Signaling by lymphotoxin (LT) and TNF is essential for the organogenesis of secondary lymphoid tissues in systemic and mucosal compartments. In this study, we demonstrated that the progeny of mice treated with fusion protein of LTβR and IgGFc (LTβR-Ig) or LTβR-Ig plus TNFR55-Ig (double Ig) showed significantly increased numbers of isolated lymphoid follicles (ILF) in the large intestine. Interestingly, double Ig treatment accelerated the maturation of large intestinal ILF. Three-week-old progeny of double Ig-treated mice showed increased numbers of ILF in the large intestine, but not in the small intestine. Furthermore, alteration of intestinal microflora by feeding of antibiotic water did not affect the increased numbers of ILF in the large intestine of double Ig-treated mice. Most interestingly, mice that developed numerous ILF also had increased levels of activation-induced cytidine deaminase expression and numbers of IgA-expressing cells in the lamina propria of the large intestine. Taken together, these results suggest that ILF formation in the large intestine is accelerated by blockage of LTβR and TNFR55 signals in utero, and ILF, like colonic patches, might play a role in the induction of IgA response in the large intestine.

The gut-associated lymphoid tissues are characterized as the initiation sites for the induction of IgA-mediated immunity and mucosally induced tolerance (1, 2). The mucosal immune system possesses a network of lymphoid organs that are composed of inductive sites (e.g., Peyer’s patches (PP))3 and effector sites (the intraepithelial and the lamina propria (LP) region) (1, 2). It had been believed that PP is the major inductive site for the initiation of Ag-specific IgA responses to a variety of exogenous Ag (3, 4); however, we and others have demonstrated that PP contribute to, but are not essential for, the induction of Ag-specific mucosal IgA responses (5, 6, 7). A recent study revealed the existence of isolated lymphoid follicles (ILF) in the small intestine that resemble PP in terms of architecture and cellular composition (8). The fact that ILF possess germinal centers and an overlying follicle-associated epithelium (FAE) containing M cells suggests their possible role as mucosal inductive sites (8).

Lymphotoxin (LT), a TNF family member, can be found in two forms: a membrane-bound heterotrimer and a soluble homotrimer (9, 10). The membrane-bound heterotrimer is comprised of two β-chains and one α-chain (LTα1β2) and is a ligand for LTβR, while the soluble homotrimer (LTα3) is ligand for both TNFR55 and TNFR75 (11, 12). Unlike the LTα trimer and TNF, which are secreted proteins, LTαβ remains membrane bound and is expressed on the restricted hemopoietic lineage, particularly by T cells, B cells, and NK cells (13). The interaction of LTαβ with LTβR is the critical molecular event triggering secondary lymphoid organ genesis and controlling spleen organization. For example, congenital lack of LTα, LTβ, or LTβR genes disrupted PP and lymph node (LN) organogenesis, and altered splenic architecture as characterized by the absence of distinct T and B cell areas and disruption of the marginal zone (14, 15, 16, 17). Furthermore, administration of LTβR-Ig fusion protein to mice during the selected time window of embryogenesis disrupted LN and PP formation in the progeny (18, 19), suggesting that the molecular interaction of membrane-bound LT with LTβR during the gestational period is essential for the initiation of LN and PP development. In contrast to PP, a recent study demonstrated that ILF formation was not influenced by the blockage of LTβR signaling with LTβR-Ig fusion protein during gestation (8). However, no ILF was found in LTα−/− or aly/aly mice, implying that ILF do require signals dependent on LT and NF-κB-inducing kinase, a critical downstream signaling molecule associated with LTβR, postgestation (8). It has recently been confirmed that, unlike PP formation, ILF formation requires LT-LTβR interaction in adulthood, as well as TNFR55-mediated signaling for their maturation (20).

An additional component of the gut immune system is the colonic patch (CP). The cytoarchitectural components and immune functions of CP and PP were remarkably similar, despite differences in the surrounding environment of mucosa and luminal microbial exposure (21). The presence of organized lymphoid tissue with M cells and germinal centers in CP suggests that Ag uptake and recognition can take place in the rectum (22, 23). Similar to the PP, in utero treatment with LTβR-Ig fusion protein depleted CP formation in progeny (23). These results suggested that PP and CP were developmentally and functionally related components of the small intestine and large intestinal (colonic) immune systems, respectively. In addition to CP, it was shown that ∼50 ILF were dispersed throughout the large intestine of BALB/c mice (8). Recently, it was reported that in utero treatment of mice with LTβR-Ig and TNFR55-Ig fusion proteins caused an increase in the number of submucosal lymphoid patches in the large intestine (24). This suggests that ILF in the small and large intestine are developmentally similar, although little else is known about these immunological structures and functions. In particular, the function of ILF in the large intestine and the precise contribution of LTβR and TNFR55 for their genesis, maturation, and the subsequent induction of IgA responses remain to be elucidated.

In this study, we provide several new findings regarding the unique contribution of the inflammatory cytokines LT and TNF in the genesis and function of ILF in the large intestine. In particular, the tissue genesis signals provided by the cytokine receptors of LTβR and TNFR55 are essential for the postnatal development of large intestinal ILF (LI-ILF). Our present findings suggest that the receptors behave as negative regulators for the genesis of LI-ILF because the blockage of prenatal LT/LTβR and TNF/TNFR55 signaling cascades accelerated the formation and maturation of ILF in the large intestine. Secondly, environmental factors, such as microflora-associated Ags, did not affect the formation and maturation of ILF in the large intestine. Finally, ILF in the large intestine play an important role for IgA+ B cell development.

Timed pregnant BALB/c mice were purchased from Japan CLEA. These mice were maintained in the experimental facility under pathogen-free conditions in the Research Institute for Microbial Diseases at Osaka University and International Vaccine Institute and received sterilized food (certified diet MF; Oriental Yeast) and tap water ad libitum. TNF and LTα double-knockout (TNF/LTα−/− mice; 129 × C57BL/6) mice were kindly provided by H. Bluethmann (Roche Center for Medical Genomics, Basel, Switzerland) (25). Germfree mice (BALB/c Yit) were kindly provided by H. Funabashi (Yakult Central Institute for Microbiological Research, Japan).

Proteins comprised of the extracellular domain of either murine TNFR55 or LTβR fused to the hinge, CH2, and CH3 domains of human IgG1 (LTβR-Ig, TNFR55-Ig, and LFA-3-Ig, respectively) were used in our studies, as described elsewhere (19, 26, 27). Timed pregnant mice were injected i.v. with 200 μg of LTβR-Ig and/or 200 μg of TNFR55-Ig on gestational days 14 and 17, as described previously (5, 19). In some experiments, progeny of mice treated i.v. with LTβR-Ig and TNFR55-Ig on gestational period were further injected i.p. with 20 μg of LTβR-Ig, TNFR55-Ig, or human IgG1 (control) at weekly intervals from 7 days after birth and 50 μg of each Ig fusion protein from 4 wk old until age of 6 wk.

The mononuclear cells from CP and ILF of the large intestine were obtained with modified method, as described previously (8). In brief, the large intestine was opened longitudinally along the mesenteric wall, and mucus and feces were vigorously washed in the RPMI 1640 medium and wiped with filter paper. Subsequently, a section of intestine ∼30 mm long was pasted on a plastic culture dish. The structures of CP and ILF are both circular in appearance, but the CP are larger than triple diameter of the ILF. Morphologically, the center of CP forms protruding configuration, and thus CP appears as dome-shaped tissue under a transillumination seromicroscope (Olympus TH3). In contrast, the ILF are recognized as flat shape. Blind test was conducted to count the number of ILF by three independent investigators by a transillumination seromicroscope. Then CP were taken with two sharp forceps and isolated, and a tiny fragment of ILF was isolated using a sharp needle (23 gauge; inner diameter, μm). CP and ILF were separately digested with collagenase (type IV, 0.5 mg/ml in RPMI 1640 including 2% FBS; Sigma-Aldrich) for 20 min in a 37°C incubator. This step was repeated until the architecture of tissue was totally disrupted. The single cell suspensions were pooled, washed, and placed on a discontinuous 40 and 70% Percoll gradient (Pharmacia). After centrifugation for 20 min at 600 × g, the cells were collected from the interface (8). To isolate the LP lymphocytes from the large intestine, mononuclear cells were dissociated using the collagenase digestion method after removal of CP and ILF, as described previously (28).

A single lymphoid cell suspension was incubated with anti-FcγRII/III mAb (BD Pharmingen) and stained with FITC- or PE-conjugated anti-B220, CD3, IgD, IgM, or IgA mAbs (BD Pharmingen). The other aliquots of cells were incubated with each isotype control mAb, including rat IgG2b or rat IgG2a (BD Pharmingen). The profiles were analyzed using FACScan with CellQuest software (BD Biosciences). Reactivity with peanut agglutinin (PNA) was demonstrated using biotinylated PNA (Vector Laboratories), followed by streptavidin-PE.

For the H&E staining, the large intestine was fixed in 4% paraformaldehyde and embedded in paraffin. The tissues were cut into 5-μm sections and stained with H&E (28). The sections were mounted and viewed under ×20 optics using a digital light microscope. Each of the images was analyzed with Photoshop (Adobe Systems). For the immunohistochemical study, freshly obtained large intestine was rapidly frozen in OCT embedding medium (Tissue-Tek) and stored at −80°C until processing (28). Cryostat sections (5 μm) were fixed in ice-cold acetone for 10 min, dried, and preblocked with anti-FcγRII/III mAb (BD Pharmingen) in PBS. Cells were stained with FITC-conjugated anti-CD11c mAb (BD Pharmingen) and PE-conjugated anti-B220 or CD3 mAbs (BD Pharmingen). The other aliquots of cells were incubated with each isotype control mAb, including hamster IgG, rat IgG2b, or rat IgG2a (BD Pharmingen). IgA-containing cells were visualized by FITC-conjugated anti-IgA mAb (BD Pharmingen). The sections were mounted and viewed under a dual red/green filter by confocal microscopy (Bio-Rad). Each of the images was analyzed with Photoshop (Adobe Systems) in a consistent manner, followed by overlying of the green and red images in the screen mode.

For scanning electron microscopy analysis, large intestinal fragments of the double Ig-treated mice were cleaned of mucus and fixed in 2% glutaraldehyde and 2% paraformaldehyde in PBS containing 100 mM HEPES for 1 h at room temperature. After being washed with PBS, specimens were treated with 1% osmium tetroxide for 1 h at room temperature and then dehydrated in graded ethanol solution. Dehydrated tissues were critical point dried with CO2, and sputter coated and observed with a scanning electron microscope (Hitachi).

For the antibiotic treatment, each group of 6-wk-old progenies was given antibiotics in drinking water for a period of 4 wk. The antibiotic water contained 500 mg/L ampicillin, 1 g/L neomycin sulfate, and 2 g/L of streptomycin (29).

A standard quantitative RT-PCR protocol was used for the study for GAPDH-based quantitative RT-PCR (30). Total RNA was extracted from mononuclear cells isolated from LP of the large intestine or CP of naive mice or ILF of double Ig-treated mice by using the RNeasy mini kit (Qiagen), according to the manufacturer’s protocol. A total of 1 μg of total RNA was reverse transcribed into cDNA using Taq Man reverse transcription kit (Applied Biosystems). Activation-induced cytidine deaminase (AID) mRNA levels were measured by real-time quantitative PCR method performed on the ABI PRISM 7500 (Applied Biosystems). For each treatment, two distinct amplifications were conducted in parallel to amplify AID cDNA and GAPDH cDNA. The amplification reactions were performed in 25 μl vol containing 100 ng of cDNA per treatment, 12.5 μl of 2× TaqMan Universal PCR Master Mix (Applied Biosystems), and 1.25 μl of 20× Assays-on-Demand Gene Expression probe for AID (Applied Biosystems) or TaqMan GAPDH probe (Applied Biosystems). AID mRNA levels from each treatment were normalized to the corresponding amount of GAPDH mRNA levels. Water controls and samples without PCR mixtures were set up to eliminate the possibility of significant DNA contamination.

An ELISPOT assay was adopted to detect total numbers of IgA Ab-forming cells in the large intestine, as described previously (31).

The data are expressed as the mean ± SE and compared using t test in Microsoft Excel Program.

Using LTβR- and TNFR55-Ig fusion proteins as soluble antagonists, we tested whether LTαβ- and TNF-mediated signals influenced the formation of organized lymphoid tissues in the large intestine. For this purpose, mice were treated with TNFR55-Ig or LTβR-Ig or LTβR-Ig plus TNFR55-Ig (double Ig) fusion protein at gestational days 14 and 17. The exposure to LTβR-Ig or double Ig during the gestation period disrupted CP formation in the progeny, which confirmed our previous finding (23). Unexpectedly, however, careful light microscopy and H&E staining analysis revealed that numerous ILF existed throughout the mucosa of the large intestine of 6-wk-old mice treated in utero with LTβR-Ig or with double Ig fusion proteins (Fig. 1,A). These changes in ILF development were not seen in the small intestine of these treated mice (6 wk old), but were unique to the large intestine. Interestingly, the progeny of mice treated with double Ig fusion protein in utero possessed more numerous ILF of larger size in the large intestine than those treated with LTβR-Ig alone. When the total number of ILF in the whole large intestine was counted under light microscopy, ∼250 ILF were found in the progeny of mice treated with double Ig during gestation (Fig. 1,B). In contrast, the numbers of ILF in control Ig-treated mice were 50 per large intestine (Fig. 1 B). When mice were treated with TNFR55-Ig alone, no significant changes were seen in the total number of ILF in the large intestine. In all mice, ILF were preferentially located in the distal region of the large intestine (data not shown). An average of total recovered cell numbers of ILF isolated from the large intestine of the progeny of mice treated with double Ig fusion protein was 10-fold higher than control Ig-treated mice (3.0 × 106 vs 0.3 × 106) and 3-fold higher than LTβR-Ig-treated mice (3.0 × 106 vs 1.0 × 106). These observations suggest that ILF formation in the large intestine was accelerated by blockage of prenatal LTβR-mediated signals, and their maturation was further enhanced by the coblockage of TNFR-mediated signals during the selected gestational period.

FIGURE 1.

Effects of prenatal blockage of LT/LTβR and TNF/TNFR55 signal on the formation of ILF in the large intestine. Timed pregnant BALB/c mice were injected i.v. with TNFR55-Ig or LTβR-Ig and/or TNFR55-Ig (double IgG1) on gestational days 14 and 17. Purified human Ig was used as control Ig. A, Morphology of colonic patch (red arrow) and ILF (blue arrow) in the large intestine of each 6-wk-old progeny. The large intestine was opened longitudinally along the mesenteric wall, and a ∼30-mm-long length of intestine was pasted onto the plastic culture dish. The picture was taken under the stereomicroscopy (upper). For the H&E staining, the large intestine was fixed in 4% paraformaldehyde and embedded in paraffin. The sections were viewed under ×20 optics with a digital light microscope (below). The bar indicates 50 μm. B, Total number of ILF in a whole large intestine was counted under a light microscope (left). Each ILF was then taken off with a sharp needle, and mononuclear cells were isolated and counted (right). The results are expressed as the mean ± SE from three mice per group and from a total of three experiments. ∗, p < 0.05, and ∗∗, p < 0.01 when compared with the number of ILF in the large intestine of control Ig-treated mice.

FIGURE 1.

Effects of prenatal blockage of LT/LTβR and TNF/TNFR55 signal on the formation of ILF in the large intestine. Timed pregnant BALB/c mice were injected i.v. with TNFR55-Ig or LTβR-Ig and/or TNFR55-Ig (double IgG1) on gestational days 14 and 17. Purified human Ig was used as control Ig. A, Morphology of colonic patch (red arrow) and ILF (blue arrow) in the large intestine of each 6-wk-old progeny. The large intestine was opened longitudinally along the mesenteric wall, and a ∼30-mm-long length of intestine was pasted onto the plastic culture dish. The picture was taken under the stereomicroscopy (upper). For the H&E staining, the large intestine was fixed in 4% paraformaldehyde and embedded in paraffin. The sections were viewed under ×20 optics with a digital light microscope (below). The bar indicates 50 μm. B, Total number of ILF in a whole large intestine was counted under a light microscope (left). Each ILF was then taken off with a sharp needle, and mononuclear cells were isolated and counted (right). The results are expressed as the mean ± SE from three mice per group and from a total of three experiments. ∗, p < 0.05, and ∗∗, p < 0.01 when compared with the number of ILF in the large intestine of control Ig-treated mice.

Close modal

To establish an exact role of TNFR and LTβR signaling after birth on the accelerated formation of ILF in the large intestine, the progenies of double Ig-treated mice on gestational period were further injected i.p. with LTβR-Ig or TNFR55-Ig or control Ig at weekly intervals from 1 wk after birth to age of 6 wk. Although the numbers of ILF were slightly lower, no significant difference was detected on the numbers of ILF in the large intestine of TNFR55-Ig-treated progenies compared with the control Ig-treated one. Interestingly, however, the large intestine of progenies postnatal treated with LTβR-Ig did not possess any ILF (Fig. 2). Furthermore, we have assessed the presence of ILF in the large intestine of TNF/LTα double-knockout mice that lack signaling pathways through both TNFR55 and LTβR. Interestingly, there are no ILF in the large intestine of those mice (Fig. 2). These findings indicate that the tissue genesis signaling pathway through LTβR, but not TNFR55, is required for the formation of ILF in the large intestine.

FIGURE 2.

Effects of postnatal blockage of LT/LTβR or TNF/TNFR55 signal on the formation of ILF in the large intestine of the progenies treated in utero with the LTβR-Ig and TNFR55-Ig (double Ig). Progeny of mice treated i.v. with double Ig on gestational days 14 and 17 were further injected i.p. with 20 μg of LTβR-Ig or TNFR55-Ig or control Ig at weekly intervals from 7 days after birth and 50 μg of each Ig fusion protein from 4 wk old. Total numbers of ILF in a whole large intestine were counted under the stereomicroscope. The results are expressed as the mean ± SE from three mice per group and from a total of two experiments. N.D., not detectable.

FIGURE 2.

Effects of postnatal blockage of LT/LTβR or TNF/TNFR55 signal on the formation of ILF in the large intestine of the progenies treated in utero with the LTβR-Ig and TNFR55-Ig (double Ig). Progeny of mice treated i.v. with double Ig on gestational days 14 and 17 were further injected i.p. with 20 μg of LTβR-Ig or TNFR55-Ig or control Ig at weekly intervals from 7 days after birth and 50 μg of each Ig fusion protein from 4 wk old. Total numbers of ILF in a whole large intestine were counted under the stereomicroscope. The results are expressed as the mean ± SE from three mice per group and from a total of two experiments. N.D., not detectable.

Close modal

To clarify the cell population in ILF of the large intestine, flow cytometric and immunohistochemical analyses were conducted using 6-wk-old progeny of mice treated in utero with control Ig or double Ig. Similar to the ILF in the small intestine (8, 20), immunohistochemical study revealed that LI-ILF were enriched with B cells (B220+), with a limited frequency of dendritic stromal cells (CD11c+) and T cells (CD3+) (Fig. 3). Their cell population is phenotypically similar to the CP, but LI-ILF possesses more B220+ cells and less CD3+ cells than CP (Fig. 3). A major population of B220+ cells in LI-ILF belongs to the IgD+ and IgM+ cells with some B220+IgA cells (Fig. 3). Similar to the phenotype of CP, LI-ILF possess detectable levels of B220+IgA+ cells, implying a role of LI-ILF as an inductive site for mucosal IgA responses. Interestingly, LI-ILF of mice treated with double Ig fusion protein in utero showed a higher density of CD3+ cells when compared with those of control Ig-treated progeny (Fig. 3). Although the frequency of CD3+ cells is increased in the treated LI-ILF, the follicle is still considered a B cell-enriched tissue because the frequency of B220+ cells outnumbers CD3+ cells (Fig. 3). In addition, flow cytometry analysis showed that LI-ILF of control Ig-treated progeny contained low numbers of germinal center-forming PNA+ B cells; however, blockage of LTβR and TNFR55 signals during gestation enhanced the formation of germinal center-forming PNA+ B cells (Fig. 3). These findings suggest that LI-ILF, similar to ILF in the small intestine, is one of key mucosal inductive tissues for initiation of IgA responses. Furthermore, LTβR- and TNFR55-mediated signal play a critical role in the control of LI-ILF development.

FIGURE 3.

Characterization of mononuclear cells in the LI-ILF and CP of mice in utero treated with LTβR-Ig and TNFR55-Ig fusion proteins, as described in Fig. 1 legend. Immunofluorescence staining (upper picture) and FACS analysis of mononuclear cells isolated from CP and ILF of the large intestine (table). For the immunohistochemical study, the large intestine of 6-wk-old progenies was rapidly frozen, and cryostat sections were stained with FITC-conjugated anti-CD11c mAb and PE-conjugated anti-B220 mAb or PE-conjugated anti-CD3 mAb. The FACS results are expressed as the mean ± SE from three mice per group and from a total of two experiments. The bar indicates 50 μm.

FIGURE 3.

Characterization of mononuclear cells in the LI-ILF and CP of mice in utero treated with LTβR-Ig and TNFR55-Ig fusion proteins, as described in Fig. 1 legend. Immunofluorescence staining (upper picture) and FACS analysis of mononuclear cells isolated from CP and ILF of the large intestine (table). For the immunohistochemical study, the large intestine of 6-wk-old progenies was rapidly frozen, and cryostat sections were stained with FITC-conjugated anti-CD11c mAb and PE-conjugated anti-B220 mAb or PE-conjugated anti-CD3 mAb. The FACS results are expressed as the mean ± SE from three mice per group and from a total of two experiments. The bar indicates 50 μm.

Close modal

Because we found large numbers of ILF in the large intestine of 6-wk-old progeny following the gestational blockage of LTβR- and TNFR55-mediated signals, we further examined the effect of double Ig treatment on kinetics of their development. For this purpose, we compared ILF number in the small and large intestine of progeny of mice treated with control Ig or double Ig at age of 3, 10, and 24 wk. When the control Ig-treated mice were examined, the numbers of ILF in both small and large intestine were gradually increased (Fig. 4). Interestingly, maximum numbers of LI-ILF were already reached as early as 3 wk old among progeny of mice treated with double Ig (Fig. 4). The total numbers of ILF were then maintained up to 24 wk old. In contrast, ILF in the small intestine of progeny treated with control Ig or double Ig fusion protein gradually developed and reached the maximum numbers at the age of 24 wk old. It was also noted that total numbers of small intestinal ILF were higher in mice treated with the double Ig in utero when compared with the control Ig-treated mice. The fact that the accelerated ILF formation in the large intestine by gestational blockage of LTβR and TNFR55 signals occurred very early after birth implied that their formation and development might not be influenced by exogenous environmental factors, such as gut microflora.

FIGURE 4.

Chronological influences on the increased LI-ILF development in mice in utero treated with LTβR-Ig and TNFR55-Ig (double Ig). Numbers of ILF in the small and large intestines (SI and LI, respectively) of mice treated in utero with control Ig or double Ig fusion protein under the stereomicroscopy, as described in Fig. 1 legend. For the H&E staining, the sections were viewed under ×20 optics with a digital light microscope. The progenies were sacrificed, and the total number of ILF in a whole small and large intestine at 3, 10, and 24 wk old was counted. The results are expressed the mean ± SE from three mice per group and from a total of two experiments. ∗, p < 0.05, and ∗∗, p < 0.01 when compared with the number of ILF of the 3-wk-old progeny.

FIGURE 4.

Chronological influences on the increased LI-ILF development in mice in utero treated with LTβR-Ig and TNFR55-Ig (double Ig). Numbers of ILF in the small and large intestines (SI and LI, respectively) of mice treated in utero with control Ig or double Ig fusion protein under the stereomicroscopy, as described in Fig. 1 legend. For the H&E staining, the sections were viewed under ×20 optics with a digital light microscope. The progenies were sacrificed, and the total number of ILF in a whole small and large intestine at 3, 10, and 24 wk old was counted. The results are expressed the mean ± SE from three mice per group and from a total of two experiments. ∗, p < 0.05, and ∗∗, p < 0.01 when compared with the number of ILF of the 3-wk-old progeny.

Close modal

Several previous studies have demonstrated a critical role of gut microflora on the ILF formation in the small intestine (8, 20, 29). To determine the potential involvement of intestinal microflora on the ILF hyperplasia in the large intestine following the prenatal blockage of LTβR and TNFR55 signals, we altered the microflora by oral administration of antibiotic water, which reduced both aerobic and anaerobic bacteria (29). Both groups of 6-wk-old progenies were fed antibiotic water for 4 wk and sacrificed at 2 and 4 wk after feeding. To see the effectiveness of the antibiotic treatment, the bacteria load was determined in the feces of control Ig- and double Ig-treated mice before and after antibiotic treatment. Essentially, no bacteria were remaining after the feeding of antibiotic water (data not shown). As shown in Fig. 5,A, the antibiotic water treatment drastically decreased ILF formation in the small intestine in both groups of mice treated in utero with control Ig or double Ig. However, strikingly, LI-ILF formation in the double Ig-treated progeny was not influenced by the antibiotic treatment, and maximum numbers of LI-ILF were not changed even after prolonged antibiotic water treatment (Fig. 5,A). Furthermore, the numbers of LI-ILF were comparable in the germfree mice to those of wild-type mice (Fig. 5 B). These results indicate that development of ILF in the large intestine is not influenced by the gut bacterial flora.

FIGURE 5.

Effects of gut microflora on the formation and development of ILF in the small and large intestine. A, Both groups of 6-wk-old mice treated with control or LTβR-Ig and TNFR55-Ig (double Ig) fusion protein in utero were fed for 4 wk with antibiotic water containing ampicillin, neomycin sulfate, and streptomycin. At 2 wk (8 wk old) or 4 wk (10 wk old) after antibiotic treatment, the total number of ILF in a whole small and large intestine was counted under light microscope, as described in the Fig. 1 legend. B, The total number of ILF in the large intestine of germfree mice was counted. The results are expressed as the mean ± SE from three mice per group and from a total of three experiments. ∗, p < 0.05 when compared with non-antibiotic-treated mice.

FIGURE 5.

Effects of gut microflora on the formation and development of ILF in the small and large intestine. A, Both groups of 6-wk-old mice treated with control or LTβR-Ig and TNFR55-Ig (double Ig) fusion protein in utero were fed for 4 wk with antibiotic water containing ampicillin, neomycin sulfate, and streptomycin. At 2 wk (8 wk old) or 4 wk (10 wk old) after antibiotic treatment, the total number of ILF in a whole small and large intestine was counted under light microscope, as described in the Fig. 1 legend. B, The total number of ILF in the large intestine of germfree mice was counted. The results are expressed as the mean ± SE from three mice per group and from a total of three experiments. ∗, p < 0.05 when compared with non-antibiotic-treated mice.

Close modal

To determine whether LI-ILF possess the ability of Ag uptake from the lumen of intestine, we examined the presence of M cells on the FAE of LI-ILF from the double Ig-treated mice. As indicated in Fig. 6 A, LI-ILF revealed a hallmark feature of M cells, i.e., a depressed surface with short and irregular microvilli. These results suggest that LI-ILF might play as Ag sampling site as like as the other organized mucosa-associated lymphoid tissues, e.g., colonic patches.

FIGURE 6.

A, The presence of M cells on the FAE of LI-ILF of mice prenatally treated with LTβR-Ig and TNFR55-Ig (double Ig) fusion protein. B and C, The mRNA expression of AID and the number of IgA Ab-secreting cells in the large intestine of mice treated with control Ig or double Ig fusion protein in utero. A quantitative RT-PCR was performed using mononuclear cells isolated from LP in the large intestine of control Ig-treated mice and from ILF in the large intestine of double Ig-treated mice. As positive control for a quantitative RT-PCR, CP of control Ig-treated mice were adopted. For the immunohistochemical study (C, left), the large intestine of 6-wk-old progenies of both groups were rapidly frozen, and cryostat sections were stained with FITC-conjugated anti-IgA. ELISPOT assay was adopted to further confirm the number of IgA Ab-forming cells in the LP of the large intestine of mice treated with control Ig (▦) or double Ig (▪) fusion protein in utero (C, right). The results are expressed as the mean ± SE from five mice per group and from a total of two experiments. The bar indicates 10 μm (A, left) and 2.5 μm (A, right).

FIGURE 6.

A, The presence of M cells on the FAE of LI-ILF of mice prenatally treated with LTβR-Ig and TNFR55-Ig (double Ig) fusion protein. B and C, The mRNA expression of AID and the number of IgA Ab-secreting cells in the large intestine of mice treated with control Ig or double Ig fusion protein in utero. A quantitative RT-PCR was performed using mononuclear cells isolated from LP in the large intestine of control Ig-treated mice and from ILF in the large intestine of double Ig-treated mice. As positive control for a quantitative RT-PCR, CP of control Ig-treated mice were adopted. For the immunohistochemical study (C, left), the large intestine of 6-wk-old progenies of both groups were rapidly frozen, and cryostat sections were stained with FITC-conjugated anti-IgA. ELISPOT assay was adopted to further confirm the number of IgA Ab-forming cells in the LP of the large intestine of mice treated with control Ig (▦) or double Ig (▪) fusion protein in utero (C, right). The results are expressed as the mean ± SE from five mice per group and from a total of two experiments. The bar indicates 10 μm (A, left) and 2.5 μm (A, right).

Close modal

To assess the ability of IgA class switching in the LI-ILF, the expression levels of AID mRNA, which plays an essential role in class switching recombination and somatic hypermutation of Ig genes, were determined (32). The mononuclear cells isolated from LI-ILF of mice treated with double Ig in utero expressed AID mRNA (Fig. 6,B). In contrast to ILF, AID mRNA was not detectable in the diffused LP region of the large intestine (Fig. 6,B). These findings suggest that the levels of μ to α class switching are increased in the large intestine of mice treated with double Ig in utero due to the maximum increase of ILF numbers. Thus, one can predict the subsequent elevation of IgA-producing cells in the large intestine of these mice. The numbers of IgA-producing cells were therefore assessed, using the large intestine of the 6-wk-old mice treated in utero with control Ig or double Ig. Interestingly, increased numbers of IgA-expressing cells were detected in the LP region of the large intestine of in utero double Ig-treated mice when compared with the control mice (Fig. 6,C, left). The results were also confirmed by the analysis of single cells using ELISPOT assay. The numbers of IgA-producing cells were increased in mononuclear cells isolated from the large intestine of mice treated with double Ig in utero when compared with the control Ig-treated mice (Fig. 6 C, right). These results further support a notion that ILF may play a role for induction of IgA+ B cells in the large intestine.

In general, a family of inflammatory cytokine-mediated signals provided via LTβR and TNFR55 is considered to be critical for the organogenesis of secondary lymphoid tissue (18, 33). One can consider the organogenic steps that take place during development as a form of programmed inflammation, given what we now know about the steps underlying the genesis of tertiary lymphoid structures in chronic disease settings (34). In this study, we demonstrated that prenatal blockage of LT/LTβR signaling cascade resulted in the acceleration of ILF formation in the large intestine, and further that prenatal blockage of TNF/TNFR55 signal enhanced their maturation. The hyperplasia of LI-ILF was not due to stimulation of exogenous environmental stimuli, such as microflora Ags. Most interestingly, LI-ILF expressed AID mRNA, and therefore might be critically involved in the generation of IgA-committed B cells. Thus, our present study is the first one to show the unique characteristics of LI-ILF as IgA-inductive sites, whose development is accelerated by the blockage of LTβR and TNFR55 in utero.

Normal numbers of ILF were present in the small intestine of mice treated with LTβR-Ig fusion protein in utero, but were absent in the LTα−/− mice and aly/aly−/− mice, suggesting that ILF formation in the small intestine did not require gestational LT/LTβR-dependent event, but needed the postnatal signals (8). In addition, a recent study demonstrated that, unlike PP, postnatal LT/LTβR signals are required for ILF formation in the small intestine, and additional TNF/TNFR55 signal and exogenous stimuli were needed for their maturation (20). In contrast, limited information is currently available on the immunological function and tissue genesis of ILF (or solitary lymphoid aggregates) in the large intestine. For example, in utero treatment with LTβR-Ig ablated the formation of CP, but scattered B cell aggregates in the mucosal layer of large intestine were retained (23). Furthermore, the progeny of mice treated with LTβR-Ig plus TNFR55-Ig fusion protein were reported to have some submucosal lymphoid patches in the large intestine (24). In the present study, we demonstrate that prenatal blockage of LT/LTβR signal enhanced the formation of LI-ILF, and prenatal blockage of LT/LTβR plus TNF/TNFR signals further accelerated this phenomenon and resulted in extreme hyperplasia of LI-ILF. It is clear that LI-ILF used LT/LTβR and TNF/TNFR signals in a different manner than other gut-associated lymphoid tissues such as PP and CP. An interesting possibility is that LTβR- and TNFR-mediated signaling might behave as a negative regulation for the LI-ILF genesis in the gestational period.

Several recent studies demonstrated a critical involvement of gut flora on the development of ILF in the small intestine. It has been reported that normal numbers of ILF were detected in the small intestine of germfree mice (8). However, another study showed that the development of ILF in the small intestine did not occur in germfree mice, but, if germfree mice were conventionalized, modest number of mature ILF developed (20). Further augmentation of ILF formation in the small intestine was induced by in utero treatment with LTβR-Ig fusion protein (20). Fagarasan et al. (29) recently revealed that alteration of bacteria flora by antibiotic treatment abolished ILF hyperplasia and germinal center enlargement in the small intestine that was provoked by the genetic deficiency of AID. Furthermore, the number of anaerobic bacteria was 100-fold increased in the small intestines of AID−/− mice than AID+/−; however, no significant changes were detected in the large intestine of AID−/− mice (29). Those results are consistent with our present results showing that the total number of ILF in the small intestine of naive mice and mice treated with double Ig in utero was significantly reduced by antibiotic treatment. However, the formation of LI-ILF of both naive and the double Ig-treated mice was not influenced by the antibiotic treatment. The fact that extreme hyperplasia of ILF in the large intestine of double Ig-treated mice was not due to stimulation of gut microflora suggests that bacterial Ag may not be involved in the development and maturation of LI-ILF. Hence, although the immunological nature of ILF in the small and large intestine seems identical in terms of cell populations and morphologic features, regulatory factors associated with programmed inflammation for their tissue genesis and maturation might be significantly different due to the different exogenous environments. We are investigating the exact regulatory factors that are specifically involved in the development of LI-ILF. In particular, an understanding of the identification and ontogeny of the cells required to induce local LI-ILF formation will be critical, as it appears these cells require LTβR and TNFR signals during embryogenesis that regulate their activity. Furthermore, the identity and regulation of the specific molecules critical for induction of LI-ILF need further study, as it is likely that these are further involved in the regulation of colon-associated diseases.

We have also demonstrated that ILF are abundantly developed in the large intestine in the absence of CP. These plentiful ILF that developed in CP-null large intestine could be a key site for the continuous generation of IgA-committed B cells. Interestingly, our present study showed mononuclear cells isolated from CP and ILF of normal mice, but not from intestinal LP, expressed high levels of AID mRNA that play an essential role for isotype switching recombination and somatic hypermutation of Ig. Furthermore, enhanced numbers of IgA-producing cells were noted in the LP region of large intestine of progenies with numerous numbers of ILF, but no CP by the gestational blockage of LT/LTβR and TNF/TNFR55 signaling pathways. Based upon their cell phenotype and micro- and macromorphology, it is reasonable to classify LI-ILF as an inductive site for intestinal IgA production in addition to CP. Therefore, LI-ILF and CP may form a reciprocal inductive network in mucosal immunity. Thus, the increased IgA response seen in the CP-null mice could be explained as a compensatory response in the absence of CP. Taken together, both ILF and CP in the large intestine are integrated inductive tissues that can compensate each other for the induction of mucosal IgA responses.

A possible contribution of LI-ILF to the development of colon-restricted inflammatory disease is just beginning to be understood. A previous study described that elimination of PP and CP, but not scattered aggregates of B cells by in utero treatment with LTβR-Ig fusion protein, resulted in the prevention of trinitrobenzenesulfonic acid-induced Th2 cell type colitis development (23). In contrast, another group found a more severe type of dextran sodium sulfate-induced colitis in PP- and mesenteric lymph node-null mice generated by the in utero treatment with LTβR-Ig and TNFR55-Ig fusion proteins (24). Although these mice did not possess peripheral lymphoid tissue, they found that there were submucosally located lymphoid follicles in the large intestine consisting of B and T cell areas. Furthermore, dextran sodium sulfate-induced colitis accelerated additional formation of these lymphoid follicles (24). In all cases of ulcerative colitis patients, the colonic LP contain numerous basal lymphoid aggregates composed of T and B lymphocytes and dendritic cells (35). Furthermore, these lymphoid aggregates increase in number and size with severity of disease (35). Overall, it seems likely that LI-ILF could be site for the generation of regulatory and/or pathogenic lymphocytes. Thus, both in human patients and in mouse disease models, LI-ILF hyperplasia is associated with colonic disease. Furthermore, whether ILF induces regulatory type responses or pathogenic type responses may depend on surrounding environmental and immunological conditions. To further address these questions, our murine model of LI-ILF hyperplasia will be useful to clarify their precise role in the control of large intestine-restricted diseases.

The authors have no financial conflict of interest.

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

1

This work is supported by grants from Core Research for Engineering, Science, and Technology of Japan Science and Technology Agency; the Ministry of Education, Science, Sports, and Culture; and the Ministry of Health and Welfare in Japan; and supported by Special Coordination Funds of the Ministry of Education, Culture, Sports, Science, and Technology; the Japanese Government; as well as a Science Research Center fund to the Immunomodulation Center at University of Ulsan from the Korean Science Engineering Foundation and the Korean Ministry of Science and Technology.

3

Abbreviations used in this paper: PP, Peyer’s patch; AID, activation-induced cytidine deaminase; CP, colonic patch; FAE, follicle-associated epithelium; ILF, isolated lymphoid follicle; LI-ILF, large intestinal ILF; LN, lymph node; LP, lamina propria; LT, lymphotoxin; PNA, peanut agglutinin.

1
Kiyono, H., J. Bienenstock, J. R. McGhee, P. B. Ernst.
1992
. The mucosal immune system: features of inductive and effector sites to consider in mucosal immunization and vaccine development.
Reg. Immunol.
4
:
54
.
2
McGhee, J. R., H. Kiyono.
1999
. The mucosal immune system. W. E. Paul, ed.
Fundamental Immunology
4th Ed.
909
. Lippincott-Raven Publishers, New York.
3
Keren, D. F., P. S. Holt, H. H. Collins, P. Gemski, S. B. Formal.
1978
. The role of Peyer’s patches in the local immune response of rabbit ileum to live bacteria.
J. Immunol.
120
:
1892
.
4
Craig, S. W., J. J. Cebra.
1971
. Peyer’s patches: an enriched source of precursors for IgA-producing immunocytes in the rabbit.
J. Exp. Med.
134
:
188
.
5
Yamamoto, M., P. Rennert, J. R. McGhee, M. N. Kweon, S. Yamamoto, T. Dohi, S. Otake, H. Bluethmann, K. Fujihashi, H. Kiyono.
2000
. Alternate mucosal immune system: organized Peyer’s patches are not required for IgA responses in the gastrointestinal tract.
J. Immunol.
164
:
5184
.
6
Kang, H. S., R. K. Chin, Y. Wang, P. Yu, J. Wang, K. A. Newell, Y. X. Fu.
2002
. Signaling via LTβR on the lamina propria stromal cells of the gut is required for IgA production.
Nat. Immunol.
3
:
576
.
7
Fagarasan, S., K. Kinoshita, M. Muramatsu, K. Ikuta, T. Honjo.
2001
. In situ class switching and differentiation to IgA-producing cells in the gut lamina propria.
Nature
413
:
639
.
8
Hamada, H., T. Hiroi, Y. Nishiyama, H. Takahashi, Y. Masunaga, S. Hachimura, S. Kaminogawa, H. Takahashi-Iwanaga, T. Iwanaga, H. Kiyono, et al
2002
. Identification of multiple isolated lymphoid follicles on the antimesenteric wall of the mouse small intestine.
J. Immunol.
168
:
57
.
9
Smith, C. A., T. Farrah, R. G. Goodwin.
1994
. The TNF receptor superfamily of cellular and viral proteins: activation, costimulation, and death.
Cell
76
:
959
.
10
Ware, C. F., T. L. VanArsdale, P. D. Crowe, J. L. Browning.
1995
. The ligands and receptors of the lymphotoxin system.
Curr. Top. Microbiol. Immunol.
198
:
175
.
11
Vandenabeele, P., W. Declercq, B. Vanhaesebroeck, J. Grooten, W. Fiers.
1995
. Both TNF receptors are required for TNF-mediated induction of apoptosis in PC60 cells.
J. Immunol.
154
:
2904
.
12
Gruss, H. J., S. K. Dower.
1995
. Tumor necrosis factor ligand superfamily: involvement in the pathology of malignant lymphomas.
Blood
85
:
3378
.
13
Browning, J. L., I. D. Sizing, P. Lawton, P. R. Bourdon, P. D. Rennert, G. R. Majeau, C. M. Ambrose, C. Hession, K. Miatkowski, D. A. Griffiths, et al
1997
. Characterization of lymphotoxin-αβ complexes on the surface of mouse lymphocytes.
J. Immunol.
159
:
3288
.
14
De Togni, P., J. Goellner, N. H. Ruddle, P. R. Streeter, A. Fick, S. Mariathasan, S. C. Smith, R. Carlson, L. P. Shornick, J. Strauss-Schoenberger.
1994
. Abnormal development of peripheral lymphoid organs in mice deficient in lymphotoxin.
Science
264
:
703
.
15
Alimzhanov, M. B., D. V. Kuprash, M. H. Kosco-Vilbois, A. Luz, R. L. Turetskaya, A. Tarakhovsky, K. Rajewsky, S. A. Nedospasov, K. Pfeffer.
1997
. Abnormal development of secondary lymphoid tissues in lymphotoxin β-deficient mice.
Proc. Natl. Acad. Sci. USA
94
:
9302
.
16
Futterer, A., K. Mink, A. Luz, M. H. Kosco-Vilbois, K. Pfeffer.
1998
. The lymphotoxin β receptor controls organogenesis and affinity maturation in peripheral lymphoid tissues.
Immunity
9
:
59
.
17
Koni, P. A., R. Sacca, P. Lawton, J. L. Browning, N. H. Ruddle, R. A. Flavell.
1997
. Distinct roles in lymphoid organogenesis for lymphotoxins α and β revealed in lymphotoxin β-deficient mice.
Immunity
6
:
491
.
18
Rennert, P. D., J. L. Browning, P. S. Hochman.
1997
. Selective disruption of lymphotoxin ligands reveals a novel set of mucosal lymph nodes and unique effects on lymph node cellular organization.
Int. Immunol.
9
:
1627
.
19
Rennert, P. D., J. L. Browning, R. Mebius, F. Mackay, P. S. Hochman.
1996
. Surface lymphotoxin α/β complex is required for the development of peripheral lymphoid organs.
J. Exp. Med.
184
:
1999
.
20
Lorenz, R. G., D. D. Chaplin, K. G. McDonald, J. S. McDonough, R. D. Newberry.
2003
. Isolated lymphoid follicle formation is inducible and dependent upon lymphotoxin-sufficient B lymphocytes, lymphotoxin β receptor, and TNF receptor I function.
J. Immunol.
170
:
5475
.
21
Owen, R. L., A. J. Piazza, T. H. Ermak.
1991
. Ultrastructural and cytoarchitectural features of lymphoreticular organs in the colon and rectum of adult BALB/c mice.
Am. J. Anat.
190
:
10
.
22
Dohi, T., K. Fujihashi, P. D. Rennert, K. Iwatani, H. Kiyono, J. R. McGhee.
1999
. Hapten-induced colitis is associated with colonic patch hypertrophy and T helper cell 2-type responses.
J. Exp. Med.
189
:
1169
.
23
Dohi, T., P. D. Rennert, K. Fujihashi, H. Kiyono, Y. Shirai, Y. I. Kawamura, J. L. Browning, J. R. McGhee.
2001
. Elimination of colonic patches with lymphotoxin β receptor-Ig prevents Th2 cell-type colitis.
J. Immunol.
167
:
2781
.
24
Spahn, T. W., H. Herbst, P. D. Rennert, N. Lugering, C. Maaser, M. Kraft, A. Fontana, H. L. Weiner, W. Domschke, T. Kucharzik.
2002
. Induction of colitis in mice deficient of Peyer’s patches and mesenteric lymph nodes is associated with increased disease severity and formation of colonic lymphoid patches.
Am. J. Pathol.
161
:
2273
.
25
Eugster, H. P., M. Muller, U. Karrer, B. D. Car, B. Schnyder, V. M. Eng, G. Woerly, M. Le Hir, F. di Padova, M. Aguet, et al
1996
. Multiple immune abnormalities in tumor necrosis factor and lymphotoxin-α double-deficient mice.
Int. Immunol.
8
:
23
.
26
Force, W. R., B. N. Walter, C. Hession, R. Tizard, C. A. Kozak, J. L. Browning, C. F. Ware.
1995
. Mouse lymphotoxin-β receptor: molecular genetics, ligand binding, and expression.
J. Immunol.
155
:
5280
.
27
Miller, G. T., P. S. Hochman, W. Meier, R. Tizard, S. A. Bixler, M. D. Rosa, B. P. Wallner.
1993
. Specific interaction of lymphocyte function-associated antigen 3 with CD2 can inhibit T cell responses.
J. Exp. Med.
178
:
211
.
28
Kweon, M. N., M. Yamamoto, M. Kajiki, I. Takahashi, H. Kiyono.
2000
. Systemically derived large intestinal CD4+ Th2 cells play a central role in STAT6-mediated allergic diarrhea.
J. Clin. Invest.
106
:
199
.
29
Fagarasan, S., M. Muramatsu, K. Suzuki, H. Nagaoka, H. Hiai, T. Honjo.
2002
. Critical roles of activation-induced cytidine deaminase in the homeostasis of gut flora.
Science
298
:
1424
.
30
Bunda, S., N. Kaviani, A. Hinek.
2005
. Fluctuations of intracellular iron modulate elastin production.
J. Biol. Chem.
280
:
2341
.
31
Fujihashi, K., J. R. McGhee, M. N. Kweon, M. D. Cooper, S. Tonegawa, I. Takahashi, T. Hiroi, J. Mestecky, H. Kiyono.
1996
. γ/δ T cell-deficient mice have impaired mucosal immunoglobulin A responses.
J. Exp. Med.
183
:
1929
.
32
Muramatsu, M., K. Kinoshita, S. Fagarasan, S. Yamada, Y. Shinkai, T. Honjo.
2000
. Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA editing enzyme.
Cell
102
:
553
.
33
Rennert, P. D., D. James, F. Mackay, J. L. Browning, P. S. Hochman.
1998
. Lymph node genesis is induced by signaling through the lymphotoxin β receptor.
Immunity
9
:
71
.
34
Kratz, A., A. Campos-Neto, M. Hanson, N. Ruddle.
1996
. Chronic inflammation caused by lymphotoxin is lymphoid neogenesis.
J. Exp. Med.
183
:
1461
.
35
Yeung, M. M., S. Melgar, V. Baranov, A. Oberg, A. Danielsson, S. Hammarstrom, M. L. Hammarstrom.
2000
. Characterization of mucosal lymphoid aggregates in ulcerative colitis: immune cell phenotype and TcR-γδ expression.
Gut
47
:
215
.