AJP - GI AJP: Regulatory, Integrative and Comparative Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Gastrointest Liver Physiol 293: G780-G787, 2007. First published August 2, 2007; doi:10.1152/ajpgi.00010.2007
0193-1857/07 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
293/4/G780    most recent
00010.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Meng, D.
Right arrow Articles by Nanthakumar, N. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Meng, D.
Right arrow Articles by Nanthakumar, N. N.

MUCOSAL BIOLOGY

Bacterial symbionts induce a FUT2-dependent fucosylated niche on colonic epithelium via ERK and JNK signaling

Di Meng,1 David S. Newburg,1 Cheryl Young,1 Amy Baker,1 Susan L. Tonkonogy,2 R. Balfour Sartor,3 W. Allan Walker,1 and N. Nanda Nanthakumar1

1Developmental Gastroenterology Laboratory, Pediatric Gastroenterology and Nutrition Unit, Department of Pediatrics, Massachusetts General Hospital for Children and Harvard Clinical Nutrition Research Center, Harvard Medical School, Boston, Massachusetts; 2Department of Microbiology, Pathology and Parasitology, North Carolina State University College of Veterinary Medicine, Raleigh, North Carolina; and 3Center for Gastrointestinal Biology and Disease, Division of Gastroenterology and Hepatology, University of North Carolina, Chapel Hill, North Carolina

Submitted 5 January 2007 ; accepted in final form 29 July 2007


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
The intestinal epithelium of the adult gut supports a complex, dynamic microbial ecosystem and expresses highly fucosylated glycans on its surface. Uncolonized gut contains little fucosylated glycan. The transition toward adult colonization, such as during recovery from germ-free status or from antibiotic treatment, increased expression of fucosylated epitopes in the colonic epithelium. This increase in fucosylation is accompanied by induction of fut2 mRNA expression and {alpha}1,2/3-fucosyltransferase activity. Colonization stimulates ERK and JNK signal transduction pathways, resulting in activation of transcription factors ATF2 and c-Jun, respectively. This increases transcription of fut2 mRNA and expression of {alpha}1,2/3-fucosyltransferase activity, resulting in a highly fucosylated intestinal mucosa characteristic of the adult mammalian gut. Blocking the ERK and JNK signaling cascade inhibits the ability of colonization to induce elevated fut2 mRNA and fucosyltransferase activity in the mature colon. Thus pioneer-mutualist symbiotic bacteria may utilize the ERK and JNK signaling cascade to induce the high degree of fucosylation characteristic of adult mammalian colon, and we speculate that this fucosylation facilitates colonization by adult microbiota.

bacterial colonization; fucosylation

THE ADULT HUMAN GASTROINTESTINAL tract is colonized by a community of more than 450 interdependent species of bacteria that are the major components of a stable but dynamic microbial ecosystem known as the gut microbiota (10, 11, 15, 23). These symbiotic bacteria had long been classified as commensals, i.e., organisms that obtain food, protection, and other benefits without damaging or benefiting another (the human host), but they are now becoming recognized as part of a mutualist relationship in which both the organisms and humans benefit from the symbiosis (7, 20, 22). Mutualist association requires reciprocal communication, exemplified by the interactions of legumes with their nitrogen-fixing bacteria (2, 25, 27). Mutualist bacteria induce species-specific signals in legume roots that stimulate production of Nod receptors on the cell surface that are bound by specific bacterial Nod ligands. This accounts for close pairing of each legume with a specific mutualist for development of root nodules (14, 25, 27). We have identified an analogous mechanism whereby the microbiota of the mature mammalian gut may initially communicate with the gut epithelium at the genesis of colonization. One or more pioneer bacterial species induce expression of fucosylated cell surface glycans by mammalian gut epithelium; these induced glycans may bind additional fucose-adherent bacteria, directing the succession of microbial colonization that results in a normal adult intestinal ecosystem.

Mammals are born with an aseptic gut where mucosa is relatively high in sialic acid-containing cell-surface glycans and relatively low in fucosylated glycans; it is rapidly colonized by a suckling microbial ecosystem (9, 15, 18). At weaning of the mucosa, the ratio of fucosylated to sialylated glycans increases abruptly as the microbial ecosystem rapidly transforms to the adult pattern of microbiota (4, 5, 18, 19, 24). Fucosylated glycans are characteristic of the cell-surface intestinal mucosa in adult mouse and human, and some mutualist and pathogenic bacteria bind specifically to these fucosylated glycans (4, 8, 12, 18, 21). Because the changes in gut glycan expression and microbiota composition normally occur simultaneously, it was not obvious which event was primary.

Germ-free mice were used to dissociate the endogenous control of intestinal mucosal ontogeny from any exogenous control exerted by bacterial signaling. The ontogeny of fucosylated glycans on the mucosal surface is a direct expression of changes in expression of gut fucosyltransferase activity (3, 4, 16, 18, 26). Transition to adult patterns of fucosyltransferase gene transcription, expression of fucosyltransferase activity, and adult patterns of fucosylated glycans in the mucosa occur in conventional mice at weaning, whereas in germ-free mice they do not occur at weaning but rather only upon exposure to typical adult gut microbiota (4, 16, 18). We hypothesized that the presence of microbiota triggers a transmembrane signaling cascade in colonocytes that induces expression of fucosyltransferases, the enzymes that synthesize fucosylated cell-surface glycans essential for establishing the adult-pattern mutualist microbiota. A transspecies, transmembrane, transcellular mechanism that permits mutualist bacteria to cause host mucosal cells to express fucosylated glycans was discovered.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Animals were treated and monitored according to an approved animal protocol of the research Animal Care Committees of the Massachusetts General Hospital and North Carolina State University, as well as the guidelines published by the American Physiological Society.

Germ-free mice. Three groups of BALB/c mice (conventionally colonized, germ free, and ex-germ free) were housed with a 12-h light/dark cycle and had access to mouse chow and water ad libitum at North Carolina State University College of Veterinary Medicine (Raleigh, NC). Conventionally colonized mice were housed under conventional conditions and euthanized at the ages of 6 and 8 wk. Germ-free mice of the same strain were maintained in a germ-free environment until immediately before being euthanized at 6 and 8 wk of age. Ex-germ-free mice were produced by removing germ-free mice from their germ-free environment at the ages of 4 and 6 wk, inoculating them with a slurry of fresh fecal and cecal contents from age-matched conventional control mice (contents of 1 conventional mouse for inoculation of 5 germ-free mice) (4, 18) directly through both orogastric intubation and addition to their drinking water, and keeping them in the same cage as conventional mice. The ex-germ-free mice were euthanized 2 wk later along with the age-matched germ-free and conventional mice. All animals were euthanized between 12 PM and 3 PM to avoid circadian influences. Data obtained from both 6- and 8-wk-old mice were almost identical; data from 8-wk-old mice are shown in Fig. 1A. Similar results were observed in Black Swiss mice (not shown).


Figure 1
View larger version (10K):
[in this window]
[in a new window]

  		Fig. 1. Regulation of colonic fucosylation in bacteria-depleted (BD) and repleted (XBD) mice. A: adult germ-free (GF) mice (Balb/C) express significantly less {alpha}1,2/3-fucosyltransferase activity than conventional (CONV) mice (*P < 0.001), but activity rises to control levels upon colonization [i.e., ex-germ-free mice (XGF)]. B: BD mice express significantly less {alpha}1,2/3-fucosyltransferase activity than CONV mice (*P < 0.001), but activity returns to control levels upon repletion of microbiota (XBD). C: following depletion of luminal bacteria, there is a precipitous decline in levels of fut2 mRNA (*P < 0.0001) that is reversed by recolonization. The mRNA for fut1and the housekeeping gene GAPDH remained low and unchanged in these mice (data not shown). All measurements were made at 8 wk of age. This indicates that fucosylation of the adult intestinal mucosa is dependent upon bacterial colonization.

 
Luminal bacteria-depleted mice. Timed-pregnant Black Swiss mice were purchased from Taconic Farms (Germantown, NY) at 16–18 days gestation, housed individually in opaque polystyrene cages with a 12:12-h light/dark cycle in the animal facility at Massachusetts General Hospital, and fed mouse chow and water ad libitum. On the due date, cages were checked every 4 h for the presence of pups. The date of birth of the pups was designated as day 0. On the following day (day 1), litters were reduced to nine pups per dam.

Antibiotic and inhibitor treatment. Three groups of 4-wk-old Black Swiss mice consisting of five pups each were treated as follows. Bacteria-depleted and bacteria-repleted groups were fed an antibiotic cocktail in their drinking water for 2 wk whereas the conventional control group received untreated water. At 6 wk of age, the bacteria-repleted group received cecal/colonic bacteria from age-matched conventionally reared mice in their drinking water, as described previously (4, 18). Animals were euthanized at 8 wk of age, and their colons were harvested for analysis.

For experiments involving inhibitors of ERK and JNK, each of the three groups (conventional, bacteria-depleted, and bacteria-repleted) consisted of 20 mice. Every 5 mice of each group received the following in their drinking water: PD98059, an ERK inhibitor at 4 µg·g body wt–1·day–1; SP600125, a JNK inhibitor at 20 µg·g body wt–1·day–1; both PD98059 and SP600125 (ERK + JNK inhibitors); and untreated water containing neither inhibitor. Animals were euthanized at 8 wk of age, and their colons were harvested for analysis (1, 17). Average water consumption was measured for each group of mice to calculate the appropriate concentrations of inhibitors in their water.

Adult mice (Black Swiss, C3H/OuJ, BALB/c, and C57/B6) were depleted of luminal bacteria by consuming the antibiotic cocktail (100 µl antibiotic cocktail·mouse–1·day–1) added to their drinking water for 2 wk, from age 4 wk through 6 wk, then commensal bacteria were introduced for 2 wk whereupon the mice were euthanized and the fucosyltransferase activity and fut2 mRNA levels of each colon were measured.

Antibiotic cocktail. Kanamycin (8 mg/ml), Gentamicin (0.7 mg/ml), Colistin (34,000 U/ml), Metronidazole (4.3 mg/ml), and Vancomycin (0.9 mg/ml) comprised the antibiotic cocktail. After the antibiotic cocktail was introduced, fresh fecal samples were collected from mice every day and assayed for the presence of bacteria in 5 different culture media, including both aerobic and anaerobic conditions, as described by Julia et al. (13). Concentrations of antibiotics in the water were calculated based on the average water consumed by age group.

Microsomal fraction. Colons were removed and thoroughly flushed with ice-cold 0.9% NaCl, then placed on a glass plate sitting on ice; the mucosa was harvested by scraping with a microscope glass slide. A 10% mucosal homogenate in 0.1 M Tris·HCl buffer (pH 7.4) was prepared over ice and centrifuged at 1,000 g for 15 min at 4°C to remove nuclei and cellular debris. The supernatant was centrifuged at 105,000 g for 1 h at 4°C in a Beckman L8-80M Ultracentrifuge with a SW41 Ti rotor. The resulting microsomal pellets were resuspended in the homogenization (Tris) buffer plus 1% Triton X-100 and 1% glycerol, aliquoted, frozen, and stored at –80°C or used immediately for the fucosyltransferase enzyme assay, as described previously (18).

Protein determination. Protein was determined by the bicinchoninic acid protein assay (Pierce, Rockford, IL) modified for 96-well microtiter plates according to the manufacturer's protocol. To each protein sample of 10 µl, 200 µl of working reagent was added followed by incubation at 37°C for 30 min. Absorbance at 560 nm was measured on a microtiter reader (BT 2000 Microkinetics Reader Spectrophotometer, Fisher Biotech, Pittsburgh, PA). The concentration of each protein sample was calculated using a standard curve produced with bovine serum albumin (BSA) (9, 18).

Fucosyltransferase assay. Enzyme activity for {alpha}1,2/3-fucosyltransferase was measured using asialofetuin type-1 as the exogenous acceptor, as previously described (18). Briefly, the reaction mixture (total volume of 0.1 ml) contained 5 mM ATP, 10 mM L-fucose, 25 mM MnCl2, 50 mM NaCl2, 50 mM HEPES (pH 7.0), 60 nmol GDP-fucose, 10 nCi GDP-[14C] fucose, and an aliquot of the microsomal fraction from the colon of each pup to contain 50 µg protein. The concentration of GDP-fucose was sufficient to maintain the enzyme activity at Vmax throughout the assay. The assay mixture was incubated for 30 min at 37°C, and the product filtered through a 24-mm glass microfiber grade B filter. The radioactivity of the filtered material was determined by scintillation counting of beta emissions.

Epithelial cell isolation. Mice were euthanized, and each colon was removed and flushed for ~5 min with 37°C Hanks' buffer containing 30 mM EDTA, as described previously (6). The colon was everted gently and secured to a glass rod with a loop of thread at the upper end of the segment of the gut. Each colon was transferred into a plastic flask containing 15 ml of warm (37°C) EDTA in Hanks' buffer and shaken for 10 min, and the solution was collected and replaced three times. The cells were centrifuged at 3,000 rpm (4°C) for 10 min, and the pellet was washed with 20 ml of ice-cold Hanks' buffer twice and dissolved in cell lysis buffer (9, 18).

Lectin staining with UEA-1. Expression of fucosyl glycoconjugates on the mucosal surface was measured on frozen tissue sections using FITC-conjugated Ulex europaeus agglutinin-1 (UEA-1) (Vector Laboratories, Burlingame, CA). The middle 1 cm of the colon was fixed for 4 h at 4°C (9, 18) in 4% paraformaldehyde, washed in ice-cold PBS containing 30% sucrose overnight at 4°C, and embedded in optimal cutting temperature compound. Frozen sections (6–7 µm thick) were blocked with PBS containing 2% BSA and then stained with labeled lectin for 1 h (10 µg/ml). Sections were then washed three times in ice-cold PBS, mounted using Anti-Fade (Vector Laboratories), and analyzed by confocal microscopy (9).

SDS PAGE analysis. Protein samples (30 or 50 µg) mixed with SDS sample buffer were loaded on 10–20% SDS Tris·HCl ready gels and transferred to Immun-Blot polyvinylidene difluoride membranes (Bio-Rad, Hercules, PA). Subsequently, membranes were blocked in blot A, 5% (wt/vol) Carnation nonfat dry milk (Nestlé, Solon, OH) in Tris-buffered saline supplemented with 0.05% Tween 20 at room temperature for 1 h, then incubated overnight at 4°C with antibody. Anti-ERK1/2, anti-p38, anti-Toll-like receptor-4, anti-E-cadherin, anti-I{kappa}B-{alpha}, and anti-I{kappa}B-beta were from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-p-ERK1/2, anti-p-p38, anti-p-ATF2, anti-ATF2, anti-p-c-Jun73, anti-c-Jun73, anti-p-JNK, anti-JNK, anti-p-MKK4, anti-MKK4, anti-p-MEK1/2, and anti-MEK1/2 were from Cell Signaling Technology (Beverly, MA). Blots were washed 3 times for 10 min each in blot A then incubated with horseradish peroxidase-conjugated secondary antibody for 1 h at room temperature. After two 10-min washes in blot A and three 10-min washes in Tris-buffered saline were performed, blots were developed via an enhanced chemiluminescence system (Supersignal; Pierce), as described previously (6, 9).

Quantitative RT-PCR. The RNA RNeasy Mini kit (Quigen, Valencia, CA) allowed extraction of total RNA from homogenized tissue. RNA was reverse transcribed with random hexamers using a GeneAmp RNA PCR kit (Applied Biosystems, Foster City, CA), and the cDNA was amplified using iQ SYBR Green Supermix (Bio-Rad) and 5 µM of each primer specified in the next paragraph. GAPDH primers were amplified in all samples. Duplicate cDNA samples were amplified 40 cycles for fut2 and 42 cycles for fut1 for both 1 min at 95°C and 1 min at 72°C. The threshold cycle (CT) was the cycle number at which fluorescence of the amplified product crossed a specified threshold value during exponential amplification. Mean CT values of each transcript were normalized by subtracting the mean CT value for the GAPDH transcript of that sample. The change in normalized transcript level was expressed relative to the control sample with a change of n in CT representing a 2n -fold difference, as described previously (9, 18).

Primer sequences. Primer sequences used for RT-PCR are: GAPDH, sense, 5'-CCTGCACCACCAACTGCTTA-3', and antisense, 5'-ATGACCTTGCCCACAGCCT-3'; FUT-2, forward, 5'-AGTCTTCGTGGTTACAAGCAAC-3', and reverse, 5'-TGGCTGGTGAGCCCTCAATA-3'; FUT-1, forward, 5'-CAGCTCTGCCTGACATTTCTG-3', and reverse, 5'-AGCAGGTGATAGTCTGAACACA-3'.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Glycans are fucosylated by {alpha}1,2-, {alpha}1,3-, {alpha}1,4-, and {alpha}1,6-fucosyltransferases; {alpha}1,2- and {alpha}1,3-fucosyltransferases are principal contributors to the total fucosyltransferase activity of adult colonic epithelium. We used a single assay for total {alpha}1,2/3-fucosyltransferase activities to explore various conditions and mechanisms that induce changes in fucosyltransferase activity in the colonic mucosa (13) and identified changes in expression of specific fucosyltransferases by measuring induction of specific fucosyltransferase mRNAs. The first objective was to compare a new murine model of bacterial induction of fucosylated glycan expression (bacteria depleted) with the established germ-free model (13, 14). {alpha}1,2/3-fucosyltransferase was measured in germ-free mice before and after colonization by adult microflora and in mice of the same strain that had been conventionally colonized since birth. Levels of {alpha}1,2/3-fucosyltransferase activity activity were significantly lower in the germ-free mouse gut, especially in the colon (the site of highest bacterial colonization), than in the same strain of conventionally colonized mice; when germ-free mice were inoculated with adult microbiota, intestinal fucosyltransferase rose to that of conventionally colonized mice by 2 wk (Fig. 1A), consistent with previous reports (13, 14). Conversely, when conventional mice were treated with a cocktail of broad-spectrum antibiotics that depleted their microbiota (bacteria-depleted mice), the levels of {alpha}1,2/3-fucosyltransferase decreased significantly within 2 wk (Fig. 1B). However, repletion of bacteria resulted in recovery of the {alpha}1,2/3-fucosyltransferase activity. Although different strains of mice express different basal levels of {alpha}1,2/3-fucosyltransferase activities in their intestinal mucosa, the significant reduction upon bacterial depletion seen in Fig. 1B occurred consistently in all mouse strains tested [Black Swiss, C3H, C57/B6, and BALB/c (not shown)]. In both the germ-free and bacteria-depleted models, these changes in {alpha}1,2/3-fucosyltransferase activity mirrored changes in transcription of fut2 mRNA (Fig. 1C) but not fut1 (the other {alpha}1,2-fucosyltransferase) mRNA or fut11 (one of the {alpha}1,3-fucosyltransferases) mRNA (data not shown), as measured by quantitative RT-PCR. Thus the changes in {alpha}1,2/3-fucosyltransferase activity is primarily mediated by changes in fut2 mRNA levels. The concomitant decreases in fut2 mRNA and {alpha}1,2/3-fucosyltransferase activity were also accompanied by corresponding decreases in expression of fucosylated glycans on the mucosal surface measured as reduced binding by Ulex europaeus I, a lectin that strongly binds {alpha}1,2-linked fucose epitopes (data not shown). These differences were most pronounced in the colon, where the {alpha}1,2/3-fucosyltransferase activities are highest in colonized mice (13, 15) and where staining of fucosylated glycans is greatest on the epithelial cell surface and mucus of goblet cells. Thus the data from both the germ-free model and the bacteria-depleted model concur that elevated adult levels of fucosyltransferase activity in mammalian gut require the presence of luminal microbiota.

To investigate specific signaling pathways that might mediate bacteria-induced fut2 mRNA and {alpha}1,2/3-fucosyltransferase expression in intestinal epithelium, we measured levels of intermediates of intestinal signaling pathways known to be inducible by bacteria (28–30). Several signaling intermediates had levels that were high under conditions of high fucosyltransferase expression in the presence of microbiota (conventionally colonized, ex-germ free, and bacteria repleted) and low under conditions of low fucosyltransferase expression in the absence of microbiota (germ free and bacteria depleted). These are known mediators of bacteria-induced pathways (Fig. 2D) and were investigated as potential mediators of fucosyltransferase gene expression.


Figure 2
View larger version (42K):
[in this window]
[in a new window]

  		Fig. 2. Activation of signal transduction pathways in the colonic epithelium during colonization. A: phosphorylation of MEK1/2 kinase, ERK1/2 kinases, and ATF2 (nuclear transcription factor 2) indicates activation of the ERK signaling pathway (D) specifically during colonization of mature GF mice. B: phosphorylation of MKK4 kinase, JNK kinases, and transcription factor c-Jun indicate activation of the JNK signaling pathway (D) specifically during colonization. These two pathways display low basal activation in CONV and GF mice but strong transient activation in XGF mice that gradually decreases to that of CONV mice by 4 wk after recolonization (not shown). C: the levels of I{kappa}B-{alpha} and -beta protein were unchanged in GF, BD, and XBD mice, indicating that the NF{kappa}B pathway (D) is not activated by colonization. D: these data suggest activation of the ERK and JNK signal transduction pathways by newly colonizing microbes.

 
MEK1/2 and ERK1/2, representatives of the ERK signaling pathway, became phosphorylated upon colonization by microbiota (Figs. 2 and 3); their functional activation was confirmed in both germ-free and bacteria-depleted animal models (Figs. 2A and 3A) by the accompanying phosphorylation of ATF2, which is downstream from MEK1/2 and ERK1/2 activation (30, 31). Similarly, MKK4 and JNK, representatives of the JNK pathway, were phosphorylated upon colonization of germ-free mice (Fig. 2B) and bacteria-depleted mice (Fig. 3B) by normal symbiotic microbiota, and their functional activation was confirmed by increased phosphorylation of c-Jun, which is downstream from MKK4 and JNK activation. Mice devoid of microbiota did not produce these phosphorylated signaling intermediates, suggesting that this signal is induced by bacterial stimulation. Conventional mice with unperturbed stable microbiota also did not have elevated levels of phosphorylated ATF2 and c-Jun, consistent with this activation being a transient trigger in response to the initial presence of adult microbiota in the adult gut. Other bacteria-activated signaling pathways of the intestinal epithelium did not change in concert with these microbiota manipulations in a manner that would suggest involvement in these transient signals. For example, the NF-{kappa}B-proinflammatory signaling pathway (29, 33, 34), represented by I{kappa}B levels, was not activated by the presence or absence of microbiota (Fig. 2C). No evidence of phosphorylation of I{kappa}B{alpha}, I{kappa}Bbeta, or any change in the level of the p65 subunit of NF-{kappa}B was apparent by Western blot analysis. Furthermore, immunohistochemistry did not produce any evidence of nuclear translocation of NF-{kappa}B subunits (not shown).


Figure 3
View larger version (42K):
[in this window]
[in a new window]

  		Fig. 3. In CONV and BD mice, expression of both ERK and JNK signaling pathways were low but were activated upon reintroduction of microbiota in the XBD mice. A: activation of the ERK signal transduction pathway is prevented by the inhibitor of ERK activation, PD98059. B: activation of the JNK pathway is inhibited by SP600125, a specific inhibitor of JNK activation. These manipulations do not affect expression of E-cadherin, a gene irrelevant to these 2 signal transduction pathways.

 
However, the coincidence of activation of ERK and JNK signaling transducers with induction of fut2 mRNA transcription does not prove ERK and JNK are specifically required for fut2 mRNA induction. To address this issue, readily available specific inhibitors of ERK and JNK signaling pathways were administered to conventional, bacteria-depleted, and bacteria-repleted mice (24, 25). The presence of ERK inhibitor did not alter expression of high levels of epithelial {alpha}1,2/3-fucosyltransferase activity by conventionally colonized mice or expression of low levels of {alpha}1,2/3-fucosyltransferase activity by bacteria-depleted mice (data not shown). However, inhibition of ERK activation in bacteria-repleted mice blocked phosphorylation of ATF2 (Fig. 3A), recovery of fucosyltransferase activity to mature levels, and induction of fut2 mRNA but not fut1 mRNA (Fig. 4, A–C), demonstrating that induction of fucosyltransferase activity and fut2 mRNA upon bacterial repletion specifically requires ERK activation. Similarly, inhibition of JNK activation in bacteria-repleted mice blocked phosphorylation of JNK kinase and the phosphorylation of c-Jun transcription factor (Fig. 3B). Inhibition of the JNK pathway also significantly attenuated induction of epithelial {alpha}1,2/3-fucosyltransferase activity and fut2 mRNA but not fut1mRNA (Fig. 4, A–C). This observation demonstrates that the colonization-dependent induction fucosyltransferase activity and fut2 mRNA require activation of the JNK signaling pathway. When these two signal transduction pathways were inhibited simultaneously by the combined administration of ERK and JNK inhibitors, induction of fucosyltransferase activity (Fig. 4A) and fut2 mRNA (Fig. 4B) in response to initial colonization was completely suppressed, with values indistinguishable from those of bacteria-depleted mice. Thus colonization-dependent induction of fucosyltransferase activity and fut2 mRNA is dependent on both ERK and JNK activation in colonocytes.


Figure 4
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 4. ERK and JNK inhibitors attenuate recolonization-induced elevation of {alpha}1,2/3-fucosyltransferase activity and fut2mRNA in XBD animals. A: {alpha}1,2/3-fucosyltransferase recovery in XBD mice is attenuated by treatment with either PD98059 or SP600125 or both inhibitors. Each data point was pooled from 2 mice (n = 4). *P < 0.001 vs. the CONV group. **P < 0.05 vs. the CONV group. B: the induction of fut2 mRNA in XBD mice is attenuated by treatment with ERK inhibitor, JNK inhibitor, and especially both (n = 6). C: the level of fut1 mRNA remains low and unchanged by these treatments. D: these data indicate that both the ERK and JNK signal transduction pathways are essential to recolonization-induced activation of fut2 mRNA, perhaps mediated by an activating protein-1 transcriptional control element in the fut2 gene.

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
The data presented herein suggest a chain of events essential for initiating colonization of the mature gut. Small baseline inocula of pioneer bacterial species bind to a critical receptor on the mucosal surface. Upon activation, this receptor phosphorylates MEK1/2 and MKK4 kinases in the cytoplasm. Activated MEK1/2 kinase (p-MEK1/2) phosphorylates ERK1/2 kinase (p-ERK1/2); activated ERK kinase (p-ERK) induces phosphorylation of ATF2. Activated MKK4 kinase (p-MKK4) phosphorylates JNK kinase (p-JNK); activated JNK kinase phosphorylates c-Jun to p-c-Jun. These activated nuclear transcription factors, p-ATF2 and p-c-Jun, induce transcription of fut2 mRNA. This induction results in an increased translation of {alpha}1,2-fucosyltransferase (fucosyltransferase activity II) protein, increased {alpha}1,2-fucosyltransferase enzyme activity, and full expression of {alpha}1,2-fucosylated glycans on the mucosal cell surface. Highly fucosylated mucosa is associated with a fully colonized gut.

Many promoter/enhancer elements for induced gene expression during cell growth, differentiation, and apoptosis respond to external stimuli such as cytokines, growth factors, stress signals, or bacterial and viral infections (33, 35). Those elements controlling immediate early genes are rapidly but transiently induced directly by intracellular signaling cascades in response to extracellular effectors, especially bacteria (36). Many of these genes are induced by nuclear control elements such as AP1 and AP2. Many glycosyltransferases are also regulated by the nuclear control elements AP1 and AP2. For example, transcription of a galactosyl {alpha}2,6-sialyltransferase is under control of both AP1 and AP2 (37), and transcription of fut4 and fut1 are controlled by AP2 (38, 39). The promoter region of the fut2 gene contains three copies of the consensus sequence for AP1 binding (tgact) at –1,800, +50, and +3,320 bp from the translational start site. Transcription factors that bind to the AP1 control element are heterodimers of the Fos, Jun, and ATF family of proteins (35, 40, 41). In our studies, the concurrent phosphorylation of c-Jun and ATF2, upon colonization or recolonization of mature gut seen in Figs. 2 and 3, was coincident with the induction of fut2 mRNA but not fut1 mRNA; this suggests that induction of mammalian gut fucosylation by bacterial colonization could be mediated through an AP1 control element of fut2. Control of transcription by AP1 is sensitive to context, including cell type, stage of development, microenvironment, and the nature of the stimulus (35, 40). Thus it is reasonable to conclude that in mature intestinal epithelial cells that are not yet colonized with fucose-binding microbiota, fucose-binding bacteria could specifically induce fut2 mRNA via activation of its AP1 promoter (Fig. 5).


Figure 5
View larger version (56K):
[in this window]
[in a new window]

  		Fig. 5. Based on the data presented in previous figures, we hypothesize that during colonization, pioneer bacteria (1) initially bind and activate a putative receptor on the membrane that (2) stimulates intracellular ERK and JNK signaling pathways, which (3) activate c-Jun and ATF2 nuclear transcription factors (perhaps to form a potential AP1 promoter complex) that (4) induce fut2 mRNA transcription, which in turn (5) elevates fucosyltransferase activity in the Golgi, thereby (6) increasing expression of fucosylated glycans on the cell surface of the colonic mucosa that (7) facilitate colonization of the adult colon by fucose-binding mutualistic symbionts.

 
The resulting fut2 mRNA transcript, after transport from the nucleus, would be translated into the fucosyltransferase activity-II protein. This induction of {alpha}1,2-fucosyltransferase activity would account for the increased {alpha}1,2/3-fucosyltransferase activity that we observe upon colonization of the gut (Fig. 1) (13). When in the Golgi apparatus, fucosyltransferase activity-II transfers fucose from GDP-fucose to glycans. The resulting fucosylglycans would be transported to, and inserted into, the extracellular surface of the intestinal mucosal cell (in goblet cells, induced fucosylglycans would also include the highly fucosylated mucus apparent in Fig. 1D). This could promote gut colonization by fucose-binding bacteria. We postulate that this initial wave of colonization by pioneer fucose-binding bacteria is the critical first step of succession of gut microbiota that evolves into the complex interdependent community that colonizes the adult gut. In this regard, such bacteria might resemble Bacteroides thetaiotaomicron, a major component of the adult gut flora that utilizes fucose and elicits induction of fucosyltransferase expression in germ-free mouse gut (5, 14, 19) or Bacteroides fragilis (18, 42), another fucose-utilizing bacteria found in human and mouse gut that we find stimulates induction of fucosyltransferase activity in bacteria-depleted mice (Meng D and Nanthakumar N, unpublished observations). This inducible signaling system could also participate in the shift in microbiota during weaning.

Mutualism is increasingly being recognized as a common, and perhaps universal, theme in biology. Establishing successful symbiosis in legume/rhizobia symbiont pairs in root nodules requires reciprocal communication between the plant root and the bacterium. Likewise, cross talk between microbiota and mammalian gut induces changes in gene expression and function in both bacteria and intestine. For example, mutantBacteroides fragilis that are unable to metabolize fucose are less able to colonize mammalian gut (18), implying that exogenous fucose from the mammalian cell surface plays an important role in intracellular signaling and metabolism within the bacterial symbionts. Our observation that bacteria stimulate intracellular signaling in mammalian intestinal mucosa, inducing expression of specific cell-surface glycans by the gut, represents the complementary component of reciprocal communication (cross talk) between mammalian and bacterial mutualists.

The use of a common pathway for signaling the presence of mutualists and for immunosurveillance by the innate mucosal immune system is consistent with the known active surveillance that characterizes the mammalian relationship with its microbiota (6, 29). Indeed, introduction of bacteria early in life is thought to produce stimulation that is necessary for the ontogeny of a balanced mucosal immune system (46). This activation of mucosal immunity may help explain the phenomenon whereby oral consumption of benign microbes (probiotics) protect against pathogens (46, 47). Consistent with the essential nature of mutualist communication, mutant mice unable to synthesize GDP-fucose (substrates for fucosyltransferase enzymes) exhibit colitis and failure to thrive after their third postnatal week (23).

Bacterial induction of fucosyltransferase expression is mediated by activation of ERK and JNK signal transduction pathways in intestinal mucosa. This pathway is recognized as central to many of the innate responses to pathogens by the intestinal mucosa but has not been previously recognized as playing a role in the communication (cross talk) between mutualists. We suggest that such communication by normal mixed adult microbiota is the essential transspecies transcellular communication that initiates gut colonization by adult microbiota and that this up-regulation of fucosylation of the mucosal surface promotes colonization by fucose-binding bacteria that may act as the pioneer species for succession of gut microbiota to the adult pattern. Thus intestinal exposure to adult microbiota is an essential event in the ontogeny of normal adult intestinal cell surface fucosylated glycans, and expression of these fucosylated glycans is essential to normal colonization by an adult pattern of microbiota.

The high fucosylation that we find to be characteristic of adult human and murine intestinal mucosa is mediated by fut2 mRNA; its binding to UEA lectin confirms that its structure contains {alpha}1,2 fucose moieties. The expression of these {alpha}1,2-fucosylated glycans in gut requires the continuous presence of the adult microbiota; thus treatment with antibiotics reverts gut mucosa to a state associated with lack of adult colonization, and this state may be more vulnerable to external insult. The recovery to the recolonized state requires ERK and JNK signaling, an integral component of reciprocal communication between mutualistic symbionts and intestinal epithelium.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported by National Institutes of Health Grants HD12437, HD31852, HD13021, DK33506, DK34987, DK40561, DK53347, RR018603; the Crohn's and Colitis Foundation; Wyeth Nutritionals International; and Fogarty Foundation/National Institute of Child Health and Human Development Training Grant D43 TW1265 to D. Meng.


   ACKNOWLEDGMENTS
 
We thank Bobby Cherayil for critical reading of the manuscript and members of the Developmental Gastroenterology Laboratory for helpful discussion and suggestions during the course of the work. The technical expertise of Wei-Shu Zhu and excellent preparation of the manuscript by Kathryn Newburg are greatly appreciated.


   FOOTNOTES
 

Address for reprint requests and other correspondence: N. Nanthakumar, Developmental Gastroenterology Laboratory, Massachusetts General Hospital-East, 114 16th St., Rm 3650, Charlestown, MA 02129 (e-mail: nanthaku{at}helix.mgh.harvard.edu)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Assi K, Pillai R, Gomez-Munoz A, Owen D, Salh B. The specific JNK inhibitor SP600125 targets tumor necrosis factor-alpha production and epithelial cell apoptosis in acute murine colitis. Immunology 118: 112–121, 2006.[CrossRef][Web of Science][Medline]
  2. Bastida A, Latorre M, Garcia-Junceda E. In vivo chaperone-assisted folding of alpha-1,6-fucosyltransferase from Rhizobium sp. Chembiochem 4: 531–533, 2003.[CrossRef][Web of Science][Medline]
  3. Becker DJ, Lowe JB. Fucose: biosynthesis and biological function in mammals. Glycobiology 13: 41R–53R, 2003.[Abstract/Free Full Text]
  4. Bry L, Falk PG, Midtvedt T, Gordon JI. A model of host-microbial interactions in an open mammalian ecosystem. Science 273: 1380–1383, 1996.[Abstract]
  5. Chu SH, Walker WA. Developmental changes in the activities of sialyl- and fucosyltransferases in rat small intestine. Biochim Biophys Acta 883: 496–500, 1986.[Medline]
  6. Claud EC, Lu L, Anton PM, Savidge T, Walker WA, Cherayil BJ. Developmentally regulated I{kappa}B expression in intestinal epithelium and susceptibility to flagellin-induced inflammation. Proc Natl Acad Sci USA 101: 7404–7408, 2004.[Abstract/Free Full Text]
  7. Comstock LE, Coyne MJ. Bacteroides thetaiotaomicron: a dynamic, niche-adapted human symbiont. Bioessays 25: 926–929, 2003.[CrossRef][Web of Science][Medline]
  8. Coyne MJ, Reinap B, Lee MM, Comstock LE. Human symbionts use a host-like pathway for surface fucosylation. Science 307: 1778–1781, 2005.[Abstract/Free Full Text]
  9. Dai D, Nanthakumar NN, Savidge TC, Newburg DS, Walker WA. Region-specific ontogeny of {alpha}2,6-sialyltransferase during normal and cortisone-induced maturation in mouse intestine. Am J Physiol Gastrointest Liver Physiol 282: G480–G490, 2002.[Abstract/Free Full Text]
  10. Eckburg PB, Bik EM, Bernstein CN, Purdom E, Dethlefsen L, Sargent M, Gill SR, Nelson KE, Relman DA. Diversity of the human intestinal microbial flora. Science 308: 1635–1638, 2005.[Abstract/Free Full Text]
  11. Gaskins HR. Microbial modulation of innate defense: goblet cells and the intestinal mucus layer. In: Gastrointestinal Microbes and Host Interactions, edited by Mackie RI, White BA, and Isaacson RE. New York: Chapman & Hall, 1997, p. 537–587.
  12. Hooper LV, Xu J, Falk PG, Midtvedt T, Gordon JI. A molecular sensor that allows a gut commensal to control its nutrient foundation in a competitive ecosystem. Proc Natl Acad Sci USA 96: 9833–9838, 1999.[Abstract/Free Full Text]
  13. Julia V, McSorley SS, Malherbe L, Breittmayer JP, Girard-Pipau F, Beck A, Glaichenhaus N. Priming by microbial antigens from the intestinal flora determines the ability of CD4+ T cells to rapidly secrete IL-4 in BALB/c mice infected with Leishmania major. J Immunol 165: 5637–5645, 2000.[Abstract/Free Full Text]
  14. Kalsi G, Etzler ME. Localization of a Nod factor-binding protein in legume roots and factors influencing its distribution and expression. Plant Physiol 124: 1039–1048, 2000.[Abstract/Free Full Text]
  15. Ley RE, Peterson DA, Gordon JI. Ecological and evolutionary forces shaping microbial diversity in the human intestine. Cell 124: 837–848, 2006.[CrossRef][Web of Science][Medline]
  16. Lin B, Hayashi Y, Saito M, Sakakibara Y, Yanagisawa M, Iwamori M. GDP-fucose: beta-galactoside {alpha}1,2-fucosyltransferase, MFUT-II, and not MFUT-I or -III, is induced in a restricted region of the digestive tract of germ-free mice by host-microbe interactions and cycloheximide. Biochim Biophys Acta 1487: 275–285, 2000.[Medline]
  17. Marandi S, De Keyser N, Stilmant C, Saliez A, Sokal EM, Guiot Y, Buts JP. Ontogeny of MAP kinases in rat small intestine: premature stimulation by insulin of BBM hydrolases is regulated by ERKs but not by p38 MAP kinase. Pediatr Res 52: 180–188, 2002.[Medline]
  18. Nanthakumar NN, Dai D, Newburg DS, Walker AW. The role of indigenous microflora in the development of murine intestinal fucosyl- and sialyltransferases. FASEB J 17: 44–46, 2003.[Abstract/Free Full Text]
  19. Pang KY, Newman AP, Udall JN, Walker WA. Development of gastrointestinal mucosal barrier. VII In utero maturation of microvillus surface by cortisone. Am J Physiol Gastrointest Liver Physiol 249: G85–G91, 1985.[Abstract/Free Full Text]
  20. Rakoff-Nahoum S, Paglino J, Eslami-Varzaneh F, Edberg S, Medzhitov R. Recognition of commensal microflora by toll-like receptors is required for intestinal homeostasis. Cell 118: 229–241, 2004.[CrossRef][Web of Science][Medline]
  21. Ruiz-Palacios GM, Cervantes LE, Ramos P, Chavez-Munguia B, Newburg DS. Campylobacter jejuni binds intestinal H(O) antigen (Fuc{alpha}1, 2Gal beta1, 4GlcNAc), and fucosyloligosaccharides of human milk inhibit its binding and infection. J Biol Chem 278: 14112–14120, 2003.[Abstract/Free Full Text]
  22. Salyers AA. Energy sources of major intestinal fermentative anaerobes. Am J Clin Nutr 32: 158–163, 1979.[Free Full Text]
  23. Savage DC. Microbial ecology of the gastrointestinal tract. Annu Rev Microbiol 31: 107–133, 1977.[CrossRef][Web of Science][Medline]
  24. Shub MD, Pang KY, Swann DA, Walker WA. Age-related changes in chemical composition and physical properties of mucus glycoproteins from rat small intestine. Biochem J 215: 405–411, 1983.[Web of Science][Medline]
  25. Smit P, Raedts J, Portyanko V, Debelle F, Gough C, Bisseling T, Geurts R. NSP1 of the GRAS protein family is essential for rhizobial Nod factor-induced transcription. Science 308: 1789–1791, 2005.[Abstract/Free Full Text]
  26. Smith PL, Myers JT, Rogers CE, Zhou L, Petryniak B, Becker DJ, Homeister JW, Lowe JB. Conditional control of selectin ligand expression and global fucosylation events in mice with a targeted mutation at the FX locus. J Cell Biol 158: 801–815, 2002.[Abstract/Free Full Text]
  27. Vassileva VN, Kouchi H, Ridge RW. Microtubule dynamics in living root hairs: transient slowing by lipochitin oligosaccharide nodulation signals. Plant Cell 17: 1777–1787, 2005.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
293/4/G780    most recent
00010.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Meng, D.
Right arrow Articles by Nanthakumar, N. N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Meng, D.
Right arrow Articles by Nanthakumar, N. N.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2007 by the American Physiological Society.