Cell Tissue Res (1992) 270:273-279
Cell&Tissue Research © Springer-Verlag 1992
Characterization of macrophage-like cells in the external layers of human small and large intestine H.B. Mikkelsen and J.J. Rumessen Anatomy Department C, The Panum Institute, University of Copenhagen, Blegdamsvej 3, DK-2200, Copenhagen N, Denmark Received March 26, 1992 / Acceted June 29, 1992
Summary. In the external layers of h u m a n small and large intestine macrophage-like cells were characterized by immunohistochemical, histochemical and electronmicroscopical methods. Using immunohistochemistry and a n u m b e r of monoclonal antibodies, the presence and distribution of phenotypic subpopulations of macrophages were evaluated. In all locations macrophagelike cells were identified with antibody EBM11, which recognizes CD68 antigen, C3bi which recognizes C D I lb, and partly with an antibody which recognizes protein 150,95 ( C D l l c ) . Macrophage-like cells in the external muscle layer were H L A - D R - p o s i t i v e (expressing the M H C class-II antigen), in contrast to macrophage-like cells in the subserosa and submucosa. Macrophage-like cells in the external muscle layer were mostly acid phosphatase-negative, and at the electron-microscopic level they were found to have features of macrophages: prim a r y lysosomes, coated vesicles and pits. However, very few secondary lysosomes were present. Birbeck granules were not observed. It is concluded that in the external muscle layer of h u m a n small and large intestine numerous m a c r o p h a g e s of a special type are present. It is discussed whether this cell type plays a role in gastrointestinal motility and/or has an immunological function.
a wide range of molecules, phagocytosis and antigen processing, we found it of interest to investigate if macrophages were numerous also in the h u m a n external muscle layer and to analyse a possible correlation of surface antigen and functional properties. As studies on macrophages in the external muscle layer of h u m a n intestine have not been previously carried out, and because of the possibility of a regulatory function of macrophages in the external muscle layer and their possible role in pathogenesis of gastrointestinal diseases, we have performed immunohistochemical, histochemical and ultrastructural studies. Our findings m a y serve as a basis for future pathological studies.
Key words: Macrophages Small i n t e s t i n e - Large intestine - External muscle layer Immunohistochemistry - Histochemistry Electron microscopy - M a n
Monoclonal antibodies against CD68, CDllb and CDtlc were used as macrophage markers and to demonstrate complement receptor antigens (Beller et al. 1982; Franklin et al. 1986; Kelly et al. 1988; Myones et al. 1988; Hinglais et al. 1989; Micklem et al. 1989; Malizia et al. 1991), anti-CD1 la to demonstrate lymphocyte function antigen (Hynes 1987; Springer et al. 1987), and anti-HLA-DR to show antigen presenting cells (expressing MHC class-II antigen) (Spencer et al. 1987; 0liver et al. 1988). In addition, anti-CD3 and anti-CD22 were used to identify T- and B-lymphocytes, respectively. The immunohistochemical techniques have previously been described (Mikkelsen et al. i990, 1991). In short, frozen sections (8 gin), fixed for 4 rain in acetone, were covered with solutions of rabbit IgG (Dako) 1 : 5 for 30 rain, and incubated for 2 h with dilutions of anti-CD68 (Dako-Macrophage, EBM11) (1:100), antiCD 11a (Dako-CDl 1a) (1 : 20), anti-CD 11b (Dako-C3bi) (1 : l 0), antiCDllc (Dako-p150,95) (1:10), anti-HLA-DR (Dako-HLA-DR) (1:10), anti-CD22(l:20), (Dako) and anti-leu4 (CD3)(1:I00)
In previous papers we have described a constant presence of large numbers of regularly distributed macrophage-like cells ( M L C ) in the external muscle layer of mouse and guinea-pig small intestine (Mikkelsen et al. 1985, 1988a, b). The functions of this m a c r o p h a g e system of the external muscle layer are still unknown. Considering the functional heterogeneity of m o n o n u c l e a r phagocytes as expressed in other tissues: secretion of
Correspondence to: H.B. Mikkelsen
Materials and methods Freshly resected, uninvolved specimens of human duodenum, jejunum, ileum and colon (ascending, transverse and sigmoid) were obtained from 10 adult patients (ages: 47-81 years, median 66 years) undergoing surgery due to gastric ulcer, volvulus or carcinoma of the pancreas, colon or rectum. The patients had no other gastrointestinal disease. Patients with peritoneal carcinosis were excluded.
Immunohistochemistry
274
Fig. 1. Ileum, CD68 immunoreactivity of cells in subserosa (SS), in the longitudinal muscle layer (LM), at Auerbach's plexus (AP), in the circular muscle layer (CM), and in submucosa (SM). x 90, bar: 200 gm
Figs. 2, 3. Ileum, serial sections. Fig. 2. CD68-positive cells in the outer part of the longitudinal muscle layer (LM). The reaction is very strong in cells lining the longitudinal muscle layer and in septa (arrows) and weaker in cells in intralamellar septa (arrowhead). Fig. 3. HLA-DR-positive cells, which all appear to be CD68positive, x 400, bar: 20 pm
(Becton Dickinson). Control sections were incubated with monoclonal mouse anti-rat IgG (Zymed, 5 gg/ml). Mostly conventional serial sections were used to compare the antibody distribution, but also "turned sections" were applied, i,e., the first section was placed cut-surface down, and the next section was placed cut-sur-
face up. The upper portions were then treated with different antibodies, which means that the two immediate adjacent surfaces were stained. Both the PAP method and the alkaline phosphatase antialkaline phosphatase (APAAP) technique were used as previously described in detail (Mikkelsen et al. 1990, 1991).
275
Histochemistry Acid phosphatase activity was detected by using naphthol AS-TR phosphate (Sigma) as substrate and hexazotized pararosalinine as coupling agent (Lojda et al. 1976). The incubation was performed at pH 5.2 with 0.1 M acetate buffer at room temperature for 6080 rain. The incubation medium contained 10% polyvinyl alcohol (PVA G18/40 Wacher) in order to prevent enzyme diffusion.
Electron microscopy After resection, 1-2 mm pieces of intestine were cut and immersion fixed in 2% glutaraldehyde (Taab, Berkshire, U.K.) and 2% paraformaldehyde (Merck, Darmstadt, FRG) in 0.1 tool/1 phosphate buffer, pH 7.3, at 20° C for more than 2 h. After immersion, 1 x 2 mm pieces were cut and left in fixative overnight and postfixed in 2% OsO4 in 0.1 mol/1 phosphate buffer, pH 7.3, for 2 h. The tissue was dehydrated in alcohol, block stained with alcoholic uranyl acetate and embedded in Epon 812 R (Merck). Two-pro thick sections were stained with toluidine blue for light-microscopical investigation. Ultrathin sections (70 nm) were cut on an LKB U1tratome and contrasted with uranyl acetate and lead citrate. Electron microscopy was carried out with a Philips 400 microscope, The study was performed in accordance with the Helsinki Declaration II and approved by the ethical committee of Copenhagen.
Results Immunohistochemistry In the duodenum, jejunum, ileum and colon m a n y cells showed immunoreactivity with the antisera used. Differences between small intestine and colon were not obvious. Anti-CD68 stained most M L C and also gave the strongest reaction. CD68-positive cells were located in the subserosa, the muscle layers and in the submucosa (Figs. 1, 2, 4, 6, 8). Some o f the positive subserosal cells appeared to line the longitudinal muscle layer (Figs. 1, 4, 8). CD68-positive cells occurred in septa between the main muscle lamellae, in smaller intralamellar septa between the muscle cells, and at the level o f Auerbach's plexus where they lined the ganglia (Figs. 1, 2, 4, 6, 8). The M L C were evenly distributed throughout the external muscle layer. Most stained cells displayed cell processes. C D l l b - p o s i t i v e cells appeared to have similar morphology and localization as CD68-positive cells, but the reaction was weaker (Fig. 5). H L A - D R - p o s i t i v e cells showed a m o r p h o l o g y similar to that of CD68-positive cells and C D l l b - p o s i t i v e cells. However, they differed in localization. The n u m b e r of stained cells was scarce in subserosa and submucosa but seemed to be the same in the muscle layers, when the above-mentioned antibodies were applied (Figs. 2, 3). In addition, m a n y endothelial cells exhibited a positive immunoreaction. CD11 c-positive cells were few, and all CD11 c-positive cells appared to be CD68-positive. However, their distribution differed as C D l l c - p o s i t i v e cells were sparse inside muscle lamellae and in the submucosa. Some cells were H L A - D R - p o s i t i v e , but none of these cells were C D 11 a-positive.
The n u m b e r of C D I l a-positive cells was rather small in the muscle layers and in the submucosa (Fig. 7) where some positive cells overlapped with CD68- and H L A DR-positive cells. The m o r p h o l o g y and localization of CD3-positive ceils differed f r o m cells stained with the above-mentioned antibodies. These cells were round and devoid of cell processes. Their n u m b e r was limited in the subserosa, muscle layers and the submucosa, but they were numerous in the lamina propria. In the muscle layers, the CD3-positive cells were situated in intralamellar septa and in septa between the muscle lamellae. CD3-positire cells also lined the ganglia, but did not appear to overlap with m a c r o p h a g e antibody-positive cells, C D I 1a-positive cells and H L A - D R - p o s i t i v e cells. CD22-positive cells were present only in lymph nodules of the lamina propria but were very rarely observed in the remaining layers.
Histochemistry Acid-phosphatase activity was present in the surface epithelium just apical to the nucleus. In the lamina propria rather large cells were positive. Both in the submucosa and in the muscle layer ganglion cells contained heavily stained granules. A few cells in the submucosa, in septa between muscle lamellae, in cells lining ganglia and in the subserosa were positive (Fig. 9).
Figs. 4, 5. Ileum, "turned" sections. Fig. 4. CD68-positive cells (arrows) in the subserosa and the longitudinaI muscle layer. Fig. 5. The same ceils (arrows) less intensively stained by anti-CDllb. × 180, bar: 100 gm Figs. 6, 7. Colon, serial sections. Fig, 6. CD68-positive cells (arrows) in the circular muscle layer (CM) and the submucosa (SM). Fig. 7. shows rather few CDI 1a-positive cells (arrows). × 150, bar: I00 pm Figs. 8, 9. Colon. Fig. 8. Section with CD68-positive cells in the subserosa and septa of the longitudinal muscle layer. Fig. 9. Acid phosphatase-positive cells (arrows). x 175, bar. 100 pm Fig. 10. Ileum. Electron-micrograph of subserosa (SS) with a vein (V), and the longitudinal muscle layer (LM). A macrophage-like cell (34) is situated partly in the subserosa and partly in a septum (S) between the muscle cells. × 1500, bar: 10 pm Fig. 11. Ileum. Higher magnification of the macrophage-like cell in Fig. 10. The nucleus (N) is irregular, with prominent heterochromatin. Several types of vesicles are apparent: dense vesicles (D), light vesicles (L), intermediate vesicles (/), and coated vesicles (C); mitochondrion (Mi). Fibroblast (F). × 12000, bar: 1 ~tm Fig. 12. Ileum. Subserosal macrophage-like cell containing numerous dense vesicles (D), secondary lysosomes (SL), light vesicles (L), intermediate vesicles (/), and coated vesicles (C). The cell is surrounded by elastic fibers (E) and collagen fibers (Co). × 8600, bar." I pm Fig. 13. Ileum. Macrophage-like cell in intralamellar septum, surrounded by muscle cells (Mu). Naked nerve terminals (N) are close to the cell; a density of the presynaptic membrane is apparent (arrow). The maerophage-like cell contains many dense- (D), light(L), and coated (C) vesicles, x 11000, bar." I pm
276
277
278
Electron microscopy MLC were observed both in the subserosa (Figs. 10, 11, 12), in inter- and intralamellar septa of the muscle layers (Fig. 13), around Auerbach's plexus and in the submucosa. The MLC in the muscle layer were rather large cells with irregular contours and long processes. A basal lamina was not observed (Fig. 13). In the cytoplasm several types of vesicles could be distinguished: (1) dense vesicles with the features of primary or secondary lysosomes, the number of secondary lysosomes being rather small; (2) light vesicles (large irregular-shaped vesicles and small round or oval vesicles); (3) vesicles of intermediate appearance; and (4) coated vesicles and coated pits. Bundles of intermediate filaments were sometimes present. Birbeck granules were not observed. In the subserosa and submucosa the MLC were richer in secondary lysosomes. In addition, they were surrounded by collagen and elastic fibers (Fig. 12), whereas MLC in the muscle layers (in septa between muscle lamellae and intralamellar septa) were in close contact with fibroblasts and occasionally with nerves (Figs. 11, 13). Specific cell contacts were not seen. Discussion
In smaller mammals, mice and guinea pigs, MLC in the muscularis externa of the small intestine were regularly found in two locations: in the subserosa and at the level of Auerbach's plexus (AP). In addition, a few solitary cells were present at the level of the plexus muscularis profundus. At the level of AP, the MLC occurred as a constant and regularly distributed cell population with intimate associations with interstitial cells of Cajal (Thuneberg 1982; Mikkelsen et al. 1985, 1988b). Furthermore, MLC located between the muscle layers were found, in mice, to be constitutively M H C class-II antigen positive; they expressed the macrophage markers F4/80 and M 1/70 (CD 1 lb) (Mikkelsen et al. 1988 a), and were acid phosphatase-negative (unpublished observation). In the external muscle layer of the human gut, a regularly distributed, large cell population of macrophages was not obvious. The MLC were not in close contact with the interstitial cells of Cajal (Rumessen and Thuneberg 1991 ; Rumessen et al. 1992), but often in close contact with fibroblasts. Moreover, they showed only limited endocytotic activity (secondary lysosomes). By use of immunohistochemistry MLC were identified in all localizations with the pan-macrophage antibodies antiCD68, anti-CDllb, and partly with anti-CDllc. MLC in the human external muscle layer showed the M H C class-II antigen (HLA-DR-positive) in contrast to MLC in subserosa and submucosa, where presence of the MHC class-II antigen was not a prominent feature. This diversity in antigenic properties may reflect variation in a single macrophage cell line, or the co-existence of several distinct populations of macrophages with quite different functions. These findings support the concept of a special population of intramuscular MLC (Mikkelsen et al. 1988 a) also in the human intestine.
In a recent review of mucosal T-lymphocyte and macrophage subpopulations in the human intestine, Harvey and Jones (1991) described two major types of dendritic accessory cells: (i) non-phagocytotic dendritic cells, which show constitutive expression of M H C class-II antigen, and (ii) mononuclear phagocytes, which are involved in ingestion of debris and microbial defense, and which are capable of facultative M H C class-II expression. The nonphagocytotic dendritic cells are further subdivided into (i) follicular dendritic cells, which facilitate B-cell responses, and (ii) interdigitating reticulum cells (IDRC), which facilitate T-cell responses and occur diffusely throughout the lamina propria. The phenotype of the IDRC is R F D I , CD1 lc, CD68, and M H C class-II positive and acid phosphatase-negative. Expression of HLA-DR-antigens on gastrointestinal mucosal macrophages has been reported in several previous studies (Selby etal. 1983; Hume 1985; Wilders etal. 1985; Hume et al. 1987). In addition, in Crohn's disease M H C class-II positive cells were observed between the muscle fibers of the main layer (Wilders et al. 1984) or transmurally (Mayer et al. 1991). It is unlikely that the function of intramuscular MLC under normal circumstances should be the T-cell response, because the number of T-cells in the muscle layers is very low, and the lymphocyte function antigen (CDlla), which has been described to be involved in T-cell recognition, antigen presentation, and T-cell activation (Malizia et al. 1991), is scarce. However, the constant presence of MHC class-II positive MLC in the external muscle layer might indicate interactions between the immune system and functions of the external muscle layer. Arizona et al. (1990) have shown that, in the rat, mast cells and other bone marrow-derived cells in the mucosa of the small intestine occur in close proximity to neural processes. They suggested that the likelihood of functional interactions between nerves and a particular cell type may increase in direct relationship to physical proximity, whether such proximity is due to chance or to other factors. As MLC occasionally were situated close to nerve fibres, communication between the MLC and enteric nerves is a possibility. The lack of secondary lysosomes in MLC does not support a prominent scavenger role for these cells, although they may take part in a possible production and/ or catabolism of collagen and elastin fibers as well as turnover of muscle cells. Macrophages are able to produce a large variety of secretory products and to express surface molecules which give them the potential to interact with different cells and molecules in their local environment (Nathan 1987). The characteristic morphology, localization and surface antigens of this system of human intramuscular MLC might point to a possible regulatory role in human gut motility based on these properties.
Acknowledgements. Supported by The Michelsen Foundation and The Novo Foundation. The authors wish to thank Marianne Juhl Christensen and Birgit Risto for skilled technical assistance, and the staff of Surgical Department C, Rigshospitalet, for helpful cooperation.
279
References Arizono N, Matsuda S, Hattori T, Kojima Y, Maeda T, Galli SJ (1990) Anatomical variation in mast cell nerve associations in rat small intestine, heart, lung, and skin. Lab Invest 62:626634 Beller DI, Springer TA, Schreiber RD (1982) Anti-Mac-1 selectively inhibits the mouse and the human type three complement receptor. J Exp Med 156:1000 1009 Franklin WA, Mason DY, Pulford K, Falini B, Bliss E, Gatter KC, Stein H, Clarke LC, McGee JO'D (1986) Immunohistological analysis of human mononuclear phagocytes and dendritic cells by using monoclonal antibodies. Lab Invest 54:322-335 Harvey J, Jones DB (1991) Review - Human mucosal T-lymphocyte and macrophage subpopulations in normal and inflamed intestine. Clin Exp Allergy 21 : 549-560 Hinglais N, Kazatchkine MD, Mandet C, Appay MD, Bariety J (1989) Human liver Kupffer cells express CR1, CR3, and CR4 complement receptor antigens. An imnmnohistochemical study. Lab Invest 61:509-514 Hume DA (1985) Immunohistochemical analysis of murine mononuclear phagocytes that express class II major histocompatibility antigens. Immunobiology 170:381-389 Hume DA, Allan W, Hogan PG, Doe WF (1987) Immunohistochemical characterization of macrophages in human liver and gastrointestinal tract: Expression of CD4, HLA-DR, OKM1, and the mature macrophage marker 25F9 in normal and diseased tissue. J Leukoc Biol 42:474-484 Hynes RO (1987) Integrins: A family of cell surface receptors. Cell 48 : 54%554 Kelly PMA, Bliss E, Morton JA, Burns J, McGee JO'D (1988) Monoclonal antibody EBM//1 : high cellular specificity for human macrophages. J Clin Pathol 41:5/0-515 Lojda Z, Grossrau R, Schiebler TH (1976) Enzymhistochemische Methoden. Springer, Berlin Heidelberg New York Malizia G, Calabrese A, Cottone M, Raimondo M, Trejdosiewicz LK, Smart CJ, Oliva L, Pagliaro L (1991) Expression of leukocyte adhesion molecules by mucosal mononucIear phagocytes in inflammatory bowel disease. Gastroenterology 100 : 150-159 Mayer L, Eisenhardt D, Salomon P, Bauer W, Plous R, Piccinini L (1991) Expression of class II molecules on intestinal epithelial cells in humans. Differences between normal and inflammatory bowel disease. Gastroenterology 100:3-12 Mickiem K J, Rigney E, Cordell J, Simmons D, Stross P, Turley H, Seed B, Mason DY (1989) A human macrophage-associated antigen (CD68) detected by six different monoclonal antibodies. Br J Haematol 73 : 6-11 Mikkelsen HB, Thuneberg L, Rumessen JJ, Thorball N (1985) Macrophage-like cells in the muscularis externa of mouse small intestine. Anat Rec 2/3:77-86
Mikkelsen HB, Mirsky R, Jessen KR, Thuneberg L (1988a) Macrophage-like cells in muscularis externa of mouse small intestine: Immunohistochemical localization of F4/80, M1/70, and Ia-antigen. Cell Tissue Res 252: 301-306 Mikkelsen HB, Thuneberg L, Wittrup IH (1988b) Selective double staining of intestitial cells of Cajal and macrophage-like cells in small intestine by an improved supravital methylene blue technique combined with FITC-dextran uptake. Anat Embryol (Berl) 178:191-195 Mikkelsen HB, Rumessen IT, Thuneberg L (1990) Prostaglandin H synthase immunoreactivity localized by immunoperoxidase technique (PAp) in the small intestine and kidney of rabbit and guinea-pig. Histochemistry 93 : 363-367 Mikkelsen HB, Rumessen JJ, Qvortrup K (1991) Prostaglandin H synthase immunoreactivity in human gut. An immunohistochemical study. Histochemistry 96:295-299 Myones BL, DalzeI1 JG, Hogg N, Ross GD (1988) Neutrophil and monocyte cell surface p150,95 has iC3b-receptor (CR4) activity resembling CR3. J Clin Invest 82:640-651 Nathan CF (1987) Secretory products of macrophages. J Clin Invest 79:319-326 Oliver AM, Thomson AW, Sewell HF, Abramovich DR (1988) Major histocompatibility complex (MHC) class II antigen (HLA-DR, DQ, and DP) expression in human fetal endocrine organs and gut. Scand J Immunol 27:731-737 Rumessen JJ, Thuneberg L (1991) Interstitial cells of Cajal in human small intestine. Ultrastructural identification and organization between the main smooth muscle layers. Gastroenterology 100:1417-1431 Rumessen J J, Mikkelsen HB, Thuneberg L (1992) Ultrastructure of interstitial cells of Caj aI associated with deep muscular plexus of human small intestine. Gastroenterology 102: 56-68 Selby WS, Poulter LW, Hobbs S, Jewell DP, Janossy G (1983) Heterogeneity of HLA-DR-positive histiocytes in human intestinal lamina propria: a combined histochemical and immunohistological analysis. J Clin Pathol 36:379-384 Spencer J, MacDonald TT, Isaacson PG (1987) Heterogeneity of non-lymphoid cells expressing HLA-D region antigens in human fetal gut. Clin Exp Immunol 67:415-424 Springer TA, Dustin ML, Kishimoto TK, Marlin SD (1987) The lymphocyte function-associated LFA-I, CD2, and LFA-3 molecules: Cell adhesion receptors of the immune system. Ann Rev Immunol 5: 223-252 Thuneberg L (1982) Interstitial cells of Cajal: intestinal pacemaker cells? Adv Anat Embryol Cell Biol 71 :/-130 Wilders MM, Drexhage HA, Kokje M, Verspaget HW, Meuwissen SGM (/984) Veiled cells in chronic idiopathic inflammatory bowel disease. Clin Exp Immunol 55:377-387 Wilders MM, Drexhage HA, Sminia T, Mullink H, Weltevreden EF, Plesh BEC, Meuwissen SGM (1985) Veiled cells in the gastrointestinal tract. Acta Chir Scand 525 [Suppl] : 93-111
Cell&Tissue Research © Springer-Verlag 1992
Characterization of macrophage-like cells in the external layers of human small and large intestine H.B. Mikkelsen and J.J. Rumessen Anatomy Department C, The Panum Institute, University of Copenhagen, Blegdamsvej 3, DK-2200, Copenhagen N, Denmark Received March 26, 1992 / Acceted June 29, 1992
Summary. In the external layers of h u m a n small and large intestine macrophage-like cells were characterized by immunohistochemical, histochemical and electronmicroscopical methods. Using immunohistochemistry and a n u m b e r of monoclonal antibodies, the presence and distribution of phenotypic subpopulations of macrophages were evaluated. In all locations macrophagelike cells were identified with antibody EBM11, which recognizes CD68 antigen, C3bi which recognizes C D I lb, and partly with an antibody which recognizes protein 150,95 ( C D l l c ) . Macrophage-like cells in the external muscle layer were H L A - D R - p o s i t i v e (expressing the M H C class-II antigen), in contrast to macrophage-like cells in the subserosa and submucosa. Macrophage-like cells in the external muscle layer were mostly acid phosphatase-negative, and at the electron-microscopic level they were found to have features of macrophages: prim a r y lysosomes, coated vesicles and pits. However, very few secondary lysosomes were present. Birbeck granules were not observed. It is concluded that in the external muscle layer of h u m a n small and large intestine numerous m a c r o p h a g e s of a special type are present. It is discussed whether this cell type plays a role in gastrointestinal motility and/or has an immunological function.
a wide range of molecules, phagocytosis and antigen processing, we found it of interest to investigate if macrophages were numerous also in the h u m a n external muscle layer and to analyse a possible correlation of surface antigen and functional properties. As studies on macrophages in the external muscle layer of h u m a n intestine have not been previously carried out, and because of the possibility of a regulatory function of macrophages in the external muscle layer and their possible role in pathogenesis of gastrointestinal diseases, we have performed immunohistochemical, histochemical and ultrastructural studies. Our findings m a y serve as a basis for future pathological studies.
Key words: Macrophages Small i n t e s t i n e - Large intestine - External muscle layer Immunohistochemistry - Histochemistry Electron microscopy - M a n
Monoclonal antibodies against CD68, CDllb and CDtlc were used as macrophage markers and to demonstrate complement receptor antigens (Beller et al. 1982; Franklin et al. 1986; Kelly et al. 1988; Myones et al. 1988; Hinglais et al. 1989; Micklem et al. 1989; Malizia et al. 1991), anti-CD1 la to demonstrate lymphocyte function antigen (Hynes 1987; Springer et al. 1987), and anti-HLA-DR to show antigen presenting cells (expressing MHC class-II antigen) (Spencer et al. 1987; 0liver et al. 1988). In addition, anti-CD3 and anti-CD22 were used to identify T- and B-lymphocytes, respectively. The immunohistochemical techniques have previously been described (Mikkelsen et al. i990, 1991). In short, frozen sections (8 gin), fixed for 4 rain in acetone, were covered with solutions of rabbit IgG (Dako) 1 : 5 for 30 rain, and incubated for 2 h with dilutions of anti-CD68 (Dako-Macrophage, EBM11) (1:100), antiCD 11a (Dako-CDl 1a) (1 : 20), anti-CD 11b (Dako-C3bi) (1 : l 0), antiCDllc (Dako-p150,95) (1:10), anti-HLA-DR (Dako-HLA-DR) (1:10), anti-CD22(l:20), (Dako) and anti-leu4 (CD3)(1:I00)
In previous papers we have described a constant presence of large numbers of regularly distributed macrophage-like cells ( M L C ) in the external muscle layer of mouse and guinea-pig small intestine (Mikkelsen et al. 1985, 1988a, b). The functions of this m a c r o p h a g e system of the external muscle layer are still unknown. Considering the functional heterogeneity of m o n o n u c l e a r phagocytes as expressed in other tissues: secretion of
Correspondence to: H.B. Mikkelsen
Materials and methods Freshly resected, uninvolved specimens of human duodenum, jejunum, ileum and colon (ascending, transverse and sigmoid) were obtained from 10 adult patients (ages: 47-81 years, median 66 years) undergoing surgery due to gastric ulcer, volvulus or carcinoma of the pancreas, colon or rectum. The patients had no other gastrointestinal disease. Patients with peritoneal carcinosis were excluded.
Immunohistochemistry
274
Fig. 1. Ileum, CD68 immunoreactivity of cells in subserosa (SS), in the longitudinal muscle layer (LM), at Auerbach's plexus (AP), in the circular muscle layer (CM), and in submucosa (SM). x 90, bar: 200 gm
Figs. 2, 3. Ileum, serial sections. Fig. 2. CD68-positive cells in the outer part of the longitudinal muscle layer (LM). The reaction is very strong in cells lining the longitudinal muscle layer and in septa (arrows) and weaker in cells in intralamellar septa (arrowhead). Fig. 3. HLA-DR-positive cells, which all appear to be CD68positive, x 400, bar: 20 pm
(Becton Dickinson). Control sections were incubated with monoclonal mouse anti-rat IgG (Zymed, 5 gg/ml). Mostly conventional serial sections were used to compare the antibody distribution, but also "turned sections" were applied, i,e., the first section was placed cut-surface down, and the next section was placed cut-sur-
face up. The upper portions were then treated with different antibodies, which means that the two immediate adjacent surfaces were stained. Both the PAP method and the alkaline phosphatase antialkaline phosphatase (APAAP) technique were used as previously described in detail (Mikkelsen et al. 1990, 1991).
275
Histochemistry Acid phosphatase activity was detected by using naphthol AS-TR phosphate (Sigma) as substrate and hexazotized pararosalinine as coupling agent (Lojda et al. 1976). The incubation was performed at pH 5.2 with 0.1 M acetate buffer at room temperature for 6080 rain. The incubation medium contained 10% polyvinyl alcohol (PVA G18/40 Wacher) in order to prevent enzyme diffusion.
Electron microscopy After resection, 1-2 mm pieces of intestine were cut and immersion fixed in 2% glutaraldehyde (Taab, Berkshire, U.K.) and 2% paraformaldehyde (Merck, Darmstadt, FRG) in 0.1 tool/1 phosphate buffer, pH 7.3, at 20° C for more than 2 h. After immersion, 1 x 2 mm pieces were cut and left in fixative overnight and postfixed in 2% OsO4 in 0.1 mol/1 phosphate buffer, pH 7.3, for 2 h. The tissue was dehydrated in alcohol, block stained with alcoholic uranyl acetate and embedded in Epon 812 R (Merck). Two-pro thick sections were stained with toluidine blue for light-microscopical investigation. Ultrathin sections (70 nm) were cut on an LKB U1tratome and contrasted with uranyl acetate and lead citrate. Electron microscopy was carried out with a Philips 400 microscope, The study was performed in accordance with the Helsinki Declaration II and approved by the ethical committee of Copenhagen.
Results Immunohistochemistry In the duodenum, jejunum, ileum and colon m a n y cells showed immunoreactivity with the antisera used. Differences between small intestine and colon were not obvious. Anti-CD68 stained most M L C and also gave the strongest reaction. CD68-positive cells were located in the subserosa, the muscle layers and in the submucosa (Figs. 1, 2, 4, 6, 8). Some o f the positive subserosal cells appeared to line the longitudinal muscle layer (Figs. 1, 4, 8). CD68-positive cells occurred in septa between the main muscle lamellae, in smaller intralamellar septa between the muscle cells, and at the level o f Auerbach's plexus where they lined the ganglia (Figs. 1, 2, 4, 6, 8). The M L C were evenly distributed throughout the external muscle layer. Most stained cells displayed cell processes. C D l l b - p o s i t i v e cells appeared to have similar morphology and localization as CD68-positive cells, but the reaction was weaker (Fig. 5). H L A - D R - p o s i t i v e cells showed a m o r p h o l o g y similar to that of CD68-positive cells and C D l l b - p o s i t i v e cells. However, they differed in localization. The n u m b e r of stained cells was scarce in subserosa and submucosa but seemed to be the same in the muscle layers, when the above-mentioned antibodies were applied (Figs. 2, 3). In addition, m a n y endothelial cells exhibited a positive immunoreaction. CD11 c-positive cells were few, and all CD11 c-positive cells appared to be CD68-positive. However, their distribution differed as C D l l c - p o s i t i v e cells were sparse inside muscle lamellae and in the submucosa. Some cells were H L A - D R - p o s i t i v e , but none of these cells were C D 11 a-positive.
The n u m b e r of C D I l a-positive cells was rather small in the muscle layers and in the submucosa (Fig. 7) where some positive cells overlapped with CD68- and H L A DR-positive cells. The m o r p h o l o g y and localization of CD3-positive ceils differed f r o m cells stained with the above-mentioned antibodies. These cells were round and devoid of cell processes. Their n u m b e r was limited in the subserosa, muscle layers and the submucosa, but they were numerous in the lamina propria. In the muscle layers, the CD3-positive cells were situated in intralamellar septa and in septa between the muscle lamellae. CD3-positire cells also lined the ganglia, but did not appear to overlap with m a c r o p h a g e antibody-positive cells, C D I 1a-positive cells and H L A - D R - p o s i t i v e cells. CD22-positive cells were present only in lymph nodules of the lamina propria but were very rarely observed in the remaining layers.
Histochemistry Acid-phosphatase activity was present in the surface epithelium just apical to the nucleus. In the lamina propria rather large cells were positive. Both in the submucosa and in the muscle layer ganglion cells contained heavily stained granules. A few cells in the submucosa, in septa between muscle lamellae, in cells lining ganglia and in the subserosa were positive (Fig. 9).
Figs. 4, 5. Ileum, "turned" sections. Fig. 4. CD68-positive cells (arrows) in the subserosa and the longitudinaI muscle layer. Fig. 5. The same ceils (arrows) less intensively stained by anti-CDllb. × 180, bar: 100 gm Figs. 6, 7. Colon, serial sections. Fig, 6. CD68-positive cells (arrows) in the circular muscle layer (CM) and the submucosa (SM). Fig. 7. shows rather few CDI 1a-positive cells (arrows). × 150, bar: I00 pm Figs. 8, 9. Colon. Fig. 8. Section with CD68-positive cells in the subserosa and septa of the longitudinal muscle layer. Fig. 9. Acid phosphatase-positive cells (arrows). x 175, bar. 100 pm Fig. 10. Ileum. Electron-micrograph of subserosa (SS) with a vein (V), and the longitudinal muscle layer (LM). A macrophage-like cell (34) is situated partly in the subserosa and partly in a septum (S) between the muscle cells. × 1500, bar: 10 pm Fig. 11. Ileum. Higher magnification of the macrophage-like cell in Fig. 10. The nucleus (N) is irregular, with prominent heterochromatin. Several types of vesicles are apparent: dense vesicles (D), light vesicles (L), intermediate vesicles (/), and coated vesicles (C); mitochondrion (Mi). Fibroblast (F). × 12000, bar: 1 ~tm Fig. 12. Ileum. Subserosal macrophage-like cell containing numerous dense vesicles (D), secondary lysosomes (SL), light vesicles (L), intermediate vesicles (/), and coated vesicles (C). The cell is surrounded by elastic fibers (E) and collagen fibers (Co). × 8600, bar." I pm Fig. 13. Ileum. Macrophage-like cell in intralamellar septum, surrounded by muscle cells (Mu). Naked nerve terminals (N) are close to the cell; a density of the presynaptic membrane is apparent (arrow). The maerophage-like cell contains many dense- (D), light(L), and coated (C) vesicles, x 11000, bar." I pm
276
277
278
Electron microscopy MLC were observed both in the subserosa (Figs. 10, 11, 12), in inter- and intralamellar septa of the muscle layers (Fig. 13), around Auerbach's plexus and in the submucosa. The MLC in the muscle layer were rather large cells with irregular contours and long processes. A basal lamina was not observed (Fig. 13). In the cytoplasm several types of vesicles could be distinguished: (1) dense vesicles with the features of primary or secondary lysosomes, the number of secondary lysosomes being rather small; (2) light vesicles (large irregular-shaped vesicles and small round or oval vesicles); (3) vesicles of intermediate appearance; and (4) coated vesicles and coated pits. Bundles of intermediate filaments were sometimes present. Birbeck granules were not observed. In the subserosa and submucosa the MLC were richer in secondary lysosomes. In addition, they were surrounded by collagen and elastic fibers (Fig. 12), whereas MLC in the muscle layers (in septa between muscle lamellae and intralamellar septa) were in close contact with fibroblasts and occasionally with nerves (Figs. 11, 13). Specific cell contacts were not seen. Discussion
In smaller mammals, mice and guinea pigs, MLC in the muscularis externa of the small intestine were regularly found in two locations: in the subserosa and at the level of Auerbach's plexus (AP). In addition, a few solitary cells were present at the level of the plexus muscularis profundus. At the level of AP, the MLC occurred as a constant and regularly distributed cell population with intimate associations with interstitial cells of Cajal (Thuneberg 1982; Mikkelsen et al. 1985, 1988b). Furthermore, MLC located between the muscle layers were found, in mice, to be constitutively M H C class-II antigen positive; they expressed the macrophage markers F4/80 and M 1/70 (CD 1 lb) (Mikkelsen et al. 1988 a), and were acid phosphatase-negative (unpublished observation). In the external muscle layer of the human gut, a regularly distributed, large cell population of macrophages was not obvious. The MLC were not in close contact with the interstitial cells of Cajal (Rumessen and Thuneberg 1991 ; Rumessen et al. 1992), but often in close contact with fibroblasts. Moreover, they showed only limited endocytotic activity (secondary lysosomes). By use of immunohistochemistry MLC were identified in all localizations with the pan-macrophage antibodies antiCD68, anti-CDllb, and partly with anti-CDllc. MLC in the human external muscle layer showed the M H C class-II antigen (HLA-DR-positive) in contrast to MLC in subserosa and submucosa, where presence of the MHC class-II antigen was not a prominent feature. This diversity in antigenic properties may reflect variation in a single macrophage cell line, or the co-existence of several distinct populations of macrophages with quite different functions. These findings support the concept of a special population of intramuscular MLC (Mikkelsen et al. 1988 a) also in the human intestine.
In a recent review of mucosal T-lymphocyte and macrophage subpopulations in the human intestine, Harvey and Jones (1991) described two major types of dendritic accessory cells: (i) non-phagocytotic dendritic cells, which show constitutive expression of M H C class-II antigen, and (ii) mononuclear phagocytes, which are involved in ingestion of debris and microbial defense, and which are capable of facultative M H C class-II expression. The nonphagocytotic dendritic cells are further subdivided into (i) follicular dendritic cells, which facilitate B-cell responses, and (ii) interdigitating reticulum cells (IDRC), which facilitate T-cell responses and occur diffusely throughout the lamina propria. The phenotype of the IDRC is R F D I , CD1 lc, CD68, and M H C class-II positive and acid phosphatase-negative. Expression of HLA-DR-antigens on gastrointestinal mucosal macrophages has been reported in several previous studies (Selby etal. 1983; Hume 1985; Wilders etal. 1985; Hume et al. 1987). In addition, in Crohn's disease M H C class-II positive cells were observed between the muscle fibers of the main layer (Wilders et al. 1984) or transmurally (Mayer et al. 1991). It is unlikely that the function of intramuscular MLC under normal circumstances should be the T-cell response, because the number of T-cells in the muscle layers is very low, and the lymphocyte function antigen (CDlla), which has been described to be involved in T-cell recognition, antigen presentation, and T-cell activation (Malizia et al. 1991), is scarce. However, the constant presence of MHC class-II positive MLC in the external muscle layer might indicate interactions between the immune system and functions of the external muscle layer. Arizona et al. (1990) have shown that, in the rat, mast cells and other bone marrow-derived cells in the mucosa of the small intestine occur in close proximity to neural processes. They suggested that the likelihood of functional interactions between nerves and a particular cell type may increase in direct relationship to physical proximity, whether such proximity is due to chance or to other factors. As MLC occasionally were situated close to nerve fibres, communication between the MLC and enteric nerves is a possibility. The lack of secondary lysosomes in MLC does not support a prominent scavenger role for these cells, although they may take part in a possible production and/ or catabolism of collagen and elastin fibers as well as turnover of muscle cells. Macrophages are able to produce a large variety of secretory products and to express surface molecules which give them the potential to interact with different cells and molecules in their local environment (Nathan 1987). The characteristic morphology, localization and surface antigens of this system of human intramuscular MLC might point to a possible regulatory role in human gut motility based on these properties.
Acknowledgements. Supported by The Michelsen Foundation and The Novo Foundation. The authors wish to thank Marianne Juhl Christensen and Birgit Risto for skilled technical assistance, and the staff of Surgical Department C, Rigshospitalet, for helpful cooperation.
279
References Arizono N, Matsuda S, Hattori T, Kojima Y, Maeda T, Galli SJ (1990) Anatomical variation in mast cell nerve associations in rat small intestine, heart, lung, and skin. Lab Invest 62:626634 Beller DI, Springer TA, Schreiber RD (1982) Anti-Mac-1 selectively inhibits the mouse and the human type three complement receptor. J Exp Med 156:1000 1009 Franklin WA, Mason DY, Pulford K, Falini B, Bliss E, Gatter KC, Stein H, Clarke LC, McGee JO'D (1986) Immunohistological analysis of human mononuclear phagocytes and dendritic cells by using monoclonal antibodies. Lab Invest 54:322-335 Harvey J, Jones DB (1991) Review - Human mucosal T-lymphocyte and macrophage subpopulations in normal and inflamed intestine. Clin Exp Allergy 21 : 549-560 Hinglais N, Kazatchkine MD, Mandet C, Appay MD, Bariety J (1989) Human liver Kupffer cells express CR1, CR3, and CR4 complement receptor antigens. An imnmnohistochemical study. Lab Invest 61:509-514 Hume DA (1985) Immunohistochemical analysis of murine mononuclear phagocytes that express class II major histocompatibility antigens. Immunobiology 170:381-389 Hume DA, Allan W, Hogan PG, Doe WF (1987) Immunohistochemical characterization of macrophages in human liver and gastrointestinal tract: Expression of CD4, HLA-DR, OKM1, and the mature macrophage marker 25F9 in normal and diseased tissue. J Leukoc Biol 42:474-484 Hynes RO (1987) Integrins: A family of cell surface receptors. Cell 48 : 54%554 Kelly PMA, Bliss E, Morton JA, Burns J, McGee JO'D (1988) Monoclonal antibody EBM//1 : high cellular specificity for human macrophages. J Clin Pathol 41:5/0-515 Lojda Z, Grossrau R, Schiebler TH (1976) Enzymhistochemische Methoden. Springer, Berlin Heidelberg New York Malizia G, Calabrese A, Cottone M, Raimondo M, Trejdosiewicz LK, Smart CJ, Oliva L, Pagliaro L (1991) Expression of leukocyte adhesion molecules by mucosal mononucIear phagocytes in inflammatory bowel disease. Gastroenterology 100 : 150-159 Mayer L, Eisenhardt D, Salomon P, Bauer W, Plous R, Piccinini L (1991) Expression of class II molecules on intestinal epithelial cells in humans. Differences between normal and inflammatory bowel disease. Gastroenterology 100:3-12 Mickiem K J, Rigney E, Cordell J, Simmons D, Stross P, Turley H, Seed B, Mason DY (1989) A human macrophage-associated antigen (CD68) detected by six different monoclonal antibodies. Br J Haematol 73 : 6-11 Mikkelsen HB, Thuneberg L, Rumessen JJ, Thorball N (1985) Macrophage-like cells in the muscularis externa of mouse small intestine. Anat Rec 2/3:77-86
Mikkelsen HB, Mirsky R, Jessen KR, Thuneberg L (1988a) Macrophage-like cells in muscularis externa of mouse small intestine: Immunohistochemical localization of F4/80, M1/70, and Ia-antigen. Cell Tissue Res 252: 301-306 Mikkelsen HB, Thuneberg L, Wittrup IH (1988b) Selective double staining of intestitial cells of Cajal and macrophage-like cells in small intestine by an improved supravital methylene blue technique combined with FITC-dextran uptake. Anat Embryol (Berl) 178:191-195 Mikkelsen HB, Rumessen IT, Thuneberg L (1990) Prostaglandin H synthase immunoreactivity localized by immunoperoxidase technique (PAp) in the small intestine and kidney of rabbit and guinea-pig. Histochemistry 93 : 363-367 Mikkelsen HB, Rumessen JJ, Qvortrup K (1991) Prostaglandin H synthase immunoreactivity in human gut. An immunohistochemical study. Histochemistry 96:295-299 Myones BL, DalzeI1 JG, Hogg N, Ross GD (1988) Neutrophil and monocyte cell surface p150,95 has iC3b-receptor (CR4) activity resembling CR3. J Clin Invest 82:640-651 Nathan CF (1987) Secretory products of macrophages. J Clin Invest 79:319-326 Oliver AM, Thomson AW, Sewell HF, Abramovich DR (1988) Major histocompatibility complex (MHC) class II antigen (HLA-DR, DQ, and DP) expression in human fetal endocrine organs and gut. Scand J Immunol 27:731-737 Rumessen JJ, Thuneberg L (1991) Interstitial cells of Cajal in human small intestine. Ultrastructural identification and organization between the main smooth muscle layers. Gastroenterology 100:1417-1431 Rumessen J J, Mikkelsen HB, Thuneberg L (1992) Ultrastructure of interstitial cells of Caj aI associated with deep muscular plexus of human small intestine. Gastroenterology 102: 56-68 Selby WS, Poulter LW, Hobbs S, Jewell DP, Janossy G (1983) Heterogeneity of HLA-DR-positive histiocytes in human intestinal lamina propria: a combined histochemical and immunohistological analysis. J Clin Pathol 36:379-384 Spencer J, MacDonald TT, Isaacson PG (1987) Heterogeneity of non-lymphoid cells expressing HLA-D region antigens in human fetal gut. Clin Exp Immunol 67:415-424 Springer TA, Dustin ML, Kishimoto TK, Marlin SD (1987) The lymphocyte function-associated LFA-I, CD2, and LFA-3 molecules: Cell adhesion receptors of the immune system. Ann Rev Immunol 5: 223-252 Thuneberg L (1982) Interstitial cells of Cajal: intestinal pacemaker cells? Adv Anat Embryol Cell Biol 71 :/-130 Wilders MM, Drexhage HA, Kokje M, Verspaget HW, Meuwissen SGM (/984) Veiled cells in chronic idiopathic inflammatory bowel disease. Clin Exp Immunol 55:377-387 Wilders MM, Drexhage HA, Sminia T, Mullink H, Weltevreden EF, Plesh BEC, Meuwissen SGM (1985) Veiled cells in the gastrointestinal tract. Acta Chir Scand 525 [Suppl] : 93-111
