in macrophages by inducing the TLR4-dependent expression of IL-6 and TNFα. Conclusions: Our findings implicate GzmA as a key regulator of the inflammatory response during abdominal sepsis and provide solid evidences about its therapeutic potential for the treatment of this severe pathology.

Keywords: Granzyme A; cecal ligation and puncture.; inflammation; peritonitis; sepsis.

Figure 1

Increased levels of extracellular GzmA are observed in patients with abdominal sepsis. Serum levels of GzmA were analysed by ELISA in healthy donors (n = 10) and compared with GzmA levels from patients with a diagnosis of abdominal sepsis (n = 10) at diagnosis (time 0 = disease onset) and during the first 72 h. Statistical analyses were performed by one-way ANOVA test with Bonferroni's post-test *P < 0.05; **P < 0.01. DL (ELISA Detection Limit).

Figure 2

The absence of granzyme A increases survival in mice during CLP and E. coli-induced sepsis. Sepsis was induced by CLP in WT and GZMA^-/- mice (A) and GZMA^-/- and GZMA^+/+ littermates (B) as described in materials and methods. After 6 h, a mixture of antibiotics, ceftriaxone (30 mg/kg) + Metronidazole (12.5 mg/kg) was administered i.p. every 24 h for 5 days. WT and GZMA^-/- sham operated mice underwent the same procedure but without the ligation and puncture of the cecum. Survival was monitored during 14 days. The data correspond to the indicated number of biological replicates (individual mice) from three independent experiments. Statistical analyse was performed using logrank and Gehan-Wilcoxon test. *p < 0.05. (C) WT and GZMA^-/- mice were infected i.p. with 2 x 10⁸ CFU of E. coli as described in materials and methods. The data corresponds to the indicated number of biological replicates from two independent experiments. Statistical analyse was performed using logrank and Gehan-Wilcoxon test. *p < 0.05. (D) An in vitro analysis was performed to determine the capacity of GzmA in the control of bacterial pathogens. M1 macrophages were differentiated as described in material and methods. Macrophages were incubated with GZMA (300 nM), infected with E. coli (MOI 1:100) and incubated for 4, 24 and 48 h at 37 °C. After incubation time cells were lysed and the number of CFU was determined. Data are presented as mean ± SEM from 3 independent experiments.

Figure 3

The absence of granzyme A reduces the level of proinflammatory cytokines during sepsis induced by CLP. Sepsis was induced by CLP in WT and GZMA^-/- mice as described in materials and methods. After 6 h, a mixture of antibiotics, ceftriaxone (30 mg/kg) + metronidazole (12.5 mg/kg) was administered every 12 h. WT and GZMA^-/- sham operated mice underwent the same procedure but without CLP. After 6 or 24 h of sepsis induction, mice were sacrificed and the levels of IL-1α, IL-1β, TNFα and IL-6 in plasma and peritoneal lavage fluids were determined by ELISA. Data are presented as mean ± SEM of 5 (sham) or at least 8 (CLP) different biological replicates (individual mice) from 3 independent experiments. Statistical analysis was performed by one-way ANOVA test with Bonferroni's post-test *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 4

GzmA is not involved in the control of bacterial pathogens during sepsis induced by CLP. Sepsis was induced by CLP in WT and GZMA^-/- mice as described in materials and methods. After 6 h, a mixture of antibiotics, ceftriaxone (30 mg/kg) + Metronidazole (12.5 mg/kg) was administered i.p. every 12 h. After 6 and 24 h of sepsis induction, a group of mice were sacrificed and the total number of CFU from aerobic bacteria was determined in peritoneal fluids, blood, spleen liver and lung (A). The most frequent strains were identified by MALDI-TOF mass spectrometry and the number of CFU of these strains was determined in peritoneal lavage fluids, blood, spleen liver and lung (B). Data are presented as mean ± SEM from 5 biological replicates (individual mice) in each group.

Figure 5

GzmA expression is increased in NK cells from septic mice. Sepsis was induced by CLP in WT and GZMA^-/- mice as described in materials and methods. Sham WT operated mice underwent the same procedure but without the ligation and puncture of the cecum. After 24 h mice were sacrificed and (A) the percentage of NK cells (NK1.1⁺CD3^-), NKT cells (NK1.1⁺ CD3⁺), CD8⁺ T cells (CD8⁺ CD3⁺) and CD4⁺ T cells (CD4⁺, CD3⁺) and (B) the intracellular expression of GzmA on NK cells, NKT cells, CD8⁺ T cells and CD4⁺ T cells were analysed in splenocytes and PBLs by flow cytometry. A representative experiment of GzmA expression in spleen is shown via dot plot (A). Numbers show the cell percentage in each quadrant. Data in graphs represent the mean ± SEM of the percentage of GzmA positive cells of each phenotype (right) and the percentage of each subtype (left) from 2 independent experiments. (B) Statistical analysis was performed by unpaired student's t test. **p < 0.01, ***p < 0.001.

Figure 6

Inhibition of extracellular GzmA improves sepsis outcome and reduces inflammation. Sepsis was induced by CLP in B6 and GZMA^-/- mice as described in materials and methods. Immediately after surgery mice were treated with 40 µg of Serpinb6b in 100 µl of PBS (10 WT and 9 GZMA^-/- mice). This treatment was repeated 6 h later and once a day during 5 days. Control mice received only 100 µl of PBS i.p. (13 WT and 6 GZMA^-/- mice). After 6 h a mixture of antibiotics, ceftriaxone (30 mg/kg) + metronidazole (12.5 mg/kg) was administered i.p. once a day for 5 days. WT sham operated mice (4 WT mice) underwent the same procedure but without the ligation and puncture of the cecum. (A) Survival was monitored during 14 days. The data correspond to the indicated number of mice combined from two independent experiments. Statistical analyse was performed using logrank and Gehan-Wilcoxon test. *p < 0.05. (B) 24 h after sepsis induction mice were sacrificed and the levels of IL-6 in plasma and peritoneal fluids was determined by ELISA. Data are presented as mean ± SEM of 4 (Sham) or 6 (CLP) biological replicates from 2 independent experiments. Statistical analysis was performed by one-way ANOVA test with Bonferroni's post-test *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 7

Active extracellular GzmA induces the expression of IL-6 and TNFα in macrophages by a mechanism dependent of TLR4. (A) WT bone marrow differentiated macrophages were stimulated with LPS 100 ng/mL, active and inactive GzmA produced in E. coli, GzmA inactivated with sepinb6b and cathepsin C. After 24 h of incubation, supernatant was collected to determine the levels of IL-6 by ELISA. (B) GzmA activity was confirmed by an activity essay incubating active and inactive GzmA produced in E. coli, GzmA inactivated with sepinb6b and cathepsin C with a specific mouse GzmA substrate. Optical density at 405 nm was measured after 4, 12 and 24 h. (C) Macrophages differentiated from WT or TLR4^-/- mouse bone marrow were stimulated with active GzmA (300 nM) produced in E. coli or in P. pastoris, GzmA inactivated with serpinb6b, LPS 100 ng/mL or S. aureus HK (1 x10⁶ CFU/mL). After 24 h of incubation, the supernatant was collected to determine the levels of IL-6 and TNFα by ELISA. Data are represented as the mean ± SEM of two independent experiments performed by duplicate. Statistical analyses were performed by one-way ANOVA test with Bonferroni's post-test, *P < 0.05, **P < 0.01; ***P < 0.001. (D) Macrophages were stimulated with LPS 1 ng/mL, GzmA (150 nM, 75 nM and 37,5 nM), inactive GzmA and serpinb6b. After 24 h incubation, supernatants were collected to determine the levels of IL-6 by ELISA. (E) Human monocytes were obtained as described in materials and methods. 5 x 10³ human monocytes were stimulated with LPS 100 ng/mL, human active GzmA and human GzmA inactivated with antithrombin III (ATIII) and heparin (HP). After 24 h incubation, supernatant was collected to determine levels of IL-6 by ELISA. Statistical analysis was performed by unpaired student's t test, ***P < 0.001. (F) Human GzmA activity was confirmed by an activity essay incubating active human GzmA (300 nM), human GzmA inactivated with antithrombin III (ATIII) and heparin (HP) with a specific human GzmA substrate. Optical density at 405 nm was measured after 4, 12 and 24 h.