Published online before print January 25, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
responses in the large intestine and increased severity of induced colitis
,1
Dalhousie Inflammation Group, Departments of Microbiology and Immunology, and
* Pathology, Dalhousie University, Halifax, Nova Scotia, Canada
1 Correspondence: Dalhousie University, Dept. of Microbiology & Immunology, Sir Charles Tupper Medical Building, Room 7C, 5850 College St., Halifax, Nova Scotia, Canada, B3H1X5. E-mail: jean.marshall{at}dal.ca
ABSTRACT
Age-associated changes in immune function and their implications for intestinal inflammation are poorly understood. Defects in innate immunity have been shown to enhance intestinal inflammation and have been demonstrated upon aging. This study aimed to determine the consequences of aging in the presence and absence of TLR2 on intestinal inflammation. Young and aged (>60 weeks), control C57Bl/6 and TLR2-deficient (TLR2–/–) mice were examined. The cecum and mid-colon were analyzed for tissue damage, cytokine profiles, and trefoil factor 3 (TFF3) expression at baseline or after 5 days of treatment with dextran sodium sulfate (DSS) and 5 or 13 days recovery. Untreated, aged TLR2–/– mice had no significant intestinal inflammation but had reduced colonic IFN-
and IL-10 compared with younger mice. Aged TLR2–/– mice developed more severe colitis than other groups, as indicated by histological examination and overall weight loss. There were significant increases in colonic IFN- following DSS treatment in young but not in aged mice. TFF3 was substantially reduced in the cecum and increased in the colon of aged but not younger TLR2–/– mice following DSS treatment. These results demonstrate that even upon aging, TLR2-deficient animals did not develop intestinal disease. However, they failed to respond appropriately to an inflammatory insult, and the consequences of this were most severe in aged animals. Cytokine and TFF3 changes associated with aging may contribute to more severe intestinal inflammation.
Key Words: mucosal immunology • rodent • inflammation • Toll-like receptors • Trefoil factors • inflammatory bowel disease
INTRODUCTION
The inflammatory bowel diseases (IBD) are considered the result of a dysregulated mucosal immune response to luminal antigens in genetically predisposed individuals [1 2 3 ]. A proportion of elderly subjects develop late-onset IBD [4 , 5 ]. Defects in innate immune function, such as polymorphisms in the nucleotide-binding oligomerization domain 2 (NOD2) and TLR4, have been associated with IBD [6 7 8 ], which in the elderly, could differ in etiology and pathogenesis from that studied in younger individuals. In the current study, we examined whether aging predisposes to or increases severity of colitis.
The innate immune system provides the body with its first line of defense against invading pathogens. The interface between the mucosal immune system and luminal microflora is highly balanced in the gastrointestinal tract. TLRs are a family of pattern-recognition receptors (PRRs) capable of recognizing conserved molecular products such as bacterial-associated LPS and bacterial lipopeptides [9
, 10
]. The cooperative actions of resident TLR-expressing cells in the intestine, including intestinal epithelial cells, dendritic cells, macrophages, and lymphocytes, provide the host with a repertoire of defense mechanisms against invasive pathogens present in the luminal microflora [11
, 12
]. Recognition of bacterial and fungal products through TLRs ultimately results in the activation of MAPKs as well as NF-
B and transcription of genes encoding various proinflammatory cytokines [9
, 10
]. This occurs via mechanisms, which are dependent on several intracellular adaptor molecules, such as MyD88. Several studies have demonstrated a protective role for MyD88-dependent TLR signaling in the gut [11
, 13
14
15
]. When TLR4-deficient or MyD88-deficient mice were administered dextran sodium sulfate (DSS), it led to increased epithelial injury, accompanied by colonic bleeding and increased mortality [11
, 13
14
15
]. These observations suggest that reduced, overall TLR signaling exposes the intestinal tract to damage from resident bacterial flora. Studies by Rakoff-Nahoum et al. [13
, 16
] suggest that the MyD88-dependent interaction between the commensal microflora and host PRRs on immune effector cells is crucial for resistance to DSS-induced injury and homeostatic processes. In TLR2-deficient animals, more severe responses to DSS colitis have been observed [13
, 17
]. It has recently been shown that epithelial barrier integrity is dependent on TLR2 stimulation, affecting tight junction-associated proteins. A deficiency in TLR2 renders mice more susceptible to intestinal mucosal injury as a result of early tight-junction disruption accompanied by PI-3K/Akt-mediated cell survival in a MyD88-dependent manner; however, cytokine responses and key molecules involved in the process of intestinal epithelial restitution were not examined [17
].
The elderly are more susceptible to certain types of bacterial infection, which can be partially attributed to their decline in aspects of immune function over time [18 19 20 21 22 ]. Expression of multiple TLRs is decreased on splenic and peritoneal macrophages isolated from aged mice, along with their ability to produce various cytokines in response to their TLR ligands [23 24 25 ]. Human peripheral blood monocyte TLR function is reduced with age with significant defects in TLR1/2-induced TNF and IL-6 production by aged monocytes compared with controls [26 ]. We hypothesized that an age-related decline in innate immune function in the large intestine might dysregulate mucosal immune responses, increasing susceptibility to mucosal injury. Such age-related changes could be of even greater importance in the absence of specific TLR function.
The consequences of mucosal injury are highly dependent on the process of restitution of the epithelium. Trefoil factors (TFFs) are a family of small, protease-resistant peptides that share one or more common trefoil motifs and facilitate intestinal epithelial cell restitution and repair [27 ]. TFF3, formerly known as intestinal TFF, is primarily produced by goblet cells [27 , 28 ]. TFF3-deficient mice have impaired epithelial cell migration and repair in the colon, as well as an increased susceptibility to DSS-induced injury [29 ]. However, little is known about the effects of TLRs or aging on TFF regulation.
In the present investigation, we examined the consequences of aging in the presence and absence of TLR2 in the large intestine. We demonstrate that aging alone does not lead to significant intestinal inflammation, even in the absence of TLR2. There were, however, age-related effects that led to increased mucosal injury following DSS treatment, accompanied by defective IFN- responses and other cytokine changes. In addition, we demonstrated a potential relationship between aging in the absence of TLR2 function and alterations in expression of TFF3 after mucosal injury.
MATERIALS AND METHODS
Mice
Dr. Shizuo Akira (Osaka University, Osaka, Japan) kindly provided male, 8- to 12-week-old TLR2–/– backcrossed greater than 10 generations onto a C57Bl/6 background, and matched C57Bl/6 control mice were obtained from Jackson Laboratory (Bar Harbor, ME, USA). Aged TLR2–/– and matched C57Bl/6 control mice were housed together for a minimum of 60 weeks prior to experimentation. There were no differences between the mean weights of aged TLR2–/– and C57Bl/6 mice prior to treatment. The animals were housed under specific pathogen-free conditions and allowed free access to regular water and food. Young mice from Jackson Laboratory were housed in our facility for a minimum of 2 weeks prior to use. All experiments were carried out in accordance with the guidelines of the Canadian Council on Animal Care (protocol #03-102).
Induction of colitis
Mice received 3% (wt/vol) DSS (MW=36,000–50,000, ICN Biomedicals Aurora, OH, USA) dissolved in sterile, distilled water ad libitum from Days 0 to 5, followed by 5 or 13 days of regular drinking water. Mice were weighed daily. Cecal and mid-colon samples were taken for histology and tissue sonication on Days 0, 10, and 18 of treatment.
Histology and sonication
The large intestine was removed, and colon length was measured from the cecum to the rectum. It was then divided into cecal and mid-colonic portions, which were opened, rinsed briefly in PBS, and cut longitudinally into two pieces. One-half was fixed in 10% neutral-buffered formalin and paraffin-embedded for histology and the other sonicated and then centrifuged (2300 g for 10 min) with supernatants collected for cytokine and TFF3 assays. Histologic sections were stained with H&E prior to evaluation.
Histologic scoring
All tissue sections were coded and assessed in a "blinded" manner. Histologic damage was scored using a modification of a protocol described previously [30
]. Entire cecal and mid-colonic sections were examined. Briefly, sections were scored based on three criteria: inflammatory cell infiltration (0–3), where 0 represents less than three inflammatory cells per field of view at 40x magnification in the lamina propria, 1 represents greater than three inflammatory cells per field of view in the lamina propria, 2 represents confluence of inflammatory cells extending into the submucosa, and 3 represents confluence of inflammatory cells present in all tissue layers. Tissue damage (0–5) is defined as 0 is normal, 1 represents damage limited to the epithelium, 2 represents focal ulceration limited to the mucosa, 3 represents focal transmural inflammation and ulceration, 4 represents extensive transmural ulceration and inflammation bordered by normal mucosa, and 5 represents extensive transmural inflammation and ulceration involving the entire section. Edema (0–2) is defined as 0 represents no edema, 1 represents focal submucosal edema, and 2 represents extensive submucosal edema involving the entire section.
ELISA
ELISAs were performed on cell-free supernatants using established assays for detection of TNF (sensitivity: 41 pg/ml), IFN- (sensitivity: 78 pg/ml), and IL-1β (sensitivity: 78 pg/ml). ELISAs were performed using the Gibco ELISA amplification system (Life Technologies, Gaithersburg, MD, USA). Antibodies used were TNF capture antibody (R&D Systems, Minneapolis, MN, USA; Cat. #AF-410-NA), TNF detection antibody (Endogen, Woburn, MA, USA; Cat. #MM-350D-B), IFN- capture antibody (PharMingen, San Diego, CA, USA; Cat. #554431), IFN- detection antibody (PharMingen; Cat. #554410), IL-1β capture antibody (R&D Systems; Cat. #MAB401), and IL-1β detection antibody (R&D Systems; Cat. #BAF401). IL-10 ELISA used Becton Dickinson OptEIATM mouse IL-10 set (BD Biosciences, San Jose, CA, USA; Cat. #555252) and was performed according to the manufacturer’s protocol.
Western blot analysis
Western blotting was performed on total protein obtained from tissue sonicants. Protein concentration was determined using a Bradford protein assay kit (Bio-Rad, Hercules, CA, USA). Equal amounts of protein from each sample (10 µg) and Odyssey® protein molecular weight markers (Li-Cor Biosciences, Lincoln, NE, USA) were treated with a reducing sample buffer, separated on a 12.5% SDS-PAGE minigel, and transferred onto a 0.2-µm nitrocellulose membrane, and membranes were blocked using Odyssey® infrared imaging system blocking buffer (Li-Cor Biosciences). For TFF3 detection, membranes were probed with a goat polyclonal anti-TFF3 (IgG ITF A-20, Santa Cruz Biotechnology, Santa Cruz, CA, USA) and detected using Alexa Fluor® 680 donkey anti-goat IgG (Molecular Probes, Eugene, OR, USA). β-Actin was detected using a rabbit polyclonal IgG (Cedarlane® Laboratories, Hornby, Ontario, Canada) and detected using IRDyeTM 800-conjugated, affinity-purified anti-rabbit IgG (Rockland, Gilbertsville, PA, USA). TFF3 antibody specificity was confirmed using a blocking peptide (Santa Cruz Biotechnology) in tenfold excess to the TFF3 primary antibody.
Statistical analyses
Differences between TLR2-deficient and wild-type mice and between young and aged mice at a given time-point were assessed using a One-way ANOVA with a Bonferroni multiple comparison post-test. Where appropriate, individual groups of mice were compared pre- and post-treatment using an unpaired t-test. A P value of <0.05 was considered significant.
RESULTS
Aging in the absence of TLR2 does not lead to the development of colitis
To test the consequences of aging on the development of colitis, TLR2–/– mice and wild-type C57Bl/6 controls were housed within the same facility for a minimum of 60 weeks, and two different anatomical portions of the large intestine were examined. No histologic signs of inflammation or the initiation of colitis were observed in the cecum (data not shown) or the mid-colon in the presence or absence of TLR2 in any of the mice. The large intestines of aged mice were similar at a histologic level to younger (8–12 weeks) TLR2–/– and wild-type mice (Fig. 1A
1B
1C
1D
).
 View larger version (119K): [in a new window] |
Figure 1. Untreated, aged TLR2–/– mice have no intestinal inflammation. Representative photomicrographs of the mid-colon (x400 original) of untreated young (8–12 weeks; A, B) and aged (76 weeks; C and D) C57Bl/6 and TLR2–/– mice (n=6–12 per group). Mid-colonic paraffin sections were stained with H&E. Untreated, young C57Bl/6 (A) and TLR2–/– (B). Untreated, aged C57Bl/6 (C) and TLR2–/– (D).
|
 View larger version (25K): [in a new window] |
Figure 2. Effect of TLR2 deficiency and aging in DSS colitis. (A) Young (8–12 weeks) and aged (60–72 weeks) C57Bl/6 and TLR2–/– mice received 3% (wt/vol) DSS in drinking water for 5 days, followed by 5 days of regular water. Percent weight change of young C57Bl/6 and TLR2–/– mice (n=18 from Days 0 to 5, and n=12 from Days 5 to 10) and aged C57Bl/6 and TLR2–/– (n=12 from Days 0 to 5, and n=6 from Days 5 to 10) mice; representative data of one of two similar experiments. Values are means±SEM. (B) Young (8–12 weeks) TLR2–/– (n=9) and C57Bl/6 (n=10) mice and aged (60–72 weeks) C57Bl/6 (n=9) mice received 3% (wt/vol) DSS in drinking water for 5 days, followed by 13 days of regular water (18 days total). On Days 10–13, five of nine TLR2–/– mice were killed as a result of loss of 20% of total body weight. The remaining four mice are represented by the hatched line (- - -). (C) Effect of DSS administration on total colon length. (D) Histologic assessment of disease activity before (n=6) and after DSS administration (n=4–12 per group). Values are means ± SEM. #, Significant difference between young and aged mice; *, significant difference between TLR2–/– and wild-type C57Bl/6 mice; , significance as a result of DSS treatment.
|
On Day 10, colon lengths were reduced from baseline (P<0.0001) in aged TLR2–/– mice, and they were significantly shorter than similarly treated, aged C57Bl/6 mice (P<0.001), with a similar trend noted in younger mice (Fig. 2C) . Notably, aged C57Bl/6 mice did not have shortened colons on Day 10 (Fig. 2C) , although they had a significant weight loss at this time-point. By Day 18, colon lengths were back above baseline lengths in young C57Bl/6 mice compared with untreated mice (P<0.05). This trend was not observed in young TLR2–/– mice. The colon length of young TLR2–/– animals did increase from Days 10 to 18 but was still significantly shorter than observed in untreated animals (P<0.05) and those of young C57Bl/6 mice (P<0.05) on Day 18. Interestingly, there was little change in the length of the colons in aged, DSS-treated C57Bl/6 mice, even by Day 18 (Fig. 2B) .
Assessment of cecal (data not shown) and mid-colon histologic sections at Day 10 following DSS treatment initiation showed that aged mice had increased mucosal ulceration and inflammation post-treatment (Figs. 2D and 3, A–D). However, TLR2 deficiency combined with aging resulted in substantial inflammatory cell infiltration into all tissue layers, extensive transmural inflammation, and mucosal ulceration and damage involving the entire tissue sections (Figs. 2D and 3D) . Significant infiltration of inflammatory cells accompanied by extensive mucosal ulceration and damage to the regular crypt architecture was also noted in young TLR2–/– mice, although it was less severe than that observed in aged TLR2–/– mice (Fig. 2D) . By Day 18, there was no significant difference in the histologic score between young and aged C57Bl/6 mice. As only a subset of less severely affected TLR2–/– mice survived to Day 18, we were unable to assess the histologic differences between this and other groups appropriately at this time-point. In contrast to previous reports [13 , 17 ], no mortality was observed in younger or aged TLR2–/– mice following DSS treatment.
 View larger version (130K): [in a new window] |
Figure 3. Aged TLR2–/– mice have severe intestinal inflammation after DSS treatment. Representative photomicrographs of the mid-colon (x400 original) of young (8–12 weeks; A and B) and aged (60 weeks; C and D) C57Bl/6 and TLR2–/– mice from two separate experiments where mice were treated with 3% (wt/vol) DSS in drinking water for 5 days, followed by 5 days of water alone (n=6–12 per group). Mid-colonic paraffin sections were stained with H&E. Day 10 young C57Bl/6 (A) and TLR2–/– (B). Day 10 aged C57Bl/6 (C) and TLR2–/– (D).
|
 View larger version (36K): [in a new window] |
Figure 4. Effects of aging and TLR2 deficiency on tissue cecal and mid-colon TNF and IL-1β content before and after DSS treatment. Young (8–12 weeks) and aged (60 weeks) C57Bl/6 and TLR2–/– mice received 3% DSS in drinking water for 5 days, followed by 5 or 13 days of regular water. Tissue samples were ultrasonicated, and supernatants were assayed for total tissue cytokine levels using ELISA. Values are means ± SEM of four to nine mice per time-point. #, Significance between young and aged mice; *, significance between TLR2–/– and wild-type C57Bl/6 mice; , significance as a result of DSS treatment.
|
Aging and the absence of TLR2 led to altered baseline IFN- and IL-10 patterns in the cecum and mid-colon
In view of the observed differences in weight loss and histologic readouts between young and aged TLR2–/– and wild-type mice, two cytokine responses considered critical to the regulation of intestinal inflammation, IL-10 and IFN-, were examined [16
, 31
32
33
]. Aged TLR2–/– mice had reduced baseline colonic IFN- levels (P<0.001) and cecal IFN- levels (P<0.05) compared with younger TLR2–/– mice. Aged C57Bl/6 mice had significantly lower levels of basal colonic IL-10 when compared with younger mice (P<0.05), but this was not associated with significant histological damage or the development of spontaneous colitis (Fig. 1C)
. There were no significant baseline differences in levels of IFN- or IL-10 in colonic or cecal tissue between untreated, young TLR2–/– and C57Bl/6 mice (Fig. 5
).
 View larger version (38K): [in a new window] |
Figure 5. Aged TLR2–/– mice have altered IFN- and IL-10 responses before and after DSS treatment. Young (8–12 weeks) and aged (60 weeks) C57Bl/6 and TLR2–/– mice received 3% DSS in drinking water for 5 days, followed by 5 or 13 days of regular water. Tissue samples were ultrasonicated, and supernatants were assayed for total tissue IFN- and IL-10 levels using ELISA. Values are means ± SEM of four to nine mice per time-point. #, Significance between young and aged mice; *, significance between TLR2–/– and wild-type C57Bl/6 mice; , significance as a result of DSS treatment.
|
and IL-10 expression during DSS colitis levels than that seen in younger mice. Aged TLR2–/– and wild-type mice did not have a significant IFN- response to DSS treatment in the cecum or mid-colon. However, both groups of younger mice showed significant increases in IFN- following DSS treatment in the mid-colon and cecum (P<0.001) by Day 10 but returned to baseline levels by Day 18 (Fig. 5)
. Despite the observed, extensive histological damage in aged TLR2–/– mice, they failed to shown any significant increase in IFN- in the cecum and mid-colon after DSS challenge (Day 10). Levels of IFN- measured in the cecum and mid-colon of aged TLR2–/– and wild-type mice were significantly lower than those of younger mice on Day 10 (P<0.001 for both). Even by Day 18, there was no change in the levels of IFN- in aged C57Bl/6 mice. On Day 10, young, TLR2-deficient mice also had significantly less cecal IFN- (P<0.05) compared with similarly treated wild-type mice (Fig. 5)
. Similar to younger mice, cecal IL-10 levels decreased following DSS treatment in aged TLR2–/– mice (P<0.0001) and were significantly lower than in similarly treated, aged C57Bl/6 mice (P<0.001; Fig. 5 ). Colonic IL-10 levels in aged TLR2–/– mice increased from baseline after DSS treatment (P<0.0001), but in aged C57Bl/6 mice, they remained relatively unaffected. Aged TLR2–/– and C57Bl/6 mice had significantly higher levels of IL-10 in the mid-colon compared with younger mice by Day 10 (P<0.001 and P<0.01, respectively). Young TLR2–/– and wild-type mice also showed significantly reduced colonic IL-10 levels following DSS treatment (P<0.0001 and P<0.05, respectively), with a similar trend in the cecum by Day 10. Interestingly, by Day 18, we observed a substantial increase in the level of IL-10 in the cecum and mid-colon of young TLR2–/– mice (P<0.001) and in the mid-colon of aged C57Bl/6 mice (P<0.01).
Comparison of Day 10, young TLR2–/– mice with Day 7, young C57Bl/6 mice revealed significantly higher levels of IL-10 and IFN- in the cecum and colon of young C57Bl/6 mice (data not shown).
TFF3 expression in the cecum and colon is modified by TLR2 deficiency and aging
The ability of young and aged TLR2–/– mice to express the restitution factor TFF3 in the mid-colon and cecum was examined using Western blotting. To confirm the specificity of the TFF3 analysis, Western blots were performed on cecal samples from mice in the presence and absence of a specific blocking peptide. The TFF3 antibody detected the 14-kDa dimeric [27
, 28
, 34
] TFF3 protein in cecal and mid-colonic samples, and this signal was inhibited when the blocking peptide was used (Fig. 6C
). No TFF3 signal was observed in lysates obtained from normal mouse spleen cells.
 View larger version (35K): [in a new window] |
Figure 6. Aging with a TLR2 deficiency affects cecal and colonic TFF3 expression. Western blot analysis of total TFF3 protein expression in the cecum (A) and the mid-colon (B) of young (8–12 weeks) and aged (60 weeks) TLR2–/– and C57Bl/6 mice that were untreated (–) or treated with 3% DSS for 5 days, followed by 5 days of drinking regular water (+). (C) Western blots incubated with a TFF3-blocking peptide showing the specificity of the TFF3 antibody. (1.) Untreated, young C57Bl/6 cecum; (2.) untreated, young TLR2–/– cecum; (3.) Day 10, young C57Bl/6 cecum. (D) Densitometric analysis of cecal Western blot data. (E) Densitometric analysis of mid-colon Western blot data. Densitometry values are means ± SEM of four mice per time-point. #, Significance between young and aged mice; *, significance between TLR2–/– and wild-type C57Bl/6 mice; , significance as a result of DSS treatment. Analysis of a larger group of mice (n=8–9) revealed a consistent inhibition of cecal TFF3 in aged, TLR2-deficient animals (mean intensity of 1.15/β-actin, P<0.0004) when compared with young C57Bl/6 controls. There was a consistent elevation of colonic TFF3 (mean intensity of 19.47/β-actin, P<0.0058, n=6–9) when compared with aged, untreated TLR2–/– (mean intensity of 9.19/β-actin).
|
DISCUSSION
These studies demonstrate an important role for age-related changes in regulating inflammation in the large intestine, suggesting the factors that contribute to the pathogenesis of IBD in the elderly could differ from those in younger individuals. Our study demonstrates that aged mice respond differently from younger mice to a DSS challenge. Aging had substantial impact on disease severity, cytokine levels, and expression of the important repair factor TFF3 in the large intestine. Several studies have highlighted the importance of MyD88-dependent TLR and/or cytokine signaling in protecting the intestinal mucosa from injury [13 14 15 16 17 , 35 ]. The specific role of TLR2 has been suggested to be primarily in the regulation of intestinal epithelial tight-junction integrity [17 ]. Aging has not been examined previously in this context. Young TLR2–/– mice showed no signs of mucosal inflammation in the colon or cecum and were indistinguishable from wild-type mice, consistent with previous reports [13 , 16 ]. Even TLR2-deficient animals aged for over 1 year did not spontaneously develop colitis. These results are consistent with reports over a 1.5-year period in aged, unchallenged MyD88–/– and MyD88–/–/IL-10–/– double-knockout mice [16 ]. However, MyD88–/– mice also have cytokine (IL-1 and IL-18) signaling defects. These results also suggest that even long-term defects in tight junctions as a result of TLR2 deficiency do not lead to spontaneous colitis.
The effects of aging and their relationship to TLR function are relatively unexplored. DSS treatment of mice disrupts epithelial barrier integrity, allowing the commensal microflora increased access to the host’s mucosal immune system. In the absence of TLR2, young and aged mice showed more severe inflammation following DSS-induced injury. However, no mortality was observed under our treatment conditions in contrast to previous reports [13 , 17 ], although five of nine young TLR2–/– mice had reached the maximum allowable weight loss when studied beyond Day 10. This clearly demonstrates that young TLR2–/– mice fail to recover from DSS-induced injury as efficiently as young wild-type mice. After DSS treatment, young and aged TLR2–/– mice had increased histologic scores, weight loss, and shortening of the colon when compared with wild-type controls (Fig. 2A 2B 2C 2D) . Interestingly, no shortening of the colon was observed in aged, wild-type mice after DSS treatment (Fig. 2C) . It has been demonstrated that aging impairs colonic smooth muscle contraction and delays wound contraction [36 , 37 ]. This may explain the lack of shortening of aged, wild-type colons after DSS treatment, although the mice had developed considerable colitis (Figs. 2D and 3C) . We have demonstrated for the first time increased severity of response to DSS treatment in aged TLR2–/– mice accompanied by distinct age-associated cytokine patterns when compared with similarly treated, younger TLR2–/– mice. Others have identified that late-onset ulcerative colitis can be more severe than early onset disease [38 ]. These clinical findings are consistent with this mouse model of colitis, where aged mice had more severe colitis than younger mice. Aging alone did not lead to the development of spontaneous colitis but led to more severe disease following DSS challenge.
The DSS model of colitis has been traditionally associated with a polarized Th1 response with increased TNF and IFN- [39
]. It has previously been shown that IL-10–/– mice spontaneously develop colitis, characterized by increased numbers of TNF- and IFN--producing lamina propria CD4+ T cells [16
, 31
]. TNF is a pleiotropic cytokine known to play an important role in the pathogenesis of IBD [40
]. Young, untreated TLR2–/– mice expressed slightly elevated levels of colonic TNF compared with controls (Fig. 4)
. Aged, wild-type and TLR2–/– mice had roughly tenfold the levels of basal TNF in their cecum compared with younger mice (Fig. 4)
. Regardless of these higher local levels of TNF, there were no signs of an ongoing, inflammatory response or the initiation of colitis histologically. Using a similar acute model of DSS-induced colitis, others have shown that administration of DSS to TNF–/– mice resulted in a larger inflammatory response [41
]. These data suggest a complex role for TNF in the DSS model of colitis in keeping with its complex role in regulating inflammation, aspects of which may be critical for host defense against the microflora.
Site-specific variation in response is a common feature of intestinal inflammation. Several studies have shown that mice are more susceptible to DSS-induced injury in the colon and slightly more resistant in the cecum [42 43 44 ]. The reason for these differences remains unknown, but altered proportions of tissue cell types and site variations in the endogenous microflora could contribute to variable susceptibility and cytokine levels between the cecum and the mid-colon. Changes in the microflora or resident cell types in the cecum during aging could contribute to the increased TNF levels observed in untreated, aged TLR2–/– and wild-type mice.
A recent study using IFN--deficient mice in DSS colitis demonstrated that wild-type IFN--producing mice develop more severe disease than IFN--deficient mice [32
]. In the absence of the innate PRR TLR2, young mice had increased cecal and colonic IFN- levels after the initiation of DSS colitis (Fig. 5)
. However, when TLR2–/– mice were aged and treated with a DSS regime, leading to more severe intestinal inflammation, they failed to show any significant increase in cecal and colonic IFN- levels (Fig. 5)
. These findings are not consistent with elevations in IFN- being a major contributor to intestinal, inflammatory responses in the absence of TLR2. Th1 and NK cells are potent producers of IFN-, which plays a pivotal role in inducing the antimicrobial effects of macrophages. An age-associated decline in aspects of immune function has been well-documented [18
19
20
21
22
]. Decreased splenic macrophage TLR expression and function [23
, 24
] have been observed in aged mice. A reduced ability to respond appropriately to intestinal flora could contribute to the worsened inflammation and cytokine dysregulation we observed in aged animals.
Stabilization of the mucus barrier and increased epithelial cell restitution by TFFs have been well-established [27
, 29
, 45
]. It has also been suggested that TFF2 and TFF3 play a role in modulating mucosal immune responses [45
46
47
]. However, a direct relationship between TLR signaling and regulation of any of the TFF family members has not been demonstrated. In young mice, the absence of TLR2 was not associated with altered TFF3 levels. A modest decrease in cecal TFF3 levels was observed in young C57Bl/6 mice after DSS administration, which could suggest TLR2 signaling negatively regulates TFF3 at this site. In contrast, aged, untreated C57Bl/6 mice showed significantly reduced TFF3 levels prior to treatment in the cecum and colon compared with younger C57Bl/6, suggesting that levels of this important, protective and repair factor decrease over time. Aged TLR2–/– mice had dramatically decreased levels of TFF3 expression in the cecum after DSS treatment, whereas TFF3 levels remained constant following DSS in aged, wild-type mice (Fig. 6A)
. These findings suggest a possible mechanism, whereby TLR2-deficient mice have an impaired ability to facilitate cecal epithelial restitution and mediate repair. This observation is supported at the histologic level in the cecum (data not shown) by the observation that aged TLR2–/– mice showed the most severe mucosal damage to crypts and the epithelium with a massive inflammatory cell infiltration and the largest weight loss. In this study, we found a strong correlation between increased histologic scores and decreased levels of TFF3 in the cecum (P<0.01). In contrast to the cecum, in the mid-colon, there were decreased baseline levels of TFF3 in both groups of aged mice compared with younger mice. After DSS treatment, there was a consistent increase in colonic TFF3 levels in aged TLR2–/– mice, but aged, wild-type mice remained unaffected. The reason for these site-specific differences in TFF3 regulation requires further investigation. Others have shown that cytokines such TNF and IL-1β negatively regulate TFF3 expression in an epithelial cell line by activation of NF-B, which has been demonstrated to negatively regulate transcription of TFF3 [48
49
50
]. Baseline TNF levels were increased significantly in the cecum of aged animals (Fig. 4)
. Despite these high levels of TNF, there were normal TFF3 levels in the cecum of aged mice. There were also no significant differences in the levels of cecal or colonic TNF or IL-1β between aged TLR2–/– and C57Bl/6 mice at Day 10, suggesting that these cytokines alone are not negatively regulating TFF3 expression.
Overall, this study suggests that neither the absence of TLR2 nor aging alone led to the initiation of colitis. There are, however, significant effects of aging in the context of a TLR2 deficiency on basal IL-10 and IFN- levels. This could potentially provide a microenvironment that favors the development of more severe colitis following epithelial damage. We also demonstrate that a deficiency in TLR2 combined with aging affects TFF3 regulation in the cecum and mid-colon. The demonstrated interactions among aging-associated changes, TLR deficiency, and TFF3 regulation may be of particular relevance to understanding the development of chronic intestinal inflammation in the elderly.
ACKNOWLEDGEMENTS
The Crohn’s and Colitis Foundation of Canada and The Canadian Institutes of Health Research (grant #MOP10966) supported research described in this article. E. J. A. was supported by a student research award from the Nova Scotia Health Research Foundation. We thank Yi-song Wei, Sandy Edgar, Pat Colp, and Jon Duplisea for their excellent technical assistance, Dr. S. Akira (Osaka University) for providing the TLR2-deficient animals necessary for this study, and Dr. Andrew Stadnyk for his comments about the manuscript. The authors have no conflicts of interest.
Received August 21, 2007; revised October 10, 2007; accepted December 30, 2007.
REFERENCES
, plays a major role in sustaining the chronic phase of colitis in IL-10-deficient mice J. Immunol. 161,3143-3149 is causatively involved in experimental inflammatory bowel disease in mice Clin. Exp. Immunol. 146,330-338[CrossRef][Medline]
deficient mice J. Gastroenterol. Hepatol. 18,560-569[CrossRef][Medline] and nuclear factor B inhibit transcription of human TFF3 encoding a gastrointestinal healing peptide Gut 52,1297-1303B and C/EBPβ Peptides 25,849-854[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  V. Cerovic, C. D. Jenkins, A. G. C. Barnes, S. W. F. Milling, G. G. MacPherson, and L. S. Klavinskis Hyporesponsiveness of Intestinal Dendritic Cells to TLR Stimulation Is Limited to TLR4 J. Immunol., February 15, 2009; 182(4): 2405 - 2415. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
