| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Center of Anatomy, Medical School of Hannover, Hannover, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This study was performed in the Peyer's patches of mice, because this
animal model is well established and has previously been used in a
multitude of studies concentrating on the
ultrastructure,6,8
the histochemistry,14-17
the transport function,1,6,18
and the interaction of
murine M cells with microorganisms.19-22
In addition, it
is a great advantage that M cells of the murine Peyer's patches
possess large amounts of 
1–2-linked fucose in the glycocalyx of
their apical membrane15-17
and thereby can easily be
identified by light microscopy using the Ulex europaeus
agglutinin (UEA-I). Lectins are histochemical markers, which not only
reliably discriminate between the cell types of the gut
epithelium23
but also represent remarkably sensitive tools
to describe the differentiation pathway of the intestinal
cells.24
To follow the developmental pathway of the dome
epithelial cells, we characterized, both ultrastructurally and
histochemically, the crypts that produce them, investigated the
distribution of M cells over the domes, and determined the sites of
entry of lymphocytes into the dome epithelium.
The crypts at the base of murine Peyer's patch domes were investigated using a combination of optical sections acquired by confocal laser scanning microscopy (CLSM) with histochemically labeled cryosections, semi-thin and ultrathin sections. The association of individual crypts with epithelial regions at the flanks of the domes was investigated by scanning electron microscopy (SEM) combined with low-power CLSM, a method that allows fluorescence microscopic views of whole domes to be generated with an extended depth of focus. To ascertain that the M cell marker UEA-I used in these fluorescence techniques detected all M cells of the Peyer's patches, lectin-gold labelings were performed on ultrathin sections that allow murine M cells to be identified by their typical ultrastructure and the extent of lectin binding to be correlated with the ultrastructural characteristics. In addition, the basement membrane component laminin was detected immunohistochemically to outline the crypts and to define their spatial positions. To determine the sites where lymphocytes enter the dome epithelium, we chemically removed the dome epithelium and revealed the distribution of holes in the basal membrane associated with the trespassing cells using SEM. The combination of the various techniques makes it possible to follow the differentiation pathway of intestinal M cells in the in vivo situation, to decide whether M cells develop as a separate cell line, and to locate the compartments of the gut epithelium where this differentiation takes place.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Female BALB/c mice kept under specific-pathogen-free conditions, 4 to 8 weeks old and weighing ~20 g, were anesthetized intraperitoneally with a mixture of Ketanest (WDT, Garbsen, Germany) and Rompun (Bayer, Leverkusen, Germany). The Peyer's patches of the small intestine were removed and rinsed in Ringer's solution at 4°C for a few seconds to remove feces and mucus. The specimens were cut into small pieces using a razor blade and fixed as described for the different methods (see below). The mice were killed by cervical dislocation.
Thin Section Electron Microscopy
Some of the Peyer's patches were fixed in a solution of 2% formaldehyde and 0.05% glutaraldehyde in phosphate buffer, pH 7.3, for 16 hours. After rinsing in phosphate-buffered saline (PBS) containing 1% L-lysine (Sigma, Deisenhofen, Germany) for 16 hours to block free aldehyde groups, single domes were microdissected, partially dehydrated in ethanol dilutions (30%, 50%, 70%) and transferred to LR White resin, medium grade (Plano, Marburg, Germany). The tissue blocks were enclosed in gelatin capsules and allowed to polymerize at 50°C for 48 hours. Semi-thin sections, 0.1 to 1.0 µm in thickness, were cut with glass knives and stained with toluidine blue. Ultrathin sections, 60 to 80 nm thick, were cut with a diamond knife and mounted on nickel grids.
The lectin-gold labeling procedure was performed with PBS, containing 0.15% bovine serum albumin-c (BSA-c, Biotrend, Cologne, Germany) and 0.1% sodium azide. Free aldehyde groups were blocked in a drop of PBS containing 0.7% L-lysine for 15 minutes. After rinsing in PBS containing 5% bovine serum albumin (Serva, Heidelberg, Germany), 0.1% cold water fish skin gelatin (Biotrend), and 1% normal goat serum (Sigma), the sections were incubated with a solution containing UEA-I-biotin (12.5 µg/ml; Sigma) and 0.05% Tween 20 (Serva) at 4°C overnight. After rinsing, the grids were incubated for 4 hours with a goat anti-biotin antibody, conjugated to 20-nm colloidal gold (BioCell, Cardiff, UK). Finally, the grids were treated with 2% glutaraldehyde in PBS, washed in distilled water, and stained with uranium acetate and lead citrate. The sections were observed in a Zeiss EM10 electron microscope (Zeiss, Oberkochen, Germany). Controls were carried out either by omitting the lectin or by preincubating the lectin with 0.03 mol/L L-fucose overnight. In both controls, only a negligible uniform background staining was found over all tissue components and over the pure resin, probably resulting from some nonspecific binding of the gold-labeled secondary antibody.
Immunohistochemistry for Laminin
Eight freshly excised Peyer's patches taken from four mice were frozen and stored in liquid nitrogen. Cryosections running parallel to the mucosal surface, 4 µm in thickness, were fixed in a mixture of methanol and acetone for 10 minutes at -20°C and transferred to PBS. The sections were incubated with a polyclonal anti-laminin antiserum (Sigma) derived from rabbit for 1 hour. After rinsing in PBS, the binding sites were detected using the avidin-biotin complex method and 4-chloro-1-naphthol as substrate or an anti-rabbit antibody conjugated to FITC. Controls were performed by replacing the primary antibody with buffer and resulted in unstained sections, except for some granulated cells in the lamina propria.
Scanning Electron Microscopy
Twenty-two domes taken from the Peyer's patches of four mice were microdissected as described for confocal microscopy (see below) and fixed as described for thin-section electron microscopy. The tissue samples were rinsed in PBS and dehydrated in a series of acetone dilutions. Critical-point drying was performed using a Balzers CPD 030 (Balzers, Liechtenstein) and sputter-coating using a Polaron E 5400 (Polaron, Watford, UK) and a gold-palladium target. The specimens were examined in a PSEM 500 scanning electron microscope (Philips, Eindhoven, The Netherlands) equipped with a PC-based device for recording and processing digital images at high resolution.25
To investigate the basal membrane of the dome epithelium and the distribution of trespassing lymphocytes, the dome epithelium was removed according to the method described by Low and McClugage26 and McClugage et al.27 Twenty-two unfixed Peyer's patches taken from nine mice were incubated with 1% boric acid for 24 hours, thoroughly rinsed in PBS, dehydrated in acetone, and prepared and examined as described for conventional SEM. The positions of holes in the basal membrane were marked with colored dots on digital images of complete domes using Photoshop software (version 4.0, Adobe, Edinburgh, UK). Longer incubation with boric acid (eg, for 48 hours) enlarged neither the number nor the diameter of holes.
Confocal Microscopy
For CLSM, the Peyer's patches of eight mice were fixed in pure ethanol for 16 hours, rehydrated in an ethanol series, and transferred to PBS. Single domes were microdissected under a binocular dissecting microscope so that the complete dome surface could be examined without overlying villi. Lectin labeling was performed with Ulex europaeus agglutinin conjugated with tetramethylrhodamine (UEA-I-TRITC, Sigma) at concentrations of 50 or 12.5 µg/ml for 2 hours. After rinsing in PBS, the samples were stored and examined in a Tris-buffered solution containing 30% glycerol and 0.1% NaN3. CLSM was performed with a BioRad 600 confocal laser scanner (Biorad, Munich, Germany) attached to an Axiovert 35M inverted microscope (Zeiss) and equipped with a krypton-argon mixed gas laser. Serial optical sections of the lectin-labeled domes were recorded using COMOS software (version 6.0, Biorad) and the 568-nm line for excitation of TRITC. Maximal projections providing an extended depth of focus were generated with Lasersharp software (version 1.02, Biorad) and Photoshop software (Adobe). Controls, performed by preincubation of the lectin with 0.03 mol/L L-fucose overnight, resulted in a drastic decrease of the lectin binding.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Semi-thin sections across the crypt region running parallel to the
mucosal surface revealed that the lymphoid tissue and the underlying
follicle of each Peyer's patch dome was surrounded by ~28 crypts
(range, 21 to 34) that showed two morphological types (Figure 1)
. Although some of these crypts closely
resembled ordinary crypts, the majority differed considerably from
ordinary crypts in size, symmetry, epithelial height, and cellular
composition (Figure 1b)
. These dome-associated crypts were oval or
kidney-shaped in cross section with maximal diameters of 70.6 ±
10 µm (range, 52 to 90 µm; n = 22) as compared with
51 ± 8.7 µm (range, 33 to 66 µm; n = 27)
measured for ordinary crypts. Although the crypt epithelium on the
(outer) villous side did not differ from that of ordinary crypts, the
epithelium facing the (inner) dome side was less intensely stained, its
height was considerably increased, and it only rarely contained goblet
cells (Figure 1b)
. The dome-associated crypts were in close contact
with the lymphoid tissue of the dome but contained only relatively few
lymphocytes as compared with the epithelium at the flanks of the dome.
In addition to these peculiarities in cellular composition and
structure, laminin immunohistochemistry showed that the dome-associated
crypts were separated from the ordinary crypts by a layer of smooth
muscle cells (Figure 1c)
. Serial cross and longitudinal sections of the
patches revealed that such layers surrounded each of the lymphoid
follicles and were continuous with the lamina muscularis mucosae.
|
Ultrathin sections of the dome-associated crypts showed that a
subpopulation of epithelial cells possessed thicker and more irregular
microvilli than the remaining cells (Figure 2, a and b)
. Labeling with UEA-I as a
marker for murine M cells, visualized on ultrathin sections by
colloidal gold, showed that the brush border of these cells, but not
that of the other epithelial cells, was rich in 1–2-linked fucose
residues (Figure 2, a and b)
. Most of these M cell precursors were
present at the side of the crypt that was in contact with the lymphoid
tissue and continuous with the dome epithelium, whereas only very few
of these cells were found at the villous side of the dome-associated
crypts. Although some lectin binding was seen in ordinary crypts,
mostly associated with the mucus of some of the goblet cells or the
granules of Paneth cells, no epithelial cells with the morphological
and lectin histochemical characteristics of M cell precursors were
found (Figure 2c)
.
|
Conventional SEM was performed to investigate the transition from
the crypt epithelium to the epithelium at the base of the domes. In the
cleft between dome and villi, crypt mouths were recognized that were
associated with regions at the lower flanks of the domes showing the
typical patchwork of enterocytes and interspersed M cells. The M cells
were identified by their brush border, which was shorter and darker
than that of surrounding enterocytes. Whereas most of the crypt mouths
were associated with such regions rich in M cells (Figure 3, c and e)
, a minority of crypt mouths
was observed that obviously produced epithelial cells for the dome (ie,
enterocytes, goblet cells, and brush cells) but did not produce M cells
(Figure 3d)
.
|
To determine how the regions rich or poor in M cells were
associated with the two types of crypts seen in the sectioning
techniques, low-power CLSM of whole domes labeled with UEA-I-TRITC was
used. Beside M cells, the goblet cells and the brush cells bind
UEA-I.15,17
Because these two cell types are rare in the
dome epithelium and can be distinguished from the M cells by their size
and binding intensity, they could be excluded from further evaluation.
Small bright dots, each representing an individual M cell, were most
numerous at the flanks of the domes (Figure 4)
. In the majority of the domes
investigated by low-power CLSM (12 of 20 domes), the M cells were not
randomly distributed over the flanks of the domes but arranged in
radial strips (Figures 4 and 5)
. Approximately 22 radial strips per
dome (range, 15 to 28) rich in M cells alternated with regions that
were poor in or even devoid of M cells. The M-cell-rich strips varied
in width between 24 and 88 µm and were ~250 µm long. These strips
were associated with groups of intensely labeled cells at the periphery
of the domes (Figure 5)
. The analysis of
single confocal sections running through the base of the domes revealed
that each of the strips was associated with an individual crypt. The
M-cell-rich strips were associated with crypts of ~70 µm in
diameter, representing the specialized dome-associated crypts described
above (Figure 5c)
. In contrast, radial strips poor in or devoid of M
cells were associated with crypts of ~50 µm in diameter, typically
lying more peripheral to the domes and representing ordinary crypts
(Figure 5d)
. On a minority of the domes investigated (the remaining 8
of 20), a relatively uniform distribution of M cells was seen.
|
|
The M cell population of individual domes may be heterogeneous
with respect to its lectin-binding properties (see 7 and 17
). To
exclude that, in CLSM, the M cell marker UEA-I detected only a minor
subpopulation of all M cells, the ultrastructural morphology was used
as an additional criterion for the identification of M cells. In 54
ultrathin sections, taken from 10 domes, 98 different M cells were
identified by their typical morphology, ie, short stub-like microvilli,
numerous cytoplasmic vesicles, and/or a pocket-like invagination of the
basolateral membrane (Figure 6)
. Intense
labeling by gold particles was found on the brush border and the apical
cytoplasm of 95 M cells (97%) identified by morphology but not on
ordinary enterocytes of the domes (Figure 6)
. Thus, it was excluded
that a major population of M cells existed that was not detected by the
Ulex lectin, and it must be assumed that the lectin-negative
strips seen in CLSM did indeed lack M cells.
|
After removal of the epithelium with boric acid, the sites where
lymphocytes had entered or left the dome epithelium appeared in SEM as
holes, ~5.6 µm (range, 3.6 to 9.2 µm, 5.6 ± 1.1 µm;
n = 46) in diameter (Figure 7)
. Trespassing lymphocytes were found
lying on the basal membrane near the holes or in the holes in most of
the domes examined (Figure 7c)
. The holes in the basal membrane of the
dome epithelium were evenly distrib-uted over the domes; no regular
arrangements were detected in the 110 domes studied by this
technique, and in particular no radial strips rich or poor in holes
and/or lymphocytes were found. This impression was confirmed by image
analysis of 15 randomly selected domes taken from eight mice, which
showed no preferential distribution, eg, arrangements in strips (Figure 7d)
.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Lectin histochemical studies revealed that the epithelial cells, which derive from an individual crypt of the small intestine, are of monoclonal origin29,30 and that M cells that derive from a common crypt are also monoclonal.7,31 The present study shows that the clone of epithelial cells, which derives from an individual dome-associated crypt, forms a radial strip on the flank of the dome rich in M cells. In addition, the combination of sectioning techniques with SEM and CLSM revealed not only that the specialized dome-associated crypts contribute to the dome epithelium but also that ordinary crypts produce epithelial cells that enter the dome. However, we were unable to detect M cell precursors in such ordinary crypts, and the radial strips associated with these were devoid of M cells, as demonstrated by scanning and confocal microscopy. The sensitivity of 97% determined for UEA-I as a marker for murine M cells further documents the absence of M cells in epithelial regions that derive from ordinary crypts. We therefore conclude that the differentiation of M cells is induced in and restricted to the dome side of the specialized dome-associated crypts and takes place neither in other crypts nor later on at the flanks of the domes.
This finding implies that lymphocytes of the dome epithelium do not induce the formation of M cells, as has previously been suggested.5,8,9,13,32 Our conclusion is further supported by the observation of a random distribution of holes in the basal membrane of the dome epithelium and the trespassing lymphocytes. If M cell development were induced by dome-epithelial lymphocytes via direct cell-cell interaction, the distribution of M cells over the domes would correspond with that of lymphocytes. As no such correlation was observed, it can be ruled out that M cells develop from enterocytes in the specific environment of the dome epithelium. The results of this study indicate that M cells represent a separate cell line distinct from enterocytes, goblet cells, Paneth cells, or other cell types of the intestinal epithelium. The different stages of M cell development are related to specific compartments, starting with M cell precursors in the upper crypt epithelium and ending with mature M cells at the tips of radial M-cell-rich strips. Although epithelial cells labeled by M cell markers have previously been described in the upper crypts using light microscopy,7,17 for the first time we identify and characterize M cell precursors at an ultrastructural level. The presence of such cells at the position where the initial differentiation into the main epithelial cell types of the intestine takes place4 indicates that M cells also derive directly from undifferentiated crypt stem cells.
The radial strips rich in M cells were associated with specialized M-cell-producing crypts, and the number of M-cell-rich strips per dome was similar to that of the dome-associated crypts. However, radial strips rich or poor in M cells were detected in most, but not in all, domes examined. The presence or absence of strips varied between the domes of individual patches and was not related to individual animals. As a Peyer's patch is formed by 5 to 10 follicles in mice, most of the follicles lie at the periphery of their patch and are in contact with ordinary crypts. They should consequently possess domes with M-cell-negative strips. Some follicles, however, lie in the center of the patch, are completely surrounded by other follicles, and are not in contact with ordinary crypts. They are thus homogeneously supplied with M cells and enterocytes, as was observed for some domes.
The dome-associated crypts possess a specialized epithelium at the dome side, in which the development of M cells from stem cells is induced and the generation of goblet cells is suppressed. These specialized epithelial regions are in direct contact with the lymphoid tissue of the follicle but only rarely contain lymphocytes. Based on the present observations, we postulate that the differentiation pathway of M cells is triggered by a still unknown humoral factor in the crypts rather than by direct cell-cell interaction, as suggested by findings in cell cultures.13 The percentage of M cells of all dome epithelial cells is relatively constant in and characteristic for each species and location (eg, 8% to 10% in mouse Peyer's patches15 ), but both increases and decreases of the relative number of M cells have been described under the influence of antigens, inflammation, drugs, or microbes.33-37 It is therefore speculated that the humoral factors that induce and maintain the development of M cells are regulated by the cells of the lymphoid tissue. In conclusion, the in vivo situation described in the present study enables us to follow the differentiation pathway of intestinal M cells and to state that their formation is restricted to specialized epithelial regions in the dome-associated crypts and probably not induced by direct interaction with intraepithelial lymphocytes in these crypts or on the dome. Additional studies are needed to identify the factors and the molecular mechanisms involved in the induction of M cells in the dome-associated crypts.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by the Deutsche Forschungsgemeinschaft (SFB280/C14).
Accepted for publication February 12, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. A. Knoop, N. Kumar, B. R. Butler, S. K. Sakthivel, R. T. Taylor, T. Nochi, H. Akiba, H. Yagita, H. Kiyono, and I. R. Williams RANKL Is Necessary and Sufficient to Initiate Development of Antigen-Sampling M Cells in the Intestinal Epithelium J. Immunol., November 1, 2009; 183(9): 5738 - 5747. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. K. Silva, J. D. T. da Silva, K. A. A. Torres, D. E. de Faria Filho, F. H. Hada, and V. M. B. de Moraes Humoral immune response of broilers fed diets containing yeast extract and prebiotics in the prestarter phase and raised at different temperatures J. Appl. Poult. Res., January 1, 2009; 18(3): 530 - 540. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Man, F. Lodi, E. Bertelli, M. Regoli, C. Pin, F. Mulholland, A. R. Satoskar, M. J. Taussig, and C. Nicoletti Macrophage Migration Inhibitory Factor Plays a Role in the Regulation of Microfold (M) Cell-Mediated Transport in the Gut J. Immunol., October 15, 2008; 181(8): 5673 - 5680. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Carapelli, M. Regoli, C. Nicoletti, L. Ermini, L. Fonzi, and E. Bertelli Rabbit Tonsil-associated M-cells Express Cytokeratin 20 and Take Up Particulate Antigen J. Histochem. Cytochem., October 1, 2004; 52(10): 1323 - 1332. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. H. Jang, M.-N. Kweon, K. Iwatani, M. Yamamoto, K. Terahara, C. Sasakawa, T. Suzuki, T. Nochi, Y. Yokota, P. D. Rennert, et al. Intestinal villous M cells: An antigen entry site in the mucosal epithelium PNAS, April 20, 2004; 101(16): 6110 - 6115. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Lo, W. Tynan, J. Dickerson, M. Scharf, J. Cooper, D. Byrne, D. Brayden, L. Higgins, C. Evans, and D. J. O'Mahony Cell culture modeling of specialized tissue: identification of genes expressed specifically by follicle-associated epithelium of Peyer's patch by expression profiling of Caco-2/Raji co-cultures Int. Immunol., January 1, 2004; 16(1): 91 - 99. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Gebert, I. Steinmetz, S. Fassbender, and K.-H. Wendlandt Antigen Transport into Peyer's Patches: Increased Uptake by Constant Numbers of M Cells Am. J. Pathol., January 1, 2004; 164(1): 65 - 72. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Ramirez and A. Gebert Vimentin-positive Cells in the Epithelium of Rabbit Ileal Villi Represent Cup Cells but not M-cells J. Histochem. Cytochem., November 1, 2003; 51(11): 1533 - 1544. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Lelouard, A. Sahuquet, H. Reggio, and P. Montcourrier Rabbit M cells and dome enterocytes are distinct cell lineages J. Cell Sci., January 6, 2001; 114(11): 2077 - 2083. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C NICOLETTI Unsolved mysteries of intestinal M cells Gut, November 1, 2000; 47(5): 735 - 739. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
