Interleukin 15 is essential for the development and differentiation of NK and memory CD8+ (mCD8+) T cells. Our laboratory previously showed that NK and CD8+ T lymphocytes facilitate the pathobiology of septic shock. However, factors that regulate NK and CD8+ T lymphocyte functions during sepsis are not well characterized. We hypothesized that IL-15 promotes the pathogenesis of sepsis by maintaining NK and mCD8+ T cell integrity. To test our hypothesis, the pathogenesis of sepsis was assessed in IL-15–deficient (IL-15 knockout, KO) mice. IL-15 KO mice showed improved survival, attenuated hypothermia, and less proinflammatory cytokine production during septic shock caused by cecal ligation and puncture or endotoxin-induced shock. Treatment with IL-15 superagonist (IL-15 SA, IL-15/IL-15Rα complex) regenerated NK and mCD8+ T cells and re-established mortality of IL-15 KO mice during septic shock. Preventing NK cell regeneration attenuated the restoration of mortality caused by IL-15 SA. If given immediately prior to septic challenge, IL-15–neutralizing IgG M96 failed to protect against septic shock. However, M96 caused NK cell depletion if given 4 d prior to septic challenge and conferred protection. IL-15 SA treatment amplified endotoxin shock, which was prevented by NK cell or IFN-γ depletion. IL-15 SA treatment also exacerbated septic shock caused by cecal ligation and puncture when given after the onset of sepsis. In conclusion, endogenous IL-15 does not directly augment the pathogenesis of sepsis but enables the development of septic shock by maintaining NK cell numbers and integrity. Exogenous IL-15 exacerbates the severity of sepsis by activating NK cells and facilitating IFN-γ production.

This article is featured in In This Issue, p.977

Interleukin-15 is a cytokine that is essential for maintaining the homeostasis and effector functions of NK and memory CD8+ (mCD8+) T lymphocytes. IL-15 prompts the generation of mature NK cells in the bone marrow (1) and it potently expands and activates peripheral NK cells to perform cytotoxic functions and facilitate cytokine secretion during viral and bacterial infections (2, 3). IL-15 also plays a pivotal role in the generation, cytotoxicity, and survival of CD8+ T lymphocytes, especially the mCD8+ subset (4, 5), and is essential for survival of NKT and intestinal intraepithelial lymphocytes (6, 7). Germline deletion of IL-15 in mice causes deficiency in NK, mCD8+ T, NKT cells, and intraepithelial lymphocytes (8). IL-15 is constitutively expressed by multiple types of cells including monocytes, macrophages, dendritic cells (DCs), fibroblasts, and epithelial cells (9, 10). Its expression is induced by cytokines such as type I (IFN-α/β) and type II (IFN-γ) IFNs as well as microbial products such as LPS, polyI:C, and viruses (11, 12). IL-15 is primarily presented in association with the unique high-affinity IL-15 receptor α (IL-15Rα) subunit, which is expressed on the surface of IL-15–producing cells and delivers signals to target cells that express the IL-2R β and γ receptor subunits, a process called trans-presentation. The IL-15/IL-15Rα complex can be released in soluble form after cleavage of the transmembrane domain of the receptor α (1315), and can also be generated in solution by mixing the individual components. The generated IL-15/IL-15Rα complex possesses a longer half-life and greater biological activity than free IL-15 and is thus termed IL-15 superagonist (IL-15 SA) (16).

NK and CD8+ T lymphocytes have been shown to facilitate physiological dysfunction and systemic inflammation during sepsis (1719). However, little is known about the factors that regulate the functions of NK and CD8+ T lymphocytes in the context of sepsis. IL-15 appears to be an essential proinflammatory mediator during sepsis, as IL-15 knockout (KO) mice show resistance to sepsis (20). In clinical studies, elevated blood IL-15 concentrations are associated with the development of organ injury and mortality in high risk gastrointestinal surgery patients and patients with severe sepsis, respectively (21, 22). However, the underlying mechanisms by which IL-15 facilitates the pathogenesis of sepsis have not been well characterized. In addition, IL-15 KO mice are not only deficient in IL-15, but have markedly decreased numbers of NK and CD8+ T cells, which are implicated in the pathogenesis of sepsis (1719). Therefore, it is unclear if lack of IL-15 alone or lack of IL-15–dependent NK and mCD8+ T cells in IL-15 KO mice confers protection against septic shock. In addition, it remains unclear whether IL-15 treatment exacerbates the pathogenesis of sepsis by activating NK and mCD8+ T cells.

In this paper, the role of endogenous and exogenous IL-15 in the pathogenesis of sepsis and its regulatory effect on NK and mCD8+ T cell viability and activity during sepsis was investigated. The studies were designed to address the hypotheses that IL-15 KO mice are resistant to septic shock due to an intrinsic deficiency of NK and mCD8+ T cells and treatment of wild type (WT) mice with IL-15 will facilitate the pathogenesis of septic shock by augmenting NK and mCD8+ T cell activation. The response of IL-15–deficient mice to sepsis caused by cecal ligation and puncture (CLP) or LPS challenge was fully examined. Specific endpoints included survival, organ injury, systemic cytokine production, and bacterial clearance. Whether regeneration of NK and mCD8+ T cells by treatment with IL-15 SA would alter the response of IL-15–deficient mice to sepsis was also assessed. Additional experiments examined whether sustained administration of IL-15 SA to WT mice or acute administration after the onset of sepsis altered sepsis-associated pathobiology. Studies employing NK or mCD8+ T cell depletion, and IFN-γ–deficient mice were performed to provide mechanistic insights.

Female and male, 8- to 12-wk old C57BL/6Tac were purchased from Taconic Farms (Hudson, NY). Breeding pairs of homozygous IL-15 null mice (C57BL/6NTac-IL15tm1Imx N5, IL-15 KO) were purchased from Taconic and the genotypes of offspring was verified by PCR analysis performed by Transnetyx (Memphis, TN). Female, 8- to 12-wk old homozygous IFN-γ null mice (B6.129S7-Ifngtm1Ts/J, IFN-γ KO) and WT C57BL/6J control mice were purchased from the Jackson Laboratory (Bar Harbor, ME). All studies were approved by the Institutional Animal Care and Use Committee at Vanderbilt University. In all experiments, 8- to 12-wk old WT and KO mice were used except in those in which IL-15 SA was administrated after the onset of sepsis, (Fig. 11), in which 16- to 20-wk old WT mice were used.

FIGURE 11.

High-dose IL-15 SA posttreatment accentuates lethality to septic shock. WT mice (16- to 20-wk old) received vehicle or IL-15 SA (2 μg) at 2 and 18 h after CLP challenge and survival rate was monitored over 7 d (A). Core temperature, bacterial counts in blood and peritoneal fluid as well as IFN-γ, IL-6, and TNF-α concentrations in the plasma were measured at 18 h after LPS (BG). n = 8–9 mice per group. Data are representative of two separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 compared with vehicle WT control.

FIGURE 11.

High-dose IL-15 SA posttreatment accentuates lethality to septic shock. WT mice (16- to 20-wk old) received vehicle or IL-15 SA (2 μg) at 2 and 18 h after CLP challenge and survival rate was monitored over 7 d (A). Core temperature, bacterial counts in blood and peritoneal fluid as well as IFN-γ, IL-6, and TNF-α concentrations in the plasma were measured at 18 h after LPS (BG). n = 8–9 mice per group. Data are representative of two separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 compared with vehicle WT control.

Close modal

NK cells were depleted in mice by i.p. injection with anti-asialoGM1 IgG (50 μg/mouse; Cedarlane Laboratories, Canada) at 24 h prior to initiation of IL-15 SA treatment or LPS challenge. CD8 T cells were depleted by treatment with anti-CD8α IgG (clone 53-6.7, 50 μg/mouse; eBioscience) at 24 h prior to IL-15 SA treatment or LPS challenge. Isotype-specific or non-specific IgG served as control in all Ab-induced leukocyte depletion experiments.

The protocol was described previously (23). In brief, mice were anesthetized with 2% isoflurane in oxygen. A 1 to 2 cm midline incision was made through the abdominal wall. The cecum was identified and ligated 1.0 cm from the tip using a 3-0 silk tie. A double puncture of the ligated cecum was performed using a 20-gauge needle. The incision was closed using autoclips. Buprenorphine (0.1 mg/kg) was administered s.c. for analgesia 30 min before CLP and twice daily thereafter. All mice received fluid resuscitation (Lactated Ringers solution, 1 ml, i.p.) immediately after injury and twice daily thereafter.

Ultrapure LPS-EB (LPS from Escherichia coli 0111:B4) from InvivoGen (San Diego, CA) was administered at a dose of 100 or 150 μg/mouse via i.p. injection. Measurements of rectal temperature, proinflammatory cytokine concentrations, and indices of acute organ injury were performed at 6 and 24 h after LPS challenge.

Recombinant mouse IL-15 was purchased from eBioscience (San Diego, CA, Cat. no. 34-8151-85). Mouse IL-15 Rα subunit Fc chimera (IL-15 Ra) was purchased from R&D Systems (Minneapolis, MN, Cat. no. 551-MR-100). For preparation of IL-15 SA, 20 μg of IL-15, and 90 μg of IL-15 Ra were incubated in 400 μl of sterile PBS at 37°C for 20 min to form the IL-15/IL-15 Ra complex. The complex was then diluted with sterile PBS to reach a concentration of 0.625 or 10 μg IL-15 SA per ml, then aliquoted and frozen. The IL-15 SA doses reported in the study are based on the amount of IL-15 present in the IL-15 SA complex.

To regenerate NK cells, IL-15 KO mice received i.p. injections of 0.125 μg of IL-15 SA (low dose) in 0.2 ml of PBS for 4 d (day 0–3). IL-15 KO mice that were treated with vehicle served as control. On day 4, spleens and livers were harvested for measurement of leukocyte numbers and activation. In additional experiments, after the same treatment regimen of IL-15 SA, IL-15 KO mice were subjected to CLP or LPS (150 μg) on day 4. In survival studies, IL-15 SA (0.125 μg) injection was continued daily to maintain the regenerated NK cell population. In additional experiments, WT mice received i.p. injections with a high dose (2 μg) of IL-15 SA at 30 min prior to or 2 h after challenge with 100 μg of LPS. A second dose of IL-15 SA was performed 18 h after LPS challenge.

IL-15 neutralizing Ab M96 was generously provided by Amgen (Thousand Oaks, CA) (24). WT mice received i.p. injection with 20 μg of M96 either 2 h or 4 d prior to CLP or LPS challenge. A survival study was followed for 7 d after the initiation of septic shock. When M96 was given 4 d prior to CLP or LPS, a second administration of M96 at the same dose was given at the time of septic insults to maintain the effect of M96 on depletion of NK cells.

Plasma was obtained from heparinized whole blood after centrifugation (2000 × g × 10 min). Plasma alanine aminotransferase (ALT) and aspartate aminotransferase (AST) concentrations were measured as indices of acute liver injury, and blood urea nitrogen (BUN) and creatinine concentrations as indices of renal injury. They were measured in the Translational Pathology Core Laboratory at Vanderbilt University using an ACE Alera Chemistry Analyzer (Alfa Wassermann, West Caldwell, NJ).

Concentrations of IL-6, IFN-γ, IL-12p70, IL-18, TNF-α, IL-1β, and IL-10 in plasma were measured using a Bio-Plex Multiplex Assay with the MAGPIX Multiplex Reader (Bio-Rad, Hercules, CA). Results were analyzed with Bio-Plex Manager Software 6.1. The concentration of soluble IL-15/IL-15Rα complex in plasma was measured using Mouse IL-15/IL-15R Complex ELISA Ready-Set-Go (eBioscience).

Bacterial counts were performed on aseptically harvested blood and peritoneal fluid. Blood was harvested via carotid laceration. Peritoneal lavage fluid was obtained by injection and aspiration of 2 ml sterile PBS into the peritoneal cavity. Samples were serially diluted in sterile PBS and cultured on tryptic soy agar plates. Plates were incubated at 37°C for 24 h and bacterial colonies were counted.

Splenocytes and hepatic leukocytes were isolated as previously described. Briefly, the spleen was smashed in PBS with the plunger from a 10 ml syringe and the homogenate was passed through a 70 μm cell strainer. Erythrocytes were lysed with RBC Lysis Buffer (Sigma Life Sciences, St. Louis, MO). The cells were counted using a TC20 Automated Cell Counter (Bio-Rad), centrifuged (300 × g × 5 min), and the cell pellet was resuspended in PBS. Livers were harvested after perfusion, which was achieved by the cutting of the hepatic portal vein, insertion of a 25 G needle into the left ventricle of the heart, and perfusion with 10 ml PBS. Liver homogenate was passed through a 70 μm cell strainer and then washed, resuspended with 10 ml of 37.5% Percoll Plus (GE Healthcare Life Sciences), and centrifuged (680 × g × 12 min at room temperature). The supernatant containing hepatocytes was discarded, erythrocytes were lysed, and the resulting mononuclear cells were counted using a TC20 Automated Cell Counter.

For surface marker staining, cells were suspended in PBS (1 × 107 cells/ml) and incubated with anti-mouse CD16/32 (1 μl/ml; eBioscience) for 5 min to block nonspecific Fc receptor–mediated Ab binding. One million cells were then transferred into polystyrene tubes. Fluorochrome-conjugated Abs or isotype controls (0.5 μg/tube) were added, incubated (4°C) for 30 min, and washed with 2 ml of cold PBS. After centrifugation (300 × g × 5 min), cell pellets were fixed with 250 μl of 1% paraformaldehyde. Abs used for surface marker labeling included CD3-Alex Fluor 488, NK1.1-PE-Cy7, CD8-FITC, CD4-FITC, CD44-PE-Cy5, CD19-PE, CD69-PE, CD11b-PE, and CD27-APC (eBioScience, BD Biosciences, San Diego). Appropriate isotype-specific Abs were used as controls.

For intracellular staining, 1 × 106 splenocytes were incubated with 4 μl PMA and ionomycin (cell stimulation mixture; eBioScience) in 1 ml of RPMI 1640 media with 10% FBS at 37°C for 5 h. After 1 h, the protein transport inhibitors brefeldin A and monensin (2 μl/ml; eBioScience) were added into cell cultures for the remaining 4 h. After incubation, cell suspensions were labeled with fluorochrome-conjugated Abs to surface markers as described above. Cells were then fixed and permeabilized with Cytofix/Cytoperm Plus (250 μl/tube; BD Biosciences) for 20 min at 4°C. After washing with BD Perm/Wash solution, anti-IFN-γ-PE (clone XMG1.2; eBioscience) was used to detect intracellular IFN-γ production. Fluorochrome-conjugated isotype-specific IgG served as controls. All samples were analyzed using an Accuri C6 flow cytometer (BD Biosciences). Data were analyzed using Accuri C6 software.

All values are presented as the mean ± SEM, except for body temperature, ALT, AST, BUN, and creatinine for which median values are designated. A Student t test was used to examine the difference between two experimental groups. Data from multiple group experiments were analyzed using one-way ANOVA followed by a post hoc Tukey test to compare groups. Survival data were analyzed using the Mantel-Cox log-rank test. A p value < 0.05 was considered statistically significant for all experiments.

IL-15 null mice have been reported to be deficient of NK, NKT, and mCD8+ T cells (25). Examination of our colony showed that NK and mCD8+ T cells were significantly decreased in the spleens and livers of IL-15 KO mice whereas NKT cell numbers were not significantly different in either tissue as compared with WT controls (Supplemental Fig. 1). CD4+ T, naive CD8+ T, and B cell numbers were not significantly different when comparing WT and IL-15 KO mice (data not shown).

A survival study was performed to assess mortality in IL-15 KO and WT mice during sepsis induced by CLP. A significant survival advantage was observed in IL-15 KO mice, in which 50% long-term survival and a 120 h median survival time were observed as compared with 0% survival and 36 h median survival in WT mice (Fig. 1A). Both WT and IL-15 KO mice developed sepsis-induced hypothermia (Fig. 1B). However, core body temperature was significantly higher in IL-15 KO mice at 6 and 18 h after CLP as compared with WT controls (Fig. 1B).

FIGURE 1.

IL-15 KO mice are resistant to CLP-induced septic shock. WT and IL-15 KO mice were subjected to CLP and were monitored for 7-d survival (A). Body temperature (B) was measured at 6 and 18 h after CLP. The median value is designated in (B). n = 11–14 mice per group. Data are representative of two to three separate experiments. *p < 0.05, ***p < 0.001, compared with WT mice.

FIGURE 1.

IL-15 KO mice are resistant to CLP-induced septic shock. WT and IL-15 KO mice were subjected to CLP and were monitored for 7-d survival (A). Body temperature (B) was measured at 6 and 18 h after CLP. The median value is designated in (B). n = 11–14 mice per group. Data are representative of two to three separate experiments. *p < 0.05, ***p < 0.001, compared with WT mice.

Close modal

Further studies were undertaken to assess the effect of IL-15 deficiency on proinflammatory cytokine production and bacterial clearance after CLP (Fig. 2). At 6 and 18 h after CLP, IL-15 was detected in the plasma of WT mice but not in IL-15 KO mice (Fig. 2A). Concentrations of several proinflammatory cytokines, including IL-6, TNF-α, IL-1β, IFN-γ, and IL-12, were significantly lower in the plasma of IL-15 KO mice compared with WT controls (Fig. 2B–F). Meanwhile, neutrophil recruitment into the peritoneal cavity at 6 h after CLP was not different between IL-15 KO and WT mice (Fig. 2G). The numbers of bacteria were lower in peritoneal lavage fluid of IL-15 KO mice than WT control mice at 6 h after CLP (Fig. 2H) but showed no significant difference between groups in blood or the peritoneal cavity at 18 h after CLP (Fig. 2I, 2J).

FIGURE 2.

IL-15 KO mice exhibit attenuated proinflammatory cytokine production during CLP-induced septic shock. Blood was harvested at 6 and 18 h after CLP challenge for measurement of proinflammatory cytokines (AF). Neutrophil numbers in peritoneal cavity (G) and bacterial colony forming units in blood and peritoneal fluid (HJ) were also measured at designated time points in WT and IL-15 KO mice. n = 6–8 mice per group. Data are representative of two separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001 compared with WT mice at designated time points.

FIGURE 2.

IL-15 KO mice exhibit attenuated proinflammatory cytokine production during CLP-induced septic shock. Blood was harvested at 6 and 18 h after CLP challenge for measurement of proinflammatory cytokines (AF). Neutrophil numbers in peritoneal cavity (G) and bacterial colony forming units in blood and peritoneal fluid (HJ) were also measured at designated time points in WT and IL-15 KO mice. n = 6–8 mice per group. Data are representative of two separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001 compared with WT mice at designated time points.

Close modal

In a dose-finding study, IL-15 KO and WT mice were injected with LPS at doses of 100 or 150 μg and monitored for sepsis-induced mortality over 7 d. WT mice showed dose-dependent mortality with 43% survival after 100 μg of LPS and 0% after 150 μg of LPS, whereas IL-15 KO mice showed improved survival at both doses of LPS (100% survival after 100 μg of LPS and 67% after 150 μg of LPS) (Fig. 3A, 3C). IL-15 KO mice developed significantly less hypothermia than WT controls at 24 h after LPS challenge (Fig. 3B, 3D).

FIGURE 3.

IL-15 KO mice are resistant to LPS-induced septic shock. In a dose escalation study, IL-15 KO and WT mice received i.p. injection with 100 or 150 μg of LPS and were observed for 7-d survival (A and C). Body temperature was measured at 24 h after 100 or 150 μg of LPS injection (B and D). n = 8–23 mice per group. Data are representative of two to four separate experiments. **p < 0.01, ***p < 0.001, ****p < 0.0001 when compared with WT mice.

FIGURE 3.

IL-15 KO mice are resistant to LPS-induced septic shock. In a dose escalation study, IL-15 KO and WT mice received i.p. injection with 100 or 150 μg of LPS and were observed for 7-d survival (A and C). Body temperature was measured at 24 h after 100 or 150 μg of LPS injection (B and D). n = 8–23 mice per group. Data are representative of two to four separate experiments. **p < 0.01, ***p < 0.001, ****p < 0.0001 when compared with WT mice.

Close modal

Further studies assessed organ injury and proinflammatory cytokine production after 150 μg LPS challenge. IL-15 concentration in the plasma was detected in WT, but not in IL-15 KO mice at 6 and 24 h after LPS (Fig. 4A). Meanwhile, lower concentrations of other proinflammatory cytokines in plasma were observed in IL-15 KO mice compared with WT mice (Fig. 4B–F). Also, concentrations of AST, BUN, and creatinine were lower in the plasma of IL-15 KO mice compared with WT controls at 24 h after LPS challenge, indicative of reduced acute hepatocellular and kidney injury (Fig. 4G–I).

FIGURE 4.

IL-15 KO mice have attenuated proinflammatory cytokine production and organ injury after LPS-induced septic shock. Blood was harvested at 6 and 24 h after 150 μg LPS challenge for measurement of proinflammatory cytokines in plasma (AF). AST, BUN, and creatinine (GI) concentrations in plasma were also measured at 24 h post 150 μg of LPS. n = 8 mice per group. Data are representative of two separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 when compared with WT mice.

FIGURE 4.

IL-15 KO mice have attenuated proinflammatory cytokine production and organ injury after LPS-induced septic shock. Blood was harvested at 6 and 24 h after 150 μg LPS challenge for measurement of proinflammatory cytokines in plasma (AF). AST, BUN, and creatinine (GI) concentrations in plasma were also measured at 24 h post 150 μg of LPS. n = 8 mice per group. Data are representative of two separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 when compared with WT mice.

Close modal

IL-15 KO mice were treated with IL-15 superagonist (0.125 μg/d) for 4 d and lymphocyte numbers in spleen and liver were examined. Total NK cell (CD3 NK1.1+) numbers in the spleen and liver were significantly higher in IL-15 SA–treated IL-15 KO mice compared with vehicle-treated IL-15 KO mice and were ∼1.9 fold higher in the spleen and ∼7.0 fold higher in the liver compared with WT mice (Fig. 5A, 5B). After IL-15 SA treatment, splenic mCD8+ T cells (CD8+CD44high) increased to levels equal to those observed in WT mice and hepatic mCD8+ T cells to ∼2.5 fold higher than in WT mice (Fig. 5A, 5B). NKT (CD3+ NK1.1+) cell numbers were increased in the liver, but not the spleen, of IL-15 KO mice treated with IL-15 SA compared with vehicle-treated controls and WT mice (Fig. 5A, 5B). Naive CD8+ T (CD8+CD44low), CD4+ T, and B cells were not significantly different when comparing IL-15 KO mice treated with IL-15 SA or vehicle (data not shown).

FIGURE 5.

Low-dose IL-15 SA treatment restores NK and memory CD8+ T cells in IL-15 KO mice. IL-15 KO mice were treated with 0.125 of μg IL-15 SA for four consecutive days. At 24 h following the last treatment, splenic (A) and liver (B) NK (CD3NK1.1+), NKT (CD3+NK1.1+), and memory CD8+ T (CD8+CD44high) lymphocyte numbers were measured using flow cytometry. The activation status (CD69 expression) and IFN-γ expression by NK (C and D) and memory CD8+ T cells (E and F) upon IL-15 SA treatment were also analyzed. For IFN-γ expression, splenocytes from WT mice and vehicle- and IL-15 SA–treated IL-15 KO mice were restimulated with PMA/ionomycin for 5 h to amplify intracellular cytokine signaling (for details see 2Materials and Methods section). In the representative dot plot, NK cells are divided into four subpopulations based on CD11b and CD27 expression, namely precursor (CD11blowCD27low), immature (I, CD11blowCD27high), mature proinflammatory (II, CD11bhighCD27high), and mature cytotoxic (III, CD11bhighCD27low) NK cells (G). The graph in (H) shows the relative number of splenic NK subsets among intact WT mice, vehicle- and IL-15 SA–treated IL-15 KO mice. n = 4–12 mice per group. Data are representative of two to four separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 compared with designated groups.

FIGURE 5.

Low-dose IL-15 SA treatment restores NK and memory CD8+ T cells in IL-15 KO mice. IL-15 KO mice were treated with 0.125 of μg IL-15 SA for four consecutive days. At 24 h following the last treatment, splenic (A) and liver (B) NK (CD3NK1.1+), NKT (CD3+NK1.1+), and memory CD8+ T (CD8+CD44high) lymphocyte numbers were measured using flow cytometry. The activation status (CD69 expression) and IFN-γ expression by NK (C and D) and memory CD8+ T cells (E and F) upon IL-15 SA treatment were also analyzed. For IFN-γ expression, splenocytes from WT mice and vehicle- and IL-15 SA–treated IL-15 KO mice were restimulated with PMA/ionomycin for 5 h to amplify intracellular cytokine signaling (for details see 2Materials and Methods section). In the representative dot plot, NK cells are divided into four subpopulations based on CD11b and CD27 expression, namely precursor (CD11blowCD27low), immature (I, CD11blowCD27high), mature proinflammatory (II, CD11bhighCD27high), and mature cytotoxic (III, CD11bhighCD27low) NK cells (G). The graph in (H) shows the relative number of splenic NK subsets among intact WT mice, vehicle- and IL-15 SA–treated IL-15 KO mice. n = 4–12 mice per group. Data are representative of two to four separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 compared with designated groups.

Close modal

The impact of IL-15 SA treatment on regenerated NK and mCD8+ T cell activation and function was determined by examining the expression of the early surface activation marker CD69 and production of IFN-γ. Expression of CD69 on splenic NK cells in IL-15 KO mice treated with IL-15 SA was not different compared with vehicle-treated IL-15 KO mice or WT mice (Fig. 5C). The numbers of mCD8+ T cells expressing CD69 was increased in IL-15 KO mice treated with IL-15 SA as compared with vehicle-treated IL-15 KO controls but not compared with WT control (Fig. 5D). PMA/ionomycin-induced IFN-γ production by NK cells was significantly higher in IL-15 KO mice treated with IL-15 SA as compared with those treated with vehicle and was not different from WT controls (Fig. 5E). PMA/ionomycin-induced IFN-γ production by mCD8+ T cells was significantly higher in IL-15 SA–treated mice compared with WT controls and vehicle-treated IL-15 KO mice (Fig. 5F).

NK cell subsets were characterized based on CD11b and CD27 expression and classified as precursor (CD11blowCD27low), immature (I, CD11blowCD27high), proinflammatory (II, CD11bhighCD27high), or cytotoxic (III, CD11bhighCD27low) (Fig. 5G) (26). IL-15 SA treatment increased the numbers of immature and proinflammatory NK cells in IL-15 KO mice to a level significantly higher than or comparable to WT controls but did not increase the numbers of the cytotoxic subset (Fig. 5H). Further studies were undertaken to assess CLP- or LPS-induced mortality of IL-15 KO mice treated with IL-15 SA for 4 d prior to septic challenge. IL-15 SA treatment was continued daily throughout the experimental period to maintain regenerated NK and mCD8+ T cell populations. During septic shock elicited by CLP or 150 μg of LPS, IL-15 SA reestablished mortality in IL-15 KO mice to a level comparable to WT control and significantly higher than IL-15 KO mice treated with vehicle (Fig. 6A, 6B). IL-15 KO mice treated with IL-15 SA also showed a significant decrease in core body temperature after LPS challenge compared with vehicle-treated IL-15 KO mice, which was comparable to that observed in WT mice (Fig. 6C). Production of proinflammatory cytokines, including IL-6 and IL-1β, was also increased in IL-15 SA–treated IL-15 KO mice at 24 h after LPS compared with vehicle-treated IL-15 KO mice (Fig. 6D, 6E). IL-15 SA treatment also prompted the development of acute kidney injury in IL-15 KO mice as indicated by elevated plasma BUN and creatinine concentrations after LPS challenge (Fig. 6F, 6G). However, IL-15 SA did not exacerbate liver injury as it did not increase concentrations of ALT and AST in the plasma following LPS challenge (data not shown).

FIGURE 6.

Low-dose IL-15 SA treatment restores IL15-mediated lethality to septic shock in IL-15 KO mice. IL-15 SA (0.125 μg) treatment was initiated 4 d prior to septic challenge and continued daily throughout the experimental period to maintain regenerated NK and mCD8+ T cell populations. Mice were monitored for survival rate over 7 d during CLP- or 150 μg of LPS-induced sepsis (A and B). Core temperature was measured at 24 h after 150 μg LPS challenge (C). Vehicle-treated WT and IL-15 KO mice served as control. Blood was collected from WT and IL-15 KO mice at 24 h after 150 μg LPS challenge for measurement of IL-6 and IL-1β (D and E) as well as ALT, AST, BUN, and creatinine (F and G) concentrations in plasma. n = 7–14 mice per group. Data are representative of two to four separate experiments. *p < 0.05, **p < 0.01, ****p < 0.0001 compared with designated groups.

FIGURE 6.

Low-dose IL-15 SA treatment restores IL15-mediated lethality to septic shock in IL-15 KO mice. IL-15 SA (0.125 μg) treatment was initiated 4 d prior to septic challenge and continued daily throughout the experimental period to maintain regenerated NK and mCD8+ T cell populations. Mice were monitored for survival rate over 7 d during CLP- or 150 μg of LPS-induced sepsis (A and B). Core temperature was measured at 24 h after 150 μg LPS challenge (C). Vehicle-treated WT and IL-15 KO mice served as control. Blood was collected from WT and IL-15 KO mice at 24 h after 150 μg LPS challenge for measurement of IL-6 and IL-1β (D and E) as well as ALT, AST, BUN, and creatinine (F and G) concentrations in plasma. n = 7–14 mice per group. Data are representative of two to four separate experiments. *p < 0.05, **p < 0.01, ****p < 0.0001 compared with designated groups.

Close modal

Treatment with anti-asialoGM1 prior to the initiation of IL-15 SA treatment prevented NK cell regeneration in IL-15 KO mice (Fig. 7A), and partially, but significantly, reversed endotoxin-induced mortality in IL-15 SA–treated IL-15 KO mice upon 150 μg LPS challenge (Fig. 7B). However, preventing CD8+ T cell regeneration by treatment with anti-CD8α (Fig. 7C) failed to rescue IL-15 KO mice treated with IL-15 SA during LPS-induced shock (Fig. 7D).

FIGURE 7.

Prevention of NK cell regeneration ablates IL-15 SA treatment–induced restoration of lethality to septic shock in IL-15 KO mice. IL-15 KO mice received anti-asialoGM1 or anti-CD8α at 24 h prior to the initiation of IL-15 SA (0.125 μg), and NK and memory CD8+ T lymphocyte counts were measured at 24 h after the last IL-15 SA treatment. Dotted lines represent the baseline levels of NK and memory CD8+ T cells in WT controls (A and C). A survival study was also undertaken to assess the effect of anti-asialoGM1 or anti-CD8α on the survival of IL-15 SA–treated IL-15 KO mice upon 150 μg LPS challenge (B and D). Isotype-specific IgG served as control. n = 5–15 mice per group. Data are representative of two to four separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 compared with designated groups.

FIGURE 7.

Prevention of NK cell regeneration ablates IL-15 SA treatment–induced restoration of lethality to septic shock in IL-15 KO mice. IL-15 KO mice received anti-asialoGM1 or anti-CD8α at 24 h prior to the initiation of IL-15 SA (0.125 μg), and NK and memory CD8+ T lymphocyte counts were measured at 24 h after the last IL-15 SA treatment. Dotted lines represent the baseline levels of NK and memory CD8+ T cells in WT controls (A and C). A survival study was also undertaken to assess the effect of anti-asialoGM1 or anti-CD8α on the survival of IL-15 SA–treated IL-15 KO mice upon 150 μg LPS challenge (B and D). Isotype-specific IgG served as control. n = 5–15 mice per group. Data are representative of two to four separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 compared with designated groups.

Close modal

Generally, IL-15 is retained inside the cell until it is chaperoned with the high-affinity receptor α and shuttled to the cell surface for delivery to target cells, a process called trans-presentation. But the IL-15/IL-15Rα complex in a soluble form is also present in plasma after proteolytic cleavage of the transmembrane domain of IL-15Rα from the cell surface. Plasma IL-15/IL-15Rα complex concentration was elevated in WT mice during shock induced by CLP or LPS and was neutralized by treatment with M96, an IL-15 neutralizing Ab, when given to WT mice 2 h prior to CLP or LPS (Supplemental Fig. 2). However, short-term neutralization of IL-15 by M96 failed to confer a survival benefit to WT mice compared with IgG-treated controls after CLP and 150 μg LPS challenges (Fig. 8A, 8B). Examination of the cytokine expression profile shows that the acute neutralization of IL-15 did not alter concentrations of other plasma proinflammatory cytokines except IFN-γ, which was lower compared with IgG control at 24 h after 150 μg LPS challenge (Fig. 8C–G). The level of anti-inflammatory cytokine IL-10 was lower in M96-treated WT mice at 24 h after LPS (Fig. 8H). Under the current treatment regimen, M96 did not decrease the number of NK cells until 24 h after LPS challenge. M96 did not alter the number of mCD8+ T cells or the activation of NK and mCD8+ T cells as indicated by CD69 expression at 6 and 24 h after LPS challenge as compared with IgG control (Supplemental Fig. 3).

FIGURE 8.

Acute neutralization of IL-15 does not mediate resistance to septic shock. WT mice received 20 μg of M96, an IL-15 neutralizing Ab i.p. at 2 h prior to CLP or 150 μg LPS challenge and a survival study was followed (A and B). Specific IgG serve as control. Concentrations of IFN-γ, IL-6, IL-1β, IL-12, TNF-α, and IL-10 in the plasma were measured at 6 and 24 h after 150 μg LPS challenge (CH). n = 5–10 mice per group. Data are representative of two to three separate experiments. *p < 0.05, **p < 0.01 compared with IgG control.

FIGURE 8.

Acute neutralization of IL-15 does not mediate resistance to septic shock. WT mice received 20 μg of M96, an IL-15 neutralizing Ab i.p. at 2 h prior to CLP or 150 μg LPS challenge and a survival study was followed (A and B). Specific IgG serve as control. Concentrations of IFN-γ, IL-6, IL-1β, IL-12, TNF-α, and IL-10 in the plasma were measured at 6 and 24 h after 150 μg LPS challenge (CH). n = 5–10 mice per group. Data are representative of two to three separate experiments. *p < 0.05, **p < 0.01 compared with IgG control.

Close modal

Further studies were undertaken to assess the effects of prolonged M96 pretreatment on CLP- or LPS-induced mortality in WT mice. M96 treatment was initiated 4 d prior to septic challenge and a second administration was given at the time of CLP or 150 μg LPS. As opposed to acute neutralization of IL-15, long-term IL-15 neutralization partially reversed the septic mortality of WT mice upon CLP challenge and significantly protected WT mice from shock induced by LPS (Fig. 9A, 9B). Under the current treatment regimen of M96, WT mice exhibited attenuated hypothermia at 24 h after LPS challenge (Fig. 9C). Analysis of splenic and hepatic lymphocyte populations at 4 d after M96 treatment showed depletion of splenic and hepatic NK cells (Fig. 9D). However, mCD8+, NKT, naive CD8+, CD4+, and B lymphocyte numbers were not altered by pretreatment with M96 (data not shown).

FIGURE 9.

Prolonged neutralization of IL-15 mediates resistance to septic shock. WT mice received 20 μg of IL-15 neutralizing Ab M96 or IgG i.p. at 4 d prior to (day −4) and at the time of CLP or 150 μg LPS challenge (day 0). A survival study was undertaken over 7 d (A and B). n = 8–11 mice per group. Core temperature was measured at 24 h after 150 μg LPS challenge (C). n = 9 mice per group. Splenic and hepatic NK cell number was measured at day 0 without LPS challenge (D). n = 6–9 mice per group. n = 8–11 mice per group in survival studies. Data are representative of two to three separate experiments. **p < 0.01, ****p < 0.0001 compared with IgG control.

FIGURE 9.

Prolonged neutralization of IL-15 mediates resistance to septic shock. WT mice received 20 μg of IL-15 neutralizing Ab M96 or IgG i.p. at 4 d prior to (day −4) and at the time of CLP or 150 μg LPS challenge (day 0). A survival study was undertaken over 7 d (A and B). n = 8–11 mice per group. Core temperature was measured at 24 h after 150 μg LPS challenge (C). n = 9 mice per group. Splenic and hepatic NK cell number was measured at day 0 without LPS challenge (D). n = 6–9 mice per group. n = 8–11 mice per group in survival studies. Data are representative of two to three separate experiments. **p < 0.01, ****p < 0.0001 compared with IgG control.

Close modal

Previous studies from our laboratory showed that systemic administration of IL-15 SA at 2 μg for four consecutive days caused immunotoxicity characterized by hypothermia, weight loss, liver injury, and mortality in WT mice (27). In this study, investigations were performed to examine the effect of exogenous 2 μg IL-15 SA on the survival of WT mice upon CLP or LPS (100 μg) challenge. If given 30 min prior to and 24 h after LPS challenge, IL-15 SA (2 μg) worsened mortality compared with vehicle control, whereas treatment with IL-15 SA alone did not cause any signs of toxicity or mortality (Fig. 10A, 10B). IL-15 SA treatment also exacerbated LPS-induced hypothermia and increased levels of AST and BUN in the plasma of WT mice after LPS challenge (Fig. 10C–E). Concentrations of IFN-γ, IL-6, IL-1β, and IL-12 were significantly higher in the plasma of WT mice treated with IL-15 SA at 6 and/or 24 h after LPS challenge, whereas the levels of TNF-α and IL-10 were not altered (Fig. 10F–K). IL-15 SA facilitated the rapid mobilization of NK cells out of spleens at 1 h following LPS challenge. IL-15 also promoted the activation of NK and mCD8+ T cells as indicated by the upregulation in CD69 expression as early as 1 h after LPS challenge (Supplemental Fig. 4).

FIGURE 10.

High-dose IL-15 SA pretreatment accentuates lethality to septic shock. WT mice received vehicle or IL-15 SA at 2 μg 30 min prior to and 24 h after CLP or LPS (100 μg) and survival studies were followed (A and B). Core temperature, AST, and BUN levels, and IFN-γ, IL-6, IL-1β, IL-12, TNF-α, and IL-10 concentrations in the plasma were measured at 24 h after 100 μg LPS (CK). n = 5–20 mice per group. Data are representative of two to three separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001 compared with vehicle WT control.

FIGURE 10.

High-dose IL-15 SA pretreatment accentuates lethality to septic shock. WT mice received vehicle or IL-15 SA at 2 μg 30 min prior to and 24 h after CLP or LPS (100 μg) and survival studies were followed (A and B). Core temperature, AST, and BUN levels, and IFN-γ, IL-6, IL-1β, IL-12, TNF-α, and IL-10 concentrations in the plasma were measured at 24 h after 100 μg LPS (CK). n = 5–20 mice per group. Data are representative of two to three separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001 compared with vehicle WT control.

Close modal

Additional studies were performed to determine the effect of IL-15 SA on the pathogenesis of sepsis in WT mice when given after the onset of sepsis. IL-15 SA (2 μg) or vehicle treatment was given to WT mice (16- to 20-wk old) at 2 and 18 h after CLP challenge. Posttreatment with IL-15 SA induced 100% mortality as compared with 22% mortality in vehicle-treated mice (Fig. 11A). At 18 h after CLP challenge, IL-15 SA-treated mice exhibited more hypothermia, and increased bacterial burden in the blood and peritoneal cavity as compared with vehicle controls (Fig. 11B–D). In addition, concentrations of IFN-γ, IL-6, and TNF-α were significantly higher in the plasma of WT mice treated with IL-15 SA after otherwise sublethal CLP challenge (Fig. 11E–G).

Further studies determined the contributions of NK and CD8+ T cells to IL-15 SA-exacerbated mortality caused by LPS challenge. Depletion of NK cells alone or in combination with CD8+ T cells attenuated mortality caused by the combined administration of IL-15 SA and 100 μg of LPS (Fig. 12A). Depletion of CD8+ T cells alone prolonged the time to 100% mortality but did not improve long-term survival. (Fig. 12A). Moreover, NK cell depletion attenuated hypothermia and production of some, but not all, proinflammatory cytokines in WT mice treated with IL-15 SA and LPS (Fig. 12B–H).

FIGURE 12.

Neutralization of NK cells and CD8+ T cells combined, NK cells, but not CD8 T cells alone, mediates resistance of IL-15 SA treated mice to septic shock. WT mice received anti-asialoGM1 and/or anti-CD8α i.p. at 24 h prior to IL-15 SA treatment, then were challenged with 100 μg LPS and a second dose of IL-15 SA at 24 h thereafter. Survival rate was monitored over 7 d (A). Isotype-specific IgG serve as control. Core temperature and IFN-γ, IL-6, IL-1β, IL-12, TNF-α, and IL-10 concentrations in the plasma were measured at 6 and/or 24 h after 100 μg of LPS (BH). n = 5–10 mice per group. Data are representative of two to three separate experiments. **p < 0.01, ***p < 0.001, ****p < 0.0001 compared with IgG WT mice that were treated with 2 μg of IL-15 SA.

FIGURE 12.

Neutralization of NK cells and CD8+ T cells combined, NK cells, but not CD8 T cells alone, mediates resistance of IL-15 SA treated mice to septic shock. WT mice received anti-asialoGM1 and/or anti-CD8α i.p. at 24 h prior to IL-15 SA treatment, then were challenged with 100 μg LPS and a second dose of IL-15 SA at 24 h thereafter. Survival rate was monitored over 7 d (A). Isotype-specific IgG serve as control. Core temperature and IFN-γ, IL-6, IL-1β, IL-12, TNF-α, and IL-10 concentrations in the plasma were measured at 6 and/or 24 h after 100 μg of LPS (BH). n = 5–10 mice per group. Data are representative of two to three separate experiments. **p < 0.01, ***p < 0.001, ****p < 0.0001 compared with IgG WT mice that were treated with 2 μg of IL-15 SA.

Close modal

Additional studies were performed to examine the functional importance of IFN-γ in the IL-15 SA-mediated exacerbation of septic severity. IFN-γ KO mice exhibited improved survival and attenuated hypothermia during shock induced by coadministration of IL-15 SA and LPS (Fig. 13A, 13B). At 6 and/or 24 h after LPS challenge, IFN-γ was not detected in the plasma of IFN-γ KO, whereas IL-6, IL-1β, IL-12, and TNF-α concentrations were significantly lower in the plasma of IFN-γ KO mice compared with WT controls (Fig. 13C–H). The anti-inflammatory cytokine IL-10 was significantly elevated after LPS challenge in IFN-γ KO mice after IL-15 SA treatment.

FIGURE 13.

IFN-γ KO mice are resistant to IL-15 SA-induced accentuation of septic shock. IFN-γ KO mice received i.p. injection with 2 μg of IL-15 SA 30 min prior to and 24 h after LPS (100 μg) challenge. A survival study was performed over 7 d (A). Core temperature and IFN-γ, IL-6, IL-1β, IL-12, TNF-α, and IL-10 concentrations in the plasma were measured at 6 and/or 24 h after 100 μg of LPS (BH). n = 5–10 mice per group. Data are representative of two separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 compared with WT control treated with IL-15 SA.

FIGURE 13.

IFN-γ KO mice are resistant to IL-15 SA-induced accentuation of septic shock. IFN-γ KO mice received i.p. injection with 2 μg of IL-15 SA 30 min prior to and 24 h after LPS (100 μg) challenge. A survival study was performed over 7 d (A). Core temperature and IFN-γ, IL-6, IL-1β, IL-12, TNF-α, and IL-10 concentrations in the plasma were measured at 6 and/or 24 h after 100 μg of LPS (BH). n = 5–10 mice per group. Data are representative of two separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 compared with WT control treated with IL-15 SA.

Close modal

Although previous studies have shown that IL-15 exhibits toxicity (27) and aggravates chronic inflammatory disorders (2834), there are few studies examining the role of endogenous and exogenous IL-15 in the pathogenesis of acute inflammation, which is typical of sepsis (35). This study supports the hypothesis that IL-15 plays a role in the pathogenesis of septic shock via maintenance and activation of NK cells and, to a lesser degree, CD8+ T cells. We report that mice genetically deficient in IL-15 show a significant survival benefit over WT mice during CLP- and LPS-induced septic shock. IL-15 KO mice display markedly decreased NK and mCD8+ T cell numbers and attenuated sepsis-induced hypothermia, hepatocellular, and kidney injuries and systemic proinflammatory cytokine production compared with WT mice. Similarly, prolonged neutralization of IL-15 over a 4-d period causes NK cell depletion and provides protection from septic shock. Treatment of IL-15 KO mice with low-dose IL-15 SA regenerates NK cells and reestablishes susceptibility to septic shock, whereas acute neutralization of IL-15 in WT mice fails to provide protection. In addition, treatment of WT mice with high-dose IL-15 SA prompts the activation of NK cells and worsens septic outcomes, which is prevented by NK cell depletion or IFN-γ neutralization. Thus, we conclude that IL-15 contributes to the pathogenesis of acute septic shock by maintaining and activating NK and, possibly, mCD8+ T cells. These cells facilitate systemic inflammation and organ injury during septic shock through a mechanism that is dependent on the production of IFN-γ.

Orinska et al. (20) previously showed the genetic deletion of IL-15 results in improved survival and better bacterial clearance during the early phases of septic shock induced by CLP. The present study also demonstrates improved survival in IL-15 KO mice and a small, but significant, improvement in bacterial clearance at 6 h after CLP. However, we did not observe differences in bacterial burden in the blood or peritoneal cavity among groups at 18 h after CLP. In another experimental model of septic shock elicited by injection with LPS, IL-15 KO mice also display protection, although no live bacterial infection is present. Thus, our studies suggest that the improved outcomes in IL-15 KO mice are associated with attenuated inflammation as evidenced by decreased systemic cytokine production in both the CLP and LPS models rather than alterations in bacterial clearance mechanisms.

NK cells are large granular innate lymphoid cells that play an essential role in tumor surveillance and elimination of virus-infected cells (36, 37). However, as shown in previous studies from our laboratory and others, NK cells participate in the propagation of acute inflammation and physiological dysfunctions in experimental models of polymicrobial peritonitis (38), endotoxin shock (28), pneumococcal pneumonia (29), systemic E. coli and Streptococcus pyogenes infection (30, 31), and polytrauma (32). Our laboratory previously showed that NK cells quickly migrate to sites of infection during intra-abdominal sepsis, in a manner regulated by the chemokines CXCL9 and CXCL10 (23, 33, 34, 39). During sepsis, activated NK cells increased production of IFN-γ, which is known to potentiate inflammation by activating macrophages, dendritic cells, and other innate immune cells (35, 38). In the current study, we show that loss of NK cells is a major factor by which IL-15 KO mice are protected from septic shock. Thus, using a unique model, we provide further evidence that NK cells play an important role in augmenting acute inflammation and contributing to the pathogenesis of organ injury and physiological dysfunction during septic shock.

Our laboratory also reported the contribution of total CD8+ T cells to the pathogenesis of septic shock as CD8+ T cell–deficient mice show improved survival during intraabdominal sepsis induced by CLP (17, 19, 40). However, there is some controversy about the role of CD8+ T cells in the pathogenesis of septic shock. β2-microglobin KO mice, which lack CD8+ T cells, are more susceptible to LPS-induced endotoxin shock, although these mice have been shown to be protected from CLP (40). The present study further showed Ab-mediated depletion of CD8+ T cells in WT mice fails to confer protection from LPS-induced septic shock. Preventing CD8+ T cell regeneration fails to attenuate septic mortality in IL-15 SA–treated IL-15 KO mice. Thus, we conclude that lack of CD8+ T cells does not contribute significantly to the resistance of IL-15 KO mice to septic shock. However, the specific contribution of memory CD8+ T cell subset to the pathogenesis of septic shock has not been well characterized, because it is currently not possible to selectively deplete these cells.

NKT cells are increased in IL-15 KO mice after treatment with low-dose IL-15 SA. It is possible that NKT cells also play a pathogenic role during septic shock as NKT-deficient mice are shown to be resistant to CLP-induced septic shock (41). However, currently there are no Abs available that can selectively deplete NKT cells in IL-15 SA–treated IL-15 KO mice so our exploration of the role of NKT cells in this setting is limited. Nevertheless, we did not note a significant decrease in NKT cell numbers in IL-15 KO mice. Thus, loss of NKT cells does not appear to contribute to the resistance of IL-15 KO mice to septic shock.

IL-15 KO mice lack not only NK cells but also the cytokine itself. It is unclear whether IL-15 alone plays an acute role in the pathobiology of sepsis. In this study, we showed that concentration of soluble IL-15/IL-15Rα was elevated in the plasma of septic mice compared with non-septic mice. Recent studies correlate elevated serum IL-15 concentration with the development of organ dysfunction and mortality in patients with septic shock (22). Thus, it is possible that IL-15 alone plays a detrimental role in the pathogenesis of septic shock. The current study shows that acute neutralization of IL-15 at the onset of septic shock failed to confer protection, whereas long-term neutralization of IL-15 prior to septic shock conferred protection. Further studies showed that acute neutralization of IL-15 neither depletes NK cells nor blocks NK cell activation at the early phase of septic shock. In contrast, long-term neutralization of IL-15 depleted 80.8% of splenic NK cells. Therefore, our studies suggest that endogenous IL-15 does not play an acute role in the pathogenesis of septic shock but contributes to septic mortality by maintaining NK cells, which aggravate systemic inflammation.

Lastly, we reported that treatment of WT mice with higher doses of IL-15 SA (2 μg) exacerbates physiological dysfunctions and mortality during septic shock when initiated either immediately before or shortly after CLP and/or LPS challenge. Activated NK cells primarily mediate the worsened septic lethality caused by coadministration of IL-15 SA. Another recent paper from our laboratory showed that IL-15 SA is effective in expanding NK, NKT, and mCD8+ T cells in burned mice but does not improve survival in a model of Pseudomonas burn wound infection (42). In contrast, Inoue et al. (43) reported that treatment with exogenous IL-15 after CLP challenge attenuates sepsis-induced apoptosis, reverses associated immune dysfunction and improves survival during sepsis. The reasons for the differences observed between the studies are not entirely clear but it could be due to the use of different mouse strains, dosage of IL-15 SA, delivery routes of IL-15 SA, as well as different severity of our models of sepsis. Inbred male and female mice on the C57BL/6J background were employed in the current study whereas Inoue et al. (43) employed male mice in the outbred CD-1 strain of mice. Investigators have noted differences in the response to sepsis among inbred and outbred strains of mice and these differences may partly explain the dissimilarities in our results (44). It is unlikely that gender differences played a role given that many experiments in each study were performed using male mice. A slightly higher dose of IL-15 SA (2 versus 1.5 μg) was used in our study and could have resulted in a more pronounced proinflammatory response. Finally, the severity of models in our posttreatment experiments were different. Our study employed a low lethality model (22% mortality in vehicle-treated mice) whereas a more severe model was employed in their study (75–90% mortality). Regardless of these differences, our current study indicates that IL-15 should be used with caution as an immunomodulator in subjects with acute sepsis due to its potential to augment systemic inflammatory responses. Nevertheless, it is important to consider proper dosing of IL-15 SA because lower doses than those used in either of our studies could provide therapeutic benefit and minimal toxicity.

In conclusion, the current study demonstrates the role of IL-15 in the pathogenesis of septic shock. Endogenous IL-15 does not play an acute role in aggravating septic mortality but is able to facilitate sepsis-induced systemic inflammation, hypothermia, acute organ injuries, and death by maintaining the NK cell pool. Exogenous IL-15 (IL-15 SA), at higher doses, potently activates NK cells and increases the susceptibility of mice to endotoxin shock. The results of this study and others also indicate that IFN-γ contributes significantly to NK cell–mediated inflammation and injury during septic shock as well as in IL-15–induced exacerbation of sepsis-associated pathobiology.

IL-15 neutralizing antibody M96 was generously provided by Amgen (Thousand Oaks, CA).

This work was supported by National Institutes of Health Grant R01 GM66885 and R01 GM104306. N.K.P. is supported by American Heart Association Postdoctoral Grant 16POST29920007.

The online version of this article contains supplemental material.

Abbreviations used in this article:

ALT

alanine aminotransferase

AST

aspartate aminotransferase

BUN

blood urea nitrogen

CLP

cecal ligation and puncture

KO

knockout

mCD8

memory CD8

SA

superagonist

WT

wild type.

1
Puzanov
I. J.
,
Bennett
M.
,
Kumar
V.
.
1996
.
IL-15 can substitute for the marrow microenvironment in the differentiation of natural killer cells.
J. Immunol.
157
:
4282
4285
.
2
Huntington
N. D.
,
Legrand
N.
,
Alves
N. L.
,
Jaron
B.
,
Weijer
K.
,
Plet
A.
,
Corcuff
E.
,
Mortier
E.
,
Jacques
Y.
,
Spits
H.
,
Di Santo
J. P.
.
2009
.
IL-15 trans-presentation promotes human NK cell development and differentiation in vivo.
J. Exp. Med.
206
:
25
34
.
3
Mortier
E.
,
Woo
T.
,
Advincula
R.
,
Gozalo
S.
,
Ma
A.
.
2008
.
IL-15Ralpha chaperones IL-15 to stable dendritic cell membrane complexes that activate NK cells via trans presentation.
J. Exp. Med.
205
:
1213
1225
.
4
Zhang
X.
,
Sun
S.
,
Hwang
I.
,
Tough
D. F.
,
Sprent
J.
.
1998
.
Potent and selective stimulation of memory-phenotype CD8+ T cells in vivo by IL-15.
Immunity
8
:
591
599
.
5
Schluns
K. S.
,
Williams
K.
,
Ma
A.
,
Zheng
X. X.
,
Lefrançois
L.
.
2002
.
Cutting edge: requirement for IL-15 in the generation of primary and memory antigen-specific CD8 T cells.
J. Immunol.
168
:
4827
4831
.
6
Malamut
G.
,
El Machhour
R.
,
Montcuquet
N.
,
Martin-Lannerée
S.
,
Dusanter-Fourt
I.
,
Verkarre
V.
,
Mention
J. J.
,
Rahmi
G.
,
Kiyono
H.
,
Butz
E. A.
, et al
.
2010
.
IL-15 triggers an antiapoptotic pathway in human intraepithelial lymphocytes that is a potential new target in celiac disease-associated inflammation and lymphomagenesis.
J. Clin. Invest.
120
:
2131
2143
.
7
Lodolce
J. P.
,
Boone
D. L.
,
Chai
S.
,
Swain
R. E.
,
Dassopoulos
T.
,
Trettin
S.
,
Ma
A.
.
1998
.
IL-15 receptor maintains lymphoid homeostasis by supporting lymphocyte homing and proliferation.
Immunity
9
:
669
676
.
8
Kennedy
M. K.
,
Glaccum
M.
,
Brown
S. N.
,
Butz
E. A.
,
Viney
J. L.
,
Embers
M.
,
Matsuki
N.
,
Charrier
K.
,
Sedger
L.
,
Willis
C. R.
, et al
.
2000
.
Reversible defects in natural killer and memory CD8 T cell lineages in interleukin 15-deficient mice.
J. Exp. Med.
191
:
771
780
.
9
Fehniger
T. A.
,
Caligiuri
M. A.
.
2001
.
Interleukin 15: biology and relevance to human disease.
Blood
97
:
14
32
.
10
Ma
A.
,
Koka
R.
,
Burkett
P.
.
2006
.
Diverse functions of IL-2, IL-15, and IL-7 in lymphoid homeostasis.
Annu. Rev. Immunol.
24
:
657
679
.
11
Mattei
F.
,
Schiavoni
G.
,
Belardelli
F.
,
Tough
D. F.
.
2001
.
IL-15 is expressed by dendritic cells in response to type I IFN, double-stranded RNA, or lipopolysaccharide and promotes dendritic cell activation.
J. Immunol.
167
:
1179
1187
.
12
Ahmad
R.
,
Ennaciri
J.
,
Cordeiro
P.
,
El Bassam
S.
,
Menezes
J.
.
2007
.
Herpes simplex virus-1 up-regulates IL-15 gene expression in monocytic cells through the activation of protein tyrosine kinase and PKC zeta/lambda signaling pathways.
J. Mol. Biol.
367
:
25
35
.
13
Blaser
B. W.
,
Schwind
N. R.
,
Karol
S.
,
Chang
D.
,
Shin
S.
,
Roychowdhury
S.
,
Becknell
B.
,
Ferketich
A. K.
,
Kusewitt
D. F.
,
Blazar
B. R.
,
Caligiuri
M. A.
.
2006
.
Trans-presentation of donor-derived interleukin 15 is necessary for the rapid onset of acute graft-versus-host disease but not for graft-versus-tumor activity.
Blood
108
:
2463
2469
.
14
Castillo
E. F.
,
Stonier
S. W.
,
Frasca
L.
,
Schluns
K. S.
.
2009
.
Dendritic cells support the in vivo development and maintenance of NK cells via IL-15 trans-presentation.
J. Immunol.
183
:
4948
4956
.
15
Han
K. P.
,
Zhu
X.
,
Liu
B.
,
Jeng
E.
,
Kong
L.
,
Yovandich
J. L.
,
Vyas
V. V.
,
Marcus
W. D.
,
Chavaillaz
P. A.
,
Romero
C. A.
, et al
.
2011
.
IL-15:IL-15 receptor alpha superagonist complex: high-level co-expression in recombinant mammalian cells, purification and characterization.
Cytokine
56
:
804
810
.
16
Rubinstein
M. P.
,
Kovar
M.
,
Purton
J. F.
,
Cho
J. H.
,
Boyman
O.
,
Surh
C. D.
,
Sprent
J.
.
2006
.
Converting IL-15 to a superagonist by binding to soluble IL-15Ralpha.
Proc. Natl. Acad. Sci. USA
103
:
9166
9171
.
17
Sherwood
E. R.
,
Enoh
V. T.
,
Murphey
E. D.
,
Lin
C. Y.
.
2004
.
Mice depleted of CD8+ T and NK cells are resistant to injury caused by cecal ligation and puncture.
Lab. Invest.
84
:
1655
1665
.
18
Enoh
V. T.
,
Fairchild
C. D.
,
Lin
C. Y.
,
Varma
T. K.
,
Sherwood
E. R.
.
2006
.
Differential effect of imipenem treatment on wild-type and NK cell-deficient CD8 knockout mice during acute intra-abdominal injury.
Am. J. Physiol. Regul. Integr. Comp. Physiol.
290
:
R685
R693
.
19
Tao
W.
,
Enoh
V. T.
,
Lin
C. Y.
,
Johnston
W. E.
,
Li
P.
,
Sherwood
E. R.
.
2005
.
Cardiovascular dysfunction caused by cecal ligation and puncture is attenuated in CD8 knockout mice treated with anti-asialoGM1.
Am. J. Physiol. Regul. Integr. Comp. Physiol.
289
:
R478
R485
.
20
Orinska
Z.
,
Maurer
M.
,
Mirghomizadeh
F.
,
Bulanova
E.
,
Metz
M.
,
Nashkevich
N.
,
Schiemann
F.
,
Schulmistrat
J.
,
Budagian
V.
,
Giron-Michel
J.
, et al
.
2007
.
IL-15 constrains mast cell-dependent antibacterial defenses by suppressing chymase activities.
Nat. Med.
13
:
927
934
.
21
Chung
K. P.
,
Chang
H. T.
,
Lo
S. C.
,
Chang
L. Y.
,
Lin
S. Y.
,
Cheng
A.
,
Huang
Y. T.
,
Chen
C. C.
,
Lee
M. R.
,
Chen
Y. J.
, et al
.
2015
.
Severe lymphopenia is associated with elevated plasma interleukin-15 levels and increased mortality during severe sepsis.
Shock
43
:
569
575
.
22
Kimura
A.
,
Ono
S.
,
Hiraki
S.
,
Takahata
R.
,
Tsujimoto
H.
,
Miyazaki
H.
,
Kinoshita
M.
,
Hatsuse
K.
,
Saitoh
D.
,
Hase
K.
,
Yamamoto
J.
.
2012
.
The postoperative serum interleukin-15 concentration correlates with organ dysfunction and the prognosis of septic patients following emergency gastrointestinal surgery.
J. Surg. Res.
175
:
e83
e88
.
23
Herzig
D. S.
,
Luan
L.
,
Bohannon
J. K.
,
Toliver-Kinsky
T. E.
,
Guo
Y.
,
Sherwood
E. R.
.
2014
.
The role of CXCL10 in the pathogenesis of experimental septic shock.
Crit. Care
18
:
R113
.
24
Lebrec
H.
,
Horner
M. J.
,
Gorski
K. S.
,
Tsuji
W.
,
Xia
D.
,
Pan
W. J.
,
Means
G.
,
Pietz
G.
,
Li
N.
,
Retter
M.
, et al
.
2013
.
Homeostasis of human NK cells is not IL-15 dependent.
J. Immunol.
191
:
5551
5558
.
25
Cooper
M. A.
,
Bush
J. E.
,
Fehniger
T. A.
,
VanDeusen
J. B.
,
Waite
R. E.
,
Liu
Y.
,
Aguila
H. L.
,
Caligiuri
M. A.
.
2002
.
In vivo evidence for a dependence on interleukin 15 for survival of natural killer cells.
Blood
100
:
3633
3638
.
26
Chiossone
L.
,
Chaix
J.
,
Fuseri
N.
,
Roth
C.
,
Vivier
E.
,
Walzer
T.
.
2009
.
Maturation of mouse NK cells is a 4-stage developmental program.
Blood
113
:
5488
5496
.
27
Guo
Y.
,
Luan
L.
,
Rabacal
W.
,
Bohannon
J. K.
,
Fensterheim
B. A.
,
Hernandez
A.
,
Sherwood
E. R.
.
2015
.
IL-15 Superagonist-Mediated Immunotoxicity: Role of NK Cells and IFN-γ.
J. Immunol.
195
:
2353
2364
.
28
Emoto
M.
,
Miyamoto
M.
,
Yoshizawa
I.
,
Emoto
Y.
,
Schaible
U. E.
,
Kita
E.
,
Kaufmann
S. H.
.
2002
.
Critical role of NK cells rather than V alpha 14(+)NKT cells in lipopolysaccharide-induced lethal shock in mice.
J. Immunol.
169
:
1426
1432
.
29
Christaki
E.
,
Diza
E.
,
Giamarellos-Bourboulis
E. J.
,
Papadopoulou
N.
,
Pistiki
A.
,
Droggiti
D. I.
,
Georgitsi
M.
,
Machova
A.
,
Lambrelli
D.
,
Malisiovas
N.
, et al
.
2015
.
NK and NKT cell depletion alters the outcome of experimental pneumococcal pneumonia: relationship with regulation of interferon-γ production.
J. Immunol. Res.
2015
:
532717
.
30
Badgwell
B.
,
Parihar
R.
,
Magro
C.
,
Dierksheide
J.
,
Russo
T.
,
Carson
W. E.
 III
.
2002
.
Natural killer cells contribute to the lethality of a murine model of Escherichia coli infection.
Surgery
132
:
205
212
.
31
Goldmann
O.
,
Chhatwal
G. S.
,
Medina
E.
.
2005
.
Contribution of natural killer cells to the pathogenesis of septic shock induced by Streptococcus pyogenes in mice.
J. Infect. Dis.
191
:
1280
1286
.
32
Barkhausen
T.
,
Frerker
C.
,
Pütz
C.
,
Pape
H. C.
,
Krettek
C.
,
van Griensven
M.
.
2008
.
Depletion of NK cells in a murine polytrauma model is associated with improved outcome and a modulation of the inflammatory response.
Shock
30
:
401
410
.
33
Herzig
D. S.
,
Driver
B. R.
,
Fang
G.
,
Toliver-Kinsky
T. E.
,
Shute
E. N.
,
Sherwood
E. R.
.
2012
.
Regulation of lymphocyte trafficking by CXC chemokine receptor 3 during septic shock.
Am. J. Respir. Crit. Care Med.
185
:
291
300
.
34
Bohannon
J.
,
Guo
Y.
,
Sherwood
E. R.
.
2012
.
The role of natural killer cells in the pathogenesis of sepsis: the ongoing enigma.
Crit. Care
16
:
185
.
35
Herzig
D.
,
Fang
G.
,
Toliver-Kinsky
T. E.
,
Guo
Y.
,
Bohannon
J.
,
Sherwood
E. R.
.
2012
.
STAT1-deficient mice are resistant to cecal ligation and puncture-induced septic shock.
Shock
38
:
395
402
.
36
Morvan
M. G.
,
Lanier
L. L.
.
2016
.
NK cells and cancer: you can teach innate cells new tricks.
Nat. Rev. Cancer
16
:
7
19
.
37
Jost
S.
,
Altfeld
M.
.
2013
.
Control of human viral infections by natural killer cells.
Annu. Rev. Immunol.
31
:
163
194
.
38
Etogo
A. O.
,
Nunez
J.
,
Lin
C. Y.
,
Toliver-Kinsky
T. E.
,
Sherwood
E. R.
.
2008
.
NK but not CD1-restricted NKT cells facilitate systemic inflammation during polymicrobial intra-abdominal sepsis.
J. Immunol.
180
:
6334
6345
.
39
Herzig
D. S.
,
Guo
Y.
,
Fang
G.
,
Toliver-Kinsky
T. E.
,
Sherwood
E. R.
.
2012
.
Therapeutic efficacy of CXCR3 blockade in an experimental model of severe sepsis.
Crit. Care
16
:
R168
.
40
Sherwood
E. R.
,
Lin
C. Y.
,
Tao
W.
,
Hartmann
C. A.
,
Dujon
J. E.
,
French
A. J.
,
Varma
T. K.
.
2003
.
Beta 2 microglobulin knockout mice are resistant to lethal intraabdominal sepsis.
Am. J. Respir. Crit. Care Med.
167
:
1641
1649
.
41
Hu
C. K.
,
Venet
F.
,
Heffernan
D. S.
,
Wang
Y. L.
,
Horner
B.
,
Huang
X.
,
Chung
C. S.
,
Gregory
S. H.
,
Ayala
A.
.
2009
.
The role of hepatic invariant NKT cells in systemic/local inflammation and mortality during polymicrobial septic shock.
J. Immunol.
182
:
2467
2475
.
42
Patil
N. K.
,
Luan
L.
,
Bohannon
J. K.
,
Guo
Y.
,
Hernandez
A.
,
Fensterheim
B.
,
Sherwood
E. R.
.
2016
.
IL-15 superagonist expands mCD8+ T, NK and NKT cells after burn injury but fails to improve outcome during burn wound infection.
PLoS One
11
:
e0148452
.
43
Inoue
S.
,
Unsinger
J.
,
Davis
C. G.
,
Muenzer
J. T.
,
Ferguson
T. A.
,
Chang
K.
,
Osborne
D. F.
,
Clark
A. T.
,
Coopersmith
C. M.
,
McDunn
J. E.
,
Hotchkiss
R. S.
.
2010
.
IL-15 prevents apoptosis, reverses innate and adaptive immune dysfunction, and improves survival in sepsis.
J. Immunol.
184
:
1401
1409
.
44
Fink
M. P.
2014
.
Animal models of sepsis.
Virulence
5
:
143
153
.

The authors have no financial conflicts of interest.

Supplementary data