SEARCH

SEARCH BY CITATION

Keywords:

  • Bacteroides vulgatus;
  • cytokines;
  • HLA-B27;
  • transgenic rats

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Background: HLA-B27/β2 microglobulin transgenic (TG) rats develop spontaneous colitis when raised under specific pathogen-free (SPF) conditions or after monoassociation with Bacteroides vulgatus (B. vulgatus), whereas germ-free TG rats fail to develop intestinal inflammation. SPF HLA-B27 TG rnu/rnu rats, which are congenitally athymic, remain disease free. These results indicate that commensal intestinal bacteria and T cells are both pivotal for the development of colitis in TG rats. However, it is not known if T cells are also required in the induction of colitis by a single bacterial strain. The aim of this study was therefore to investigate the role of T cells in the development of colitis in B. vulgatus–monoassociated HLA-B27 TG rats.

Methods: HLA-B27 TG rnu/rnu and rnu/+ rats were monoassociated with B. vulgatus for 8–12 weeks. CD4+ T cells from mesenteric lymph nodes (MLNs) of B. vulgatus–monoassociated rnu/+ TG donor rats were transferred into B. vulgatus–monoassociated rnu/rnu TG recipients.

Results:B. vulgatus–monoassociated rnu/+ rats showed higher histologic inflammatory scores and elevated colonic interferon-γ mRNA, cecal myeloperoxidase, and cecal IL-1β levels compared to those in rnu/rnu TG rats that did not contain T cells. After transfer of CD4+ cells from colitic B. vulgatus–monoassociated rnu/+ TG donor rats, B. vulgatus–monoassociated rnu/rnu TG recipients developed colitis that was accompanied by B. vulgatus-induced IFN-γ production by MLN cells in vitro and inflammatory parameters similar to rnu/+ TG rats.

Conclusions: These results implicate CD4+ T cells in the development of colitis in HLA-B27 TG rats monoassociated with the nonpathogenic bacterial strain B. vulgatus.

(Inflamm Bowel Dis 2007)

The important role of commensal intestinal bacteria in the pathogenesis of experimental colitis and human inflammatory bowel diseases (IBD) has been emphasized by numerous recent studies.1 This has been demonstrated in multiple animal models of experimental colitis, in which the susceptible host develops spontaneous colitis in the presence of nonpathogenic intestinal organisms.2–9 In a well-characterized animal model of inflammation, HLA-B27/β2 microglobulin transgenic (TG) rats expressing high copy numbers of the human HLA-B27 gene developed spontaneous colitis, gastritis, and arthritis by 3 months of age when raised under specific pathogen-free (SPF) conditions.2 Commensal bacteria of the intestinal tract have been extensively implicated in the development of disease. Treatment of HLA-B27 TG rats with broad-spectrum antibiotics is effective in preventing and ameliorating disease.10 In addition, when HLA-B27 TG rats are raised in a germ-free environment, they fail to develop intestinal inflammation and arthritis,3, 11 whereas colitis and gastritis develop within 1 month after transfer to an SPF environment.3 Although the results of numerous studies have been reported, the mechanisms underlying intestinal inflammation in TG rats remain unclear. Disease does not develop in SPF TG rnu/rnu rats, which are congenitally athymic and therefore lack conventional T lymphocytes, thus providing direct evidence that T cells are required for the development of colitis in this model. Even if rnu/rnu TG rats carry the disease-prone B27 transgenic locus, they are protected from disease, whereas the presence of T cells in their heterozygous rnu/+ TG littermates coincides with the development of colitis under SPF conditions.12, 13 The critical role of T cells in the induction of intestinal inflammation in this model was further emphasized by the development of colitis in SPF rnu/rnu HLA-B27 TG recipients after T cells had been transplanted, with CD4+ T cells being more efficient than CD8+ T cells in transferring disease.12

Most studies of HLA-B27 TG rats have been carried out under SPF conditions. However, monoassociation of HLA-B27 TG rats would provide a model to efficiently explore the role of T cells in the induction of colitis by a single bacterial strain. A previous study from our laboratory showed that monoassociation of HLA-B27 TG rats with Bacteroides vulgatus (B. vulgatus), a Gram-negative nonpathogenic microorganism, leads to induction and perpetuation of colitis.14 In HLA-B27 TG rats, B. vulgatus preferentially induced colitis within 1 month of monoassociation, whereas no inflammation resulted from monoassociation with Escherichia coli (E. coli) isolated from a patient with Crohn's disease.14 The ability of B. vulgatus to induce colitis was first described using carrageenan-fed guinea pigs as the model system.15 Furthermore, Bacteroides species were identified among bacteria that preferentially expanded in HLA-B27 TG rats after creation of a blind cecal loop, which resulted in development of severe cecal inflammation.16

To study the role of T cells in the chronic inflammation induced by B. vulgatus, we evaluated T-cell responses to B. vulgatus components in vitro. Although our previously published results showed that T cells from SPF TG rats produce interferon (IFN)–γ after in vitro stimulation with lysates of commensal intestinal microorganisms,17 we were unable to demonstrate B. vulgatus-specific T-cell responses in mesenteric lymph node (MLN) cells of B. vulgatus–monoassociated animals (unpublished observation). Therefore, the aim of the present study was to determine whether T cells are necessary for colitis that develops in B. vulgatus monoassociated HLA-B27 TG rats. For this purpose, we used a unique model of monoassociation in the nude HLA-B27 TG rat. To our knowledge, this is the first study presenting data derived from monoassociated nude HLA-B27 TG rats. Our results showed that B. vulgatus induced colitis in previously germ-free TG rnu/+ rats but not in rnu/rnu TG rats. Furthermore, disease developed in rnu/rnu TG recipients after transfer of activated CD4+ T cells from B. vulgatus–monoassociated rnu/+ TG rats. These data provide clear evidence that T cells are required for the induction of colitis by B. vulgatus in rnu/rnu TG rats. Moreover, the results show that CD4+ cells alone can generate the responses we observed, and CD8+ cells are not required.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Animals

We used germ-free HLA-B27 TG rats (of the 33-3 line on an F344 background)2, 18 and their nontransgenic (non-TG) littermates that were either homozygous or heterozygous for the rnu allele (rnu/rnu, or rnu/+, respectively).19, 20Rnu/rnu rats are athymic, whereas heterozygous rnu/+ rats are euthymic and have normal T-cell function.12 All germ-free animals were monoassociated with B. vulgatus in the Gnotobiotic Animal Core of the Center for Gastrointestinal Biology and Disease at the College of Veterinary Medicine, North Carolina State University (Raleigh, NC). Rats used for these experiments were between 4 and 8 months old, and animals were monoassociated with B. vulgatus for 8–12 weeks. Whether the HLA-B27 transgene was present was determined by PCR using DNA isolated from tail clippings. All studies were approved by the Institutional Animal Care and Use Committees of North Carolina State University (Raleigh, NC) and the University of North Carolina at Chapel Hill (Chapel Hill, NC).

Experimental Design

Germ-free rnu/rnu and rnu/+ HLA-B27 TG rats were monoassociated for 8–12 weeks with B. vulgatus by oral and rectal swabbing with a slurry of fecal material from B. vulgatus–monoassociated rats. In separate experiments, we intraperitoneally (i.p.) transferred 107 MLN CD4+ T cells derived from rnu/+ HLA-B27 TG rats monoassociated with B. vulgatus into B. vulgatus–monoassociated rnu/rnu TG recipients, as described below. After 7–9 weeks, the rats were killed, and MLN cells were collected for analysis of cell surface molecules by flow cytometry and for in vitro stimulation with B. vulgatus components. Cecal and colonic tissues were collected for myeloperoxidase (MPO), cecal cytokines, and histologic examination, as described below. B. vulgatus derived from a guinea pig with carrageenan-induced colitis15 was a kind gift from Dr. A. B. Onderdonk, Harvard University (Cambridge, Mass). B. vulgatus monoassociation and the absence of contaminating organisms were confirmed by anaerobic and aerobic cultures, as well as by Gram staining.

CD4+ T-Cell Transfer

Donor cells were isolated from MLNs from rnu/+ HLA-B27 TG rats monoassociated with B. vulgatus for at least 4 weeks. Single MLN cell suspensions were prepared by gentle teasing. CD4+ T cells were obtained by negative selection, as described previously,17 using antirat CD45RA and anti-rat CD8a antibody–coated microbeads and columns designed for cell depletion according to the manufacturer's instructions (Miltenyi Biotec, Auburn, Calif). Subsequently, 107 CD4+ T cells were transferred i.p. into rnu/rnu TG recipients that had been monoassociated with B. vulgatus for at least 2 weeks.

Histology

Colons and ceca were fixed in formalin and stained as previously described.3, 21 During the blinded evaluation, they were each given a validated histologic inflammatory score ranging from 0 to 4.3, 21

Preparation of Bacterial Lysates

Bacterial lysates were prepared as described previously.17, 22 Briefly, cecal contents, pooled from several B. vulgatus–monoassociated non-TG rats, were disrupted using glass beads to lyse the bacteria. The lysate was then filtered through a 0.22-μm syringe filter.17 A single colony of murine E. coli isolated from SPF normal C57BL/6 mice was cultured in Brain-Heart Infusion broth for 48 hours, and the resulting bacteria were used to prepare a lysate preparation as described above. The sterility of the lysates was confirmed by aerobic and anaerobic bacterial cultures.

Mesenteric Lymph Node Cell Cultures

MLNs were obtained from rnu/rnu and rnu/+ TG rats, and 4 × 105 cells were cultured in 96-well flat-bottomed microplates (Costar 3595), in 0.2 mL of complete medium (RPMI 1640 plus 5% heat-inactivated fetal calf serum, 2 mM L-glutamine, 1 mM sodium pyruvate, 5 × 10−5 M 2-mecaptoethanol, and 50 μg/mL gentamicin) for 3 days. Cells were stimulated with different concentrations of B. vulgatus or E. coli lysate, as indicated. Culture supernatants were collected and stored at −20°C.

Flow Cytometry

MLN cells were evaluated by flow cytometry using the following fluorochrome-labeled or unlabeled reagents: ME-1 (anti-HLA-B27) hybridoma culture supernatant (ATCC, Rockville, Md); FITC-labeled goat antimouse IgG (γ-chain-specific) antibody (Southern Biotechnology, Birmingham, Ala), to detect binding of the unlabeled ME-1; FITC-labeled goat anti-rat IgG (H+L) antibody (Kirkegaard & Perry Laboratories, Gaithersburg, Md), to identify B cells; PE-antirat CD4 (clone W3/25) and FITC anti-rat CD8 (clone OX-8), from Caltag (Burlingame, Calif); and FITC anti-CD45RC (clone OX-22) and FITC antirat TCRαβ (clone R73), from BD Biosciences Pharmingen (San Diego, Calif), to identify T-cell subpopulations.

Cytokine Measurement

We used the following commercially available ELISA reagents, as previously described, to measure IFN-γ and IL-10 in culture supernatants17: unlabeled polyclonal anti-IFN-γ antibody and biotin-labeled monoclonal anti-IFN-γ antibody (clone DB-1), from Biosource International (Camarillo, Calif); unlabeled monoclonal antirat IL-10 antibody (clone A5-7) and biotin-labeled monoclonal antirat IL-10 antibody (clone A5-6), from BD Biosciences Pharmingen (San Diego, Calif); and horseradish peroxidase–labeled streptavidin (BD Biosciences Pharmingen, San Diego, Calif). Standard curves were generated using recombinant cytokines (BD Biosciences Pharmingen, San Diego, Calif).

Cytokine mRNA Expression in Cecal Tissues

RNA was isolated from cecal and colonic tissues using TRIzol (Invitrogen, Carlsbad, Calif), and mRNA was reverse-transcribed (1 μg RNA) and amplified using primers specific for rat IFN-γ and β-actin,17 as described previously.23 The PCR products (10 μL) were subjected to electrophoresis on 2% agarose gels containing gel Star fluorescent dye (FMC, Philadelphia, Pa). Fluorescence staining was captured using an Alpha Imager 2000 (AlphaInnotech, San Leandro, Calif). The IFN-γ mRNA was quantified by densitometry, and the ratio of IFN-γ to β-actin mRNA was calculated.

Cecal IL-1β Analysis

For measurement of cecal IL-1β, tissue from the distal cecum was homogenized in the presence of protease inhibitors.24 IL-1β was determined in the supernatant according to the manufacturer's instructions (National Institute for Biological Standards and Controls, South Mimms, UK).

MPO Assay

Ceca were homogenized as described in the previous section. Subsequently, MPO activity (units per gram of cecal tissue) was quantified, as described previously.21, 25

Statistical Analysis

Concentrations of the various cytokines were calculated as means ± standard errors of triplicate culture supernatants. Data were analyzed using the paired Student t test, and a P value of less than 0.05 was considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Colitis in B. vulgatus–Monoassociated TG Rats

TG rnu/+ rats monoassociated with B. vulgatus for 8–12 weeks developed mild histologic inflammation in both the cecum and the colon (Figs. 1A and 2). This inflammation was characterized by mononuclear cell infiltration of the intestinal mucosa with crypt hyperplasia. In contrast, TG rnu/rnu rats housed under identical conditions did not show microscopic signs of colitis (Figs. 1B and 2). When CD4+ T cells derived from colitic B. vulgatus–monoassociated rnu/+ TG rats were transferred into B. vulgatus–monoassociated rnu/rnu TG rats, the recipients developed mild colitis in the cecum and colon. This inflammation was very similar in appearance and severity to that observed in the intestinal tract of TG rnu/+ rats monoassociated with B. vulgatus (Figs. 1C and 2).

thumbnail image

Figure 1. Representative photomicrographs of tissue sections (×10) from cecal tissue of 6- to 8-month-old (A) B. vulgatus–monoassociated TG rnu/+ rats; (B) B. vulgatus–monoassociated TG rnu/rnu rats; (C) B. vulgatus–monoassociated TG rnu/rnu rats after CD4+ T-cell transfer.

Download figure to PowerPoint

thumbnail image

Figure 2. Blinded cecal and colonic inflammatory scores in B. vulgatus–monoassociated TG rnu/+ and TG rnu/rnu rats and in B. vulgatus–monoassociated TG rnu/rnu rats after CD4+ T-cell transfer. Monoassociated rats were killed 8–12 weeks after B. vulgatus monoassociation. Values represent mean ± SEM of cecal or colonic histologic scores (0–4). Each group consisted of 6 or 7 rats (*P < 0.001 versus TG rnu/rnu rats).

Download figure to PowerPoint

Evaluation of MLN Cell Populations

MLNs from B. vulgatus–monoassociated TG rnu/+ rats with colitis were larger and contained more than 5 times as many cells as those from TG rnu/rnu rats without disease (Table 1). MLNs from TG rnu/+ rats contained 5 times as many CD4+ T cells as CD8+ T cells (Table 1). Most TG rnu/rnu MLN cells expressed immunoglobulin, a surface marker of B cells, whereas TG rnu/+ MLNs contained a 2- to 3-fold smaller percentage of B cells than those from rnu/rnu TG rats.

Table 1. Cell Numbers and Cell Subpopulations in Mesenteric Lymph Nodes of B. vulgatus–Monoassociated rnu/rnu, rnu/+, and rnu/rnu Recipients of CD4 T Cells from rnu/+ HLA-B27 B. vulgatus–Monoassociated Transgenic Rats
 B. vulgatus–Monoassociated HLA-B27 Transgenic Rats
rnu/rnu (n = 7)rnu/+ (n = 7)rnu/rnu Recipients of TG rnu/+ CD4 T Cells (n = 6)
  • Values represent mean total number of cells (± SEM) or mean percentage of lymphoid cell subpopulations (± SEM).

  • a

    P < 0.001

  • b

    P <0.01, versus rnu/rnu transgenic rats.

Total cells (× 106)19 ± 2100 ± 8a48 ± 6b
TCRαβ1.9 ± 0.663.0 ± 3.0a9.7 ± 1.4b
CD42.8 ± 0.351.1 ± 1.9a9.6 ± 1.4b
CD81.5 ± 0.110.2 ± 0.7a1.3 ± 0.2
Surface Ig93.5 ± 0.736.4 ± 3.2a81.7 ± 1.2a
HLA-B2797.3 ± 0.798.1 ± 0.196.5 ± 0.7

Donor MLN cells, enriched for CD4+ T cells, were 98.1% ± 0.2% CD4-positive; expression of CD45RC, the phenotype of activated T cells, was found on 71.6% ± 0.6% of these CD4 cells, indicating that the majority of donor CD4+ T cells had been activated in vivo.26 After adoptive transfer of CD4+ T cells, the total number of unseparated MLN cells and the proportion of CD4+ cells were significantly higher and the proportion of surface Ig-expressing cells was lower in rnu/rnu TG recipients than in rnu/rnu TG rats that did not receive CD4+ T cells. Moreover, expression of CD45RC was low in 89.2% ± 3.2% of recipient CD4+ MLN cells. All animals identified as HLA-B27 transgenic by PCR analysis of tail DNA also expressed HLA-B27 on MLN cells.

Mucosal Cytokine and MPO Production

IL-1β levels in cecal homogenates reflected the severity of cecal inflammation in rnu/+ TG rats. Compared to that from B. vulgatus–monoassociated rnu/rnu TG rats, cecal tissue from rnu/+ rats and from rnu/rnu CD4+ T-cell recipients contained significantly more IL-1β (Fig. 3A), with the latter group producing higher levels of cecal IL-1β than the former (P < 0.05).

thumbnail image

Figure 3. (A) IL-1β and (B) myeloperoxidase (MPO) in cecal homogenates from B. vulgatus–monoassociated TG rnu/+ and TG rnu/rnu rats and from B. vulgatus–monoassociated TG rnu/rnu rats after CD4+ T-cell transfer. Data represent mean ± SEM of values obtained from 6 or 7 rats per group (*P < 0.05 versus TG rnu/rnu rats; **P < 0.05 versus TG rnu/+ rats).

Download figure to PowerPoint

Cecal MPO showed the same pattern as cecal IL-1β with approximately twice the MPO in cecal tissue from TG rnu/+ rats and CD4+ T-cell recipients than in that from TG rnu/rnu rats and higher MPO levels in the CD4+ T-cell rnu/rnu cell recipients than in TG rnu/+ rats (Fig. 3B).

Cecal and Colonic IFN-γ mRNA Expression

The IFN-γ/β actin mRNA ratio was significantly higher in colons from TG rnu/+ rats and from rnu/rnu recipients of CD4 T cells than in those from TG rnu/rnu rats (P < 0.05; Fig. 4A,B). The cecal IFN-γ/β actin mRNA ratio in rnu/rnu recipients of CD4+ T cells was also significantly higher than that of cecal tissue from rnu/rnu animals.

thumbnail image

Figure 4. (A) Expression of IFN-γ mRNA in cecal and colonic tissues from B. vulgatus–monoassociated TG rnu/+ and TG rnu/rnu rats and from B. vulgatus–monoassociated TG rnu/rnu rats after CD4+ T-cell transfer. Results are expressed as the ratio of IFN-γ mRNA to β-actin (*P < 0.05 versus TG rnu/rnu rats). (B) IFN-γ PCR products from cecal and colonic samples of 1 representative rat from each group. Lanes: 1, standard; 2, cecum of TG rnu/rnu rat; 3, cecum of TG rnu/+ rat; 4, cecum of TG rnu/rnu recipient of TG rnu/+ CD4+ T cells; 5, colon of TG rnu/rnu rat; 6, colon of TG rnu/+ rat; 7, colon of TG rnu/rnu recipient of TG rnu/+ CD4+ T cells; 8, negative.

Download figure to PowerPoint

MLN Cell Cytokine Production

MLN cells isolated from either TG rnu/rnu or TG rnu/+ rats did not produce detectable levels of IFN-γ after stimulation with cecal lysates prepared from B. vulgatus–monoassociated rats (Fig. 5A). Interestingly, when CD4+ TG T cells were transferred into rnu/rnu TG recipients monoassociated with B. vulgatus, MLN cells stimulated with B. vulgatus lysate secreted significant amounts of IFN-γ (Fig. 5A). However, when MLN cells were stimulated with E. coli lysates, no IFN-γ was detected (data not shown), suggesting that the IFN-γ responses observed are specific for components of B. vulgatus. As shown in Figure 5B, MLN cells from all 3 groups of rats produced IL-10 after stimulation with B. vulgatus lysate, suggesting that the MLN cells obtained from each group of rats are capable of responding in vitro to components of the bacterial lysate and that T cells are not the cell type that produces IL-10 in this system. Moreover, these results indicate that intestinal inflammation develops even though MLN cells from TG rnu/rnu recipients of CD4+ T produce IL-10.

thumbnail image

Figure 5. IFN-γ and IL-10 production in mesenteric lymph node cell cultures from B. vulgatus–monoassociated TG rnu/+ and TG rnu/rnu rats and from B. vulgatus–monoassociated TG rnu/rnu rats after CD4+ T-cell transfer. Rats were killed 8–12 weeks after B. vulgatus monoassociation, and mesenteric lymph node cells were stimulated for 3 days with cecal bacterial lysate from B. vulgatus–monoassociated non-TG rnu/+ rats. (A) IFN-γ and (B) IL-10 were measured in triplicate culture supernatants by ELISA. Values represent mean values (pg/mL) ± SEM (*P < 0.005 versus TG rnu/rnu and TG rnu/+ rats).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

B. vulgatus has been implicated in the induction and perpetuation of colitis. It is an anaerobic Gram-negative bacterial strain that resides in the colon in high numbers. B. vulgatus is a nonpathogenic bacterium; this common commensal enteric bacterial species does not induce colitis in wild-type rats. However, genetically susceptible HLA-B27 TG littermates develop intestinal inflammation after monoassociation with B. vulgatus.14

In the present study, we used a unique model of monoassociated rnu/rnu HLA-B27 TG rats. We showed that T cells are required for the induction of intestinal inflammation in rnu/rnu TG rats monoassociated with nonpathogenic commensal B. vulgatus. This was demonstrated by histologic scores, cecal MPO and IL-1β levels, and colonic IFN-γ mRNA levels that were higher in rnu/+ TG rats than in rnu/rnu TG rats. The level of histologic inflammation in the rnu/+ TG rats detected in the present study is similar to the histologic scores reported by Rath et al for TG rats that did not carry a nude allele.14 In that study, B. vulgatus–monoassociated TG rats with normal T-cell development showed average cecal histologic scores of 1.7 ± 0.2, compared to the 1.4 ± 0.1 we found for the rnu/+ TG rats in the present study. Interestingly, we observed that rnu/+ HLA B27 TG rats housed under SPF conditions had consistently higher histologic inflammation scores than did rnu/+ TG rats monoassociated with B. vulgatus (data not shown). The complex intestinal microflora that colonizes the distal intestinal tract under SPF conditions provides a strong antigenic stimulus for a genetically susceptible host, resulting in a more severe colonic inflammation than that in B. vulgatus–monoassociated TG rats, with likely synergistic effects between multiple bacterial strains. Monoassociation with B. vulgatus induces only mild inflammation, probably because there are no other disease-inducing microorganisms.

Breban et al. showed that rnu/rnu HLA-B27 TG rats housed under SPF conditions do not develop colitis.12 The results of the present study extend those findings, confirming that T cells are also a critical component of the potentially less complex set of factors involved in the development of colitis after colonization with a single microorganism. In several other animal models, T cells have been shown to be crucial to the development of inflammation. For example, severe-combined immunodeficient (SCID) mice develop colitis after transfer of naive T cells (CD4+ T cells with high expression of CD45RB in the mouse).27 The equivalent population of naive CD4+ T cells in the rat, cells that express CD45RC at a high level, induced inflammation in the stomach and pancreas and also wasting disease in athymic rats, whereas CD4+ T cells with a low level of CD45RC expression were considerably less pathogenic.26 T cells have also been shown to play a pivotal role in the colitis that develops in IL-10-deficient mice,28 IL-2-deficient mice,29 and Tgε26 transgenic mice.30 However, all studies mentioned have been performed under SPF conditions. The requirement for T cells has not been demonstrated in a model of monoassociation with a nonpathogenic bacterial strain.

It has been clearly shown that CD4 cells, not CD8 cells, are crucial for the induction of disease in HLA-B27 TG rats. CD4+ T cells were much more efficient than CD8+ T cells in transferring disease into nude HLA-B27 rats.12, 31 In the present study, we observed that an inoculum of CD4+ T cells more than 70% of which were activated (expressing CD45RC at a low level) and less than 30% of which were naive (expressing CD45RC at a high level) induced colitis in B. vulgatus–monoassociated rnu/rnu TG rats that was similar in histologic severity to that of B. vulgatus–monoassociated rnu/+ TG rats. We recently demonstrated that IFN-γ is produced by cecal bacterial lysate-stimulated MLN cells derived from SPF HLA-B27 TG rats with severe colitis.17 Depletion of CD4+ cells but not CD8+ cells resulted in complete loss of IFN-γ secretion by MLNs from TG rats. In addition, CD4+-enriched MLN cells from TG rats but not from non-TG rats produced IFN-γ after stimulation with cecal bacterial lysate-pulsed antigen-presenting cells.17 Combined, these results indicate that cecal bacteria and their products stimulate MLN-derived CD4+ T cells to produce IFN-γ in SPF TG rats. In the present study, MLN cells from B. vulgatus–monoassociated rnu/+ TG rats stimulated with B. vulgatus lysates did not produce detectable IFN-γ in vitro. In contrast, after transfer of CD4+ T cells into B. vulgatus–monoassociated rnu/rnu TG rats, MLN cells from recipients secreted IFN-γ. The reason for this discrepancy is unclear. However, a possible explanation comes from studies of lymphoid cell homeostasis.32 In a variety of model systems in which lymphocytes were transferred into lymphopenic hosts such as SCID mice, nude mice, nude rats, and RAG-deficient mice, the responses of the transferred cells were exacerbated compared to the responses that would be seen in hosts with a normal complement of lymphocytes. One explanation for the apparently heightened responses of the transferred cells is that CD4+ T cells proliferate until the lymphoid system reaches homeostasis, in part because of the lack of competition for cytokines and other factors that the cells require.33 As shown in Table 1, rnu/rnu rats are indeed lymphopenic; moreover, 7–9 weeks after i.p. injection of 107 CD4+ cells, the mean total number of MLN cells of the recipients increased by more than 2-fold. We postulate therefore that the lack of homeostatic control of lymphocytes in the MLNs of the B. vulgatus–monoassociated rnu/rnu recipient rats allows proliferation and differentiation of the transferred rnu/+ CD4+ cells. These B. vulgatus–reactive MLN cells then produce IFN-γ after in vitro restimulation with B. vulgatus components.

Although we did not detect IFN-γ after B. vulgatus lysate stimulation of MLN cells from monoassociated rnu/+ rats, colonic tissue isolated from these rats showed expression of IFN-γ mRNA. It is possible that resident intraepithelial lymphocytes, a heterogeneous population of cells including NK cells and TCRγδ-expressing T cells that have been shown to express IFN-γ mRNA and to produce this cytokine, are the source of this activity.34

An immunoregulatory role for cytokine IL-10 has been well established.35 In the present study, we detected essentially equivalent amounts of IL-10 in the supernatants of in vitro–stimulated MLN cells from rnu/rnu TG rats and from rnu/rnu TG recipients of rnu/+ TG CD4-positive cells (Fig. 5B), yet the latter animals developed intestinal inflammation and their MLN cells produced IFN-γ in response to components of B. vulgatus. In separate studies we have demonstrated that TG rat MLN cells are refractory to the inhibitory effects of IL-10.36, 37 Therefore, it is possible that IL-10 is unable to regulate the inflammation that develops in B. vulgatus–monoassociated HLA-B27 transgenic rats.

In conclusion, the results of the present study show that T cells are required for colitis to develop in HLA-B27 TG rats monoassociated with nonpathogenic commensal B. vulgatus. Furthermore, colitis can be transferred by CD4+ T cells into rnu/rnuB. vulgatus–monoassociated TG recipients, indicating that CD4-expressing cells are a critical component for generating the responses we measured. MLN cells from rnu/rnu HLA-B27 TG rats that received CD4+ T cells responded to B. vulgatus lysates in a bacterial species-specific manner. Studies of disease-susceptible gnotobiotic rodents monoassociated with disease-inducing commensal intestinal bacteria will help to unravel the complex microbe–host interactions that lead to chronic immune-mediated intestinal inflammation.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The authors thank Erin Kelly, Emily Kushner, Desmond McDonnell, and Lisa Wiltron for expert technical support at the College of Veterinary Medicine, North Carolina State University, Raleigh, North Carolina. The authors also thank Charlotte Walters, Immunoassay Core, Center of Gastrointestinal Biology and Disease at the University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, and Donna Kronstadt, Gnotobiotic Animal Core, Center of Gastrointestinal Biology and Disease at North Carolina State University, Raleigh, North Carolina, for their expert assistance. The authors are also grateful to Dr. Joel Taurog, University of Texas Southwestern Medical Center, Dallas, Texas, for supplying us with the rnu/rnu HLA-B27 rats.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES