Department of Internal Medicine I, University of Regensburg, 93042 Regensburg, Germany
Correspondence: Gerhard Rogler, M.D., Ph.D., Department of Internal Medicine I, University of Regensburg, 93042 Regensburg, Germany. E-mail: gerhard.rogler{at}klinik.uni-regensburg.de
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: surface antigen expression • interleukin-1ß transcription • inflammatory bowel disease
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Intestinal macrophages represent one of the largest compartments of the mononuclear-phagocyte system in the body [3 ]. They are localized preferentially at the sites of antigen entry, e.g., in the subepithelial region of the intestinal lamina propria and domes of Peyer’s patches [4 5 6 ]. Resident macrophages constitute 10–20% of the mononuclear cells (MNCs) in the lamina propria, as determined by immunohistochemistry and tissue disaggregation experiments [4 , 7 8 9 10 ].
From immunohistochemical and flow-cytometric studies, it is known that intestinal macrophages express CD44, CD68, acid phosphatase, and nonspecific esterase [2 , 11 , 12 ]. A number of surface markers such as CD4, HLA-DR, 25F9 (specific for phagocytic macrophages), EBM/11, 3C10, RFD7 (for tissue macrophages), and Y1/82A are expressed heterogeneously by these cells [13 14 15 ].
Less than 10% of the macrophages from normal colonic mucosa express the typical monocyte/macrophage-specific surface markers CD14, CD16, CD11b, and CD11c [2 , 16 , 17 ]. In addition, the level of expression of the T-cell costimulatory molecules B7-1 (CD80) and B7-2 (CD86) on intestinal macrophages is low [16 , 18 ]. In normal alveolar macrophages, a similar phenotype was described, with low expression of CD14 and CD16 [19 20 21 22 23 ]. Under normal conditions, CD14 expression is also low in liver macrophages [24 25 26 ].
Intestinal macrophages show important functional differences compared with monocytes or in vitro-differentiated macrophages. Interleukin-1ß (IL-1ß), a cytokine released predominantly by mononuclear phagocytes, is an important mediator in inflammation [27 , 28 ]. The expression and induction of IL-1ß is best characterized in lipopolysaccharide (LPS)-stimulated peripheral blood monocytes or in vitro-differentiated macrophages. In the inflamed mucosa typical of inflammatory bowel disease, large amounts of IL-1ß are produced [29 , 30 ]. Studies on the expression of IL-1ß in intestinal mucosa [31 ] and in isolated mucosal cells [32 , 33 ] have shown that IL-1ß is produced almost exclusively by intestinal macrophages during inflammation. Cells isolated from normal colonic mucosa produce very little IL-1ß [34 ] despite stimulation with LPS [32 ].
Several models have been established for the in vitro differentiation of macrophages. A well-established model system is the culturing of isolated monocytes in plastic dishes or Teflon® bags (Biofolie 25; In Vitro Systems, Osterode, Germany) with 2% human AB serum for 7 days [34 , 35 ]. Under these conditions, differentiation of macrophages is associated with high surface expression of CD14 [36 ], but macrophages induced in serum-free cultures with high amounts of albumin, immunoglobulin, or macrophage-colony stimulating factor also showed high expression of CD14 [36 ].
The differentiation and maturation of macrophages have also been studied in three-dimensional culture systems including multicellular spheroids (MCSs) [37 , 38 ]. MCSs were used as a three-dimensional model for nonvascularized metastasis of tumor cells or cell lines [38 ]. In these three-dimensional aggregates, conditions such as the O2 gradient, pH, extracellular-matrix production, and cell proliferation are similar to those found in vivo [39 40 41 ]. Konur and coworkers showed that, in several bladder carcinoma cell lines, no down-regulation of the typical macrophage antigens CD14 and CD11c occurred [42 ].
Until now, the typical phenotype of intestinal macrophages from normal mucosa could not be induced in vitro. It is of great importance for the better understanding of normal gut physiology to elucidate the factors that can induce this very distinct macrophage phenotype. We used the MCS technique to induce close contact of monocytes with IECs in a three-dimensional environment. Cell-cell and cell-matrix contacts occur in MCS cells, similar to those found in vivo [43 , 44 ]. Epithelial cells secrete an extracellular matrix that provides a milieu for invading cells which is similar to that of the in vivo situation.
A recent report states that the application of different electron-microscopic techniques showed that the mucosal basement membrane is perforated with numerous small pores in vivo, through which cells in the lamina propria communicate with enterocytes [45 ]. It was further demonstrated that macrophages specifically cluster beneath the hyperporous areas of the basement membrane, with thick cellular processes penetrating it. These authors suggested that the spatially regulated pattern of perforation of the epithelial basement membrane indicates phase-specific interactions of lamina propria macrophages with IECs [45 ]. Those interactions with epithelial cells and contact with three-dimensional structures may be crucial for intestinal macrophage differentiation. Recently Randolph and coworkers showed that monocytes could differentiate into CD14-negative dendritic cells in only 2 days when they migrated and remigrated across an endothelial model [46 ]. In contrast, monocytes that remained in the subendothelial matrix became macrophages, retaining CD14 expression [46 ].
To investigate possible influences of three-dimensional structures and extracellular matrix on the specific differentiation of intestinal macrophages, MCSs from IECs and control cell lines were cocultured with elutriated and purified blood monocytes.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell lines
Eight IEC lines were tested for their ability to form spheroids.
The cell lines CaCo-2, SW-480, SW-620, T-84, and HUTU-80 were not
capable of forming MCSs and could not be used for the experiments.
Spheroids could be generated with three of the tested colonic IEC cell
lines (HT-29, WiDr, and LS174T). Two spheroid-forming human carcinoma
cell lines of a nonintestinal origin (J82 and RT4) and the
nonepithelial L87/4 cell line were used as controls.
The well-established HT-29 colon carcinoma cell line was derived from a primary tumor of a 44-year-old female [47 ]. WiDr cells were originally isolated from a primary adenocarcinoma of the rectosigmoid colon of a 78-year-old female [48 ]. LS174T is a trypsinized variant of LS180, both lines being derived from a 58-year-old female with Dukes type B adenocarcinoma of the colon [49 ].
J82 cells were derived from a poorly differentiated, invasive urothelial carcinoma [47 ]. RT4 cells were derived from explants of a recurring papillary tumor of the bladder [50 ]. The cell line L87/4 was originally derived from human bone marrow stroma and was immortalized by simian virus 40 transformation [51 ].
WiDr, LS174T, J82, RT4, and L78/4 cells were grown in RPMI 1640 medium (PAN Systems, Aidenbach, Germany) supplemented with 10% fetal calf serum (PAN Systems), 1% penicillin-streptomycin mixture (PAN Systems), and 1% nonessential amino acids (Biochrom, Berlin, Germany). HT-29 cells were grown in Dulbecco’s modified Eagle’s medium (Biochrom) supplemented with 10% fetal calf serum, 1% penicillin-streptomycin mixture, 1% nonessential amino acids, and 1% sodium pyruvate (Biochrom).
All cell lines were cultured under standard tissue culture conditions.
Generation of MCSs
MCSs were generated according to the liquid overlay culture
technique [52
] (Fig. 1
). Cells (4 x 103) suspended in 0.2 mL of
medium/well were seeded in agarose-coated wells of 96-well plates and
cultured under static conditions. The cells formed small aggregates
within 3 days, which further enlarged during culture. After 3 days,
half of the medium was replaced by fresh medium. After 3 more days of
culture, MCSs had formed and were used for experiments.
 View larger version (19K): [in a new window] |
Figure 1. Schematic illustration of the generation of MCSs using the liquid
overlay technique.
|
Flow cytometry
Flow cytometry was performed using a FACScan (Becton Dickinson,
San Jose, Ca) or an EPICS® XL-MCL (Coulter, Krefeld, Germany), both
of which are equipped with an argon ion laser with an excitation power
of 15 mW at 488 nm. The fluorescence of 10,000 cells was measured by
each flow-cytometric method. A four-decades log-scale forward light
scatter and a linear-scale right-angle scatter were used. Analysis
gates were set around debris and intact cells on forward light scatter
versus linear-scale right-angle scatter dot plots. The fluorescence dot
plots were generated using the gated data. Data acquisition and
analysis were performed automatically using LYSIS TM II software,
version 1.1 (Becton Dickinson) or WIN-MDI software
(http://facs.scripps.edu/help/html/read1ptl.htm).
Immunostaining for flow cytometrical analysis
For flow-cytometric analysis of cocultures of IECs and
monocyte/macrophages, MCSs were harvested after 24 h, 3 days, and
7 days and disaggregated with 2 mM EDTA (Merck, Darmstadt, Germany) for
5 min. Cells were resuspended in phosphate-buffered saline (PBS). For
each staining, 0.2 mL of the cell suspension were placed into 1.5-mL
polypropylene tubes. For three-color analysis, cells were stained with
a fluorescein isothiocyanate-conjugated anti-EP-4 antibody (clone
Ber-EP4; Dako, Hamburg, Germany), a phycoerythrin (PE)-conjugated
anti-CD33 (monocyte/macrophage-marker) antibody (clone My9; Coulter),
and a TRICOLOR-conjugated anti-CD14 (clone TüK4) or anti-CD16
antibody (clone 3G8; both from Caltag Laboratories, Burlingame, CA).
monocytes were also stained with a fluorescein
isothiocyanate-conjugated anti-CD11b antibody (clone BEAR1) and the
PE-conjugated anti-CD33 antibody (clone MY9; both from Coulter) for
characterization. Incubation was performed in the dark for 20 min on
ice to ensure specific staining. Cells were washed twice with cold PBS
and resuspended in 0.5 mL of PBS.
Immunohistochemistry
The cocultures of MCSs and monocytes were harvested from the
agarose-coated wells at the indicated time points, embedded in medium
for freeze sections (Miles, Elkhart, IN), and frozen in liquid
nitrogen. Embedded specimens were cut with a cryostat (Reichert,
Heidelberg, Germany) at 5-µm thickness, mounted on glass slides, and
dried for 24 h.
A single immunohistochemical-staining procedure was carried out according to a standard alkaline phosphatase anti-alkaline phosphatase (APAAP) technique [53 ]. Described briefly, sections were fixed in acetone for 10 min. Blocking with Tris-buffered saline-1% bovine serum albumin (BSA) for at least 30 min was followed by incubation with the mouse monoclonal antibody against the monocyte/macrophage antigen CD68 (clone KP-1; Dako), which is not expressed by IECs or with the monoclonal antibody against the LPS receptor CD14 (clone Rmonocyte52, Immunotech, Hamburg, Germany). After rinsing with Tris-buffered saline, rabbit anti-mouse immunoglobulin G [immunoglobulins, mainly immunoglobulin G, were isolated from mouse serum (Dako)] at a final concentration 0.175 mg/mL was used as bridging antibody, followed by incubation with APAAP complex (Dianova, Hamburg, Germany) for 30 min each time. Subsequent incubation with Fast-Red® substrate (Biogenex Laboratories, San Ramon, CA) resulted in a red precipitate. Finally the sections were counterstained with hematoxylin and mounted.
Immunohistochemical double staining
To prove the monocyte/macrophage character of the stained cells,
a double-staining procedure was applied. Sections were fixed in
ice-cold acetone and incubated with 0.3% H2O2
to inactivate endogenous peroxidase, followed by blocking in PBS-1%
BSA. After incubation with anti mouse CD68 (clone KP1; Dako), a
biotinylated secondary antibody (goat anti mouse; Dianova) was applied
followed by incubation with the avidin-biotin-peroxidase complex
(Vector Laboratories, Burlingame, CA). Subsequent incubation with
diaminobenzidine in PBS-0.003% H2O2 (Sigma,
Steinheim, Germany) resulted in a brown staining of the CD68-positive
cells.
For the second staining, sections were again incubated in 0.3% H2O2 to block peroxidase activity of the first staining step. After blocking unspecific binding in PBS-1% BSA, sections were incubated with antibodies against different monocyte/macrophage antigens: anti-CD11b (clone BEAR1), anti-CD11c (clone BU15), anti-CD14 (clone Rmonocyte52), and anti-CD16 (clone 3G8) (all from Immunotech; 2-ng/mL final concentration) followed by the biotinylated secondary antibody (goat anti-mouse; Dianova) and avidin-biotin-peroxidase complex (30 min each). Incubation with a benzidine dihyrochloride-sodium nitroprusside solution (both from Sigma) resulted in a blue precipitate. Finally sections were dehydrated and mounted.
Quantification of IL-1ß RNA
For the determination of IL-1ß RNA in cells obtained from
MCS-monocyte/macrophage cocultures, the MCSs of three 96-well plates
were disaggregated by EDTA after 24 h, 3 days, and 7 days of
culture. Cells were seeded into six-well plates and stimulated with 1
or 50 ng/mL of LPS (Sigma). After 2 h, total RNA was isolated from
the cells with the RNeasy mini kit (Qiagen, Hilden, Germany). Same
amounts of RNA (2 µg) were reverse transcribed in a 20-µL reaction
(Retroscript kit; Ambion, Wiesbaden, Germany). PCR for IL-1ß was
carried out with the Ambion competitive quantitative reverse
transcription (RT)-PCR kit (Ambion). A RNA-Competicon was used as an
exogenous competitive template for quantitative RT-PCR experiments. The
"competicon" contains the same primer-binding sites as the
endogenous target but can be distinguished from the endogenous target
due to a difference in size. Competicon was added to the RT samples in
different dilutions. During PCR, a competition between competicon and
template DNA takes place. Due to the size difference, competicon and
template DNA can be separated by electrophoresis. As the copy number of
the added competicon becomes known, the amount of IL-1ß copies in the
template DNA can be determined by comparison of the intensity of the
two signals in the agarose gel.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Flow-cytometrical analysis of monocyte/macrophage surface antigens
The phenotype of the freshly elutriated monocytes was determined
by flow cytometry (Fig. 2.
). monocytes were positive for the monocyte/macrophage marker CD33
and the surface antigens CD14 (Fig. 2A)
, CD16 (Fig. 2B) , and CD11b
(Fig. 2C)
as expected. Monocyte/macrophages did not change this
phenotype when cultured without epithelial cells in Teflon® bags for
7 days (Fig. 2D
2E
2F)
.
 View larger version (29K): [in a new window] |
Figure 2. Flow-cytometric analysis of freshly isolated monocytes and in vitro
differentiated macrophages. Cells were triple-stained for EP4-FITC,
CD33-PE, and CD14- or CD16-TRICOLOR or double stained for
CD11b-FITC and CD33-PE at day 0 (freshly elutriated) and
after 7 days of culture in Teflon® bags. (A) Monocytes at day 0 were
positive for CD33 and CD14. (B) Coexpression of CD33 and CD16 was
detected in freshly elutriated monocytes. (C) At day 0, cells were also
positive for CD33 and CD11b. Monocyte/macrophages did not change their
phenotype during a 7-day culture period in Teflon® bags. Cells were
still positive for CD33/CD14 (D), CD33/CD16 (E) and CD33/CD11b (F).
|
|
View this table: [in a new window] |
Table 1. Percentage of CD33++ Cells (Monocyte/Macrophage) in MCS Cocultures
during the 7-Day Culture Period
|
|
View this table: [in a new window] |
Table 2. Antigen Expression of Monocyte/Macrophages at Indicated Time Points
after Culture with MCSs of IEC Lines and the Control Cell Line
|
 View larger version (39K): [in a new window] |
Figure 3. Flow-cytometric analysis of monocyte/macrophages cocultured with
spheroids of HT-29 cells. (A, B, C) Dot plot of CD14-TRICOLOR
fluorescence vs. CD33-PE fluorescence signal. (D, E, F) Dot plot of
CD14-TRICOLOR fluorescence vs. EP-4-FITC fluorescence signal. After
24 h of coculture, 86% of the CD33-positive cells showed
expression of CD14 (A). After 3 days of coculture, the percentage of
CD33/CD14-positive cells was 59% (B). After 7 days of coculture, the
situation had changed with only 11% of CD33-positive cells also
expressing CD14 (C). Expression at 24 h (D), 3 days (E), and 7
days (F) also showed the decrease of CD14-positive cells during the
7-day culture period. CD14-positive cells showed no coexpression of the
epithelial cell marker EP-4 and were therefore identified as
monocyte/macrophages.
|
LS174T-MCSs also induced intestinal-like differentiation of invading monocytes (data not shown) with similar kinetics.
In identical experiments carried out with spheroids of the control cell line J82, this specific differentiation of invading monocyte/macrophages could not be observed. After 24 h of coculture, 92% of the monocyte/macrophages were positive for CD14 (Table 2) . This percentage did not change during the culture period, with 94% of macrophages found to be CD14 positive after 3 and 7 days (Table 2) . Similar results were obtained for CD16 expression (Table 2) .
Immunohistochemical analysis of antigen expression in
IEC-monocyte/macrophage cocultures
For immunohistochemical analysis, the well-established
intracellular macrophage-antigen CD68 [54
,
55
] was used for the positive detection of
monocyte/macrophages in the spheroids. Analogous experiments, carried
out with spheroids without monocytes, showed that CD68 was not
expressed by all cell lines used (data not shown). CD68 was used
because an intracellular antigen and a surface antigen are better
suited for double stainings than two surface antigens.
In cocultures of IEC-MCSs with freshly isolated monocytes, a relatively low number of the initially added monocytes attached to the three-dimensional aggregates and infiltrated the MCSs within 24 h. To determine the phenotype of the macrophages, we analyzed the presence of the typical macrophage surface antigens CD14, CD16, CD11b, and CD11c after 24 h, 3 days, and 7 days of coculture of MCSs with monocytes. Experiments were carried out with three different IEC lines (HT-29, WiDr, and LS174T) and three control cell lines (J82, RT4, and L78/4). Results are from three representative experiments with monocytes of different donors (n = 3).
In MCSs of the IEC line HT-29 cocultured with monocytes for 24 h, CD68 (Fig. 4A )- and CD14 (Fig. 4B) -positive cells could be detected by the APAAP method. Immunohistochemical double staining showed that CD68-positive cells also expressed CD14, CD16, CD11b, and CD11c (Fig. 4C 4D 4E 4F) , confirming the FACS data.
 View larger version (145K): [in a new window] |
Figure 4. Expression of antigens after 24 h of coculture. Spheroids of the
cell line HT-29 were cocultured with freshly isolated
monocyte/macrophages for 24 h. Antigen expression was determined
by immunohistochemistry (APAAP-method and double staining). After
24 h of coculture, monocyte/macrophages were preferentially
located at the MCS surface. (A) Adherent and invading
monocyte/macrophages could be detected with the
monocyte/macrophage-specific marker CD68 (APAAP-method, magnification
x200). (B) Invading cells were also positive for CD14 (APAAP-method,
magnification x200). Adherent and invading cells were positive for
CD14 (C), CD16 (D), CD11b (E), and CD11c (F) (double staining;
magnification, x1,000).
|
 View larger version (142K): [in a new window] |
Figure 5. Expression of antigens after 3 days of coculture. Spheroids of the cell
line HT-29 were cocultured with freshly isolated monocyte/macrophages
for 3 days. Antigen expression was determined by immunohistochemistry
(APAAP-method and double staining). Expression of CD68 (A) and CD14 (B)
was clearly detectable over the whole cross-section of the spheroids
(APAAP method; magnification, x200). Expression of CD14 (C), CD16 (D),
CD11b (E), and CD11c (F) was randomly distributed in the spheroids
(double staining, magnification x1,000).
|
 View larger version (153K): [in a new window] |
Figure 6. Expression of antigens after 7 days of coculture. Spheroids of the cell
line HT-29 were cocultured with freshly isolated monocytes for 7 days.
Antigen expression was determined by immunohistochemistry (APAAP method
and double staining). (A) CD68-positive cells were clearly stained
after 7 days of coculture (APAAP method; magnification, x200). (B) No
CD14-positive cells were detected by the APAAP method (magnification,
x200). CD14 (C), CD16 (D), CD11b (E), and CD11c (F) were no longer
detectable on monocyte/macrophages inside the spheroids (double
staining; magnification, x1,000).
|
Identical experiments carried out with the three control cell lines did not show specific intestinal differentiation of the monocyte/macrophages inside the MCSs. No change in expression of the tested antigens was detectable during the 7-day culture period of MCSs with monocyte/macrophages. Figure 7 shows monocyte/macrophages inside MCSs of the cell line J82 after a culture period of 7 days. There was strong staining of CD68 as seen in experiments with the IEC-MCSs (Fig. 7A) . In contrast to IEC-MCSs, however, CD14 (Fig. 7B 7C) , CD16, CD11b, and CD11c could also be clearly detected at this time point (Fig. 7D E F ).
 View larger version (146K): [in a new window] |
Figure 7. Expression of antigens in MCSs of the control cell line J82 co-cultured
with monocyte/macrophages. After 7 days of coculture, expression of
CD68, CD14, CD16, CD11b, and CD11c was determined by
immunohistochemistry (APAAP method and double staining). CD68- (A) and
CD14-positive (B) cells were clearly detectable in spheroids of the
control cell line J82 (APAAP-method; magnification, x200). In contrast
to monocyte/macrophages cultured in IEC MCSs, invading cells in control
experiments still showed expression of CD14 (C), CD16 (D), CD11b (E),
and CD11c (F) after 7 days of coculture (double staining;
magnification, x1000).
|
In monocyte/macrophages cultured in spheroids of the IEC line HT-29, a decrease in the IL-1ß RNA copy numbers could be detected during the 7-day culture period. In a representative experiment under stimulation with 1 ng/mL of LPS (Fig. 8A ), the IL-1ß RNA copy number went down from 5 x 107 copies at day 1 of coculture to 1 x 107 copies at day 3. At day 7, the IL-1ß copy number in monocyte/macrophages cultured inside the HT-29 spheroids was 1 x 106 (n = 3). This represents a clear down-regulation of IL-1ß transcription in response to LPS and cannot be explained by the slight decrease in monocyte/macrophage number during the 7-day culture period. Similar results could be obtained with monocyte/macrophages differentiated in spheroids of the IEC line WiDr (data not shown). Stimulation with 50 ng/mL of LPS in HT-29 spheroids had no influence on the down-regulation of IL-1ß RNA copy numbers. After 7 days of monocyte/MCS co-culture, the copy number had decreased from 5 x 107 copies at day 1 to 5 x 105 at day 7 (Fig. 8B) . Experiments carried out with the control cell line J82 showed different results. The IL-1ß RNA copy number was constant over the 7-day coculture period (1 x 107) (Fig. 8C) . IL-1ß mRNA was lower or not detectable in unstimulated cultures (data not shown).
 View larger version (43K): [in a new window] |
Figure 8. Quantitative RT-PCR for IL-1ß in LPS-stimulated, disaggregated
spheroids. (A) HT-29 spheroids cultured with monocytes were
disaggregated by EDTA after 24 h, 3 days, and 7 days of coculture.
Cell suspensions were seeded into six-well plates and stimulated with 1
ng/mL of LPS for 2 h. Total RNA was isolated, and a competitive
quantitative RT-PCR for IL-1ß was performed. After 24 h of
coculture, the IL-1ß RNA copy number was 5 x 107
(lane 1). After 3 days of coculture, the IL-1ß copy number went down
to 1 x 107 (lane 7). At day 7 of coculture, a copy
number of 1 x 106 (lane14) for IL-1ß was
detectable. (B) HT-29 spheroids cultured with monocytes were
disaggregated by EDTA after 24 h, 3 days, and 7 days. Cell
suspensions were stimulated with 50 ng/mL of LPS for 2 h. Total
RNA was isolated, and IL-1ß RNA was quantified by competitive
quantitative RT-PCR. At day 1, the determined copy number for IL-1ß
mRNA again was 5 x 107 (lane 1); at day 3, the copy
number was 1 x 107 (lane 8). In this representative
experiment, the copy number for IL-1ß mRNA went down to 5 x
105 (lane 15) copies after 7 days of coculture. (C) Same
experiments carried out with spheroids of the control cell line J82.
The IL-1ß copy number was constant (1x107) over the
whole culture period. (D) Same experiments carried out with HT-29
spheroids cultured without monocyte/macrophages. No signal for IL-1ß
RNA could be detected. (E) Control experiment carried out with
monocyte/macrophages cultured without MCSs for 7 days. The copy number
for IL-1ß RNA was still >5 x 107.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Recently we showed that the phenotype of intestinal macrophages from normal, not inflamed mucosa is CD33+, CD44+, CD68+, CD14-, CD16-, CD11b-, CD11c-, HLA-DRlow, CD80-, and CD86- [16 ]. The observed lack of typical "activation-associated" monocyte/macrophage markers indicates anergy to luminal pathogens by normal intestinal macrophages. The reduced responsiveness to LPS stimulation found after monocyte/macrophage coculture with MCS of IEC lines in the experiments presented correlates well with the concept of an anergic macrophage type. Which factors, molecules, or components of cell aggregates could be responsible for this process of tissue-specific differentiation? The induction of monocyte/macrophage differentiation is not mediated by secretion of soluble factors by the IECs, as shown by transwell coculture experiments and experiments with conditioned medium (data not shown). In experiments with conditioned medium from either IEC monolayers or spheroids, monocytes also did not show any differentiation. Therefore, other forms of interaction between monocyte/macrophages and IECs seem relevant.
Until now, only a few studies have investigated direct interactions between macrophages and IECs. Martin and coworkers showed that macrophages were able to adhere to IECs and exchange a fluorescent dye when adhesion was present [56 ]. In normal intestinal mucosa, macrophages form a layer in the subepithelial region of the lamina propria separated from the epithelial cells by the basement membrane [2 , 4 ]. Pavli and co-workers recently demonstrated that, in the intestinal mucosa, macrophages are concentrated in a band immediately beneath the luminal epithelium [57 ]. Takahashi-Iwanaga and coworkers showed that the mucosal basement membrane is perforated with numerous small pores in vivo and that a direct interaction of subepithelial macrophages, sending cell protrusions through those pores, and of IECs is likely [45 ]. The reasons for those interactions and the functional consequences for both cell types have not yet been elucidated. In addition, it has been shown that macrophages or dendritic cells are capable of transmigration and reverse transmigration across basement membranes [58 ]. The basement membrane is also permeable for large molecules, as for example complement factors [59 ] which might mediate cell communication.
It has also been demonstrated that components of the basement membrane and extracellular matrix are involved in cell differentiation [60 ]. Gels of the extracellular matrix support cell differentiation in vitro. Many cell types form multicellular aggregates on matrix gels with increased cell-cell contacts as compared with regular monolayers on rigid matrix substrates. Hohn and co-workers demonstrated that cell-cell communication appears to play a subordinate role for cytodifferentiation in cell aggregates on matrix gels, so that substrate anchorage and physical properties of the substrate may be the decisive factors [60 ]. Basement membrane molecules and extracellular matrix are involved in heterotypic cell interactions leading to IEC differentiation [61 ]. Coculture models, hybrid intestines, and antisense approaches have shown that assembly of the basement membrane after close contacts between epithelial and fibroblastic cells is necessary for the expression of differentiation markers [61 ].
The same could be true for intestinal macrophages. Of course, MCSs do not resemble the in vivo situation with a layer of macrophages separated from the epithelial layer by the basement membrane. However, MCSs have been shown to contain large amounts of extracellular matrix [62 63 64 ] including typical basement membrane proteins such as collagen I, collagen III and IV, laminin, or fibronectin. Because cell-cell contact and extracellular matrix are thought particularly to regulate the expression of a subset of integrin molecules, it is important that the expression of these molecules in MCSs is very similar to the in vivo situation [65 , 66 ]. In fact, data from the colon carcinoma cell lines HRT-18 and CX-2 have shown that integrin expression of tumor cells depends on the culture system and that integrin expression in multicellular tumor spheroids is more similar to the in vivo situation in nude mouse tumors [66 ].
MCSs have recently been used for the study of the interaction of cell lines and immunocompetent cells [38 ]. However, in most cases these experiments were focused on the interaction of macrophages and tumor cells. The effects described in this manuscript were also observed with tumor cell lines because primary cells were not able to form spheroids [67 ]. It should be kept in mind that an explanation for our findings also might be the down-regulation of macrophage function as self-protection by tumor cells. However, because we showed that tumor-derived spheroids from other epithelial cells do not have this effect, it could be a specific property of these particular intestinal cells.
In cocultures of different epithelial cell lines of other than intestinal origin, no decrease of CD14 was observed [37 ]. Also in our control experiments, CD14, CD16, CD11b, and CD11c expression remained constant in the nonintestinal cell lines used. This demonstrates that the decrease of CD14, CD16, CD11b, and CD11c is specific for IECs at least for the three investigated cell lines forming MCSs. Therefore, it can be concluded that the described model offers a powerful tool for the study of tissue-specific monocyte/macrophage differentiation mechanisms.
The IL-1ß response to LPS stimulation was reduced in the monocyte/macrophages cocultured with IEC MCSs, which functionally resembles the normal intestinal macrophage. Rugtveit and coworkers showed that LPS stimulation had no effect on normal mucosal macrophages or CD14-depleted lamina propria MNCs from inflamed mucosa [68 ]. They suggested that, in chronically inflamed mucosa, macrophages with a monocyte-like phenotype expressing CD14 are primed for the production of tumor necrosis factor and IL-1. The mucosa-resident macrophages, in contrast, are unresponsive to LPS. Our differentiation model induced a CD14-negative macrophage phenotype with decreased response to LPS challenge similar to in vivo findings by Rugtveit and coworkers [68 ]. We cannot exclude that other events besides differentiation reduced IL-1ß mRNA in our cocultures. However, the number of copies remained constant in the J82 control cell line, indicating that at least the intestinal epithelial character of the HT-29 and the WiDr cell lines had influence on IL-1ß expression in the cocultured monocytes.
Several questions arise considering these findings. The mechanism of the induction of this specific differentiation must be elucidated. It is probably not only relevant for the intestinal mucosa but for macrophages in other tissues of endodermal origin. The identification of this (or these) differentiation factor(s) could be of therapeutic relevance for the treatment of inflammatory mucosal disorders in which the persistent presence of a CD14-positive, reactive macrophage population is of pathophysiological relevance.
Received September 4, 2000; revised March 31, 2001; accepted April 5, 2001.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  N. Kamada, K. Maeda, N. Inoue, T. Hisamatsu, S. Okamoto, K. S. Hong, T. Yamada, N. Watanabe, K. Tsuchimoto, H. Ogata, et al. Nonpathogenic Escherichia coli Strain Nissle 1917 Inhibits Signal Transduction in Intestinal Epithelial Cells Infect. Immun., January 1, 2008; 76(1): 214 - 220. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E Lippert, W Falk, F Bataille, T Kaehne, M Naumann, M Goeke, H Herfarth, J Schoelmerich, and G Rogler Soluble galectin-3 is a strong, colonic epithelial-cell-derived, lamina propria fibroblast-stimulating factor Gut, January 1, 2007; 56(1): 43 - 51. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Desnues, M. Ihrig, D. Raoult, and J.-L. Mege Whipple's Disease: a Macrophage Disease Clin. Vaccine Immunol., February 1, 2006; 13(2): 170 - 178. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Hausmann, F. Bataille, T. Spoettl, K. Schreiter, W. Falk, J. Schoelmerich, H. Herfarth, and G. Rogler Physiological Role of Macrophage Inflammatory Protein-3{alpha} Induction during Maturation of Intestinal Macrophages J. Immunol., August 1, 2005; 175(3): 1389 - 1398. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K Schreiter, M Hausmann, T Spoettl, U G Strauch, F Bataille, J Schoelmerich, H Herfarth, W Falk, and G Rogler Glycoprotein (gp) 96 expression: induced during differentiation of intestinal macrophages but impaired in Crohn's disease Gut, July 1, 2005; 54(7): 935 - 943. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
