Lactobacillus casei downregulates commensalsʼ inflammatory signals in Crohnʼs disease mucosa

ORIGINAL ARTICLE

Lactobacillus casei Downregulates Commensals’ Inflammatory Signals in Crohn’s Disease Mucosa Marta Llopis, PhD,* Maria Antolin, PhD,* Monica Carol, PhD,* Natalia Borruel, MD,* Francesc Casellas, MD,* Cristina Martinez, MSc,* Eloy Espı´n-Basany, MD,† Francisco Guarner, MD,* and Juan-R. Malagelada, MD*

Background: The interaction of commensal bacteria with the intestinal immune system is an essential factor in the development of inflammatory bowel disease (IBD). The study of isolated commensal bacteria’s effects on the mucosal immune response might be relevant for a better understanding of pathophysiological mechanisms in IBD.

Methods: We investigated the immune responses to signals from the commensal Escherichia coli ATCC 35345 and the probiotic Lactobacillus casei DN-114 001 in Crohn’s disease (CD) mucosa. Ileal specimens were obtained during surgery from CD patients. Mucosal explants were incubated with L. casei or its genomic DNA; TNF-␣, IFN-␥, IL-2, IL-6, IL-8, and CXCL1 were measured in the supernatant. Second, tissue expression of key proinflammatory cytokines (IL-6, TGF-␤, IL-23p19, IL-12p35, IL-17F), and chemokines (IL-8, CXCL1, CXCL2) was evaluated after incubation with L. casei or E. coli. Finally, combination experiments were carried out by incubating both strains with mucosal explants at different timepoints.

Results: Live L. casei significantly decreased secretion of TNF-␣, IFN-␥, IL-2, IL-6, IL-8, and CXCL1 by CD mucosa, but the effect was not reproduced by L. casei DNA. Second, live L. casei downregulated expression of IL-8, IL-6, and CXCL1 and did not modify expression of IL-23p19, IL-12p35, and IL-17F. In contrast, E. coli significantly upregulated expression of all these cytokines. Interestingly, combination experiments revealed the ability of L. casei to prevent and counteract the proinflammatory effects of E. coli. Conclusions: Live L. casei can counteract the proinflammatory effects of E. coli on CD inflamed mucosa by specific downregulation of key proinflammatory mediators. (Inflamm Bowel Dis 2009;15:275–283)

Received for publication July 16, 2007; Accepted July 21, 2008. From the *Digestive System Research Unit, Centro de Investigacio´n Biome´dica en Red de Enfermedades Hepa´ticas y Digestivas (Ciberehd), † Department of Surgery, Hospital Universitari Vall d’Hebron, Autonomous University of Barcelona, Spain. Supported in part by grants from Generalitat de Catalunya (RE: 2001SGR00389) and Ministerio de Educacio´n y Ciencia (SAF 2007-64411).Ciberehd is funded by the Instituto de Salud Carlos III (Madrid, Spain). Reprints: Maria Antolin, Digestive System Research Unit, Fundacio´ Institut de Recerca Hospital Universitari Vall d’Hebron, Psg.Vall d’Hebron 119-129, Barcelona 08035, Spain (e-mail: [email protected]). Copyright © 2008 Crohn’s & Colitis Foundation of America, Inc. DOI 10.1002/ibd.20736 Published online 6 October 2008 in Wiley InterScience (www.interscience. wiley.com).

Inflamm Bowel Dis

●

Volume 15, Number 2, February 2009

Key Words: IBD, commensals, inflammation, probiotics, cytokines, chemokines

M

ucosal immune responses to bacteria require accurate recognition to distinguish between commensals and pathogens. In order to maintain intestinal homeostasis in the gut, the immune system must tightly regulate cellular responsiveness and maintain a balance between active immunity and tolerance. Some disease states have been associated with an imbalance of the intestinal microbiota and dysregulated recognition of commensal bacteria by the immune system, as is the case with inflammatory bowel disease (IBD). Although it is generally accepted that intestinal bacteria provide the key stimuli triggering mucosal inflammation both in animal models and in patients with IBD,1 some specific bacteria have been shown to provoke protective antiinflammatory responses. In fact, administration of specific probiotic bacteria attenuates development of colitis in animal models2,3 and has demonstrated efficacy in the remission of chronic pouchitis in humans.4 Lactobacillus casei is a commensal bacterium that can be found in the gut of healthy people.5 Some strains of L. casei have been isolated and are being used in clinical and experimental studies for their immunomodulatory properties. Milk fermented with L. casei strain DN-114 001 was able to modulate the number of lymphocytes and CD56 cells in subjects under academic examination stress.6 In addition, ingestion of L. casei DN-114 001 for 8 weeks increased oxidative burst capacity of monocytes and NK cells tumoricidal activity in healthy middle-age people, reinforcing their innate immune defenses.7 Furthermore, the same strain reduced antigen-specific skin inflammation by controlling the size of the CD8⫹ effector pool in an experimental model of allergic contact dermatitis.8 In experimental models of IBD, the L. casei strain DN-114 001 has been shown to reduce mucosal injury, inflammatory response, and bacterial translocation.2,9 In vitro, this strain displayed beneficial properties on inflamed mucosal explants from Crohn’s disease (CD) patients by inhibiting overproduction of TNF-␣ and reducing the number of activated T helper lymphocytes.10 –12 These effects were not observed in noninflamed CD mucosa nor in healthy ileal mucosa from control individuals.10 Moreover, L.

275

Inflamm Bowel Dis

Llopis et al

casei has been demonstrated to downregulate the transcription of proinflammatory CXCL1 and CXCL2 chemokine genes induced after invasion of cultured intestinal epithelial cells by Shigella flexneri.13 However, the ability of L. casei to modulate mucosal responses to stimuli by other commensal bacteria has not been investigated. This setting of bacterial interaction and competition would be closer to the in vivo situation in CD, where intervention trials with probiotics have shown limited clinical benefit so far. The mechanisms underlying the effects of probiotics are still being outlined. It is accepted that highly conserved motifs of luminal flora are recognized by a collection of membrane and intracellular receptors named Toll-like receptors (TLRs). Activation of those receptors by recognition of bacterial components activates intracellular pathways that induce specific patterns of gene expression, shaping innate and adaptive immunity. In recent years, CpG motifs composed of an unmethylated CpG dinucleotide flanked by two 5⬘ purines and two 3⬘ pyrimidines have become increasingly interesting due to their immunomodulatory properties via TLR recognition. CpG motifs have been claimed as the immunostimulatory component of bacterial DNA,14 and its effects on recognition by the immune system are controversial. Some reports support the idea that the antiinflammatory properties of probiotics in experimental animal models of colitis are mediated by their DNA.15,16 Besides, in vitro studies on PBMC and epithelial cells have also reported antiinflammatory properties of DNA from probiotics.17,18 Nevertheless, no previous attempt has been made to compare the effect of a viable probiotic against its own genomic DNA on human CD mucosa. The aims of the current study were 1) to investigate the role of genomic DNA as mediator of the antiinflammatory effect of L. casei DN-114 001 in CD mucosa; 2) to further characterize the in vitro effects of the L. casei strain on mucosal immune responses by focusing on the expression of key chemokines and cytokines, with special attention to the recently described Th17 pathway; and 3) to test whether these effects would prevail in the presence of a nonpathogenic commensal E. coli strain.

MATERIALS AND METHODS Patients Samples of intestinal mucosa were obtained during surgery from 16 patients with CD (9 male and 7 female, age 31 ⫾ 4, mean SEM) who underwent ileal resection for stricture unresponsive to conventional medical treatment. At the time of surgery all patients were receiving oral corticosteroids (prednisolone 0.5–1.0 mg/kg/day) and/or azathioprine (2–2.5 mg/kg/day). None of the patients were or had been on anti-TNF therapy for at least 2 months prior to surgery. The diagnosis of CD had been previously established by routine clinical, radiological, and endoscopic criteria and

276

●

Volume 15, Number 2, February 2009

was confirmed by histological evaluation of the surgical specimen. Ileal stricture was diagnosed on the basis of clinical symptoms and imaging studies. Macroscopic examination confirmed ileal stricture in all patients and histological examination demonstrated transmural inflammation, intense fibrosis, lymphoid aggregates in the submucosa, and granulomas. All patients received similar preparation for colonic surgery including gut lavage with electrolyte-polyethylene glycol solution and broad spectrum antibiotic therapy.

Bacteria Preparations Lactobacillus casei DN-114 001 was provided by Danone Vitapole (Palaiseau, France). Escherichia coli (ATCC 35345, Ecor-26), a nonpathogenic strain from the Ochman–Selander collection of standard reference E. coli strains isolated from healthy populations,19 was kindly provided by Prof. Juan Aguilar (Biochemistry, Faculty of Pharmacy, University of Barcelona). All reagents were from Difco (Detroit, MI) unless indicated. Bacteria were grown 18 hours at 37°C in a 5% CO2 atmosphere incubator: L. casei in Man-Rogosa-Sharpe liquid medium (MRS) and E. coli in 1% tryptone peptone medium, 0.5% NaCl, and 0.5% yeast extract. At the end of the exponential phase bacteria were harvested and cell counts were estimated by optical density at 600 nm absorbance. Afterward, bacteria were added to the tissue culture wells at the appropriate dilution to reach a final concentration of 106 CFU/mL.

Genomic DNA Reagents used for DNA extraction were from Sigma (St. Louis, MO) unless indicated. Genomic DNA of L. casei was extracted from 150 mL of 48-hour cultures. Harvested cells were pelleted and incubated with 5.4 mL enzymatic lysis buffer (20 mM Tris, 2 mM EDTA, 1.2% Triton X-100) and 3.6 mL lysozyme (50 mg/mL) for 30 minutes at 37°C. Homogenates were then further lysated overnight at 70°C with 6 mL digestion buffer AL and proteinase K (Qiagen, Chatsworth, CA). Proteins were removed by twice adding an equal volume of phenol:chlorophorm:isoamyl-alcohol (25: 24:1), and a chloroform v:v (Merck, Darmstadt, Germany) washing step was carried out. Nucleic acids were precipitated from the aqueous phase with cold ethanol (Merck) plus sodium acetate 3 M, pH 5.2. After centrifugation the supernatant was discarded and the pellet was washed twice with cold ethanol 70% to remove salts. Finally, pellets were allowed to dry and suspended in buffer TE, pH 8 (10 mM Tris-ClH, 1 mM EDTA). The purity of DNA was confirmed by measuring the UV 260/280 absorbance ratio (⬎1.8). Extracted DNA was tested for endotoxin content by limulus amebocyte lysate (BioWhitaker, Walkersville, MD) and only preparations under 0.05 U/mL were used. We checked the equivalence of the live inoculum in terms of DNA concentration.

Inflamm Bowel Dis

●

Volume 15, Number 2, February 2009

Organ Culture Organ culture assays were performed as we previously described.10,11 Briefly, full-thickness ileal wall specimens, including areas with macroscopic lesions, were collected at surgery. After intensive rinsing and washing with sterile saline the specimens were immediately transferred to our laboratory in sterile saline at 4°C. Multiple mucosal samples weighing 25–35 mg were separated from underlying tissue and placed on culture filter plates (15 mm diameter wells with 500 ␮m bottom mesh; Netwell Culture Systems, Costar, Cambridge, MA) and orientated so that the epithelial surface was uppermost. Filters were suspended over wells containing 1500 ␮L medium, consisting of RPMI 1640 (CanSera, Rexdale, Ontario, Canada) supplemented with 10% FCS (Gibco BRL, Eggenstein, Germany), 16 ␮g/mL vancomycin (Lilly, Indianapolis, IN), 2500 U/mL colistin (Pharmax, Kent, UK), and 50 ␮g/mL gentamycin (Normon, Madrid, Spain). Samples were preincubated with antibiotics for 3 hours at 37°C in a humidified 5% CO2 atmosphere to eradicate most of the indigenous flora. Thereafter, the medium was replaced by fresh 10% FCS-RPMI with or without bacterial stimulation (Blank) and incubated 4, 18, or 24 hours at 37°C in a humidified 5% CO2 chamber (see below). Finally, supernatants and tissues were collected and stored at – 80°C. Analysis included media pH measurement and lactate dehydrogenase activity (LDH) in supernatants and tissues as a viability measurement of the tissue at the end of the experiments.

Bacterial Stimulation Lactobacillus casei genomic DNA, 106 CFU/mL culture of live L. casei or E. coli, or both strains were added in the different experimental sets. In the combination assays, 4 patients were included and conditions were run in triplicate. In competition assays E. coli and L. casei strains were cocultured together for 24 hours with the inflamed mucosa, and in probiotic pretreatment assays L. casei was cocultured for 24 hours and E. coli added for the last 8 hours. Conversely, in posttreatment probiotic assays E. coli was cocultured for 24 hours and L. casei was added for the last 8 hours. Release of TNF-␣ (enzyme-linked immunosorbent assay [ELISA], eBioscience, San Diego, CA), IL-8, and CXCL-1 (DuoSet, R&D Systems, Minneapolis, MN) was measured in supernatants at the end of the experiments.

L. casei DNA Activity Assay Lactobacillus casei DNA was extracted from the whole bacterial suspension as described above. Immune response to L. casei DNA was compared to that of the live bacteria in inflamed CD mucosa using the organ culture method. The following conditions were run in parallel: Blank with no bacteria nor DNA, live L. casei (106 CFU/mL), and L. casei DNA (50 ␮g/mL). Commercial CpG rich sequences (KCpG2006, and Control KCpG2006, MWG-Biotech, Ger-

L. casei Downregulates E. coli Signals

many) were also tested at the same concentration as DNA, as a reference control situation. After 18-hour incubation, supernatants and tissues were collected and stored at ⫺80°C. Analysis included quantification of Th1 released proinflammatory cytokines; TNF-␣ (ELISA), IFN-␥, IL-2, and IL-6 by cytometric bead assay (Simplex CBA, BD) and also chemokines IL-8 and CXCL1 following manufacturer’s instructions. Six patients were included in this assay and conditions were run in triplicate. The relevance of these mediators in the inflammatory CD process and the previously demonstrated potential of L. casei in their downregulation in other experimental approaches in vitro10,12,13 led to their inclusion in the study.

Gene Expression Modulation by Commensal Bacteria In gene expression organ culture assays, shorter incubation times were required to ensure optimal tissue RNA integrity. CD mucosa was incubated with no bacteria (Blank), L. casei, or E. coli (106 CFU/mL) for 4 hours. After incubation, tissues were harvested in RNAlater (Ambion, Austin, TX) and stored at ⫺80°C for later RNA isolation. Total RNA was isolated from mucosal samples using RNeasy mini kit (Qiagen) following the manufacturer’s instructions. RNA quantity and integrity were evaluated using RNA 6000 Nano Chip in a Bioanalyzer 2100 (Agilent Technologies, Palo Alto, CA) and was judged to be acceptable if the 28S/18S ribosomal fragments ratio was over 1.5 and the RNA integrity number (RIN) was over 6. RNA samples were stored at – 80°C until measurement. Cytokine mRNA expression was quantified by real-time polymerase chain reaction (PCR). First, 1 ␮g of total RNA was used for the reverse transcription reaction to synthesize the first strand of cDNA following the high capacity cDNA archive kit instructions (Applied Biosystems, Foster City, CA). Relative quantification (RQ) of cytokines and chemokines was performed using commercially available Taqman gene expression assays on an ABI PRISM 7500 real-time PCR system (Applied Biosystems). IL-6, TGF-␤, IL-23p19, IL-12p35, and IL-17F were quantified to explore the recently discovered Th17 inflammation pathway for its implication in autoimmune disorders such as CD. IL-8 and CXCL1 were determined for their important role as chemoattractant chemokines in the first steps of the innate immunity process. Data were obtained as threshold cycle (Ct) values. Gene expression levels for each individual sample were normalized relative to the PPIA gene, Peptidylprolyl Isomerase A (Cyclophilin A). Six patients were included in this study and the conditions were run in duplicate.

LDH Activity Assessment Viability was assessed in all tissues through the release of LDH into the supernatant according to Finnie et al.20 The ratio of LDH activity in the culture supernatant over total

277

Inflamm Bowel Dis

Llopis et al

●

Volume 15, Number 2, February 2009

TABLE 1. Effect of L. casei DNA on TNF-␣ release from CD inflamed mucosa

TNF-a

Blank

DNA L. casei 5 ␮g/mL

DNA L. casei 50 ␮g/mL

CpG 2006 5 ␮g/mL

Control CpG 2006, 5 ␮g/mL

4.95 ⫾ 1

2.91 ⫾ 1.23

2.79 ⫾ 1.95

4.61 ⫾ 0.9

4.82 ⫾ 0.63

Values are mean ⫾ SEM pg/mg tissue of TNF-␣ released from CD mucosa after 18 hours incubation with L. casei DNA or commercial CpG-rich sequences (n ⫽ 3). Concentrations of 5 and 50 ␮g/mL of L. casei DNA similarly modulated TNF-␣ secretion versus Blank (no added bacteria). Levels of TNF-␣ after incubation with CpG2006 or control CpG2006 were similar to the Blank condition.

LDH activity in tissue homogenates was calculated and used to estimate the percentage of viable tissue. Tissue samples were homogenized in Tris/HCl (100 mmol/L pH 7.4) and LDH activity was analyzed with the spectrophotometric method (pyruvate-lactate) recommended by the Scandinavian Society of Enzymes.

Ethical Considerations Prior informed consent was obtained from each patient and the study was approved by the Ethics Committee of our institution (Comite` d’E`tica i Investigacio´ Clı´nica, Hospital Universitari Vall d’Hebron, Barcelona, Spain).

Statistical Analysis Results are expressed as the mean and SEM. Statistical differences were determined using repeated-measures analysis of variance (ANOVA; GraphPad Instat, San Diego, CA) and the Bonferroni method as the posttest.

RESULTS Genomic L. casei DNA Does Not Reproduce the Antiinflammatory Effect of the Live Strain DNA activity assays aimed to determine the contribution of probiotic genomic DNA molecules to the potent anti-TNF-␣ effect induced by L. casei DN-114 001 on CD inflamed mucosa.10,11 As previously observed, LDH release as an index of tissue viability was not modified after coincubation with bacterial products and maintenance of the pH range at 7.4 ⫾ 0.1 was confirmed in all organ culture experimental conditions. Lactobacillus casei genomic DNA immunomodulatory activity was compared to that of viable bacteria on inflamed CD mucosal explants. For this purpose we confirmed that the amount of genomic DNA in live inocula at 106 CFU/mL was equal to 5 ␮g/mL. However, as previous reports testing the effects of bacterial genomic DNA used 50 ␮g/mL DNA, we tested both 5 and 50 ␮g/mL concentrations in preliminary studies. The results show that both concentrations displayed similar effects on TNF-␣ release (Table 1). Incubation of commercial CpG-rich sequences and control CpG molecules with CD mucosal explants as a reference situation did not have any effect on TNF-␣.

278

As shown in Figure 1, coincubation of inflamed mucosa with viable L. casei significantly reduced secretion of the Th1 cytokines TNF-␣, IFN-␥, IL-2, and IL-6 and chemokines IL-8 and CXCL1 in the supernatant (P ⬍ 0.05 compared to Blank, where no bacteria were added), while genomic DNA did not induce significant changes. Although DNA of L. casei induced some decrease in TNF-␣ secretion, this effect did not reach statistical significance. Levels of IL-2 secretion by CD explants were low but over the detection limit of the procedure (1 pg/mL).

Modulation of Cytokine and Chemokine Gene Expression Since the recently discovered Th17 inflammatory pathway seems to play a crucial role in autoimmune disorders, including IBD, we aimed to characterize the expression of cytokines related to that pathway in the ileal mucosa of CD patients. The results of gene expression on CD inflamed mucosa versus healthy ileal mucosa demonstrated an upregulation of cytokines IL-6, TGF-␤, IL23p19, IL-12p35, and IL-17F involved in the Th17 pathway and the chemokines IL-8, CXCL1, and CXCL2 involved in leukocyte recruitment (Table 2). Second, we aimed to investigate the influence of L. casei on the expression of cytokines related to this pathway and to compare this effect with that of a nonprobiotic commensal E. coli. Whereas exposure of inflamed mucosa to E. coli tended to increase the expression levels of IL-6 (Fig. 2), exposure to L. casei resulted in a significant inhibition of this gene expression compared to the Blank condition, where no bacteria had been added (P ⬍ 0.05). On the other hand, mucosal expression of the antiinflammatory cytokine TGF-␤ was not modified by the 2 bacterial strains tested. Interestingly, exposure of CD mucosa to the nonpathogenic E. coli strain induced a significant increase in IL-23p19, IL-12p35, and IL-17F expression compared to Blank (P ⬍ 0.05), while L. casei did not induce changes (Fig. 2). Finally, we determined gene expression of chemokines IL-8, CXCL1, and CXCL2 as key factors involved in leukocyte recruitment. Exposure of inflamed mucosa to E. coli increased the expression levels of the tested chemokines, being statistically significant for CXCL2 compared to Blank

Inflamm Bowel Dis

●

L. casei Downregulates E. coli Signals

Volume 15, Number 2, February 2009

FIGURE 1. Lactobacillus casei DNA does not reproduce the antiinflammatory effect of the live strain. Cytokine and chemokine secretion was measured in supernatants after organ culture experiments where CD mucosa had been incubated 18 hours with L. casei (106CFU/mL) or its genomic DNA (50 ␮g/mL). The results are expressed as mean ⫾ SEM (n ⫽ 6 patients, each condition was run in triplicate). Incubation with L. casei significantly decreased the release of TNF-␣, IFN-␥, IL-2, IL-6, IL-8, and CXCL1 compared to the Blank condition (*P ⬍ 0.05). However, L. casei genomic DNA did not reproduce the effect of live bacteria on cytokine secretion, although it induced some decrease on TNF-␣ secretion that did not reach statistical significance.

TABLE 2. Fold Changes of Cytokine and Chemokine Expression in CD Inflamed Ileal Versus Healthy Mucosa

Fold change

IL-6

TGF-␤

IL-23p19

IL-12p35

IL-17F

IL-8

CXCL1

CXCL2

31 ⫾ 5.9

1.3 ⫾ 0.2

5.1 ⫾ 1.5

1.6 ⫾ 0.1

10.6 ⫾ 6.8

20.1 ⫾ 6.9

4.7 ⫾ 1.2

5 ⫾ 0.8

Expression of genes related to Th17 inflammatory pathway and chemoattractant chemokines were assessed by relative quantification RT-PCR. Data obtained from n ⫽ 3 healthy and n ⫽ 6 CD inflamed ileal mucosa. High level of expression was observed for the cytokines IL-6, IL-23p19, IL-17, and the chemokines IL-8, CXCL1, and CXCL2.

279

Llopis et al

Inflamm Bowel Dis

●

Volume 15, Number 2, February 2009

FIGURE 2. Differential cytokine and chemokine expression by E. coli and L. casei. Graphs show mucosal gene expression of cytokines and chemokines after coculture of inflamed CD mucosa with L. casei or a commensal E. coli strain for a period of 4 hours. The results of relative quantification are expressed as mean ⫾ SEM (n ⫽ 6 patients). L. casei significantly downregulated IL-6 (*P ⬍ 0.05), as compared to the Blank. Expression of TGF-␤ was not modified by any of the strains, while expression of IL-23p19, IL-12p35, and IL17F was significantly upregulated by E. coli as compared to the Blank condition (*P ⬍ 0.05). Expression of chemokines IL-8, CXCL1, and CXCL2 was reduced by L. casei, being statistically significant for IL-8 and CXCL1. CXCL2 expression was upregulated by E. coli as compared to the Blank (*P ⬍ 0.05).

(P ⬍ 0.05). On the contrary, L. casei decreased chemokine gene expression, reaching statistical differences for IL-8 and CXCL1 (P ⬍ 0.05 compared to Blank).

Live L. casei Counteracts Proinflammatory Effects of E. coli To further investigate the observed differences in the mucosal immune response to the 2 tested commensal strains in terms of cytokine and chemokine expression profile, we tested combinations of both strains to determine the predominant response of CD mucosa. In the competition assays, when both strains were incu-

280

bated together for 24 hours, the levels of TNF-␣ were similar to those in L. casei control and significantly lower than in E. coli control (Fig. 3). Similarly, in the probiotic pretreatment assays, when mucosa specimens were preexposed to L. casei for 16 hours prior to coincubation with E. coli, the levels of TNF-␣ in supernatant were significantly lower than in E. coli control (P ⬍ 0.05) and similar to the L. casei control. Finally, in the probiotic posttreatment assays, when the tissue had been incubated with E. coli for 16 hours before addition of L. casei, a reduction in TNF-␣ levels similar to those in the L. casei control was consistently detected, although the changes did not achieve statistical significance.

Inflamm Bowel Dis

●

Volume 15, Number 2, February 2009

FIGURE 3. Lactobacillus casei prevents and reverses E. coli proinflammatory effect. TNF-␣, IL-8, and CXCL 1 release was measured in the incubation medium after coculture of inflamed CD mucosa with combinations of L. casei and E. coli. Each strain was also cocultured in parallel as a reference control. Results are expressed as mean ⫾ SEM, n ⫽ 4 patients. In competition assay, when both strains were incubated together for 24 hours, TNF-␣ levels were similar to those in L. casei and significantly lower than in E. coli control (P ⬍ 0.05). In probiotic pretreatment, when inflamed mucosa was preexposed for 16 hours to L. casei prior to the addition of E. coli, TNF-␣ secretion was significantly lower than in E. coli (P ⬍ 0.05) and similar to L. casei control. In probiotic posttreatment, changes did not achieve statistical significance. Chemokine release modulation of IL-8 and CXCL1 in competition assays followed a similar profile. Mediator release in the Blank condition (not exposed to bacteria) was: TNF-␣: 14.3 ⫾ 3.4 pg/mg of tissue, IL-8: 6.3 ⫾ 1.0 ng/mg of tissue, and CXCL1: 475 ⫾ 85.5 pg/mg of tissue.

IL-8 and CXCL1 chemokine release modulation in the competition assays followed a similar profile. When both strains were incubated together for 24 hours, levels of chemokines were similar to those in the L. casei control and significantly lower than in the E. coli control.

DISCUSSION The discovery of new mechanisms that permit a finetuning of the cytokine signaling regulatory circuit21 has shed light on the potential role of commensal bacteria in disease states. It is accepted that in IBD there is a situation of dysregulated intestinal homeostasis where commensal bacteria seem to play a deleterious role, inducing pathological inflammatory responses and contributing to the chronicity of

L. casei Downregulates E. coli Signals

the disease. This report analyzed the immune responses by human intestinal mucosa to signals from isolated and combined commensal bacterial strains. We confirm that commensal bacteria have an impact on immune responses by inflamed CD mucosa that appears to be bacteria-specific, and that certain proinflammatory signals from a commensal bacterium such as E. coli are counteracted by the presence of L. casei. There are conflicting data concerning response of the immune system to DNA molecules.15–17,22,23 Genomic CpG motifs have been shown to possess immunostimulatory properties,14 exacerbating colitis in experimental models.23 Conversely, several reports have demonstrated the immunomodulatory effects of genomic DNA extracted from probiotic bacteria, thus ameliorating experimental colitis.16,17 Indeed, it has been suggested that viable bacteria might not be necessary for specific beneficial effects, DNA being responsible for its immunomodulatory properties. We aimed to determine whether genomic DNA from L. casei would mediate the antiinflammatory effect of viable bacterial cells, previously described in our organ culture model, using explants of inflamed intestinal mucosa.10,11 Our experiments confirm that L. casei inhibited mucosal secretion of TNF-␣ and describe as well, for the first time in CD inflamed mucosa, that the secretion of the proinflammatory cytokines IFN-␥, IL-6, and IL-2 and the chemokines IL-8 and CXCL1 is inhibited by the presence of viable L. casei DN-114 001. However, none of these effects was completely reproduced by incubation of tissue with bacterial DNA, although some partial effect was observed in TNF-␣. Commercial CpG-rich sequences failed at modulating TNF-␣ secretion in our preliminary study. This might be attributed to the fact that different sequence motifs of rich CpG sequences have been shown to dramatically induce different profiles and kinetics in immune activation.14 Additional studies with other rich CpG content motifs would be necessary to confirm this point. We detected low levels of IL-2 secretion in our study that may be a consequence of azathioprine and corticosteroids long-term therapy in CD patients.24,25 Thus, our studies did not find support for the hypothesis that genomic DNA is responsible for the complete antiinflammatory properties of this particular probiotic. This statement is reinforced by the evidence that only TNF-␣, but not other major proinflammatory cytokines and chemokines, might be partially decreased by L. casei DNA coincubation. Discrepancies between the effects of live probiotic bacteria and its genomic DNA have been described by other groups in terms of intestinal barrier integrity following hemorrhagic shock, which agrees with the fact that DNA does not fully reproduce the effects of the viable bacteria.26 Considering the potent immunomodulatory effect of the live probiotic on the secretion of Th1 cytokines, we examined the specificity of this effect. Therefore, we investigated the involvement of this probiotic strain versus a com-

281

Llopis et al

mensal nonpathogenic E. coli on the induction of cytokines involved in the Th17 inflammatory pathway, recently related to autoimmune inflammatory disorders including IBD. Extensive effort has been concentrated recently in the IL-12-related IL-23 cytokine for its potential key role in mechanisms of cell-mediated tissue damage that takes place in autoimmune disorders and in severe immune responses triggered by infections. Furthermore, high levels of IL-23 transcripts have been correlated with severity of endoscopic lesions in CD-inflamed mucosa.27 IL-23 is produced in response to inflammatory danger signals by macrophages and dendritic cells,28 and seems to play a crucial role in survival, expansion, and further differentiation of Th17 cells.29 –33 Besides, IL-6 in combination with TGF-␤ are considered the main inducers of Th17 cells. Our results confirm that IL-23 and IL-6 cytokines are highly upregulated in CD versus healthy mucosa controls. Interestingly, coincubation of CD explants with a commensal E. coli sharply increased the expression of IL-23, whereas coincubation with L. casei did not affect the expression of this cytokine. On the other hand, L. casei significantly decreased expression of IL-6, while E. coli tended to increase it. This differential immune response to these 2 nonpathogenic strains may be crucial in a setting of autoimmunity, i.e., immune-mediated damage of its own tissues, as is the case in CD. Our results clearly support the notion that gut commensals can regulate expression of cytokines in different ways, thus influencing outcome responses in proinflammatory patterns. Recent studies reported elevated expression of IL-12 cytokine upon stimulation of macrophages of CD.34 In concordance, our results show an elevated expression of IL-12 cytokine on inflamed CD compared to healthy mucosa. Interestingly, incubation of CD explants with the commensal E. coli further increased levels of IL12p35, while incubation with the probiotic L. casei did not modify its expression. This underlines the fact that 2 commensal bacteria of different natures specifically modulate innate immune responses by antigen presenting cells.35 The outcome of Th17 priming and effector lymphocytes in the mucosa is secretion of proinflammatory cytokines IL-6, TNF-␣, and IL-17, which is characteristic of this T-cell subset. Large numbers of T cells and monocytes/macrophages expressing IL-17 have been found in inflamed mucosa of CD patients but not in healthy individuals,36 and our results confirm this finding. We investigated modulation of IL-17F cytokine gene expression by L. casei and E. coli and found that, while E. coli significantly increased expression of this cytokine, L. casei did not modify it. The production of chemokines such as IL-8, CXCL1, and CXCL2 by intestinal epithelial cells is critical to the development of inflammatory infiltrates in the intestine. IL-8 is an important mediator in the innate immune system response and dysregulation of IL-8 production has been pro-

282

Inflamm Bowel Dis

●

Volume 15, Number 2, February 2009

posed as a major contributor to intestinal inflammation in IBD patients, increasing with the degree of histological inflammation regardless of diagnosis.37,38 Our results confirm the high expression of IL-8 and CXCL1 in CD mucosa. Moreover, exposure of inflamed mucosa to E. coli further increased expression of IL-8 and CXCL1, while L. casei significantly downregulated these chemokines. As both chemokines are involved in chemoattraction of polymorphonuclear cells, the effect of L. casei appears as a potential mechanism to diminish the recruitment of inflammatory cells and thus a possible contributor to the antiinflammatory effect. This result is in agreement with previous results from our group in the TNBS model of colitis, when we reported an abrogation of ICAM-1 upregulation by this probiotic strain, resulting in an attenuation of leukocyte recruitment, and therefore, an improvement in the severity of colitis.9 Moreover, a recent article demonstrated that L. casei is able to downregulate the transcription of proinflammatory CXCL1, CXCL2, and CCL20 chemokines genes induced after invasion of cultured intestinal epithelial cells by Shigella flexneri.13 In our system we demonstrated that L. casei downregulates chemokines transcription and secretion from already inflamed CD mucosa without additional stimulus of other commensal or pathogen bacteria. This effect was bacteria-specific because commensal E. coli showed upregulation of these chemokine genes. Taking into account the differential recognition of the 2 tested bacteria by the mucosal immune system of CD patients, it appeared necessary to test combinations of both strains in order to investigate the immune response of CD mucosa. We investigated whether the antiinflammatory signals induced by L. casei were maintained in the presence of a commensal nonprobiotic bacterium as the tested E. coli, choosing the secretion of TNF-␣ as a key cytokine involved in primary immune responses and the polymorphonuclear chemoattractant chemokines IL-8 and CXCL1. Therefore, inflamed CD mucosa was exposed to a combination of both bacteria L. casei and E. coli at different timepoints for a period of 24 hours. Our results clearly show that L. casei is able to overcome the exaggerated reaction of the immune system against E. coli, both in a preventive and competitive way, as demonstrated by a significant reduction of TNF-␣ and chemokine secretion, suggesting that the composition of commensals might drive an overall immune response. In summary, our study suggests that viable bacteria induce specific immune responses by inflamed intestinal mucosa. Hypothetically, adequate combination of bacterial signals may influence the final outcome of the immune response, shifting from self-tissue damage mediated by autoimmunity toward regulatory tolerogenic responses that could favor mucosal healing. The impact of such stimuli is difficult to achieve in the in vivo situation due to the complexity of the microbial ecosystem. However, the arrival of new technolo-

Inflamm Bowel Dis

●

Volume 15, Number 2, February 2009

gies to detect nonculturable bacteria, and the development of more efficient intervention systems with probiotics/prebiotics to impact the microbial composition of the gut ecosystem, should provide better opportunities to investigate this hypothesis in the near future.

ACKNOWLEDGMENTS The authors thank Mrs. Montserrat Casellas, Mrs. Milagros Gallart, and Mrs. Carmen Alastrue for technical assistance in the analytical procedures. We also thank Dr. Mar Guilarte for suggestions concerning the Th17 pathway and Mrs. Fidelma Greaves for assistance in the English version of the article.

REFERENCES 1. Sartor RB. Therapeutic manipulation of the enteric microflora in inflammatory bowel diseases: antibiotics, probiotics, and prebiotics. Gastroenterology. 2004;126:1620 –1633. 2. Llopis M, Antolin M, Guarner F, et al. Mucosal colonisation with Lactobacillus casei mitigates barrier injury induced by exposure to trinitronbenzene sulphonic acid. Gut. 2005;54:955–959. 3. Madsen K, Cornish A, Soper P, et al. Probiotic bacteria enhance murine and human intestinal epithelial barrier function. Gastroenterology. 2001; 121:580 –591. 4. Gionchetti P, Rizzello F, Venturi A, et al. Oral bacteriotherapy as maintenance treatment in patients with chronic pouchitis: a doubleblind, placebo-controlled trial. Gastroenterology. 2000;119:305–309. 5. Rochet V, Rigottier-Gois L, Sutren M, et al. Effects of orally administered Lactobacillus casei DN-114 001 on the composition or activities of the dominant faecal microbiota in healthy humans. Br J Nutr. 2006;95: 421– 429. 6. Marcos A, Warnberg J, Nova E, et al. The effect of milk fermented by yogurt cultures plus Lactobacillus casei DN-114 001 on the immune response of subjects under academic examination stress. Eur J Nutr. 2004;43:381–389. 7. Parra D, De Morentin BM, Cobo JM, et al. Monocyte function in healthy middle-aged people receiving fermented milk containing Lactobacillus casei. J Nutr Health Aging. 2004;8:208 –211. 8. Chapat L, Chemin K, Dubois B, et al. Lactobacillus casei reduces CD8⫹ T cell-mediated skin inflammation. Eur J Immunol. 2004;34:2520 – 2528. 9. Angulo S, Llopis M, Antolin M, et al. Lactobacillus casei prevents the up-regulation of ICAM-1 expression and leukocyte recruitment in experimental colitis. Am J Physiol Gastrointest Liver Physiol. 2006;291: G1155–1162. 10. Borruel N, Carol M, Casellas F, et al. Increased mucosal tumour necrosis factor alpha production in Crohn’s disease can be downregulated ex vivo by probiotic bacteria. Gut. 2002;51:659 – 664. 11. Borruel N, Casellas F, Antolin M, et al. Effects of nonpathogenic bacteria on cytokine secretion by human intestinal mucosa. Am J Gastroenterol. 2003;98:865– 870. 12. Carol M, Borruel N, Antolin M, et al. Modulation of apoptosis in intestinal lymphocytes by a probiotic bacteria in Crohn’s disease. J Leukoc Biol. 2006;79:917–922. 13. Tien MT, Girardin SE, Regnault B, et al. Anti-inflammatory effect of Lactobacillus casei on Shigella-infected human intestinal epithelial cells. J Immunol. 2006;15:1228 –1237. 14. Krieg AM. Immune effects and mechanisms of action of CpG motifs. Vaccine. 2000;19:618 – 622. 15. Rachmilewitz D, Karmeli F, Takabayashi K, et al. Immunostimulatory DNA ameliorates experimental and spontaneous murine colitis. Gastroenterology. 2002;122:1428 –1441.

L. casei Downregulates E. coli Signals

16. Rachmilewitz D, Katakura K, Karmeli F, et al. Toll-like receptor 9 signalling mediates the anti-inflammatory effects of probiotics in murine experimental colitis. Gastroenterology. 2004;126:520 –528. 17. Jijon H, Backer J, Diaz H, et al. DNA from probiotic bacteria modulates murine and human epithelial and immune function. Gastroenterology. 2004;126:1358 –1373. 18. Lammers KM, Brigidi P, Vitali B, et al. Immunomodulatory effects of probiotic bacteria DNA: IL-1 and IL-10 response in human peripheral blood mononuclear cells. FEMS Immunol Med Microbiol. 2003;38:165–172. 19. Ochman H, Selander RK. Standard reference strains of Eschericha coli from natural populations. J Bacteriol. 1984;157:690 – 693 20. Finnie IA, Dwarakanath AD, Taylor BA, et al. Colonic mucin synthesis is increased by sodium butyrate. Gut. 1995;36:93–99. 21. Taganov KD, Boldin MP, Chang KJ, et al. NF-kappaB-dependent induction of microRNA miR-146, an inhibitor targeted to signaling proteins of innate immune responses. Proc Natl Acad Sci U S A. 2006;103: 12481–12486. 22. Katakura K, Sato Y, Sato N, et al. CpG motifs in plasmid DNA exacerbate inflammation in experimental colitis. Gastroenterology. 2001;120(suppl 1):A518. 23. Obermeier F, Dunger N, Deml L, et al. CpG motifs of bacterial DNA exacerbate colitis of dextran sulfate sodium-treated mice. Eur J Immunol. 2002;32:2084 –2092. 24. Chan GL, Pharm D, Canafax DM, et al. The therapeutic use of azathioprine in renal transplantation. Pharmacotherapy. 1987;7:165–177. 25. Pinkston P, Saltini C, Mu¨ller-Quernheim J, et al. Corticosteroid therapy suppresses spontaneous interleukin 2 release and spontaneous proliferation of lung T lymphocytes of patients with active pulmonary sarcoidosis. J Immunol. 1987;139:755–760. 26. Luyer MD, Buurman WA, Hadfoune M, et al. Strain-specific effects of probiotics on gut barrier integrity following hemorrhagic shock. Infect Immun. 2005;73:3686 –3692. 27. Schmidt C, Giese T, Ludwig B, et al. Expression of interleukin-12related cytokine transcripts in inflammatory bowel disease: elevated interleukin-23p19 and interleukin-27p28 in Crohn’s disease but not in ulcerative colitis. Inflamm Bowel Dis. 2005;11:16 –23. 28. Gocke AR, Cravens PD, Ben LH, et al. T-bet regulates the fate of Th1 and Th17 lymphocytes in autoimmunity. J Immunol. 2007;178:1341–1348. 29. Aggarwal S, Gurney Al. IL-17: prototype member of an emerging cytokine family. J Leukoc Biol. 2002;71:1– 8. 30. Bettelli E, Carrier Y, Gao W, et al. Reciprocal developmental pathways for the generation of pathogenic effector TH17 and regulatory T cells. Nature. 2006;441:235–238. 31. Cua DJ, Kastelein RA. TGF-beta, a double agent in the immune pathology war. Nat Immunol. 2006;7:557–559. 32. Mangan PR, Harrington LE, O’Quinn DB, et al. Transforming growth factor-beta induces development of the T(H)17 lineage. Nature. 2006; 441:231–234. 33. Veldhoen M, Hocking RJ, Atkins CJ, et al. TGF-beta in the context of an inflammatory cytokine milieu supports de novo differentiation of IL-17-producing T cells. Immunity. 2006;24:179 –189. 34. Fuss IJ, Becker C, Yang Z, et al. Both IL-12p70 and IL-23 are synthesized during active Crohn’s disease and are down-regulated by treatment with anti-IL-12 p40 monoclonal antibody. Inflamm Bowel Dis. 2006;12:9 –15. 35. McKenzie BS, Kastelein RA, Cua DJ. Understanding the IL-23-IL-17 immune pathway. Trends Immunol. 2006;27:17–23. 36. Fujino S, Andoh A, Bamba S, et al. Increased expression of interleukin 17 in inflammatory bowel disease. Gut. 2003;52:65–70. 37. Katsuta T, Lim C, Shimoda K, et al. Interleukin-8 and SDF1-alpha mRNA expression in colonic biopsies from patients with inflammatory bowel disease. Am J Gastroenterol. 2000;95:3157–3164. 38. Stallmach A, Giese T, Schmidt C, et al. Cytokine/chemokine transcript profiles reflect mucosal inflammation in Crohn’s disease. Int J Colorectal Dis. 2004;19:308 –315.

283