M cells are a kind of intestinal epithelial cell in the follicle-associated epithelium of Peyer's patches. These cells can transport antigens and microorganisms into underlying lymphoid tissues. Despite the important role of M cells in mucosal immune responses, the origin and mechanisms of differentiation as well as cell death of M cells remain unclear. To clarify the mechanism of M cell differentiation, we established a novel murine intestinal epithelial cell line (MIE) from the C57BL/6 mouse. MIE cells grow rapidly and have a cobblestone morphology, which is a typical feature of intestinal epithelial cells. Additionally, they express cytokeratin, villin, cell-cell junctional proteins, and alkaline phosphatase activity and can form microvilli. Their expression of Musashi-1 antigen indicates that they may be close to intestinal stem cells or transit-amplifying cells. MIE cells are able to differentiate into the M cell lineage following coculture with intestinal lymphocytes, but not with Peyer's patch lymphocytes (PPL). However, PPL costimulated with anti-CD3/CD28 MAbs caused MIE cells to display typical features of M cells, such as transcytosis activity, the disorganization of microvilli, and the expression of M cell markers. This transcytosis activity of MIE cells was not induced by T cells isolated from PPL costimulated with the same MAbs and was reduced by the depletion of the T cell population from PPL. A mixture of T cells treated with MAbs and B cells both from PPL led MIE cells to differentiate into M cells. We report here that MIE cells have the potential ability to differentiate into M cells and that this differentiation required activated T cells and B cells.
- M cell
- MIE cell
- follicle-associated epithelium
- Peyer's patch
- small intestine
the gastrointestinal tract has four principal epithelial cell lineages, each of which has a specific location in the mucosal epithelium: absorptive enterocytes in the small intestine comprise the principal epithelial cell lineage of the intestinal mucosa; mucin secreting cells are a type of goblet cell in the small intestine and colon; enteroendocrine cells secrete peptide hormones; and Paneth cells in the small intestine contain large apical secretory granules and express specific proteins, including lysozyme, tumor necrosis factor, and the anti-bacterial cryptdin molecules (3, 5, 41, 54). Another less common cell lineage is the M (microfold) cell. These functional intestinal epithelial cells are clonal populations derived from a single stem cell in the crypts (5, 9, 41). Wnt/β-catenin (33, 49) and Notch (22, 64, 67) were reported to the signaling pathways related to the intestinal epithelial differentiation. It is likely that an improved understanding of the molecular pathways that regulate the proliferation and differentiation of intestinal epithelial cells will provide a clearer insight into the location and behavior of intestinal epithelial cells. The crucial mechanisms determining the fate of these cells has yet to be elucidated completely, however.
To clarify the differentiation mechanisms of functional intestinal epithelial cells, many intestinal cell lines have been established. These include, for example, human colon adenocarcinoma cell lines (HT29, HRA-19, T84, LIM1863, and Caco-2) (32, 40, 66), normal human intestinal epithelial crypt cells (HIEC) (48), a rat intestinal cell line (IEC-6) (52), porcine intestinal epithelial cell line (IPEC-J2) (58), and transimmortalized mouse intestinal cells (m-ICcl2) (2). Although HIEC and IEC-6 have normal phenotypic properties, most of the established cell lines arise from carcinoma tissue and are immortalized. Therefore, these models are constrained by this theoretical limitation.
M cells, the fifth intestinal epithelial cell lineage, are highly specialized cells within the follicle-associated epithelium (FAE) of the intestinal Peyer's patches (PP). They can transport antigens and microorganisms into the underlying lymphoid tissues and have an important role in mucosal immune responses. In this way, antigen-presenting cells in the mucosal immune system encounter a variety of antigens that enter the body through the gut mucosa (34). Despite the important role of M cells in mucosal defense, the origin, differentiation, and cell death mechanisms of M cells remain unclear.
Some investigators have proposed two hypotheses with respect to the differentiation of M cells. One is that M cells arise and segregate directly from stem cells of the crypt in the same way as other epithelial cells (14, 36, 42), and the other is that mature enterocytes convert to the M cell phenotype under the influence of lymphocytes or microorganisms (4, 30). We previously reported that the proliferating M cell marker-positive cells were observed in FAE crypts, which were not associated with lymphocytes under the basolateral membrane (42). The number of M cells has also been reported to decrease at the apical end of the FAE (42, 60). These investigators have suggested the possibility that M cells revert to enterocytes at a position near to the FAE apex, because they failed to detect apoptotic M cells. A specific marker for M cells in BALB/c mice is α-l-fucose, which can be detected by lectin Ulex europaeus type I (UEA I) (6, 16, 59). M cells in rats, pigs, rabbits, and calves are detected by cytoskeletal proteins, such as cytokeratin 8 (35, 53), cytokeratin 18 (15), vimentin (23), and actin and villin (25), respectively. The lack of alkaline phosphatase (ALP) activity on M cells has also been used to identify them in some species (24, 46, 47, 61). However, a general marker for M cells for all species has not been established. This has possibly led to the controversies surrounding the origin and mechanism of differentiation of M cells.
Recently, an in vitro M cell model has been established to clarify the mechanism of M cell differentiation. After coculture with Peyer's patch lymphocytes (PPL) or a B cell line, Caco-2 cells and m-ICcl2 cells were able to differentiate into M cells (12, 30). Various experiments have been done using this in vitro M cell model (18, 37). However, the crucial factors for M cell differentiation remain unclear. In this study, we have established a novel murine intestinal epithelial cell line (MIE) and have used it to investigate the characteristics and normal phenotypic properties of MIE cells. We report here that MIE cells have the potential ability to differentiate into M cells and that they exhibit the ontogeny and characteristics of M cells.
MATERIALS AND METHODS
Male C57BL/6 mice (Japan SLC, Shizuoka, Japan) were used. The mice were kept in cages under a cycle of 12 h light and 12 h dark and were maintained under specific pathogen-free conditions. The experiments were permitted by the Institutional Office and conducted in accordance with the Guidelines for Animal Experimentation in Tohoku University, which were permitted by the Government Committtee.
Establishment of MIE cells.
Small intestines were removed from mice (10 wk old) after slaughter, washed with a sterile 0.1 M phosphate-buffered saline (PBS, pH 7.4), and transferred to serum-free Dulbecco's modified Eagle medium (DMEM, GIBCO, Grand Island, NY) supplemented with penicillin (10 U/ml) and streptomycin (10 mg/ml). Intestinal tissues were cut finely, washed three times with serum-free DMEM, and centrifuged at 200 g for 7 min. The tissue pellet was resuspended in DMEM supplemented with 10% fetal bovine serum (10% FBS-DMEM) and seeded into a collagen-coated flask (Sumilon, Tokyo, Japan).
All cultures were maintained at 37°C and 5% CO2 in a humidified incubator. Cells were treated with a sucrose-EDTA buffer (pH 7.5; 0.45 M sucrose, 0.36% EDTA, 0.1% BSA in PBS) for a few minutes, detached by 0.04% trypsin/PBS, and then collected. Some of the primary culture cells were diluted to 40 cells/ml in 10% FBS-DMEM, and 100 μl was aliquoted into each well of collagen-coated 96-well plates (Sumilon). Each well was microscopically checked for cell growth and monoclonal expansion at 5 days after plating. Wells for which it was certain there was a single colony of rapidly growing cells with epithelial-like morphology were marked and passaged when they reached ∼70% confluence. Thus we established three clonal MIE lines.
Growth rate of MIE cells.
MIE cells were passaged every 3 or 4 days in 10% FBS-DMEM. The passage method of MIE cells was as described above. Cells were seeded into three wells of a six-well plate (Sumilon) at 0.5 × 104 cells/cm2. Cell proliferation was determined in triplicate by the Trypan blue-dye exclusion method with a hemocytometer on each of 5 days after plating. Doubling time (DT) was calculated according the formula DT = (t − t0) log 2/(log N − log N0), where t − t0 is the period of time for the cell growth and N − N0 is the increase in the number of cells.
Mouse anti-pan cytokeratin (CK) monoclonal antibody (MAb) (clone c-11, Sigma, Saint Louis, MO, 1:100), mouse anti-CK 20 MAb (clone Ks20.8, Nichirei, Tokyo, Japan, 1:1), goat anti-villin polyclonal antibody (pAb) (Santa Cruz Biotechnology, Santa Cruz, CA), rabbit anti-zonula occludens-1 (ZO-1) pAb (Zymed Laboratories, San Francisco, CA, 1:50), rabbit anti-claudin 1 pAb (Zymed Laboratories, 1:50), mouse anti-β-catenin MAb (Sigma, 1:100), goat anti-annexin V pAb (Santa Cruz Biotechnology, 1:50), rabbit anti-Sgne-1 (7B2, Quartett, Berlin, Germany), and biotinylated rat anti-Musashi-1 MAb (clone 14H1, 1:200) (25) were used for immunocytochemistry. Mouse anti-pan-CK MAb (1:1,000), anti-CK 20 MAb (1:1) and biotinylated rat anti-Musashi-1 MAb (1: 500) were also used for the immunohistochemistry.
MIE cells were cultured in Sonicseal slide well chambers (Nunc, Roskilde, Denmark) and Transwell inserts (3-μm Transwell filters, Corning, NY). Cells were washed with cold PBS and fixed in methanol-acetone for 7 min at −20°C. Cells were treated with PBS containing 1.5% normal goat serum for 20 min and then incubated with the primary antibodies for 14 h at 4°C. Cells were washed three times with PBS and incubated with the secondary antibodies suitable to each of the primary antibodies for 1 h at room temperature. FITC-conjugated anti-mouse IgG (Sigma, diluted to 1:400), FITC-conjugated anti-rabbit IgG (Sigma, diluted to 1:400) and Alexa Fluor 488 or 594-conjugated chicken anti-goat IgG (Molecular Probes, Leiden, The Netherlands, 1:200) were used as the secondary antibodies. For the visualization of the biotinylated antibody, fluorescein streptavidin (Vector Laboratories, Burlingame, CA, diluted to 1:200) was used. Finally, cells were counterstained with propidium iodide for 5 min. These were examined by the confocal laser microscope (MRC-1024, Bio-Rad, Richmond, CA). To determine the specificity of the immunostaining, negative controls were run in which the primary antibody was omitted.
Intestinal tissues were obtained from 6-wk-old male C57BL/6 mice and fixed with 4% paraformaldehyde solution in 0.1 M phosphate buffer (PB, pH 7.4) for 24 h at 4°C. Tissues were embedded in paraffin. The paraffin sections, 4 μm thickness, were mounted on silane-coated slide glasses and were dewaxed in xylene and rehydrated in a series of graded ethanol solutions and transferred to PBS. The sections were treated with 3% H2O2 in methanol for 5 min to quench endogenous peroxidase activity. For the staining of Musashi-1, the sections were heated in a microwave oven in 0.01 M citrate buffer (pH 6.0) for 10 min to facilitate antigen retrieval. After blocking with 3% BSA-PBS, the sections were incubated with primary antibodies for 14 h at 4°C in a moist chamber and developed by use of an ABC-PO kit (Vector Laboratories) for 1 h at room temperature. For the staining of the cytokeratin, the sections were incubated with 0.05% proteinase (P8038, Sigma) in Tris·HCl buffer (pH 7.6) for 5 min at 37°C for antigen retrieval. After the incubation with the primary antibodies, the sections were reacted with N-histofine MOUSESTAIN KIT (Nichirei). Finally, the sections were visualized by 3,3′-diaminobenzidine tetra-hydrochloride (Dojin Laboratories, Kumamoto, Japan) and then counterstained with Mayer's hematoxylin for 30 s.
Detection of ALP activity.
MIE cells were cultured in the upper chamber of Transwell inserts and fixed with methanol-acetone for 7 min at −20°C. After fixation, cells were rinsed in Tris·HCl buffer (pH 9.5) and incubated with NBT/BCIP substrate at 37°C.
MIE cells were cultured in the upper chamber of Transwell inserts and fixed with 2.5% glutaraldehyde in 0.1 M PB for 1 h. After being washed with 0.1 M PB, cells were dehydrated with a series of graded ethanol solutions and substituted with t-butyl alcohol. The cells were freeze dried and coated with platinum-palladium and examined by scanning electron microscopy (SEM) (S4200, Hitachi, Tokyo, Japan).
The fixed cells were washed with 0.1 M PB and then postfixed with 1% osmium tetroxide in 0.1 M PB. Following dehydration with a series of graded ethanol solutions and substitution with n-butyl glycidyl ether, the cells were embedded in Epon 812 resin (TAAB, Berkshire, UK), sectioned into 80-nm thickness by use of an ultramicrotome and stained with uranyl acetate and lead acetate. The sections were examined by transmission electron microscopy (TEM) (H8100, Hitachi).
TER and paracellular permeability.
MIE cells (9 × 103, 104, and 105 cells) were seeded into the apical chamber of Transwell inserts and cultured for 11 days. Transepithelial electrical resistance (TER) and paracellular permeability were measured everyday, and microsphere transport was measured every other day. TER was measured by use of Millicell-ERS (Millipore, Billerica, MA). The unit of measurement was ohms per centimeter squared, and values were determined by subtracting the blank well TER from the test well TER and dividing this by the surface area of the well. Paracellular permeability was determined by using FITC-dextran (FD-20S, Sigma). FITC-dextran/10% FBS-DMEM (1.0 mg/ml) was added into the apical chamber of the Transwell inserts. After incubation for 3 h, both apical and basal chamber media were collected and their excitations were measured by use of Fluoroskan Ascent (Thermo, Summerland, BC, Canada). Paracellular permeability was quantified by the apical-to-basal flux rates of the FITC-dextran.
Preparation of intestinal lymphocytes and PPL.
Male C57BL/6 (6 wk-old) mice were used. For the preparation of intestinal lymphocytes, small intestines containing PP were removed and rinsed in a sterile PBS. The luminal side of the intestines was washed three times with a 20-ml syringe. Intestines were cut finely, transferred to PBS containing 1 mg/ml collagenase (3410533, Wako, Osaka, Japan) and 1 mg/ml hyaluronidase (H3757, Sigma), and then incubated at 37°C for 10 min in a shaking water bath. The digested tissues were washed with serum-free DMEM and filtered through a 40-μm stainless steel mesh to separate the cells from the tissue fragment. Finally, intestinal lymphocytes were isolated by a one-step gradient centrifugation (Lympholyte, Cedarlane, Hornby, ON, Canada). Intestinal lymphocytes were resuspended at a concentration of 2 × 107 cells/ml in 10% FBS-RPMI (RPMI1640, GIBCO).
For the preparation of PPL, PP were dissected from small intestines, washed in sterile PBS, and cut finely. After centrifugation, the PP were collected and digested by enzymes, as described above. The digested PP were washed with 2% FBS-RPMI (GIBCO) and passed through a mesh to obtain a single cell suspension. PPL were resuspended at a concentration of 2 × 107 cells/ml in 10% FBS-RPMI. For the activation of PPL, 3 μg/ml anti-CD3 MAb (145-2C11, R&D Systems, Minneapolis, MN) and 3 μg/ml anti-CD28 MAb (PV-1, Southern Biotechnology Associates, Birmingham, AL) were added to the cultures of PPL (20).
Freshly isolated intestinal lymphocytes and PPL were collected by centrifugation, washed with PBS containing 1% BSA and 0.1% sodium azide, and immunostained by the direct immunofluorescence method with the following MAbs: FITC anti-mouse CD3 (145-2C11, BioLegend, San Diego, CA) and PE anti-mouse CD45R/B220 (RA3-6B2, BioLegend). After immunostaining, the cells were assessed by flow cytometry on a FACScalibur flow cytometer (Becton Dickinson, San Jose, CA). All cytometry data were analyzed with CellQuest software (Becton Dickinson).
Isolation of T cells and B cells from PPL.
Peyer's patch-T cells (PP-T) and -B cells (PP-B) were isolated by using T cell and B cell isolation kits (Miltenyi Biotec, Bergisch Gladbach, Germany), respectively. Freshly isolated PPL were incubated with a cocktail of biotin-conjugated MAbs for 10 min on ice. Following incubation with a superparamagnetic MicroBeads conjugated-anti-biotin MAb, the cells binding MAbs were removed by Auto MACS magnetic columns (Milteny Biotec). These negative-selected cells were used as PP-T cells or PP-B cells. The positive-selected non-T cells were used as T cell-depleted PP cells (T-d-PP). The purity of these cells was assessed by flow cytometry on a FACScalibur flow cytometer (Becton Dickinson).
Induction of M cell differentiation.
MIE cells (9 × 105 cells) were seeded on the Transwell inserts and cultured in 12-well plates for 3, 5, and 7 days. After culture, each Transwell was inverted, and silicon tubes were attached to the bottom of the Transwell inserts. Intestinal lymphocytes, PPL (107 cells), PP-T cells (2.0 × 106), T-d-PP cells (107), or a mixture of PP-T cells (2.0 × 106) and PP-B cells (8.0 × 106) were added to the attached silicon tubes in a volume of 500 μl. In the case of the monocultured MIE cells, 500 μl of 10% FBS-DMEM was added into the silicon tubes. The cultures were incubated in plastic culture boxes (Sumilon) at 37°C, in 5% CO2 for 1–160 h. After coculture with lymphocytes, silicon tubes were detached from the Transwell inserts and they were set into 12-well plates. To determine the transcytosis activity of the MIE cells, yellow-green fluorescent microspheres (1.0 μm, F8823, Molecular Probes, Carlsbad, CA) were added to the apical chamber of MIE cells and incubated for 9 h. After incubation, basal media were collected, and the transported microspheres were quantified by FACScalibur flow cytometer (Becton Dickinson). All experiments were done in quintuplicate.
Establishment of a MIE cell line.
We have established three clonal MIE cell lines. One clone was selected on the basis of its ability to proliferate and was used for the following experiments. When MIE cells were cultured, they assumed a monolayer, cobblestone, epithelial-like morphology, with close contact between cells (Fig. 1A). All MIE cells were stained strongly for pan-CK and villin, which provided the evidence of the intestinal epithelial nature of MIE cells (Fig. 1, B and C). Immunocytochemical analysis was performed using antibodies for ZO-1, claudin 1, and β-catenin (Fig. 1, D–F). All of these proteins were localized at the cell-cell contact region and in the cytoplasm. The morphological characterizations of MIE cells were examined by SEM. Some short and irregular microvilli were observed on the surface of MIE cells when cultured for 1 day (Fig. 1G). These microvilli increased day by day and formed tightly after 15 days of culture (Fig. 1, G–I). Alkaline phosphatase (ALP) activity is a marker for intestinal epithelial cells. Some MIE cells expressed ALP activity on the first day of culture, and this activity increased with time in culture (Fig. 1, J–L). Fig. 2 shows the growth curve of MIE. This cell line grew stably with a population doubling time of 14.6 h between the first and fourth days and became confluent by 5 days after seeding (Fig. 2).
MIE cells were strongly positive for pan-CK (clone C-11) (Fig. 1B and Fig. 3A). Pan-CK-positive cells in the small intestine were observed in the villus epithelium, just above the Paneth cells of the crypt and in the FAE (Fig. 3, B–D). We also investigated the expression of another CK (CK20) in MIE cells and the small intestine. The upper part of the villus epithelium and FAE expressed CK20 (Fig. 3, F and H). In contrast, the expression of CK20 was not detected in MIE cells and the cells in the crypt base of the small intestine (Fig. 3, E and G). MIE cells expressed Musashi-1, a putative intestinal epithelial stem cell marker (Fig. 3I). Musashi-1 expression was observed in a few cells just above the Paneth cells where the stem cells are thought to be located in the murine small intestine (Fig. 3J). In contrast, Paneth cells were completely devoid of immunoreactivity for Musashi-1. Musashi-1 immunoreactivity was not detected in villus epithelium or in the upper part of the crypt. These data suggest that MIE cells may originate from crypt cells, such as stem cells or transit-amplifying cells.
TER and paracellular permeability of MIE cell monolayer.
TER is widely used to describe the permeable properties of tight junctions between epithelial cells (8). FITC-dextran flux rate is also an indicator of paracellular permeability (7). Therefore, we examined the permeability to FITC-dextran and the TER of monolayer MIE cells cultured on Transwell inserts. After 1 day of culture, MIE cells plated at a high cell density (9.0 × 105 cells/well) already had a maximally high TER value. MIE cells plated at medium and low cell densities reached this maximum TER value at 2 days and 11 days after seeding, respectively (Fig. 4A). The paracellular permeability in the middle and high cell density culture was very low on the first day of culture (Fig. 4B). This low permeability in the low cell density culture reached a minimum by 7 days. The microspheres transported into basal area were effectively zero in the low cell density wells after 5 days in culture (Fig. 4C). MIE cells at a cell density of 9.0 × 105 cells/well transported few microspheres and their paracellular permeability became very low and remained at this level after 3 days in culture. These results indicate that MIE cells (9.0 × 105 cells/well) can form a monolayer with tight junctions at 3 days after seeding.
Transcytosis activity of MIE cells.
The composition of isolated lymphocytes was assessed by flow cytometry. Intestinal lymphocytes and PPL were gated as R1 in Fig. 5, A and D, respectively. Intestinal lymphocytes contained equal numbers of T cells (50%) and B cells (50%) (Fig. 5, B and C), whereas PPL had few T cells (10%) and were mostly B cells (87%) (Fig. 5, E and F). The purity of the T cells in PP-T was 95%, and that of B cells in PP-B was 99% (Fig. 5, G and H). The number of T cells in T-d-PP was less than 1% (Fig. 5I). These isolated cells were used to induce the differentiation of MIE cells into M cells.
MIE cells were precultured for 3, 5, or 7 days and then cocultured with intestinal lymphocytes for 64 h (Fig. 6A) to investigate their ability to differentiate into M cells. MIE cells precultured for 3 days were able to transport microspheres to the basal chamber of the insert; the total number transported was ∼18 × 104 particles by 9 h after their addition. However, their ability to transport particles was much reduced when they were precultured for 5 or 7 days. In the monocultured MIE cells without intestinal lymphocytes, a few microspheres were found to be transported.
Secondly, we examined the effective length of coculture of MIE cells with intestinal lymphocytes on their differentiation (Fig. 6B). MIE cells were precultured on Transwell inserts for 3 days and then cocultured with intestinal lymphocytes. The transcytosis activity of MIE cells was observed at 8 h and reached a peak by 16 h after coculture. This transcytosis activity gradually decreased as the coculture time then increased further. However, monocultured MIE cells were not able to transport microspheres into the basal area. Because PPL has been reported to induce M cell differentiation (23), we also cocultured MIE cells with the PPL cells (Fig. 7A). PPL could not induce any transcytosis activity in MIE cells. However, PPL costimulated with anti-CD3/CD28 MAbs (PPL + CD3/CD28) enabled MIE cells to transport microspheres by 64 h after coculture. By 160 h, a large number of microspheres were transported under these conditions by MIE cells (Fig. 7A).
Thirdly, we determined which cell types from the PPL cells were essential for the induction of transcytosis activity in MIE cells. The activity of PPL costimulated with CD3/CD28 MAbs to induce transcytosis in MIE cells was much reduced by the depletion of the T cell population (T-d-PP cells + CD3/CD28). In addition, PP-T cells costimulated with MAbs (PP-T cells + CD3/CD28) also failed to induce any transcytosis activity of MIE cells (Fig. 7B). A mixture of PP-T cells costimulated with MAbs and PP-B cells (PP-T, PP-B cell + CD3/CD28) was able to induce transcytosis activity in MIE cells to nearly the same extent as the original PPL cells costimulated with MAbs (Fig. 7C). These data suggest that activated T cells and B cells are both essential for the induction of transcytosis activity in MIE cells.
The uptake and internalization of microspheres in MIE cells was observed by using a confocal laser microscope. Many microspheres were observed in MIE cells cocultured with intestinal lymphocytes (Fig. 8B). The x-z section revealed that microspheres were incorporated into the cytoplasm of MIE cells and detected around and under their nuclei (Fig. 8D). This level of incorporation was also observed in MIE cells cocultured with PPL + CD3/CD28. In contrast, no microspheres were observed in monocultured MIE cells (Fig. 8, A and C). It has been reported that murine M cells express annexin V (65) and Sgne-1 (19). Therefore, we also analyzed the expression of annexin V and Sgne-1 in M cell-differentiated MIE cells. MIE cells cocultured with PPL + CD3/CD28 incorporated many microspheres and strongly expressed annexin V and Sgne-1 (Fig. 8, F and H). In contrast, monocultured MIE cells only weakly expressed annexin V or Sgne-1 (Fig. 8, E and G).
SEM analysis revealed that MIE cells monocultured for 16 and 112 h developed some short and irregular microvilli (Fig. 9, A and C). Following coculture with intestinal lymphocytes or PPL treated with MAbs, the microvilli disappeared from the surface of the MIE cells, took on a smooth appearance coincident with an enhanced ability to incorporate the microspheres (Fig. 9, B and E). However, PPL not stimulated with antibodies did not bring about a change in the microvilli morphology of the MIE cells (Fig. 9D). These data indicate that MIE cells might be able to differentiate into M cells following coculture with intestinal lymphocytes or PPL costimulated with MAbs, along with the ultrastructural changes on the cell surface. TEM shows that microspheres are incorporated into the cytoplasm of cocultured MIE cells, but not in monocultured MIE cells (Fig. 9, F and G).
We have established a clonal murine intestinal epithelial cell line (MIE) from the C57BL/6 mouse. This cell line has undergone over 50 passages with no detectable loss of phenotypic properties. MIE cells expressed villin for the formation of microvilli, some junctional proteins, CK, and ALP enzyme activity. We report here that MIE cells also express Musashi-1 antigen. Musashi-1, a neural RNA binding protein, has been identified as a mammalian homologue of the Drosophila Musashi protein that is required for asymmetric division of sensory neural precursor cells (43, 44) and is a marker for neural stem cells (45, 55, 56). It has also been reported that Musashi-1-positive cells account for up to 30–50 cells per crypt in its lower part, including stem cells and transit-amplifying cells (1). Furthermore, it has been reported that Musashi-1 is a marker of stem cell and early progenitor cells in murine intestinal crypts (27, 50, 57). IEC-6 cells, a rat small intestinal epithelial cell line, express Musashi-1 (27) and have the distinctive morphology of epithelial cells such as microvilli and a number of typical characteristics of intestinal crypt cells (52). These data suggest that MIE cells have many characteristics of and may be close to intestinal epithelial stem cells or transit-amplifying cells.
M cells are normally found in close contact with cells of hematopoietic origin: dendritic cells, macrophages, and lymphocytes. All of these cell types play a fundamental role in M cell differentiation (13). In particular, the crucial role of lymphocytes in M cell differentiation has been demonstrated in vitro by M cell models using Caco-2 cells and mICcl2 cells (12, 30). Caco-2 cells differentiate into cells resembling M cells during coculture with PPL or Raji B cells. These M-like cells have some of the in vivo characteristics, such as transcytosis activity, a decrease of sucrase-isomaltase and villin expression, disorganization of microvilli and a change of α5β1 integrin localization (30, 63).
In this study, we report that MIE cells are able to differentiate into M cells following coculture with intestinal lymphocytes or PPL costimulated with anti-CD3/CD28 MAbs. Following this, these cells display features typical of M cells, such as transcytosis activity, the disorganization of microvilli (Fig. 8). Several reports show that in vivo M cells have disorganized microvilli (34) and that this morphological change on M cells is regulated by the distribution of microvillar proteins such as actin and villin (25, 29). Therefore, the disorganization of microvilli in M cells is not only a typical morphology of this cell type but also essential for the transcytosis of macromolecules. In addition, M cell-differentiated MIE cells were strongly positive for annexin V and Sgne-1, which are thought to be markers for M cells (19, 65). It has been reported that annexin V involves many membrane-related events such as endocytosis, exocytosis, membrane scaffolding, and the regulation of ion fluxes across membranes (28) and that M cell-specific expression of annexin V contributes to the vesicular transport of macromolecules and to the formation and maintenance of the pocketlike structure in M cells. Although the biological function of Sgne-1 in M cells remains unclear, it is surmised to contribute to the intracellular membrane trafficking of the mucosal antigens of M cells. These data suggest that the expressions of annexin V and Sgne-1 may support the transport of microspheres in M cell differentiated MIE cells.
Without the presence of the anti-CD3/CD28 MAbs, PPL were not able to induce M cell differentiation from MIE cells. Costimulation with anti-CD3/CD28 MAbs has been used widely for T cell activation in vitro (31). These data suggest that the activated T cells are essential for M cell differentiation of MIE cells. To determine whether activated T cells could induce M cell differentiation themselves, we cocultured MIE cells with T cells isolated from PPL (PP-T) (Fig. 6B). PP-T cells costimulated with anti-CD3/CD28 MAbs could not induce M cell differentiation in MIE cells. The mixture of PP-T cells costimulated with MAbs and PP-B cells could induce M cell differentiation in MIE cells as well as PPL costimulated with MAbs (Fig. 6C). These data indicate that both of the activated PP-T cells and PP-B cells are essential for the M cell differentiation of MIE cells.
Interestingly, the inductive effect of intestinal lymphocytes on the ability of MIE cells to bring about transcytosis occurred immediately (Fig. 5B). On the other hand, it took a long time for PPL costimulated with MAbs to induce M cell differentiation in MIE cells (Fig. 6A). The rapid effect of intestinal lymphocytes might be due to the presence of activated T cells in intestinal lymphocytes. In addition, the length of time that transcytosis activity was maintained in the presence of intestinal lymphocytes was shorter than that in the presence of PPL in the presence of the MAbs. Considering the report on the necessity of mature lymphocytes for M cell differentiation (10), these differences may be due to differences in the specific populations of cells between the intestinal lymphocytes and PPL.
Since PPL is mainly composed of B cells, T cells, and dendritic cells (34), these cells are thought to be important for the development of PP and the FAE. In B cell-deficient mice (μMT−/−, Igh6, JHD), PP are reduced in number and size (17). An in vitro model of M cell differentiation has shown that the conversion of Caco-2 cells into M cells occurs in the presence of B cells (29). These data indicate that B cells are important for the development of PP and FAE. PP are completely lacking in the lymphotoxin (LT) -α, -β, or -βR knockout (KO) mice (11). However, LT and the tumor necrosis factor produced by B cells are thought to be dispensable for the maintenance of the FAE (62). On the other hand, because TCRβ and TCRδ KO mice have normally developed PP and FAE, the lack of T cells might not have any significant influence on the development of PP and FAE. However, FAE were even more difficult to find in Rag-1−/− mice than in B cell-deficient mice (17), and the M cell rate in FAE of these KO mice is higher than that of normal mice (10, 51). Although mature PPL including T cells and B cells are important for the development of FAE, other factors may also be needed for M cell differentiation.
FAE secretes some chemokines such as CCL20/Mip3α (39), CXCL16 (20), and CCL9 (68). Dendritic cells, B cells, and T cells in murine PP express CCR6, a receptor for CCL20/Mip3α, and migrate to the subepithelial dome (SED). In addition, the CCR6−/− mice show a decrease of the numbers of M cells and CD4+ T cells in PP (39). CXCR6, a receptor for CXCL16, is highly upregulated in PP-T cells after the in vitro costimulation with anti-CD3/CD28 MAbs. After transplantation, these activated T cells react to CXCL16 secreted by FAE and migrate to the SED (20). The interaction of CD4+ T cells and FAE has been demonstrated under inflammatory conditions, and the indomethacin-induced inflammation increases the M cell number and the counts of M cell associated B cells and CD4+ T cells in rats (38). We consider that activated T cells can influence the microenvironment for M cell differentiation and that a specific microenvironment produced by T cells and B cells rather than any one cell type may induce M cell differentiation. We consider that our in vitro M cell model will be able to contribute to the elucidation of the mechanisms by which M cells differentiate.
This study was supported by a Grant-in-Aid for Scientific Research (18658104) from the Ministry of Education, Culture, Sports, Science and Technology, a Research Fellowships for Young Scientists Program from Japan Society for the Promotion of Science (JSPS), and two grants (BSE Control Project, and Secure and Healthy Livestock Farming Project) from the Ministry of Agriculture, Forestry and Fisheries.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society