JOURNAL MENUFIND ISSUES
GET ACCESS
ABOUT THIS JOURNAL
Marie-B&n&dicte Romond1, Catherine Mulli&2, Michel Colavizza1, Fran&oise Revillion3, Jean-Philippe Peyrat3 andDaniel Izard1DOI:&10.1111/j.08.00501.x
FEMS Immunology & Medical Microbiology pages 85&92, Author Information1
Laboratoire de Bact&riologie-Virologie (EA3610), Facult& des Sciences Pharmaceutiques et Biologiques, Lille, France2
Facult& de Pharmacie, Amiens, France3
Laboratoire d'Oncologie Mol&culaire Humaine, Centre Oscar Lambret, Lille, France*Correspondence: Catherine Mulli&, Facult& de Pharmacie, 1 rue des Louvels, 80037 Amiens Cedex, France. Tel.: +33 322 827 637; fax: +33 322 827 469; e-mail: Publication HistoryIssue online: 6 JAN 2009Version of Record online: 9 DEC 2008Received 4 February 2008; revised 6 November 2008; accepted 6 November 2008.First published online 8 December 2008.
ARTICLE TOOLS
Bifidobacterium;bacregulationThis study aimed at determining the contribution of intestinal bifidobacteria to the immune system activation using widely distributed galectins as markers of immune cell homoeostasis. In human flora-associated mice, bacteria were enumerated in the gut, blood, spleen, liver and lungs, while the expression of galectin-1 (Gal-1) and galectin-3 (Gal-3) was estimated by PCR in the intestine and real-time quantitative PCR in the other organs. Gal-1 and -3 were rarely expressed in the intestine. In blood, only Gal-1 was expressed while both galectins were expressed in all other organs. A high prevalence of colonic bifidobacteria was associated with a lower expression of both pulmonary galectins, whose levels negatively correlated with bifidobacterial counts. Caecal bifidobacterial counts also negatively correlated with pulmonary Gal-3 mRNA levels. The spleen was the only organ showing an upregulation of Gal-1 expression related to its bacterial contamination. However, this upregulation was only observed when bifidobacteria were not detected in the colon. A putative mechanism explaining the reduced expression of galectins when bifidobacteria highly colonize the mouse intestine could be that, by reducing the bacterial translocation, bifidobacteria also lead to a decreased blood concentration of substances produced by intestinal bacteria.Bacterial translocation (BT) is defined as the transmucosal passage of viable and nonviable microorganisms and their byproducts, such as endotoxins, across an intact intestinal barrier.BT in the normal human appendix was demonstrated to parallel the development of the local immune system (). It is considered to help the gut to elaborate a controlled local immune response to keep alimentary antigens away from the internal milieu. Septic complications might arise from BT only when the host's immune defences are overwhelmed or when the gut mucosa are impaired, for example in patients undergoing elective abdominal surgery, organ donors or those with intestinal obstruction, colorectal cancer, ischaemia&reperfusion injury shock or pancreatitis (). Strong evidence was provided that septic complications were more prevalent in patients who showed BT in their mesenteric lymph nodes (). After small bowel transplantation in children, BT was also reported in about half the patients and was increased by the presence of a colon allograft. A substantial percentage of acute rejection was associated with BT episodes (). Failure of the gut barrier and endotoxaemia have also been implicated in multiple organ failure syndromes and mortality in prematures ().We recently demonstrated that the composition of the intestinal flora is a major factor regulating BT, especially the passage of bacteria through the enterohepatic route (). A reduction in BT was related to the gut colonization prevalence with bifidobacteria and counts in bifidobacteria. It could thus be tempting to promote intestinal bifidobacteria to reduce BT in populations at risk of developing infection such as prematures, children suffering from short bowel syndrome or patients who have undergone an intestinal transplantation. Manipulation of the gut flora to reduce BT could, however, coincide with a pro-inflammatory status of the host, especially in internal organs where the clearance of bacteria is highly active.Indeed, colonization of the gut by bifidobacteria can affect the immune system response to a foreign antigen at the distance of the gut likely by priming a Th1 response. In infants, the increase in intestinal bifidobacteria was shown to contribute to an enhanced IgA response to parenteral poliovirus vaccination (). In gnotobiotic mice, colonization with bifidobacteria elicited a higher anti-rotavirus immunoglobulin A (IgA) response following rotavirus infection (). It was also shown that colonization of germ-free mice with bifidobacteria triggers a T-cell response that allows for the regulation of Escherichia coli translocation to internal organs ().Evaluation of the inflammatory status of organs under physiological conditions can be carried out using ubiquitously expressed proteins involved in the regulation of inflammation. Galectins are a family of carbohydrate-binding proteins that function as master regulators of immune cell homoeostasis and inflammation (). Among them, galectin-1 (Gal-1) and galectin-3 (Gal-3) are widely distributed in tissues () and bind distinct cell surface glycoprotein receptors to control T cell death (). Extracellular Gal-1 causes the death of activated, but not resting T cells (), while intracellular Gal-3 blocks T cell death (). In contrast to the antiapoptotic function of intracellular Gal-3, extracellular Gal-3 directly induces the death of T cells. Because the expression of Gal-1 and -3 transcripts is primarily detected after activation in T cells, B cells, macrophages and endothelial cells (), quantification of their transcripts in internal organs (e.g. the lungs, kidneys, liver or spleen) with a reduced bacterial contamination reflects the stimulation of these organs. Indeed, a single Corynebacterium kutscheri infection induced the upregulation of Gal-3 in the lungs of rats () while the synthesis and accumulation of Gal-3 was shown in the cytosol of phagocytic leucocytes during streptococcal pneumonia in mice, and was associated with proinflammatory cytokine expression (). In the spleen, Gal-3 accumulated in dendritic cells following infection with Trypanosoma cruzi (). Gal-1 expression was also substantially increased in the spleen of mice infected with T. cruzi (). Therefore, Gal-1 and Gal-3 seem relevant markers of bacteria- and/or parasite-driven inflammation.In the present study, we address the issue of the regulation of intestinal and extraintestinal expressions of Gal-1 and Gal-3 according to the establishment of intestinal bifidobacteria as well as other anaerobic members that were shown to play a role in BT such as Bacteroides sp. and clostridia. To this end, we quantified their colonization prevalence in the gut and the expression of the galectins in different tissues of a previously described mouse model associated with a human flora (HF) (). In addition to bacteria from the gut, bacteria that temporarily contaminate internal organs through BT were also enumerated in the lungs, liver, spleen and kidneys, as they might also modulate galectin gene expressions. Analysis of pro-[tumour necrosis factor & (TNF-&) and interleukin 6 (IL-6)] and anti-inflammatory (IL-10) cytokine expression was also carried out to further characterize the possible inflammatory status of internal organs. Gal-1 and Gal-3 expressions were thereafter analysed according to intestinal and extraintestinal bacterial populations.AnimalsAll experiments were conducted following the guidelines of the French Ministry of Agriculture. Adult germ-free C3H mice (CDTA, CNRS, Orl&ans, France) were associated with a complex HF and afterwards kept in sterile isolators. To reproduce a normal immune response to commensal intestinal bacteria, only generations descendant from these formerly germ-free mice were used for the assays, the pups being in contact with intestinal bacteria from birth. These pups, born to HF dams, were also kept in sterile isolators (La Calh&ne, V&lizy, France) with free access to sterile RO3 pellets (UAR, Epinay-sur-Orge, France) and water to avoid any oral intake of foreign and/or environmental bacteria that might break the balance of the implanted HF. At birth, the bifidobacterial load in pups thus depended on dams' bifidobacterial carriage. This model mimics the colonization process in humans. At weaning, no difference in the intestinal colonization of pups with bifidobacteria was observed, irrespective of the bifidobacterial load of the dam. After weaning, a follow-up of intestinal bifidobacteria showed that the kinetic of bifidobacterial shedding was mainly affected by the size of this bifidobacterial load. Hence, to describe the influence of microbial community composition on gene expression, experiments were conducted only in 12&18-week-old adult mice, corresponding to the beginning of bifidobacterial shedding and a stable level of galectin expression in organs ().Sample collectionMice were anaesthetized by chloroform inhalation after a 24-h fast. Fractions of the spleen, liver, lungs, ileum (proximal, median and distal specimens), caecum and colon were collected under aseptic conditions for immediate bacterial counting. The remaining fractions of organs were also collected under aseptic conditions, perfused with phosphate-buffered saline containing an RNAse inhibitor (Applied Biosystems, Courtaboeuf, France) and placed in an RNAlater& stabilizing solution (Qiagen, Courtaboeuf, France). They were then stocked at &20&&C until use.Bacterial enumerationOrgan samples were transferred to an anaerobic chamber (La Calh&ne), homogenized in a prereduced sterile 1&:&4 diluted Ringer solution with cysteine&HCl (0.03%) and serially diluted. Appropriate dilutions of heart blood, fractions of the spleen, liver and lungs, and Peyer's patches were plated onto horse blood agar (Columbia agar base, Oxoid, Dardilly, France) supplemented with glucose (0.5%) and cysteine&HCl (0.03%). Appropriate dilutions of the ileal fragments, caecum and colon were plated onto Beerens medium () and Bile Bacteroides Esculin medium for enumeration of bifidobacteria and Bacteroides, respectively. Clostridial spores were enumerated by plating dilutions of the intestinal fractions heated for 10&min at 70&&C on Columbia agar base (Oxoid) supplemented with glucose (0.5%) and cysteine&HCl (0.03%). Plates were incubated in an anaerobic chamber for 5 days and colonies were counted. Bifidobacteria were identified at the species level using multiplex PCR () and the other bacterial groups using API systems (Biom&rieux, Marcy l'Etoile, France).Isolation of total RNATotal RNA was isolated from organ samples (10&30&mg) using the RNeasy Mini Kit (Qiagen). The disruption and homogenization of samples were performed using a Retsch Homogeniser MM200 (Haan, Germany). The amount of RNA extracted was quantified by measuring A260&nm. Its purity was checked by the A260&nm/A280&mn ratio and ranged between 1.904 and 2.627. PCR primers (Gal-1 forward: tcaatcatggcctgtggtctg, reverse: Gal-3 forward: gatttcaggagagggaatgatgttg, reverse: cagttattgtcctgcttcgtgttaca) and TaqMan fluorogenic probes (Gal-1: tcgccagcaacctgaatctcaaacctg, Gal-3: ttccactttaacccccgcttcaatgagaac) were designed using the primer express software program (version 1.5, Applied Biosystems). For cytokine semi-quantitative assessment, amplification of &-actin was used as control and the primers for cytokines were as follows: IL-6 forward: gatttacataaaatagtccttcc, reverse: tgc TNF-& forward: gagtagctgagccagcgcgcc, reverse: g and IL-10 forward: ttgggttgccaagccttatcgga, reverse: ttcacctgctccactgccttgct. The gene specificity of the sequences chosen for primers and probes was confirmed using the blastn programSemi-quantitative reverse transcriptase (RT)-PCRReverse transcription and PCR were performed in two steps on total RNA extracts from intestinal fragments, blood and internal organs. The first reaction mixture (40&&L final volume) contained 50&ng total RNA (5&&L), 10 & PCR II buffer (4&&L) (Applied Biosystems), MgCl2 (1.25&mM), 16.6&U RNAse inhibitor, Moloney Murine Leukaemia Virus reverse transcriptase (12.5&U), dNTPs (500&&M) and poly dT (2.5&&M). Reverse transcription was performed at 42&&C for 90&min, followed by a 7-min incubation at 90&&C after an intermediate cooling step at +4&&C for 10&min. The reaction mixture for PCR (50&&L final volume) contained 5&&L DNA transcript, 10 & PCR II buffer (5&&L), MgCl2 (1.75&mM), 2.5&U Ampli Taq Gold (Applied Biosystems), dNTPs (250&&M) and 500&nM forward and reverse primers. For Gal-1 and Gal-3, after heating at 95&&C for 8&min, PCR (30&s at 95&&C, 30&s at 50&&C and 90&s at 72&&C for 40 cycles) was carried out. The last elongation period was of 10&min at 72&&C. For cytokines, after heating at 95&&C for 8&min, PCR conditions were: 94&&C for 1&min, 40&&C for 1&min and 72&&C for 2&min (40 cycles). PCR-amplified samples were separated on a 2% agarose gel and visualized by ethidium bromide staining.Real-time quantitative PCR (RT-qPCR)Total RNA was isolated from the spleen, liver and lungs of four mice maintained under the same conditions as the control group. The RNA extracts were then pooled according to the organ they were extracted from and used to calibrate the method (pooled extract subsequently referred to as a calibrator). Reverse transcription and PCR were performed in a one-step methodology using the ABI PRISM 7700 Sequence Detection System (Applied Biosystems). The reaction mixture (50&&L final volume) contained 50&ng total RNA, 10 & TaqMan buffer (5&&L), MgCl2 (5&mM), 20&U RNAse inhibitor, 12.5&U Moloney Murine Leukaemia Virus reverse transcriptase, 1.25&U AmpliTaq Gold DNA polymerase (Applied Biosystems), 300&&M dNTPs, 200&nM probe and 200&nM forward and reverse primers (except for Gal-3 in the liver: 300&nM of each primer). After a 10-min step at 65&&C, reverse transcription was performed at 42&&C for 30&min. The activation of the Taq DNA Polymerase, 10&min at 95&&C, was followed by PCR (15&s at 95&&C and 60&s at 61&&C for 40 cycles). A nontemplate control was included in each experiment. All nontemplate controls and samples were assayed in duplicate. The relative quantification of Gal-1 and Gal-3 gene expressions was performed using the comparative cycle threshold (CT) method (Applied Biosystems, User bulletin #2, 1997). The expression of the target gene (i.e. Gal-1 and Gal-3) was normalized against the expression of the &-actin housekeeping gene (forward primer: t reverse primer: gggatgtttgctccaaccaac, probe: gccgtcgccttcaccgttccagttttt). The TaqMan probes of Gal-1 and Gal-3 carried a 5& FAM (6-carboxy-fluorescein) reporter dye and the TaqMan probe of &-actin carried a 5& VIC (Applied Biosystems) one.Statistical analysisVariations in the bacterial counts or galectin relative expressions (RE) were analysed using the Mann&Whitney rank test or the nonparametric Kruskal&Wallis test for independent samples. A test of multiple comparisons was carried out subsequently in case of heterogeneity to specify which group differed (). Correlation was obtained by calculating the Spearman correlation coefficient. The statistical evaluation was performed using the vassarstats website (). Statistical significance was defined as P&0.05.Intestinal microbial communityBifidobacteria were established in nearly 59% of the caeca (). Forty-one per cent of the mice jointly harboured bifidobacteria in both the caecum and the colon. The isolated bifidobacterial strains were identified as Bifidobacterium breve and Bifidobacterium longum. In animals deprived in caecal bifidobacteria, Bacteroides was detected. Isolated Bacteroides strains were primarily identified as Bacteroides thetaiotaomicron, Bacteroides ovatus and Bacteroides uniformis. Only two mice harboured bifidobacteria and Bacteroides at the same time in the caecum. Moreover, bifidobacteria and Bacteroides counts correlated negatively (rs=&0.663, P=0.00376). The highest carriage prevalence of clostridial spores was found in the caecum (59%). Most of the strains were identified as Clostridium hathewayi. A few were unspecified using Api32A, but were closely related to Clostridium bifermentans, Clostridium innocuum and Clostridium butyricum.Table&1.&
&Distribution of anaerobic bacteria along the intestine (n=17 mice) &BifidobacteriaBacteroidesClostridia&&Ileum&Proximal& (0)3.3&0.40 (6)2.4 (1)Median3.3&0.11 (2)3.2&0.26 (4)2.9&0.20 (4)Distal4.5&0.24 (7)3.7&0.51 (4)3.4&0.73 (7)&Caecum&6.1&1.15 (10)4.4&0.77 (7)4.1&0.43 (10)&Colon&6.2&0.36 (7)4.5&0.52 (6)3.2&0.39 (7)BTEnumeration of bacteria under anaerobic conditions gave positive results in 29.4&52.9% of the internal organs according to the location (). The liver and spleen were the most commonly contaminated organs. The isolated bacteria were identified as Clostridium clostridioforme, E. coli (spleen), Morganella morganii (liver), Enterococcus faecalis, Propionibacterium acnes and Staphylococcus epidermidis. A few anaerobic rods were isolated, but could not be identified using Api32A. Clostridia and enterobacteria were found in organs of mice deprived in intestinal bifidobacteria.Table&2.&
&Bacterial contamination of internal organs (n=17 mice) &Total floraClostridiaEnterobacteriaEnterococciSpleen4.6&1.19 (8)4.5&0.69 (2)5.5&1.50 (3)& (0)Liver4.3&1.28 (9)3.3&0.14 (2)4.8&0.62 (3)5.5&0.01 (2)Lungs5.3&1.45 (5)& (0)5.2&1.31 (3)7.0 (1)Blood3.9&0.74 (5)3.6&0.60(3)5.0 (1)3.7 (1)Expression of Gal-1 and Gal-3 in the intestine and bloodGal-1 mRNA was rarely detected in the ileum and caecum of the HF-associated mice (data not shown). In the colon, two-thirds of the tested colonic fragments were positive. All mice expressing Gal-1 were, however, deprived in colonic bifidobacteria. Expression of Gal-3 was rare along the intestine. In the absence of detectable clostridia, two-thirds of the animals did not express Gal-3 in any intestinal fragment. The remaining mice expressed Gal-3 only in one of the ileal fragments out of the five intestinal fragments tested. When clostridia were detected in the intestine, all animals expressed Gal-3 in the lower part of the intestine (two to three fragments out of the distal ileum, caecum and colon). In blood, Gal-3 mRNA was never detected by semi-quantitative RT-PCR despite a few clostridial bacteraemias (), whereas all samples were positive for Gal-1 mRNA.Expression of cytokines in internal organsTNF-& and/or IL-6 mRNA were detected in only half of the splenic samples while about 15% of the samples did not express both cytokines. In contrast, IL-10 mRNA was detected in 87.5% of the samples. The presence of bacteria in the spleen was not related to a high prevalence of either cytokine detection.In the lungs, TNF-& mRNA was detected in 87.5% of the samples whereas IL-6 and IL-10 were expressed in only half of them. The presence of Gram-negative bacteria was associated with a strong expression of TNF-&. Otherwise, TNF-& mRNA was only detected when bifidobacteria poorly colonized the lower part of the intestine.In the liver, TNF-&, IL-6 and IL-10 mRNA were detected in about 90% of the samples. The negative samples for cytokines corresponded to highly contaminated hepatic fractions, indicating that TNF-&, IL-6 and IL-10 expression was independent of the bacterial load.Expression of Gal-1 and Gal-3 in internal organsIn the liver, spleen and lungs of all animals, both galectins were expressed. RT-qPCR was then used to determine the effects of bacterial contamination and of the composition of the enteric flora on their expression. In the liver, the mean RE of Gal-1 mRNA was 0.92&0.43. The presence of bacteria in the liver () or the composition of the intestinal flora did not influence the hepatic expression of Gal-1. In contrast, Gal-3 mRNA was lower when intestinal Bacteroides were detected in the colon (). Moreover, Gal-3 expression negatively correlated with the colonic Bacteroides population (rs=&0.708, P=0.00146). In the spleen, Gal-1 and Gal-3 RE were high (2.92 and 1.52, respectively), indicating an upregulation. For Gal-1, this upregulation coincided with a lack of bifidobacteria in the colon and bacterial contamination of the spleen (). Moreover, Gal-1 expression positively correlated with the presence of bacteria in the organ itself (rs=0.756, P=0.02988, n=8). When bifidobacteria were detected in the colon, splenic Gal-1 mRNA contents were lower (although still high), but RE values were similar irrespective of the splenic bacterial contamination (). Neither the presence of bacteria in the spleen nor the counts in intestinal bifidobacteria modified the splenic Gal-3 expression (mean RE value=1.52&0.46). In the lungs, both galectins were downregulated in mice harbouring bifidobacteria in the colon whereas their expressions were similar irrespective of the pulmonary bacterial contamination (). Moreover, Gal-3 RE negatively correlated with caecal and colonic bifidobacterial counts (caecum: rs=&0.607, P=0.000977; colon: rs=&0.737, P=0.00074). Gal-1 mRNA also negatively correlated with bifidobacteria, but only with the colonic population (rs=&0.506, P=0.03844). However, caecal clostridia and Bacteroides had opposite effects on Gal-1 expression (downregulation for clostridia, upregulation for Bacteroides) (). Moreover, Gal-1 mRNA correlated negatively with the caecal clostridial population (rs=&0.547, P=0.02315).Figure&1. &Hepatic Gal-3 mRNA expression according to colonic colonization with Bacteroides thetaiotaomicron. *P&0.04 with vs. without B. thetaiotaomicron (Mann&Whitney test).Figure&2. &Gal-1 expression according to bacterial contamination of the spleen and colonic colonization with bifidobacteria. *P=0.007 contaminated vs. no **P=0.017 in contaminated spleens of mice with vs. without intestinal bifidobacteria (Kruskal&Wallis test).Figure&3. &Downregulation of galectin expression exerted in the lungs by colonic colonization with bifidobacteria. *P=0.017 Gal-1 mRNA expression in mice with vs. without intes **P=0.007 Gal-3 mRNA expression in mice with vs. without intestinal bifidobacteria (Mann&Whitney test).Figure&4. &Influence of caecal clostridia and Bacteroides populations on Gal-1 mRNA expression in the lungs. *P&0.02 mice with vs. witho **P&0.05 mice with vs. without caecal Bacteroides (Mann&Whitney test).The present animal model was designed to replace conventional animals because the change in hygienic standards that occurred in the late 1990s in breeding facilities led to a modification in the intestinal carriage of anaerobic bacteria characterized by a drastic decrease in counts in Bacteroides, bifidobacteria and clostridia. The rectal flora from a human donor harbouring sufficient amounts of Bacteroides, bifidobacteria and clostridia (i.e. 5 & 106&CFU&mL&1 of inoculum) was used to create the present HF model. As the immune system of these formerly germ-free mice remained partly immature (e.g. low number of Peyer's patches), they could not be used as such for the present investigations. However, the immune system of pups born to these ex-germ-free mice resembled the conventional one, making these animals suitable for the work. Additionally, Bacteroides, bifidobacteria and clostridia were shown to easily colonize the intestine of the pups. Besides, the uneven shedding of bifidobacteria provided the opportunity to study the impact of the number of intestinal bifidobacteria on the inflammatory status of the host.In a first attempt to characterize the inflammatory status of internal organs of HF-mice in response to BT, we investigated cytokines such as TNF-&, IL-6 and IL-10. However, in contrast to galectins, they were not readily expressed in internal organs. Besides, no relationship between the bacterial load in internal organs and cytokine expression was observed by semi-quantitative RT-PCR. Thus, although they are commonly used for determining the response to pathogens, they are not suitable markers for investigating the response to translocating bacteria. Their infrequent expression does not allow for analysing the effect of microbial community composition on the inflammatory status of the internal organs. Thus, we turned to galectins as alternative inflammation indicators ().The present study showed that the quantitative distribution of the ubiquitous Gal-1 and Gal-3 transcripts in HF-associated mice was comparable to the one observed in conventional animals (), except for a rare detection in the intestine. Using in situ hybridization analysis,
demonstrated the region-dependent expression of galectin subtypes 1 and 3 in the digestive tract of conventional mice. Only weak signals of Gal-1 were detected diffusely in the lamina propria mucosa and in the muscle layer. In our study, the rare expression of Gal-1 indicated that association to an HF led to an unchanged Gal-1 expression, similar to conventional mice. In contrast to the broad distribution of Gal-3 mRNA in the gastric and intestinal mucosa of conventional animals, HF mice expressed this lectin poorly. The rare expression of Gal-3 in mice poorly colonized with clostridia suggested that Gal-3 expression might reflect the response to some specific intestinal bacteria.In the liver, Gal-3 is primarily localized in Kupffer cells and its quantitative analysis based on densitometric data indicated a low relative density in the liver as compared with the lungs and spleen (). In our study, the lack of galectin overexpression in liver samples containing bacteria as compared with sterile samples could indicate an immune tolerance that resembles the anti-inflammatory response to commensal bacteria of intestinal dendritic cells (). The absence of or poor TNF-& and IL-6 mRNA detection in highly contaminated hepatic samples as compared with sterile ones also supports this hypothesis. However, a high colonization prevalence of Bacteroides in the colon affected the hepatic expression of Gal-3. This could be related to a metabolic response of the liver rather than an immune one. Indeed, rat hepatic stellate cells under high glucose concentrations show a downregulation of galectin protein (). And B. thetaiotaomicron as well as B. ovatus possess a broad range of glycosidases, involved in the breakdown of a broad array of host-derived glycans and dietary polysaccharides (). Hence, the oligosaccharides released into the colonic lumen by Bacteroides could partly be absorbed by enterocytes and transported through the portal vein route to the liver, where they could affect the expression of galectin in hepatic cells.In the lungs of adult conventional mice, Gal-3 protein was detected in the covering epithelia of the bronchioles and in some macrophages whereas Gal-1 protein was observed in the base of alveolar attachments to blood vessels and the pleura and in the tips of secondary alveolar septae (). In our study, the lack of a relationship between the expression of galectins and bacterial contamination in the lungs might result from the relatively poor number of pulmonary immune cells as compared with other cells expressing galectins. For instance, a noninfected mouse yields a total of 1 & 105&5 & 105 alveolar macrophages, which is negligible, compared with the overall number of cells expressing galectins in the lungs. Endothelial cells were shown to upregulate Gal-1 synthesis when stimulated with lipopolysaccharides and cytokines (). Therefore, some sterile lung samples with a high level of Gal-1 transcripts could contain bacterial byproducts such as lipopolysaccharides. In this context, the reduced pulmonary expression of galectin in animals whose colon harboured bifidobacteria would not only reflect the rarer contact with bacteria as demonstrated previously in the lungs of these animals () but also the seldom exposure to bacterial byproducts. The effect of the caecal Bacteroides further supports this hypothesis. The colonization of the caecum by Bacteroides was related to an increased translocation of live bacteria to the lungs () and, in the present study, it favoured a higher level of Gal-1 mRNA in the organ. A putative mechanism supporting this increase is that lipopolysaccharides from Bacteroides were transferred to the lungs and locally stimulated endothelial cells. The downregulation of Gal-1 transcription in animals whose caecum was highly colonized with clostridia could also indicate that the lungs were infused with compounds from the bacterial cell wall. Although we never isolated clostridia in the lungs, likely because of an efficient killing related to the high pulmonary redox potential, we may assume that endothelial cells were more often in close contact with clostridial compounds such as lipoteichoic acid (LTA), a major cell wall component of Gram-positive bacteria, when clostridia highly colonized the intestine. LTA inhibits lipopolysaccharide-induced adhesion molecule expression in human microvascular endothelial cells (). In our model, the reduced expression of pulmonary Gal-1 when clostridia highly colonized the caecum could be related to an increased concentration of clostridial compounds in the microvascularization that partly inhibited the transcription of Gal-1 mRNA induced by Gram-negative bacteria (such as enterobacteria or Bacteroides)-derived compounds.Immune cells are the main constituents of the spleen. In this study, a high splenic expression of both galectins was observed that could reflect the efficient response of immune cells to any antigen reaching this organ. The correlation between splenic bacterial counts and Gal-1 expression also supported the assumption that the overexpression of Gal-1 in organs contaminated with bacteria reflected an immune response initiated by the presence of these live bacteria. Interestingly, a high colonic colonization with bifidobacteria abrogated the increase in Gal-1 mRNA related to the bacterial contamination of the spleen. Because a high colonization prevalence of the colon by bifidobacteria was also correlated with a decreased expression of galectin in the lungs, the mechanism underlying the effect in both organs could be identical (i.e. a reduced exposure to bacterial byproducts such as lipopolysaccharides).In conclusion, the prevention of BT observed when bifidobacteria highly colonized the gut of HF-associated mice did not lead to a proinflammatory status in the internal organs tested, considering the reduced expression of galectins. A putative mechanism explaining this reduction could be that the flow of bacteria and their byproducts to internal organs is reduced by the high bifidobacterial colonization of the intestine, therefore improving the first clearance steps.The authors are grateful to Fr&d&ric Huguet, Val&rie Lhotellier and Nathalie Bernard-Derensy for their technical assistance.
Baum LG, Seilhamer JJ, Pang M, Levine WB, Beynon D & Berliner JA (1995) Synthesis of an endogenous lectin, galectin-1, by human endothelial cell activation. Glycoconjugate J
12: 63&68. ,,
Beerens H (1990) An elective and selective isolation medium for Bifidobacterium spp. Lett Appl Microbiol
11: 155&157.
Bjursell MK, Martens EC & Gordon JI (2006) Functional genomic and metabolic studies of the adaptations of a prominent adult human gut symbiont, Bacteroides thetaiotaomicron, to the suckling period. J Biol Chem
281: 36269&36279. ,,
Blease K, Chen Y, Hellewell PG & Burke-Gaffney A (1999) Lipoteichoic acid inhibits lipopolysaccharide-induced adhesion molecule expression and IL-8 release in human lung microvascular endothelial cells. J Immunol
163: 6139&6147. ,,
Cicalese L, Sileri P, Green M, Abu-Elmagd K, Kocoshis S & Reyes J (2001) Bacterial translocation in clinical intestinal transplantation. Transplantation
71: 1414&1417. ,,
Foster JJ, Goss KL, George CL, Bangsund PJ & Snyder JM (2006) Galectin-1 in secondary alveolar septae of neonatal mouse lung. Am J Physiol-Lung C
291: L1142&L1149. ,,
Gebbers JO & Laissue JA (2004) Bacterial translocation in the normal human appendix parallels the development of the local immune system. Ann NY Acad Sci
1029: 337&343. ,
Hernandez JD & Baum LG (2002) Ah, sweet mystery of death! Galectins and control of cell fate. Glycobiology
12: 127R&136R. ,,
Hooper LV, Falk PG & Gordon JI (2000) Analyzing the molecular foundations of commensalisms in the mouse intestine. Curr Opin Microbiol
3: 79&85. ,,
Kim H, Lee J, Hyun JW, Park JW, Joo HG & Shin T (2007a) Expression and immunohistochemical localization of galectin-3 in various mouse tissues. Cell Biol Int
31: 655&662. ,,
Kim PK, Kim MR, Kim HJ, Yoo HS, Kim JS, Cho EH & Kim CW (2007b) Proteome analysis of the rat hepatic stellate cells under high concentrations of glucose. Proteomics
7: 2184&2188. ,,
Lowry R (2007) VassarStats [Online]. Vassar college website,
MacFie J (2004) Current status of bacterial translocation as a cause of surgical sepsis. Brit Med Bull
71: 1&11. ,
MacFie J, O'Boyle C, Mitchell CJ, Buckley PM, Johnstone D & Sudworth P (1999) Gut origin of sepsis: a prospective study investigating associations between bacterial translocation, gastric microflora, and septic morbidity. Gut 45: 223&228. ,,
MacFie J, Reddy BS, Gatt M, Jain PK, Sowdi R & Mitchell CJ (2006) Bacterial translocation studied in 927 patients over 13 years. Br J Surg
93: 87&93. ,,
Moreau MC & Gaboriau-Routhiau V (2000) Influence of resident intestinal microflora on the development and functions of the intestinal-associated lymphoid tissue. Probiotics (Fuller & Perdigon, eds), pp. 69&114. Kluwer academic publishers, the Netherlands.
Mulli& C, Odou MF, Singer E, Romond MB & Izard D (2003) Multiplex PCR using 16S rRNA gene-targeted primers for the identification of bifidobacteria from human origin. FEMS Microbiol Lett
222: 129&136.
Mulli& C, Yazourh A, Thibault H, Odou MF, Singer E, Kalach N, Kremp O & Romond MB (2004) Increased poliovirus-specific intestinal antibody response coincides with promotion of Bifidobacterium longum-infantis and Bifidobacterium breve in infants: a randomized, double-blind, placebo-controlled trial. Pediatr Res 56: 791&795.
Nio J, Kon Y & Iwagana T (2005) Differential cellular expression of galectin family mRNAs in the epithelial cells of the mouse digestive tract. J Histochem Chem
53: 1323&1334. ,,
Perillo NL, Uittenbogaart CH, Nguyen JT & Baum LG (1997) Galectin-1, an endogenous lectin produced by thymic epithelial cells, induces apoptosis of human thymocytes. J Exp Med
185: 1851&1858. ,,
Rabinovich GA, Rubinstein N & Toscano MA (2002) Role of galectins in inflammatory and immunomodulatory processes. Biochim Biophys Acta
1572: 274&284. ,,
Romond MB, Colavizza M, Mulli& C, Kalach N, Kremp O, Mielcarek C & Izard D (2008) Does the intestinal bifidobacterial colonisation affect bacterial translocation?
14: 43&48. ,,
Scherrer B (1984) Biostatistiques. Ga&tan Morin-Eska Publishing Co., Quebec, Canada.
Sharma R, Tepas JJ III, Hudak ML, Mollitt DL, Wludyka PS, Teng RJ & Premachandra BR (2007) Neonatal gut barrier and multiple organ failure: role of endotoxin and proinflammatory cytokines in sepsis and necrotizing enterocolitis. J Pediatr Surg 42: 454&461. ,
Stillman BN, Hsu DK, Pang M, Brewer CF, Johnson P, Liu FT & Baum LG (2006) Galectin-3 and galectin-1 bind distinct cell surface glycoprotein receptors to induce T cell death. J Immunol
176: 778&789. ,,
Takenaka S, Safroneeva E, Xing Z & Gauldie J (2007) Dendritic cells derived from murine colonic mucosa have unique functional and phenotypic characteristics. J Immunol
178: 7984&7993. ,,
Vray B, Camby I, Vercruysse V, Mijatovic T, Bovin NV, Riccardi-Castagnoli P, Kaltner H, Salmon I, Gabius HJ & Kiss R (2004) Up-regulation of galectin-3 and its ligands by Trypanosoma cruzi infection with modulation of adhesion and migration of murine dendritic cells. Glycobiol
14: 647&657. ,,
Wang J, Barke RA, Charboneau R & Roy S (2005) Morphine impairs host innate immune response and increases susceptibility to Streptococcus pneumoniae lung infection. J Immunol
174: 426&434. ,,
Won YS, Jeong ES, Park HJ, Lee CH, Nam KH, Kim HC, Park JI & Choi YK (2007) Up-regulation of galectin-3 by Corynebacterium kutscheri infection in the rat lung. Exp Anim
56: 85&91. ,,
Yamazaki S, Tsuyuki S, Akashiba H, Kamimura H, Kimura M, Kawashima T & Ueda K (1991) Immune response of Bifidobacterium-monoassociated mice. Bifidobacteria Microflora
10: 19&31.
Yang RY & Liu FT (2002) Galectins in cell growth and apoptosis. Cell Mol Life Sci
60: 267&276.
Z&&iga E, Gruppi A, Hirabayashi J, Kasai KI & Rabinovich GA (2001) Regulated expression and effect of galectin-1 on Trypanosoma cruzi-infected macrophages: modulation of microbicidal activity and survival. Infect Immun 69: 6804&6812. ,