Leptin Corrects Host Defense Defects after Acute Starvation in Murine Pneumococcal Pneumonia

Streptococcus pneumoniae colonizes the upper respiratory tract of susceptible hosts and is the most common cause of communityacquired pneumonia, accounting for 175,000 hospitalizations per year in the United States alone (1, 2). Individuals who are particularly susceptible to pneumococcal pneumonia include the critically ill and those with chronic diseases, such as alcoholism, chronic obstructive pulmonary disease, and cancer (3). These same individuals frequently become energy malnourished and this contributes to their increased susceptibility to infection (4-7). The mechanisms responsible for impaired innate immune responses against bacterial infections arising from energy malnutrition are poorly understood.

Acute energy malnutrition, induced by fasting, initiates a number of physiologic adaptations that conserve energy stores at the expense of the immune system and ultimately, host defense against infection (8). These changes include hypoglycemia, increases in immunosuppressive glucocorticoids, and reductions in circulating leptin (9-11). Associated with these alterations are defects in macrophage and neutrophil effector functions, decreases in serum complement, atrophy of the thymus and spleen, and reductions in circulating lymphocytes (8-10, 12).

Leptin is a pleiotropic hormone that has been identified as an important signal in the neuroendocrine response to fasting that regulates energy homeostasis and immune function (9-11, 13, 14). Under normal circumstances, blood leptin levels are positively correlated with total body fat mass, but decline dramatically during fasting disproportionately to body fat stores (15-17). We have previously reported that leptin-deficient mice exhibit an impaired host response against gram-negative pneumonia in vivo and this defect was associated with impaired macrophage and neutrophil phagocytosis of Klebsiella pneumoniae and reduced macrophage leukotriene synthesis in vitro (18, 19). All of these defects could be restored with exogenous leptin in vitro. In the current study, we asked if a physiologic stimulus to reduce circulating leptin levels (fasting for 48 h) impaired host defense against pneumococcal pneumonia. We also tested the effects of maintaining circulating leptin levels during acute energy deprivation on the host response against pneumococcal pneumonia in vivo. We observed that leptin reconstituted the fasting-induced host defense impairments against S. pneumoniae. This is the first report, to our knowledge, that demonstrates that leptin administration in vivo plays a protective role in pulmonary antibacterial host defense.

METHODS

Mice

C57BL/6J mice, 10 to 12 wk old, were obtained from Jackson Laboratories (Bar Harbor, ME) and maintained under specific pathogen-free conditions. All experiments were in compliance with the Animal Care and Use Committee of the University of Michigan.

Acute Starvation Model

Mice were randomly assigned to one of three groups: (1) ad libitum fed for 48 h (control) and given intraperitoneal injections of saline; (2) fasted for 48 h and given intraperitoneal injections of saline; or (3) fasted for 48 h and given recombinant murine leptin (Calbiochem, La Jolla, CA), 1 µg/g initial body weight, via intraperitoneal injections. Injections were given at 9:00 A.M. and 6:00 P.M. This leptin replacement protocol during fasting has been shown to achieve, 6 to 12 h postinjection, circulating leptin levels similar to those of ad libitum-fed animals (9, 11). All animals were provided with unlimited access to water.

S. pneumoniae Inoculation

S. pneumoniae serotype 3,6303, was obtained from the American Type Culture Collection (Manassa, VA), and aliquots were grown to midlogarithmic phase in Todd-Hewett broth with 0.5% yeast extract (Difco, Detroit, MI) at 37°C. The concentration of bacteria was determined spectrophotometrically (A^sub 600^; see online supplement for details). Mice were anesthetized with ketamine and xylazine (80 and 10 mg/kg of body weight, respectively), and 10^sup 5^ cfu of S. pneumoniae, suspended in 40 µl of saline, were administered via the intranasal route.

Bronchoalveolar Lavage and Peripheral Blood Cell Differential Counts

Bronchoalveolar lavage (BAL) was performed as previously described (20) and peripheral blood was obtained by orbital bleeding from uninfected mice and infected mice 24 h after S. pneumoniae challenge. The total number and differential counts of viable leukocytes in BAL fluid (BALF) and peripheral blood, isolated after red blood cell lysis (Unopette Microcollection System; Becton-Dickson, Rutherford, NJ), were determined as previously reported (18).

Determination of Lung and Blood S. pneumoniae Colony-forming Units

Twenty-four hours after 5. pneumoniae challenge, lung homogenate and blood colony-forming units were determined as previously described (21).

Cytokine, Leptin, Corticosterone, and Glucose Determinations

Murine cytokines granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin 6 (IL-6), IL-10, IL-12, IFN-?, macrophage inflammatory protein 2 (MIP-2), tumor necrosis factor a (TNF-a) (OPTEIA kits; BD Pharmingen, San Diego, CA), and leptin (Assay Designs, Ann Arbor, MI) were determined in BALF and lung homogenates after extraction as previously described (18). Leptin and corticosterone (Assay Designs) levels were also determined in blood samples by ELISA. Blood glucose was determined using a glucometer (Glucometer Elite; Bayer, Elkhart, IN).

Alveolar Macrophage Leukotriene B^sub 4^ Determination

Murine alveolar macrophages (AMs) were obtained by lavage, washed once in phosphate-buffered saline, resuspended in Roswell Park Memorial Institute (RPMI) 1640 (Invitrogen, Carlsbad, CA) to a concentration of 10^sup 5^ cells/ml, and stimulated with the calcium ionophore A23187 for 30 min at 37°C in 5% CO2 in air. The culture medium was assayed for leukotriene B^sub 4^ (LTB^sub 4^) using commercially available EIA kits (Cayman Chemical, Ann Arbor, MI).

AM Killing and Phagocytosis of S. pneumoniae In Vitro

For killing experiments, 1 × 10^sup 6^ cfu of S. pneumoniae were cultured in RPMI 1640 and 10% autologous serum with or without 5 × 10^sup 5^ AMs from mice of all treatment groups. Three samples from each tube were serially diluted and plated on soy-based blood agar plates (Difco) before (Time 0) and 90 min after incubation. After 16 h of incubation at 37°C, 5. pneumoniae colony-forming units were enumerated. A total of 2 × 10^sup 5^ AMs were incubated with 2 × 10^sup 6^ cfu of fluorescently labeled (BODIPY FL; Molecular Probes, Eugene, OR) S. pneumoniae with 2.5% autologous serum in RPMI (total volume, 1 ml) for 30 min. The percentage of AMs that had phagocytosed fluorescent 5. pneumoniae was determined by counting 200 macrophages in random fields using fluorescent microscopy (1,000×), and the data in experimental animals were expressed as a percentage of the values observed in AMs from fed animals (see online supplement for details).

Statistical Analysis

Where appropriate, mean values were compared using a paired t test, a one-way analysis of variance, or a Kruskal-Wallis test on ranks for nonparametric data. The Student-Newman-Keuls or the Dunnett's test was used for mean separation. In all cases, a p value of less than 0.05 was considered significant.

RESULTS

Effect of Fasting and S. pneumoniae Challenge on Body Weight and Leptin

To deplete circulating leptin levels, we used a 48-h fast before S. pneumoniae challenge. We also administered leptin to fasted mice to determine if this would preserve circulating leptin levels as indicated by previous reports (11, 14). As shown in Figure 1, 48 h of fasting resulted in a reduction of approximately 4 g or 18% of the initial body weight. Similar weight losses were observed in the fasted animals given leptin. By 72 h (24 h post-S. pneumoniae challenge), body weight had declined an additional 1.5 g in both the fasted and fasted animals given leptin. In the fed animals, we observed a decline of 1 g (5%) in body weight 24 h after S. pneumoniae challenge. As anticipated, blood leptin levels declined in uninfected animals after fasting. When exogenous leptin was given to fasted animals, blood leptin was maintained at levels similar to those observed in fed animals (Figure 2A). Blood leptin levels increased in all groups after S. pneumoniae challenge, were lowest in fasted animals, and were not different between fed and fasted mice given leptin. No differences in lung homogenate leptin levels were observed between treatment groups in uninfected animals. In contrast, lung homogenate leptin levels postinfection increased twofold in fed and 10-fold in fasted mice given leptin (Figure 2B).

Blood Glucose and Corticosterone Levels

As shown in Table 1, baseline blood glucose levels declined in animals that were either fasted for 48 h or fasted and given leptin. Blood glucose levels decreased 24 h after S, pneumoniae challenge in all groups and no differences were observed between mice that were fasted and fasted with leptin. As anticipated, fasting resulted in an increase in blood corticosterone levels and the provision of exogenous leptin slightly but not significantly attenuated this response. Blood corticosterone levels increased in both groups of fasted mice 24 h after S. pneumoniae challenge. Although corticosterone increased modestly (p = 0.31) in the blood of the fed animals after S. pneumoniae challenge, this value was not statistically different than that of the fasted and fasted mice given leptin. Thus, administration of leptin did not correct hypoglycemia or hypercorticosteronemia with fasting.

Impaired In Vivo Bacterial Clearance in Fasted Animals Is Restored with Exogenous Leptin

To determine if fasting suppresses pulmonary antibacterial host defense, we compared the lung bacterial burdens in each group of mice 24 h after challenge with an intranasal inoculum of 10^sup 5^ cfu S. pneumoniae. As shown in Figure 3, the S. pneumoniae burden in fasted mice was approximately 20-fold greater than that of fed animals. Leptin administration largely prevented this excess of lung colony-forming units (threefold down from a 20-fold increase). We did not observe positive cultures for 5. pneumoniae in the peripheral blood in any of the experimental groups.

Restoration of Impaired Pulmonary Polymorphonuclear Neutrophil Recruitment in Fasted Mice with Exogenous Leptin

We next determined whether the observed impairment in pulmonary bacterial clearance in fasted mice correlated with defective cellular recruitment. There were no differences between treatment groups in the total number of cells, which consisted of more than 95% AMs, in the BALF of uninfected animals (data not shown). By 24-h post-S. pneumoniae challenge, the number of polymorphonuclear neutrophils (PMNs) in the BALF of the fed mice had increased to approximately 7 × 10^sup 5^ cells/ml (Figure 4). The number of PMNs in the BALF of fasted mice was approximately 50% less than that of the fed animals. In contrast, sustaining circulating leptin levels in fasted animals entirely restored pulmonary PMN recruitment.

Exogenous Leptin Maintains Peripheral Blood Leukocytes in Fasted Animals post-S. pneumoniae Challenge

Because the impairment in PMN recruitment observed in fasted mice postinfection could reflect a reduction of these cells in peripheral blood, we performed total and differential cell counts on peripheral blood obtained from mice after fasting for 48 and 24 h post-S. pneumoniae administration. There were no differences in peripheral blood total leukocyte counts in uninfected animals. As shown in Figure 5, 24 h after 5. pneumoniae challenge there were approximately 65% less total leukocytes in the fasted (1.02 × 10^sup 6^ leukocytes/ml) as compared with the fed mice (2.95 × 10^sup 6^ leukocytes/ml). This difference can be explained by the decreased number of PMNs, lymphocytes (1.6 × 10^sup 6^ in fed vs. 8.4 × 10^sup 5^ lymphocytes/ml in fasted mice), and monocytes (3.5 × 10^sup 5^ in fed vs. 1.0 × 10^sup 5^ cells/ml in fasted mice). Administration of leptin to fasted animals nearly restored total leukocyte numbers (2.16 × 10^sup 6^ leukocytes/ml), and PMN, lymphocyte (1.32 × 10^sup 6^ cells/ml), and monocyte counts (1.3 × 10^sup 5^ cells/ml) after S. pneumoniae challenge.

Leptin Restores IL-6 and MlP-2 Levels in BALF in Fasted Mice

An effective host response against pneumococcal pneumonia requires the elaboration of proinflammatory cytokines and chemokines that promote macrophage effector functions and PMN recruitment. Therefore, we assessed, in lung homogenates and BALF of mice from the various experimental groups, levels of cytokines (GM-CSF, IL-6, IL-10, IL-12, IFN-?, MIP-2, and TNF-a) known to play an important role in host defense against bacterial pneumonia. Although we did not observe differences in GM-CSF, IL-10, IL-12, IFN-?, or TNF-a (data not shown) between our treatment groups, we did observe that fasting reduced IL-6 and MIP-2 levels in BALF after S. pneumoniae challenge. Leptin administration to fasted animals maintained the levels of these cytokines relative to those of fed animals (Figure 6).

Exogenous Leptin Reconstitutes Decreased AM LTB^sub 4^ Synthesis

Previous reports have indicated that leptin deficiency is associated with a reduction in the ability of AMs to produce leukotrienes, which are known to play an important role in the host response against bacterial pneumonia (18, 22-24). Therefore, we assessed LTB^sub 4^ synthesis in AMs after stimulation with the calcium ionophore A23187 in vitro. We chose to study the ex vivo synthetic capacity of AMs recovered from murine BALF rather than measuring LTB^sub 4^ in lung homogenate or BALF samples from infected mice. We did this because LTB^sub 4^ is the most abundant leukotriene produced by PMNs, which were reduced in the BALF of fasted mice after infection (Figure 4). As shown in Figure 7, 48 h of fasting reduced the ability of AMs to synthesize LTB^sub 4^, and the provision of exogenous leptin to fasted animals nearly maintained this ability.

Defective AM Killing of S. pneumoniae in Fasted Animals Is Restored with Exogenous Leptin

To determine the effects of fasting on AM bactericidal capacity, we compared the ability of these cells to kill S. pneumoniae in vitro. When S. pneumoniae was cultured with serum in the absence of AMs, bacterial colony-forming units increased approximately fourfold within 90 min (Figure 8A). When AMs from fed animals were incubated with bacteria, less than 10% of the bacteria survived (i.e., > 90% killing). In comparison to cells from fed animals, threefold more 5. pneumoniae survived after being cultured with AMs from fasted animals. In contrast, we did not observe an impairment in the bactericidal function of AMs harvested from mice given exogenous leptin in vivo during fasting.

AM Phagocytosis of 5. pneumoniae Is Modestly Impaired in Fasted Animals

AM phagocytosis in cells from fed, fasted, and fasted animals given exogenous leptin was assessed to determine if the observed impairments in pulmonary bacterial clearance in vivo and bacterial killing in vitro were due to defective phagocytosis. As shown in Figure 8B, phagocytosis of S. pneumoniae in AMs from fasted animals was approximately 20% less than that of the control animals (p = 0.03). The provision of exogenous leptin in vivo during fasting did not affect this response.

DISCUSSION

Previously, we observed severe host-defense impairments against gram-negative pneumonia in genetically leptin-deficient mice. In the current study, we tested the effects of a physiologic reduction in leptin induced by acute starvation on pulmonary host defense. We also studied the effects of leptin administration to fasted animals to determine the role of leptin in fasting-induced immune suppression. After an intranasal challenge with S. pneumoniae, pulmonary bacterial clearance was impaired in fasted animals and this defect was associated with reduced PMN accumulation and IL-6 and MIP-2 levels in the BALF. In addition, peripheral blood leukocyte counts did not increase after S. pneumoniae challenge in fasted animals. We also observed that LTB^sub 4^ synthesis and phagocytosis and killing of S. pneumoniae in vitro were also impaired in AMs from fasted animals. Leptin administration to fasted mice restored bacterial clearance, circulating and BALF PMN counts, IL-6 and MIP-2 levels in BALF, and AM LTB^sub 4^ synthesis and bacterial killing. These results demonstrate that leptin deficiency plays an important role in the impaired host defense against pneumococcal pneumonia in energy deprivation, and suggests a potential role for this adipokine as a therapeutic agent in the energy-deprived immunocompromised host.

Increases in both blood and lung-homogenate leptin levels were observed in both fed and fasted animals after intranasal S. pneumoniae challenge. This observation was consistent with our previous report showing that leptin levels increase in the blood and lungs of mice challenged with K. pneumoniae via the intratracheal route (18). Others have also reported elevated leptin levels in the blood and tissues during inflammation and infection (25-29). Although cells other than adipocytes, such as T cells and macrophages, are capable of leptin synthesis (26, 30), the adipocytes located in various fat depots are probably the major source of leptin produced during bacterial pneumonia that can leak into the lung interstitium and alveolar space with other plasma proteins. The mechanism that would account for the observed increases of lung leptin to supranormal levels after S. pneumoniae challenge among the fasted animals given exogenous leptin is unclear. We speculate that it may reflect increased microvascular permeability in response to pulmonary inflammation, which resulted in preferential leptin accumulation in the lung. The presence of increased leptin levels in the lung may have directly enhanced bacterial clearance by augmenting AM cytokine and LTB^sub 4^ synthesis.

IL-6, LTB^sub 4^, and MIP-2 are known to play protective roles in the host response against bacterial pneumonia, and the reduction in the levels of these mediators produced in fasted animals was associated with an increase in lung S. pneumoniae burden and lower numbers of PMNs (31, 32). When IL-6 knockout mice were challenged with S. pneumoniae, their survival was reduced and their pulmonary bacterial clearance was defective (32). The mechanism responsible for these impairments is not clear. However, it is likely that IL-6 produces both pro- and antiinflammatory effects in the lung during pneumococcal pneumonia. High doses of IL-6 enhance PMN phagocytosis and superoxide release and increase the survival of PMNs by inhibiting apoptosis (33, 34). The antiinflammatory effects of IL-6 include the inhibition of the proinflammatory cytokines TNF-a and IL-1 and the induction of an important tissue inhibitor of matrix metalloproteinase (TIMP-1). These effects may limit damage to pulmonary tissues during pneumococcal pneumonia and prevent extrapulmonary bacterial dissemination (35-37).

LTB^sub 4^ also plays an essential role in protecting the host against both Klebsiella and pneumococcal pneumonia (22, 31). LTB^sub 4^ is not only a potent chemoattractant of PMNs to sites of infection but it also augments the ability of AMs and PMNs to phagocytose and kill K. pneumoniae in vitro (20, 24, 38, 39). The PMN chemoattractant MIP-2 has also been shown to activate PMNs for improved phagocytosis and killing of Escherichia coli (40). The ability of exogenous leptin to improve bacterial clearance in fasted animals may have been due to the restoration of these mediators that are known to enhance leukocyte activation and PMN accumulation. Furthermore, leptin has been shown to augment the synthesis of IL-6, MIP-2, and LTB^sub 4^ (18, 41-43).

AMs from fasted mice exhibited a modest reduction in their ability to phagocytose S. pneumoniae in vitro and we have previously reported that phagocytosis of K. pneumoniae was attenuated in AMs and PMNs harvested from leptin-deficient mice. We have also shown that culturing these cells with leptin in vitro reconstitutes phagocytosis and that administering leptin to leptin-deficient mice in vivo restores the phagocytic defect in PMNs (18, 19). However, the provision of exogenous leptin to fasted mice did not affect AM phagocytosis of S. pneumoniae, suggesting that the impairment in bacterial clearance observed in fasted animals could not be explained by changes in AM phagocytosis. The reconstitution of pulmonary bacterial clearance in fasted animals given exogenous leptin was partially explained by the restoration of AM bactericidal function. Although the mechanism(s) by which leptin restored killing of S. pneumoniae in AMs from fasted mice is a subject for future investigation, we speculate that leptin may have augmented the elaboration of bactericidal reactive oxygen species because this effect has been observed in human PMNs treated with leptin in vitro (44).

The net accumulation of PMNs into the lungs during bacterial pneumonia reflects local lung determinants for recruitment and survival as well as the magnitude of production of these cells by the bone marrow. Therefore, the reduced number of PMNs found in the BALF of the fasted mice may be attributed to both a reduction in the elaboration of pulmonary MIP-2 and LTB^sub 4^, as well as the lower numbers of these cells found in peripheral blood samples. This latter defect cannot be explained by decreases in GM-CSF induced by fasting because we did not observe significant differences between treatment groups in the most abundant source of this cytokine during bacterial pneumonia (the lung homogenates) in infected animals (data not shown). The maintenance of peripheral blood PMNs observed in fasted animals given leptin post-S. pneumoniae challenge may have been due to the direct effects of leptin, which has been shown to stimulate the proliferation of murine myelocytic and hematopoietic progenitor cells in vitro (45). Another plausible explanation is that leptin may have promoted PMN survival as this adipokine has been shown to inhibit apoptosis in neutrophils and dendritic cells by inducing the expression of antiapoptotic proteins bcl-^sub 2^ and bcl-^sub XL^ (46, 47).

The ability of starvation to suppress immune responses and increase susceptibility to infectious disease has been recognized for a number of years (48). Physiologic adaptations to starvation include increases in glucocorticoids that ultimately act to maintain blood glucose levels by promoting gluconeogenesis through the metabolism of energy substrates from skeletal muscle and adipose tissue. Reductions in circulating lymphocytes and atrophy of lymphoid tissues also accompany acute starvation in mice (10, 14). Hypoglycemia is known to suppress the respiratory burst and, potentially, microbial killing in peripheral blood PMNs (49). Similarly, pharmacologic doses of glucocorticoids are known to decrease PMN recruitment and cytokine synthesis, and predispose the host to opportunistic infections (50, 51). We observed that acute starvation resulted in weight loss, hypoglycemia, elevations in blood corticosterone, and reduced serum leptin levels. While the provision of exogenous leptin did maintain circulating leptin levels at baseline and after S. pneumoniae challenge, it did not significantly affect blood glucose or corticosterone levels. These results indicate that the ability of leptin to reverse the fasting-induced suppression of the innate immune response against S. pneumoniae was not due to alterations in blood glucose or glucocorticoid levels.

In summary, we have demonstrated that sustaining circulating leptin levels during fasting can prevent impairments in pulmonary host defense against S. pneumoniae. This report provides novel insights into the role of leptin in the regulation of the innate immune response against bacterial infection in vivo and is the first study, to our knowledge, that demonstrates that leptin administration in vivo plays a protective role against bacterial pneumonia. These results support the hypothesis that leptin is a key link between nutritional status and innate immune function, and these findings provide a rationale for future studies exploring the role of leptin as a therapeutic agent to augment the immune response in the energy-deprived immune-suppressed host.

Conflict of Interest Statement: None of the authors have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

Acknowledgment: The authors thank Dr. John Younger for helpful insight into the discussion of data.

© 2006 American Thoracic Society Provided by ProQuest LLC. All Rights Reserved.