SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

A breakdown in intestinal barrier function and increased bacterial translocation are key events in the pathogenesis of sepsis and liver disease. Altering gut microflora with noninvasive and immunomodulatory probiotic organisms has been proposed as an adjunctive therapy to reduce the level of bacterial translocation and prevent the onset of sepsis. The purpose of this study was to determine the efficacy of a probiotic compound in attenuating hepatic and intestinal injury in a mouse model of sepsis. Wild-type and interleukin-10 (IL-10) gene–deficient 129 Sv/Ev mice were fed the probiotic compound VSL#3 for 7 days. To induce sepsis, the mice were injected with lipopolysaccharide (LPS) and D-galactosamine (GalN) in the presence and absence of the peroxisome proliferator-activated receptor gamma (PPARγ) inhibitor GW9662. The mice were killed after 6 hours, and their colons were removed for the measurement of the cytokine production and epithelial function. The functional permeability was assessed by the mannitol movement and cyclic adenosine monophosphate–dependent chloride secretion in tissue mounted in Ussing chambers. The livers were analyzed for bacterial translocation, cytokine production, histological injury, and PPARγ levels. The tissue levels of tumor necrosis factor alpha, interferon gamma, IL-6, and IL-12p35 ribonucleic acid were measured by semiquantitative reverse transcription polymerase chain reaction. Mice injected with LPS/GalN demonstrated a breakdown in colonic barrier function, which correlated with enhanced proinflammatory cytokine secretion, bacterial translocation, and significant hepatic injury. A pretreatment with oral probiotics prevented the breakdown in intestinal barrier function, reduced bacterial translocation, and significantly attenuated liver injury. The inhibition of PPARγ with GW9662 abrogated the protection induced by probiotics. Conclusion: Orally administered probiotics prevented liver and intestinal damage in a mouse model of sepsis through a PPARγ-dependent mechanism. (HEPATOLOGY 2007.)

Liver dysfunction and failure contribute to the high mortality rates seen in patients with Gram-negative sepsis. The presence of lipopolysaccharide (LPS) from Gram-negative bacteria in the systemic circulation results in the activation of the innate immune system and the secretion of high levels of proinflammatory cytokines. In animal models, LPS challenge can induce a systemic reaction resulting in a sepsis-like condition characterized by fever, hypotension, and widespread tissue damage. D-Galactosamine (GalN) increases the susceptibility of mice to LPS-induced shock by impairing liver metabolism.1 Challenging mice with low doses of LPS in conjunction with GalN results in massive liver apoptosis and increased mortality.

Tumor necrosis factor alpha (TNF-α) plays a central role in the overwhelming systemic inflammatory response to LPS.2 However, the complete blockade of TNF-α production does not improve survival in animals or humans3 The activation of nuclear factor kappa B (NF-κB) has been shown to play a key role in the pathogenesis of sepsis and is a pivotal step in the regulation of several immune and proinflammatory genes, including TNF-α.4 The modulation of NF-κB activity has been proposed as a strategy for reducing the mortality associated with sepsis. Peroxisome proliferator-activated receptor gamma (PPARγ) is a nuclear hormone receptor and transcription factor that has been shown to have a role in regulating NF-κB activity.5 High levels of PPARγ expression are found in the liver and colon,6 and studies have suggested that PPARγ may have an important role in the maintenance of the intestinal mucosal barrier.7 The activation of the PPARγ receptor by either endogenous ligands or pharmacological compounds8–10 attenuates the onset and severity of septic shock.

The gastrointestinal tract appears to play a key role in the pathogenesis of liver failure in septic models as the high systemic levels of TNF-α induce a breakdown of intestinal barrier function and allow the translocation of bacteria to the liver.11 The presence of bacteria or bacterial products, including endotoxin and bacterial DNA, in the liver causes an amplification of the systemic inflammatory response to endotoxin. Endogenous probiotic bacteria of the gut such as Bifidobacterium and Lactobacillus play a vital role in maintaining the intestinal mucosal barrier.12, 13 Probiotic bacteria found in the 8-strain preparation VSL#3 have been shown to modulate intestinal epithelial barrier function and cytokine secretion through effects on epithelial cells and modulation of the NF-κB and PPARγ pathways.12, 14, 15 The administration of probiotic bacteria has been shown to reduce the translocation of pathogenic bacteria and reduce tissue damage in various animal injury models.16, 17

On the basis of these findings, we hypothesized that the probiotic compound VSL#3 would be effective in preventing liver injury in a septic model by maintaining intestinal barrier function and attenuating proinflammatory cytokine secretion. In agreement with this hypothesis, our findings demonstrate that the oral administration of VSL#3 prevents the breakdown in colonic barrier function associated with sepsis and attenuates liver injury induced by LPS and D-galactosamine (GalN) through a PPARγ-dependent mechanism.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Animals and In Vivo Procedures.

Wild-type (WT) and interleukin-10 (IL-10)–deficient mice on a 129 Sv/Ev background were fed standard 9% fat rodent blocks and housed in individual cages under pathogen-free conditions. The facility's sanitation was verified by Health Sciences Lab Animal Services at the University of Alberta. For in vivo studies, separate groups of mice received daily oral gavage with the probiotic compound VSL#3 (Bifidobacterium longum, Bifidobacterium breve, Bifidobacterium infantis, Lactobacillus casei, Lactobacillus plantarum, Lactobacillus acidophilus, Lactobacillus delbrueckii subsp. Bulgaricus, and Streptococcus salivarius subsp. thermophilus; VSL Pharmaceuticals, Fort Lauderdale, FL; 2.8 × 108 colony-forming units in 30 μl) for 7 days. This dosage was selected on the basis of the dose per kilogram shown to be effective in human trials and was the dosage used in previous studies that showed a beneficial effect in IL-10–deficient mice.13 After 7 days of the VSL#3 treatment, the mice were subjected to an LPS (Fluka, Oakville, Canada)/GalN (Sigma, Oakville, Canada) challenge. On the day of the study, the mice were injected intraperitoneally with 40 μg/kg LPS (Escherichia coli O55:B5) and 360 mg/kg GalN in phosphate-buffered saline according to the method of Galanos et al.18 The control mice received an injection with saline. Preliminary dose-dependent studies demonstrated that mice receiving this LPS/GalN combination developed increased intestinal permeability and maximal enhanced gut cytokine secretion within 4-6 hours following injection (data not shown). Thus, a time of 6 hours following injection was chosen for the experimental measurements.

All experiments were performed according to the Canadian Council of Animal Care guidelines for the care and use of laboratory animals in research and with the permission of the local ethics committee.

Mucosal Cytokine Measurements.

Mice were euthanized by cervical dislocation, and whole colons were removed, flushed with phosphate-buffered saline, and resuspended in tissue culture plates (Falcon 3046, Becton Dickinson Labware, Lincoln Park, NJ) in RPMI-1640 media supplemented with 50 mM 2-mercaptoethanol, 10% fetal bovine serum, streptomycin (100 U/ml), and penicillin (100 U/ml). The cultures were incubated at 37°C in 5% CO2 for 24 hours. Supernatants were harvested and stored at −70°C for cytokine level quantification. The levels of TNF-α and interferon gamma (IFN-γ) in the culture supernatants were measured with enzyme-linked immunosorbent assay kits (Medicorp, Montreal, Canada) according to the manufacturer's instructions.

Determination of Cytokine Ribonucleic Acid (RNA).

Total RNA was isolated from tissue samples with Trizol (Gibco, Burlington, Canada) according to the manufacturer's instructions. RNA (1 μg) was quantified by the determination of the absorbance at 260 nm on a Bio-Rad SmartSpec 3000 with a conversion factor of the absorbance at 260 nm of 1.00, which was equal to 40.0 μg/ml RNA. The absorbance at 280 nm was also detected to determine protein contamination. The 260 nm/280 nm ratio determined the purity of the extraction, with a ratio of 1.7 determined to be clean. RNA was reverse-transcribed and amplified by a polymerase chain reaction (PCR) with a GeneAmp 2400 (Applied Biosystems, Foster City, CA). The results were standardized to β-actin.19 IL-6, IL-10, IL-12p35, TNF-α, IFN-γ, PPARγ, lipocalin-2, and β-actin sequences were obtained from GenBank and used to design intron-spanning primers. The primer sequences used for PCR are shown in Table 1.

Table 1. PCR Primer Sequences
Primer NameSequence
TNF-α forward primer5′-ATGAGCACAGAAAGCATGATC-3′
TNF-α reverse primer5′-TACAGGCTTGTCACTCGAATT-3′
IFN-γ forward primer5′-TACTGCCACGGCACAGTCATTGAA-3′
IFN-γ reverse primer5′-GCAGCGACTCCTTTTCCGCTTCCT-3′
PPARγ forward primer5′-AAACATATCACCCCCCTGCA-3′
PPARγ reverse primer5′-GCAGCAGGTTGTCTTGGATG-3′
IL-10 forward primer5′-GTGAAGACTTTCTTTCAAACAAAG-3′
IL-10 reverse primer5′-CTGCTCCACTGCCTTGCTCTTATT-3′
IL-6 forward primer5′-CACAAAGCCAGAGTCCTTCAGAG-3′
IL-6 reverse primer5′-CTAGGTTTGCCGAGTAGATCTC-3′
IL-12p35 forward primer5′-CTTTGATGATGACCCTGTGC-3′
IL-12p35 reverse primer5′-TTTGGGGAGATGAGATGTGA-3′
Lipocalin-2 forward primer5′-TTTCACCCGCTTTGCCAAGT-3′
Lipocalin-2 reverse primer5′-GTCTCTGCGCATCCCAGTCA-3′
Actin forward primer5′-CGTGGGCCGCCCTAGGCACCA-3′
Actin reverse primer5′-TTGGCCTTAGGGTTCAGGGGGG-3′

Epithelial Function.

For measurements of the permeability, a colonic segment was mounted in Ussing chambers exposing mucosal and serosal surfaces to 10 ml of oxygenated Krebs buffer (115 mM NaCl, 8 mM KCl, 1.25 mM CaCl2, 1.2 mM MgCl2, 2 mM KH2PO4, and 225 mM NaHCO3; pH 7.35). The buffers were maintained at 37°C by a heated water jacket and circulated by CO2/O2. Fructose (10 mM) was added to the serosal and mucosal sides. For measurements of the basal mannitol fluxes, 1 mM mannitol with 10 μCi [H3]mannitol was added to the mucosal side. The spontaneous transepithelial potential difference (PD) was determined, and the tissue was clamped at a zero voltage by the continuous introduction of an appropriate short-circuit current (Isc) with an automatic voltage clamp (DVC 1000, World Precision Instruments, New Haven, CT), except for 5-10 seconds every 5 minutes when PD was measured by the removal of the voltage clamp. The tissue ion conductance (G) was calculated from PD and Isc according to Ohm's Law.20 PD is expressed as millivolts, Isc is expressed as milliamps per centimeter squared, and G is expressed as millisiemens per centimeter squared. The baseline Isc and G values were measured after a 20-minute equilibration period, and increases in Isc were induced by the addition of the adenylate cyclase–activating agent, forskolin (10−5 M), to the serosal surface. The epithelial responsiveness was defined as the maximal increase in Isc to occur within 5 minutes of exposure to the secretagogue.

Bacterial Translocation.

To measure bacterial translocation across the intestinal epithelium, livers were removed aseptically, homogenized in brain-heart infusion broth, and plated on brain-heart infusion and deMan-Rogosa-Sharpe agar. Agar plates were incubated aerobically for 24 hours and anaerobically for 48 hours at 37°C. The culture results were determined by the number of colony-forming units per gram of tissue calculated from dilutions of liver homogenates.

Histopathology.

Liver and colonic tissues were obtained at the time of death for histological analysis. Liver tissue was fixed in 10% neutral-buffered formalin and embedded in paraffin. Sections 4 μm thick were stained with hematoxylin and eosin and analyzed by a pathologist who was blinded to the groups. To grade for the degree of liver injury and inflammation, the amount of hepatic parenchyma affected by necrosis or inflammation was semiquantitatively graded on a scale of 0 (absent), 1 (mild), 2 (moderate), and 3 (extensive).21

Western Blotting.

For western blot analysis, whole tissue homogenates were lysed in Mono Q buffer (1.08 g of β-glycerophosphate, 38.04 mg of ethylene glycol tetraacetic acid, 0.5 ml of Triton X-100, and 200 μl of 1 M MgCl2 per 100 ml), and 50 μg of protein was subjected to electrophoresis on 10% sodium dodecyl sulfate–polyacrylamide gels as described.19 The antibodies used were anti-PPARγ (Upstate Biotechnology, Charlottesville, VA). Secondary staining was conducted with horseradish peroxidase–conjugated goat anti–rabbit immunoglobulin (1:3000; Amersham, Baie d'Urfe, Canada) and was followed by chemiluminescent detection with a commercial reagent according to the manufacturer's instructions (Lumilight, Roche, Laval, Canada). The equivalent loading was confirmed by Ponceau's staining, and multiple exposures were performed to ensure that the film was not overexposed.

Statistical Analysis.

Bacterial translocation data are represented as medians and ranges; all other data are expressed as means ± the standard error of the mean (SEM). Data were tested for the normality of the distribution, and analyses were performed with the statistical software SigmaStat (Jandel Corp., San Rafael, CA). Comparisons between multiple groups (for enzyme-linked immunosorbent assay, PCR, and Ussing chamber results) were carried out with a 1-way analysis of variance with a post hoc test of significance between individual groups. Differences were considered significant when P was less than 0.05.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Probiotics Prevented an LPS/GalN-Induced Increase in Mucosal Cytokine Secretion.

In the WT control mice, LPS/GalN injection resulted in enhanced secretion and RNA levels of the proinflammatory cytokines TNF-α and IFN-γ in the colon (Fig. 1). In contrast, colonic tissue from mice pretreated with probiotics did not demonstrate either enhanced secretion or increased RNA levels for either cytokine in response to LPS/GalN injection (Fig. 1).

thumbnail image

Figure 1. Effects of LPS/GalN on the cytokine secretion and RNA levels in colonic tissue. In WT control mice, LPS/GalN induced a significant increase in (A) TNF-α and (B) IFN-γ secretion and (C,D) RNA levels. Probiotic administration prevented an LPS/GalN-induced increase in TNF-α and IFN-γ secretion and RNA levels. The values are means ± SEM. Sham-treated control: n = 8; LPS/GalN-treated control: n = 8; sham-treated probiotic mice: n = 8; LPS/GalN-treated probiotic mice: n = 10. *P < 0.05 compared with the sham-treated control, sham-treated probiotic mice, and LPS/GalN-treated probiotic mice.

Download figure to PowerPoint

Probiotics Maintained Colonic Barrier Function and Reduced Bacterial Translocation.

IFN-γ and TNF-α cause a functional disruption of epithelial tight junctions.22, 23 To determine if, in addition to preventing the increased mucosal secretion of proinflammatory cytokines, probiotics were effective in maintaining gut barrier function, the colonic permeability was assessed by the measurement of the mannitol flux in Ussing chambers. Preliminary experiments confirmed that following the injection of LPS/GalN, mice demonstrated a maximal enhanced colonic mannitol flux 6 hours following injection (data not shown); thus, all subsequent measurements were made at this time. As shown in Fig. 2A, LPS/GalN significantly increased the colonic mannitol flux, and this was indicative of a breakdown in the barrier function. The probiotic treatment prevented the LPS/GalN-induced breakdown in the barrier function. Mice pretreated with viable probiotics also demonstrated decreased basal mannitol flux before LPS/GalN injection in comparison with WT control mice (Fig. 2A).

thumbnail image

Figure 2. Effects of LPS/GalN on the colonic electrical parameters and barrier function. In WT control mice, LPS/GalN reduced (B) Isc, (C) the ΔIsc response to forskolin, and (D) PD. This was coupled with (A) an increase in unidirectional mannitol fluxes. Probiotic administration prevented LPS/GalN-induced changes in all parameters. The values are means ± SEM. Sham-treated control: n = 8; LPS/GalN-treated control: n = 8; sham-treated probiotic mice: n = 8; LPS/GalN-treated probiotic mice: n = 10. *P < 0.05 compared with the sham-treated control, sham-treated probiotic mice, and LPS/GalN-treated probiotic mice. +P < 0.05 compared with the sham-treated control.

Download figure to PowerPoint

Sepsis is frequently associated with bacterial translocation into other tissues. One hundred percent of the mice injected with LPS/GalN demonstrated bacterial translocation into liver tissue (Table 2). The probiotic pretreatment reduced bacterial translocation; only 2 of the 6 mice receiving probiotics had culturable bacteria in the liver, and furthermore, the bacterial numbers cultured were significantly reduced. Sham-treated animals had no bacteria cultured from liver tissue.

Table 2. Bacterial Translocation in the Liver
GroupNumber of Affected Mice/Total Number in the GroupRange (Colony-Forming Units per Gram of Liver)
  • NOTE. No bacterial translocation was detected in the sham-treated control or VSL#3-treated mice.

  • *

    P < 0.01 in comparison with the sham-treated or probiotic + LPS/GalN groups.

Sham-treated (n = 6)0/60
Probiotic sham-treated (n = 6)0/60
LPS/GalN (n = 6)6/614–234* (14, 24, 36, 58, 101, and 234)
Probiotic + LPS/GalN (n = 6)2/60–5 (0, 0, 0, 0, 1, and 5)

Probiotics Maintained Epithelial Ionic Transport Function.

In addition to disrupting the barrier function, proinflammatory cytokines also reduce epithelial ion transport.24, 25 In WT control mice, LPS/GalN injection significantly depressed both Isc (Fig. 2B) and the delta short-circuit current (ΔIsc) in response to the adenosine 3′,5′-cyclic monophosphate–dependent agonist, forskolin (Fig. 2C). PD (Fig. 2D) was also reduced by LPS/GalN injection. Mice that were pretreated with probiotics maintained their colonic epithelial transport function in the presence of systemic LPS/GalN, as evidenced by normal values for the basal Isc, forskolin secretory response, and PD.

Role of IL-10.

WT control mice injected with LPS/GalN demonstrated increased levels of colonic RNA for IL-10 in comparison with sham-treated mice (Fig. 3), and this was likely a counter-regulatory response to the enhanced levels of TNF-α induced by LPS/GalN. The oral treatment of mice with probiotics resulted in significantly enhanced basal levels of IL-10 RNA in comparison with that seen in WT control mice and did not demonstrate increased IL-10 RNA in response to LPS/GalN injection, possibly because the TNF-α levels were not increased in these animals. To determine if IL-10 was necessary for probiotic-induced protection against LPS/GalN-induced gut barrier dysfunction, IL-10 gene–deficient mice were pretreated with probiotics, and the colonic permeability and cytokine secretion in response to LPS/GalN were assessed. However, as shown in Fig. 4, IL-10–deficient mice pretreated with oral probiotics were also protected from LPS/GalN-induced barrier dysfunction. In addition, similarly to WT mice, IL-10–deficient mice responded to LPS/GalN with increased colonic TNF-α (Fig. 4C) and IFN-γ (Fig. 4D) levels, and the probiotic treatment prevented this increase. These data suggest that although oral probiotics enhance the levels of IL-10 in the colon in WT mice, IL-10 does not appear to be necessary for probiotic-induced protection in this LPS/GalN-induced injury model.

thumbnail image

Figure 3. Effects of LPS/GalN on the colonic IL-10 RNA levels. In WT control mice, LPS/GalN increased the levels of RNA for IL-10. Mice pretreated with probiotics demonstrated a significant increase in the basal levels of IL-10 RNA in comparison with WT controls. Mice that were pretreated with probiotics did not demonstrate any changes in the IL-10 RNA levels after LPS/GalN injection. The values are means ± SEM. *P < 0.05 compared with the sham-treated control. Sham-treated control: n = 8; LPS/GalN-treated control: n = 8; sham-treated probiotic mice: n = 9; LPS/GalN-treated probiotic mice: n = 10.

Download figure to PowerPoint

thumbnail image

Figure 4. Effects of LPS/GalN on the barrier function and cytokine RNA levels in IL-10–deficient mice. LPS/GalN (A) increased mannitol flux and (B) reduced Isc. This was coupled with increased RNA levels for (C) TNF-α and (D) IFN-γ. The probiotic treatment reduced mannitol fluxes in sham-treated IL-10–deficient mice and prevented LPS/GalN-induced changes in all parameters. The values are means ± SEM. Sham-treated IL-10–deficient mice: n = 8; LPS/GalN-treated IL-10–deficient mice: n = 8; sham-treated IL-10–deficient + probiotic mice: n = 8; LPS/GalN-treated IL-10–deficient + probiotic mice: n = 9. *P < 0.05 compared with the sham-treated IL-10–deficient mice and probiotic-treated mice. +P < 0.05 compared with the sham-treated control.

Download figure to PowerPoint

Effects of Probiotics on the Liver Histology.

Liver sections from mice injected with LPS/GalN demonstrated significant hepatic damage in comparison with sham-treated animals (Table 3). Liver tissue exhibited microabscesses, hemorrhaging, and extensive neutrophil and macrophage accumulation in parenchyma (Fig. 5B). Liver sections from mice that were pretreated with probiotics (Fig. 5C) had normal histology. Mice pretreated with probiotics and injected with LPS/GalN had essentially normal histology, with the occasional microabscesses noted (Fig. 5D).

Table 3. Histological Scores of Liver
GroupHistological Grade (0–3)
  • NOTE. The amount of hepatic parenchyma affected by necrosis or inflammation was semiquantitatively graded on a scale of 0 (absent), 1 (mild), 2 (moderate), and 3 (extensive).21 The values are the means ± SEM.

  • *

    P < 0.02 in comparison with sham-treated, probiotic sham-treated, and probiotic + LPS/GalN groups.

Sham-treated (n = 6)0 ± 0
Probiotic sham-treated (n = 6)0 ± 0
LPS/GalN (n = 6)1.8 ± 0.3*
Probiotic + LPS/GalN (n = 6)0.4 ± 0.2
thumbnail image

Figure 5. Histological analysis of liver sections showing (B) LPS/GalN-treated mice having severe hepatic injury with prominent hemorrhage, inflammation, and microabscess formation. (D) Probiotics protected the mice from LPS/GalN-induced severe hepatic injury (original magnification ×400). (A) The sham-treated control, (B) probiotic-treated mice, and (D) probiotic + LPS/GalN mice showed essentially normal histology (inset, original magnification ×500).

Download figure to PowerPoint

Effects of Probiotics on Hepatic TNF-α, IL-6, and IL-12p35.

Systemic LPS is captured and cleared by macrophages and hepatic Kupffer's cells.26 This activation of macrophages by LPS results in an up-regulation and secretion of TNF-α, IL-6, and IL-12p35 and subsequent liver tissue damage.18, 26 To determine if the protection associated with probiotic administration was accompanied by alterations in hepatic cytokines, TNF-α, IL-6, and IL-12p35 RNA were measured by semiquantitative reverse transcription polymerase chain reaction. Enhanced levels of TNF-α and IL-6, but not IL-12p35 RNA, were observed after LPS/GalN injection (Fig. 6). The probiotic treatment significantly reduced IL-6 expression in response to LPS/GalN injection. However, the probiotic treatment did not significantly alter either basal levels of TNF-α or the LPS/GalN-induced up-regulation. This would suggest that the protection against liver damage induced by oral probiotics could not be attributed to a probiotic-induced decrease in hepatic TNF-α levels but may be due to reduced IL-6 expression.

thumbnail image

Figure 6. LPS/GalN increased RNA levels of (A) hepatic TNF-α and (B) IL-6 but not (C) IL-12p35. The probiotic treatment did not prevent an LPS/GalN-induced increase in the hepatic TNF-α RNA levels but did prevent an increase in IL-6 RNA. The values are means ± SEM. Sham-treated control: n = 6; LPS/GalN-treated control: n = 6; sham-treated probiotic mice: n = 6; LPS/GalN-treated probiotic mice: n = 6. *P < 0.05 compared with the sham-treated control and sham-treated probiotic mice.

Download figure to PowerPoint

Role of PPARγ in Liver Injury and Intestinal Dysfunction.

The activation of the PPARγ receptor has been shown to attenuate the onset and severity of septic shock by modulating NF-κB activity.7 LPS/GalN injection resulted in a reduction in hepatic, but not colonic, protein PPARγ expression in control mice (Fig. 7). The probiotic pretreatment prevented this reduction in hepatic PPARγ expression. To determine if this modulation of PPARγ expression was reflected in PPARγ target genes, RNA levels of lipocalin-2 were measured.27 A decreased level of lipocalin-2 RNA was seen in the liver and correlated with the reduction in the PPARγ protein (Fig. 7). To determine if the effects of probiotics on the epithelial barrier and cytokine secretion were dependent on PPARγ activation, a group of mice were treated with a selective PPARγ antagonist, GW9662, 30 minutes before LPS/GalN injection.28 As shown in Fig. 8, the inhibition of PPARγ abrogated the ability of probiotics to protect against an LPS/GalN-induced breakdown in colonic barrier function and cytokine secretion.

thumbnail image

Figure 7. Analysis of colonic and hepatic PPARγ protein expression by a western blot and measurements of the RNA levels of the PPARγ target gene lipocalin-2. In the WT control mice, increased protein expression of PPARγ was seen in colons from LPS-injected mice pretreated with probiotics in comparison with sham-treated or LPS/GalN-treated mice. LPS/GalN also induced RNA levels for lipocalin in the colon. In the liver, LPS/GalN reduced the protein expression of PPARγ in WT mice but not in mice pretreated with probiotics. This correlated with a reduction in the RNA levels for lipocalin in the liver. Sham-treated control: n = 4; LPS/GalN-treated control: n = 4; sham-treated probiotic mice: n = 4; LPS/GalN-treated probiotic mice: n = 4.

Download figure to PowerPoint

thumbnail image

Figure 8. Effects of the pharmacological inhibition of PPARγ on colonic barrier function and cytokine production. Mice were injected with GW9662 30 minutes before the LPS/GalN injection. The inhibition of PPARγ abrogated the protective effect of the probiotic (n = 6) treatment on (A) the mannitol flux, (B) the ΔIsc response to forskolin, and (C) TNF-α and (D) IFN-γ RNA levels in mice treated with LPS/GalN. The values are means ± SEM. LPS/GalN-treated control: n = 8; LPS/GalN-treated probiotic mice: n = 8; LPS/GalN-treated probiotic + GW9662 mice: n = 8. *P < 0.05 compared with the LPS/GalN-treated probiotic mice.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

In this study, we have demonstrated that probiotics effectively attenuate liver damage and maintain gut barrier and epithelial function in a microbial-induced sepsis mouse model. The oral administration of probiotic bacteria reduced bacterial translocation, colonic gene expression, and tissue secretion of the proinflammatory cytokines TNF-α and IFN-γ. In addition, orally administered probiotics maintained the gene expression of hepatic PPARγ during sepsis, and the inhibition of PPARγ abrogated the ability of probiotic bacteria to maintain gut function.

The presence of LPS in the systemic circulation results in the activation of the innate immune system and a massive secretion of proinflammatory cytokines, particularly TNF-α. This systemic inflammatory response induces massive liver damage and results in high mortality.18 Our data are consistent with these findings; mice injected with LPS demonstrated extensive liver damage and enhanced gene expression and secretion of TNF-α and IFN-γ in the colon. The oral probiotic treatment prevented both the up-regulation of gene expression and the enhanced secretion of TNF-α and IFN-γ in the colon in response to LPS. The reduction in the colonic levels of proinflammatory cytokines in the presence of systemic LPS was associated with a probiotic-induced maintenance of epithelial barrier and transport function, including a reduction in bacterial translocation. This finding is in agreement with previous studies showing that the administration of probiotic bacteria diminishes bacterial translocation in models of liver disease16 and reduces bacterial infection after liver transplant.29 In a study of intensive-care patients, a breakdown in intestinal permeability preceded the development of septic shock, and a multivariate analysis revealed that the only measured parameter present on admission to the intensive-care unit that was predictive of subsequent sepsis was a breakdown in gut barrier function.30 This breakdown in the barrier has been attributed to the effects of circulating high levels of TNF-α and IFN-γ on epithelial tight junctions.22

Previous studies have shown that an LPS-induced up-regulation of TNF-α release is closely followed by enhanced secretion of IL-10 as a counter-regulatory cytokine.31 IL-10 is produced by epithelial and immune cells and has anti-inflammatory and immunosuppressive characteristics. Studies have demonstrated that IL-10 can protect mice from lethal endotoxemia when administered concomitantly or following LPS challenge.32, 33 IL-10 is also a potent inhibitor of LPS-induced TNF-α and IFN-γ secretion. In our study, an increased level of RNA for IL-10 was seen in colons from mice treated with probiotics before LPS challenge. Other studies have also reported that probiotics induce IL-10 production.34, 35 Interestingly, mice pretreated with probiotics did not demonstrate an increased IL-10 response to LPS challenge. This may be due to a primary effect of probiotics in preventing an LPS-induced rise in proinflammatory cytokine secretion, or alternatively, a pretreatment with probiotics induces a maximal up-regulation of IL-10. However, the protective effects of probiotics on gut function did not appear to be dependent on the presence of IL-10, as the administration of LPS and probiotics to IL-10 gene–deficient mice yielded a reduction of cytokine expression and maintenance of gut barrier function similar to what was seen in WT mice.

The ability of probiotics to protect the gut and liver from systemic endotoxemia may involve a probiotic-induced inhibition of NF-κB activity through effects on PPARγ.15 Numerous studies have shown that probiotics modulate NF-κB activity,14, 15 and recent studies have shown that PPARγ inhibits the activation of NF-κB.5 In this study, we have provided novel in vivo evidence that the up-regulation of the PPARγ protein in the colon occurs in response to LPS challenge, and the oral feeding of mice with probiotics prevents this enhanced expression. This suggests that the enhanced expression of PPARγ in the colon may occur as a compensatory mechanism to down-regulate the LPS-induced increase in mucosal TNF levels, which does not occur in probiotic-treated mice. PPARγ appears to be involved in mediating the probiotic-induced protection of liver and gut function, in that the PPARγ inhibitor GW9662 abolished the protective effects of probiotics after LPS challenge. Both endogenous and pharmaceutical ligands of PPARγ have been shown to be efficacious in reducing the multiple organ injury and dysfunction of LPS-injected rats.8–10 Furthermore, the inhibition of PPARγ with the antagonist GW9662 has been shown to exacerbate the extent of liver injury in a rat model of hemorrhagic shock.10

As has been reported in other sepsis models,36 the injection of LPS resulted in enhanced bacterial translocation to the liver, an up-regulation of hepatic TNF-α and IL-6 levels, and significant hepatic histological damage. Probiotic administration reduced bacterial translocation and IL-6 levels and attenuated liver damage but did not prevent an increase in hepatic TNF-α levels. However, probiotics did prevent the LPS-induced down-regulation of PPARγ protein expression in the liver. The transcriptional activity of PPARγ is regulated by ligand binding and posttranslational modifications, including phosphorylation, and the degradation of PPARγ occurs through the ubiquitin-proteasome system.37 In that we have previously shown probiotics to reduce cellular proteasomal activity,14, 38 it is possible that the maintenance of the PPARγ protein in the presence of systemic endotoxin was due to a probiotic-induced reduction in PPARγ degradation.

In conclusion, orally administered probiotics are effective in protecting the gut against an LPS-induced breakdown in barrier and epithelial function. Furthermore, this protection involves a PPARγ-dependent mechanism and extends to an attenuation of hepatic injury. These data support the hypothesis that a bacterial source of antigenic stimulation from the gut is required for a proinflammatory response in liver disease and septic shock and that altering gut microflora with probiotics represents a potential adjunctive therapeutic option for the prevention of hepatic damage in patients at risk of developing sepsis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

For their assistance with the project, we thank Francis Cheung, Sarah Martin, and Robert Penner.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References