Quantitative Assessment of Macrophages in the Muscularis Externa of Mouse Intestines.код для вставкиСкачать
THE ANATOMICAL RECORD 294:1557–1565 (2011) Quantitative Assessment of Macrophages in the Muscularis Externa of Mouse Intestines H.B. MIKKELSEN,1* J.O. LARSEN,2 P. FROH,1 AND T.H. NGUYEN1 Department of Cellular and Molecular Medicine, Faculty of Health Sciences, University of Copenhagen, Denmark 2 Department of Neuroscience and Pharmacology, Faculty of Health Sciences, University of Copenhagen, Denmark 1 ABSTRACT Quantiﬁcation of intestinal cells is challenging for several reasons: The cell densities vary throughout the intestines and may be age dependent. Some cell types are ramiﬁed and/or can change shape and size. Additionally, immunolabeling is needed for the correct identiﬁcation of cell type. Immunolabeling is dependent on both up- and down-regulation of the antigen being labeled as well as on the primary and secondary antibodies, the ﬁxation, and the enhancement procedures. Here, we provide a detailed description of immunolabeling of CD169þ cells and major histocompatibility class II antigen (MHCIIþ) cells and the subsequent quantiﬁcation of these cells using design-based stereology in the intestinal muscularis externa. We used young (5-weeks-old) and adult (10-weeksold) mice. Cell densities were higher in jejunum-ileum, when compared with colon. In jejunum/ileum, the cell densities increased in oral-anal direction in adults, whereas the densities were highest in the midpart in young animals. In colon, the cell densities decreased in oral-anal direction in both groups of animals. Except for the density of MHCIIþ cells in colon, the cell densities were highest in young animals. Densities of CD169þ and MHCIIþ cells did not differ, except in the colon of young animals where the CD169þ density was almost twice as high as the MHCIIþ density. CD169 and MHCII antigens seem to be expressed simultaneously by the same cell in jejunum/ileum. We conclude that cell densities depend on both the age of the mouse and on the location in the intestines. Anat C 2011 Wiley-Liss, Inc. Rec, 294:1557–1565, 2011. V Key words: CD1691 cells; MHCII1 cells; macrophages; stereology; mouse; intestine; young; adult Robust quantitative data are often important in cell characterization in experimental, developmental, and pathologic studies. In intestinal motility disturbances, for example, both the densities of interstitial cell of Cajal, macrophages, and mast cells may be of interest (Mikkelsen, 2010). Most studies of these cells are, however, based on semiquantitative techniques, where the investigator counts cell proﬁles in few arbitrarily chosen ﬁelds of visions in a speciﬁed region of the intestine or measure the amount of ﬂuorescence in sections. These techniques for intestinal cell quantiﬁcation are biased to variable degrees. Both the number of cell proﬁles per section area and the amount of ﬂuorescence are functions of both the size and shape of the cells and of the C 2011 WILEY-LISS, INC. V amount of intercellular tissue. As some of the cells are ramiﬁed and can change their shape and/or size, it is especially important to apply a counting rule that is Grant sponsor: Vera and Carl Johan Michaelsens Foundation. *Correspondence to: H.B. Mikkelsen, Department of Cellular and Molecular Medicine, The Panum Building 22.4, Faculty of Health Sciences, University of Copenhagen, Blegdamsvej 3, 2200 Copenhagen N, Denmark. E-mail: email@example.com Received 28 October 2010; Accepted 2 May 2011 DOI 10.1002/ar.21444 Published online 1 August 2011 in Wiley Online Library (wileyonlinelibrary.com). 1558 MIKKELSEN ET AL. number-weighted and not size- and/or shape-weighted. In addition, the densities of intestinal cells vary regionally, which necessitates random sampling of ﬁelds of visions. Also in a recent study on gastrointestinal neuromuscular pathology, a need to standardize collection, processing, and quantiﬁcation of neuronal and glial elements in enteric neuropathologic samples was emphasized (Knowles et al., 2009). Here, we present a design-based stereological sampling technique that circumvents the sampling-related problems associated with cell quantiﬁcation. In order for the actual application of this technique to be unbiased, it is necessary to be able to correctly identify all the cells of interest. Immunohistochemical methods are often used to distinguish the cells at the light microscopic level, and the applied staining techniques should be able to either immunostain all cells of interest exclusively without signiﬁcant background staining or if more than one cell type is being stained, the investigator should be able to distinguish between them on the basis of location and/or morphology or by using double staining. Mouse macrophages can be labeled with several rat monoclonal antibodies, but they possess the ability to change immunophenotype (and function) and the antibodies may not stain the macrophage cell line exclusively. F4/80 antibody directed toward a plasma membrane glycoprotein is the most commonly used macrophage marker but may to some extent be less speciﬁc because it has also been identiﬁed on cells of the following cell lines: monocytes, eosinophils, and subgroups of dendritic cells (McGarry and Stewart, 1991; Takahashi et al., 1992; Geissmann et al., 2010). Additional drawbacks are that both special ﬁxation and enhancement techniques are recommended to obtain an acceptable staining quality. Antibodies toward scavenger receptor class A (CD204) stain muscularis macrophages in outbred NMRI mice (Mikkelsen et al., 2004), but because of a polymorphism of scavenger receptor class A they are not usable in C57Bl/6 mice (Daugherty et al., 2000). CD169 antibody, however, has been recognized as a marker for metallophilic macrophages in the spleen, and has also been demonstrated to be present on macrophages in the muscularis externa (De Winter et al., 2005; Mikkelsen et al., 2008). The macrophages in the small intestine express the major histocompatibility class II antigens (MHCII) in conventionally housed adult mice, but not in newborn or germfree mice (Mikkelsen et al., 2004). This study evaluates regional differences and age related differences in the densities of MHCIIþ cells and CD169þ cells with modern stereologic sampling. A detailed description of the applied staining and quantiﬁcation protocols are provided to acquaint potential users in the ﬁeld of intestinal motility to these tools for tissue quantiﬁcation. Double staining with MHCII and CD169 antibodies is not possible, partly because they require different ﬁxation protocols and partly because they both are rat monoclonal antibodies. In the muscle layers, the macrophages, that is, the F4/80þ, CD11bþ, and class A scavenger receptorþ cells endocytose FITC-dextran (Mikkelsen et al., 1988, 2004; Mikkelsen, 2010). As most MHCIIþ cells also endocytose Fluoresceinisothiocyanatedextran (FITC-dextran) we presume them to represent the same cell type. In this study we used FITC-dextran to evaluate if the CD169þ cells (which possess an identical morphology and distribution) have similar endocytic abilities and in this way represent the same cell type. MATERIALS AND METHODS Animals Female speciﬁc pathogen-free C57Bl/6 (B6) mice (21) were used (Taconic). We used six 5-weeks-old (young) and six 10-weeks-old (adult) mice for the quantitative main study, three animals for the FITC-dextran uptake study, and additionally eight mice in a pilot study performed to examine for a potential differential shrinkage following two different ﬁxation procedures. The mice were killed by cervical dislocation. All animals were kept in a 12:12 light-dark cycle with free access to food and water. All experimental procedures were in accordance with current national regulations issued by The Danish Council on Animal Care. Antibodies The primary antibodies were rat anti-mouse MHC class II antigen (Neomarkers A3-5, RT 946-P) (1:100) and rat anti-CD169 (Serotec, MCA-884) (1:250). The secondary antibodies were biotin conj. goat anti-rat (Amersham, RPN 1005) (1:500), followed by StreptAB-complex/HRP and DAB (DakoCytomation, using the recommendations of the company or rhodamine-conjugated rabbit anti-rat antibodies (Jackson) (1:100). Controls were incubated with rat IgG2a (Serotec) and irrelevant rat antibodies. Tissue Preparation and Immunohistochemistry FITC-dextran uptake was examined in three animals by injecting FITC dextran (MW 70.000, Sigma) 0.2 mL 0.71 mM in 154 mM NaCl intraperitoneally 24 hr before sacriﬁce. In all animals, the intestines were removed to prepare whole mount preparations, that is, stretched preparations of muscularis externa, from jejunum-ileum and colon. The removed part of jejunum-ileum started at the ﬁrst Peyer’s Patch and ended 1 cm before the ileocecal junction ( 26 cm long). The colon was taken from the junction between cecum and proximal colon and as distal as it was possibly to cut it (5–6 cm long). The intestines were kept in Tris Buffered Saline on ice during the procedure, and the mucosa and submucosa were removed with ﬁne forceps and scissors under a stereomicroscope. The isolated muscle coats were placed in TBS with nifedipine 1 lmol L1 to ensure relaxation, and the muscle coats were pinned and stretched onto a Sylgard plate. The same person performed the stretching and pinning to avoid a potential bias arising from a differential stretching of the muscle coat. The muscle coat from the colon was divided into four whole mounts. The muscle coat from jejunum-ileum was divided into approximately 14–16 whole mounts that were assigned alternately (with a random start) to MHCII immunolabeling and CD169 immunolabeling, respectively. The length of the whole mounts varied from 1.5 cm to 2 cm. Whole mounts designated for MHCII immunolabeling were ﬁxed with 96% alcohol for 10 min and whole mounts designated for CD169 immunolabeling were ﬁxed with 4% paraformaldehyde, pH: 7.4 for 3 hr. After ﬁxation, the pinned whole mounts were kept in TBS at 1559 CELL DENSITIES IN MOUSE INTESTINES Fig. 1. Left: The counting rule for an isolated frame: Proﬁles completely or partly within the frame are counted provided that they do not in any way touch any neighboring frames below or to the left of the current frame, that is, proﬁles in the frame or at the (dashed) inclusion lines are counted, provided that they do not in any way touch the (solid) exclusion lines or their inﬁnite extensions. To apply this counting rule, it is necessary to inspect an area around each counting frame, that is, a guard area, to know the full extension of the proﬁle. Three proﬁles (drawn solid black) were counted in the counting frame. Right: When a 2D-region is divided into rectangular frames (numbered consecutively from the lower left corner), a proﬁle is counted the ﬁrst time it appears within a frame as one proceeds systematically through the frames (drawn solid black in the frame where it is counted). The ability to ensure correct deﬁnition of the proﬁles that belong to the area of an isolated frame permits us to draw a sample of rectangles. If one systematically inspects every second frame, that is, half the population, after randomly selecting the start within the ﬁrst two frames, 50% of the time one will count in frames: 1, 3, 5, 7, and 9 (count 10 proﬁles), and 50% one will count in frames: 2, 4, 6, and 8 (count four proﬁles). The mean of these two counts is 7, that is, the correct number of half the population. Inspired by Figure 1 in Larsen (1998). 4 C until immunostaining. All washing and incubation solutions contained 0.5% triton-X 100. The tissue was quenched in 1% H2O2 for 30 min and preincubated with 10% goat serum containing 0.5% Triton-X 100 to reduce nonspeciﬁc staining. Primary and secondary antibodies were diluted in TBS containing 0.5% Triton-X 100 and 10% goat serum. Incubations were done at 4 C; overnight for primary antibodies, 4 hr for biotin-conjugated antibodies, and 2 hr for ABC-complex. The chromogen was 0.5% diaminobenzidine in 0.035% H2O2. The whole mounts were mounted with Aquatex (Merck). Rhodamine-conjugated antibodies were applied on whole mounts from mice that had received FITC-dextran to do double labeling. A pilot study was conducted to test for a possible differential shrinkage of the whole mounts following the two ﬁxation procedures. The muscle coats through the entire intestine were divided into whole mounts about 1.5 cm long. With a random start every second, whole mount was ﬁxed with 96% alcohol for 10 min and every second with 4% paraformaldehyde, pH: 7.4 for 3 hr. The whole mounts were measured after ﬁxation and mounting. counted cells divided by the sum of counting frame areas. Stereological Analysis The areal densities (i.e., the number of cells per surface area of the muscle coat), NA, of CD169þ and MHCIIþ cells were estimated in the whole mounts from the jejunum-ileum and colon. An unbiased counting frame of area, a(frame), was positioned at coordinates of a lattice of systematic, uniformly random points. The number of cells, Q, within the counting frame was counted through the full-thickness of the whole mount, and the areal density was estimated as the sum of P NA :¼ P Q aðframeÞ Densities were calculated both locally within the individual whole mount and globally for the entire jejunumileum and for the entire colon, respectively. The counting rule and the sampling principle are shown in Fig. 1. The stereological analysis was performed on a computer monitor using computer-assisted interactive stereological test systems (The CAST-grid software, Olympus, Denmark). Live video images of the ﬁelds of vision in the microscope were transmitted by a video camera to the computer screen. The microscope was equipped with stepping motors that controlled stage movements via the software. The entire region was delineated at a low magniﬁcation (102 using an 2 PlanApo objective). Cells were counted at a ﬁnal magniﬁcation of 1,024 using a 20 UPlanApo oil immersion objective (NA ¼ 0.8) to which the 2 objective used for delineation was paracentered. At high magniﬁcation, the computer-controlled stage of the Olympus BX51 microscope was programmed to move the section systematically, random in a raster pattern within the delineated region with interactively deﬁned steps separated by a distance of 600 lm in the xand y-axes, respectively. At each point in the raster pattern, the image of an unbiased counting frame (of area 5,232 lm2) was superimposed on to the microscope image via the video-computer interface and was ‘‘moved’’ by moving the plane of focus through the entire thickness of the whole mount specimen, and all cells within the unbiased counting frame were counted. The section 1560 MIKKELSEN ET AL. Fig. 2. Immunostaining with CD169 antibody and MHCII antibody in the muscularis of jejunum-ileum in young and adult animals. A, C, E, and G express CD169. B, D, F, and H express MHCII. A, B, E, and F are from young animals and C, D, G, and H are from adult animals. A, B, C, and D are from proximal jejunum and E, F, G, and H are from the distal ileum. Bar: 50 lm. sampling fraction was 0.5, and the area sampling fraction was approximately 0.015. In the pilot study, the length of each whole mount preparation was measured using the CAST-grid ‘‘measure length’’ feature. hyde, respectively, were calculated for each animal and were compared using a two-tailed paired t-test. We found no difference in the mean lengths (P ¼ 0.91) indicating that there was no differential shrinkage in alcohol ﬁxated gut and paraformaldehyde ﬁxated gut in our protocols. Age related differences and the differences between jejunum-ileum and colon were tested using a two-way ANOVA, and the global densities of CD169þcells and MHCIIþ-cells were compared using a twotailed paired Student’s t-test. STATISTICS In the pilot study, the mean lengths of the whole mounts ﬁxated with 96% alcohol and 4% paraformalde- CELL DENSITIES IN MOUSE INTESTINES 1561 Fig. 3. A gallery of confocal micrographs taken through the thickness of the jejunal muscularis in an adult mouse. A, B, and C are from the serosa and D, E, and F are from the level of AP. G, H, and I are from the level of the DMP. A and D show cells which have taken up FITC-dextran and B and E are CD169þ cells. C and F show that the cells which are CD169þ also have taken up FITC-dextran. At the level of DMP, H and I show CD169þ oblong cells, but FITC-dextran uptake is lacking in G and I. Bar: 30 lm. RESULTS Immunohistochemistry Stereology CD169þ cells had a morphology and distribution comparably with that of MHCIIþ cells along the intestine both in young and adult mice (Fig. 2). In the small intestine, both the serosal cells and the cells at the level of Auerbach’s plexus (AP) were ramiﬁed, and both cell types contained small FITC-dextran vesicles. At the level of the deep muscular plexus (DMP), a few oblong CD169þ cells were observed. They did not contain FITCdextran (Fig. 3). In the colon, most serosal CD169þ and MHCIIþ cells were oblong cells, but small cells without ramiﬁcations were occasionally observed in scattered groups. At AP, the cells were ramiﬁed, and in the circular muscular layer, an occasional bipolar cell was observed (Fig. 4). In addition, we observed that young mice had many round to oval cells in the proximal part of colon (both CD169þ and MHCIIþ cells), whereas in the distal part only few cells were scattered and the densities were low. Global cell densities are shown in Fig. 5, and the relevant statistical data when comparing young versus adults and jejunum-ileum versus colon are given in Table 1. CD169þ cells versus MHCIIþ cells are compared in Table 2. Young animals showed signiﬁcantly higher global densities of CD169þ cells in both jejunum-ileum (35%) and in colon (31%) compared with adult mice (see Table 1). As for MHCIIþ cells, the difference between young and adult animals did not reach signiﬁcance. The group variances were nonsigniﬁcantly higher in adult animals, when compared with young animals, except for the group variances in the density of MHCIIþ cells in colon that was signiﬁcantly higher in young animals (P ¼ 0.03). For both cell types in both age groups, the cell densities were signiﬁcantly higher in jejunum-ileum compared with colon. In young animals, the differences were 16% and 57% for CD169þ cells and MHCIIþ cells, respectively, and in adult animals, the corresponding differences were 13% and 28%. The global densities of 1562 MIKKELSEN ET AL. Fig. 4. Immunostaining with CD169 antibody and MHCII antibody in the muscularis of colon in adult animals. A and C express CD169 and B and D express MHCII. A and B are from the proximal part of colon, C and D are from the distal part. Bar: 50 lm. in Fig. 5. In adult animals, both cell-types increased in density in oral-anal direction in jejunum-ileum and decreased in oral-anal direction in colon. In young animals, the decrease in oral-anal direction in colon is equally clear, whereas it appears as if the densities in jejunum-ileum are higher in the midpart of jejunum-ileum. The ﬁrst part of jejunum-ileum in adult mice seems to have the smallest inter-animal variance in cell densities (Fig. 6). DISCUSSION Fig. 5. The global cell densities of immunolabeled cells in each mouse. Circles show young mice and squares show adult mice. Solid symbols show the density of CD169þ-cells and open symbols the density of MHCIIþ-cells. The horizontal lines give the group mean. CD169þ-cells and MHCIIþ cells were apparently the same in jejunum-ileum in both age groups and in colon in the adult animals. In colon in young animals, there was a signiﬁcant higher density of CD169þ cells, when compared with MHCIIþ cells (48% higher). The local cell densities from each whole mount preparation are shown Our study shows that the numbers of both MHCIIþ and CD169þ cells vary along the intestinal tract. In studies comparing cell densities, it is therefore optimal to take several samples along the entire course of the intestine or (if that is not possible as, e.g., in biopsy studies) to at least select from the exact same region of the intestine. Furthermore, unbiased counting and sampling principles should be applied to the sampled tissue. ‘‘Cell counts’’ based on counting cell proﬁles in few arbitrarily chosen ﬁelds of visions or based on measuring the amount of ﬂuorescence in sections are biased to variable degrees. The number of cell proﬁles per section area depends on the size and shape of the cells as well as on the amount of intercellular tissue as do the amount of ﬂuorescence emitted from a section. Recent advances in automatic image analysis may when some speciﬁc requirements are fulﬁlled provide reliable data. Disadvantages are, that it cannot distinguish cells that are part of a network and that one can only count in one focal plane (and not through the entire section as we do) 1563 CELL DENSITIES IN MOUSE INTESTINES TABLE 1. Comparing the cell densities in young animals versus adult animals (horizontal) and in jejunum-ileum versus colon (vertical) CD169þ-cells in jejunum-ileum CD169þ-cells in colon P-value Difference MHCIIþ-cells in jejunum-ileum MHCIIþ-cells in colon P-value Difference Young (n ¼ 6) Adults (n ¼ 6) P-value Difference 271 (0.16) 228 (0.18) 0.0028 16% 279 (0.11) 119 (0.39) 0.00042 57% 201 (0.21) 174 (0.24) 0.00041 34.5% 30.6% Ns 21.7% 28.2% 13% 229 (0.21) 166 (0.16) 28% 2 The table gives the group means (in number/mm ) for the global cell densities in jejunum-ileum and colon. The coefﬁcient of variation (SD/mean) is given in brackets. Horizontally, the cell densities of young animals (aged 5 weeks) are compared with those of the adult animals (aged 10 weeks) and vertically the cell densities in jejunum-ileum are compared with those in colon. The difference in group mean between young and adult animals is calculated as the difference between group means divided by the group mean of the adults. The difference in group means of densities between jejunum-ileum and colon is calculated as the difference between the group means divided by the group mean of the density in jejunum-ileum. TABLE 2. Comparing CD1691-cells versus MHCII1-cells (paired t-test) CD169þ-cells Jejunum-ileum in young animals (n ¼ 6) Jejunum-ileum in adult animals (n ¼ 6) Colon in young animals (n ¼ 6) Colon in adult animals (n ¼ 6) 271 201 228 174 (0.16) (0.21) (0.18) (0.24) MHCIIþ-cells 279 229 119 166 (0.11) (0.21) (0.39) (0.16) P-value Ns (0.710) Ns (0.078) 0.0188 Ns (0.490) Difference 48% The table gives the group means of the cell densities of CD169þ-cells and MHCIIþ-cells (in number/mm2) in jejunum-ileum and colon. The coefﬁcient of variation (SD/mean) is given in brackets. The difference in group means is calculated as the difference between the group means divided by the group mean of the density of CD169þ-cells. so it is necessary that all cells are in focus in one plane. Another drawback is that the ﬂuorescence will fade in time so that the sections cannot be stored too long before the actual analysis take place. We have previously shown that macrophages in mouse small intestine are ramiﬁed in the serosa and at the level of AP, whereas a more bipolar macrophage type resides in the circular muscle layer (Mikkelsen et al., 1988; Mikkelsen, 1995, 2010). As the cells are ramiﬁed and can change their shape and/or size, it is especially important to apply a counting rule that is number-weighted and not sizeand/or shape-weighted. The counting can be performed without the special equipment used in this study. It sufﬁce to use a microscope with an unbiased counting frame put into the eyepiece or alternatively a microscope, where the ﬁeld of vision is video-transmitted to a computer screen and superimposed with an unbiased counting frame. The stage can be manually moved on the stage-knob as the sampling only needs to be simple random to be unbiased. The reason that it is an advantage to use systematic random sampling is that the precision of the estimate thus increases. One might approach the systematic randomness by painting a mark on the knob to approach uniform movements. We found that the densities of CD169þ cells and MHCIIþ cells were comparable in all regions of the small intestine and that in adult mice cell densities increased in oral-anal direction. In young animals, the densities also differed in oral-anal direction but were in that group highest in the mid-part of jejunum/ileum. These ﬁndings suggest that CD169þ cells and MHCIIþ cells represent the same macrophage subtype. However, in a previous study on small intestinal macrophages, we found that MHCIIþ cells outnumber F4/80þ cells (Mikkelsen et al., 2008) and therefore suggested the existence of (at least) two macrophage subtypes with similar morphologies. An alternative explanation could be that the macrophages express different activation state as the F4/80 receptors seem to be down-regulated on macrophages in smooth muscle tissue and dense connective tissue and up-regulated during alternative activation (Mikkelsen, 2010). MHCII seem to be expressed by most macrophages in conventionally housed mice (Mikkelsen et al., 1988, 2004, 2008; Ozaki et al., 2004; Flores-Langarica et al., 2005; Bogunovic et al., 2009) but not by macrophages in germ-free and newborn mice (Mikkelsen et al., 2004). In recent years, nonlymphoid tissue dendritic cells have been described to be present in mice (Helft et al., 2010). In the MHCII positive cell population of mouse muscularis, presence of dendritic cells has been described (Flores-Langarica et al., 2005). However, in a recent study, only one cell population was found expressing a MHCIIhigh, CD11clow CD103, CD11bþ, F4/80þ phenotype (Bogunovic et al., 2009). The cell population was described to be derived from monocytes, had a monocyte/macrophage morphology, and was responsive to M-CSF. This is in accordance with our ﬁndings, where we, in this study, have found that the CD169þ cells located in serosa and at AP show FITC-dextran endocytosis and in previous studies have demonstrated that FITC-dextran was endocytosed by all cells labeled with antibodies toward F4/80, CD11b, and class A scavenger receptor, and by most cells labeled with antibodies toward MHCII (Mikkelsen et al., 1988, 2004). We used the CD169 antibody as it appears to demonstrate most of the macrophages in muscularis externa of 1564 MIKKELSEN ET AL. mouse intestine (Mikkelsen et al., 2008) and staining with F4/80 and class A scavenger receptor antibodies differ in different mouse strains. In colon of both adult and young animals, we found lower densities of both CD169þ and MHCIIþ cells, when compared with jejunum-ileum. Furthermore, in the colon of the young animals, the density of CD169þ cells was almost twice as high as that of MHCIIþ cells. Previously studies on cell densities in the intestines are conﬂicting. In a study on MHCIIþ cells in mouse small intestine and colon, an increasing number of cells through small intestine and colon was reported (Flores-Langarica et al., 2005), whereas a study on ED-2þ macrophages in rat intestines shows a signiﬁcant higher density of macrophages in the small intestine, when compared with the colon (Kalff et al., 1998), which is in accordance with our ﬁndings in mouse colon. As MHCII is considered to be up-regulated during classical activation and germfree and newborn mice are MHCII negative, this may suggest less activation in colonic macrophages. It is surprising as the luminal content of the bacterial ﬂora of various regions (duodenum, jejunum, ileum, and large intestine) of the gastrointestinal tracts differ (Mitsuoka, 2000) and bacteria in the small intestine is considered to be less pathogenic than those of the colon (Marteau et al., 2001). In jejunum-ileum, however, there is a higher amount of antigens from ingested food, whereas there are considerably less food antigens in colon. It has also been shown that in the noninﬂamed intestinal mucosa, macrophages are noninﬂammatory but retain avid scavenger and host defense functions (Smith et al., 2005). The diversity in the number of the macrophages in small intestine/colon may also reﬂect that small intestine is more active in motor function, which may cause more damage and thereby repair processes. Altogether this indicates that the variable MHCII expression may reﬂect different activation states of the macrophages or different subtypes of cells. This study was done on a C57Bl/6 mouse strain; it is probably the most widely used laboratory mouse strain, due to the availability of congenic strains and easy breeding. It is also the most widely used ‘‘genetic background’’ for genetically modiﬁed mice. However, the results may differ in other mouse strains. We can conclude that cell densities depend on both the age of the mouse and on the location in the intestines. The higher densities in young mice may be due to agerelated changes in the intestinal microﬂora pattern (Mitsuoka, 2000) or to a higher degree of tissue remodeling as described to take place in embryonic and foetal mice (Morris et al., 1991; Hopkinson-Woolley et al., 1994) and the different densities in jejunum and colon may reﬂect different microenvironments. ACKNOWLEDGEMENTS Fig. 6. The local cell densities in each mouse. Each graph is for an individual animal where the x-axis is the section number (0 through 16 for jejunum-ileum and 0 through 4 for colon) and the y-axis the cell density in that individual section. Circles show young mice, and squares show adult mice. Solid symbols show the density of CD169þcells and open symbols the density of MHCIIþ-cells. The horizontal lines give the group mean. The authors thank Hanne Hadberg and Keld Ottesen for skilled technical assistance. LITERATURE CITED Bogunovic M, Ginhoux F, Helft J, Shang L, Hashimoto D, Greter M, Liu K, Jakubzick C, Ingersoll MA, Leboeuf M, Stanley ER, Nussenzweig M, Lira SA, Randolph GJ, Merad M. 2009. Origin of the lamina propria dendritic cell network. Immunity 31:513–525. CELL DENSITIES IN MOUSE INTESTINES Daugherty A, Whitman SC, Block AE, Rateri DL. 2000. Polymorphism of class A scavenger receptors in C57BL/6 mice. J Lipid Res 41:1568–1577. De Winter BY, Van Nassauw L, De Man JG, de Jonge F, Bredenoord AJ, Seerden TC, Herman AG, Timmermans JP, Pelckmans PA. 2005. Role of oxidative stress in the pathogenesis of septic ileus in mice. Neurogastroenterol Motil 17:251–261. Flores-Langarica A, Meza-Perez S, Calderon-Amador J, EstradaGarcia T, Macpherson G, Lebecque S, Saeland S, Steinman RM, Flores-Romo L. 2005. Network of dendritic cells within the muscular layer of the mouse intestine. Proc Natl Acad Sci USA 102: 19039–19044. Geissmann F, Gordon S, Hume DA, Mowat AM, Randolph GJ. 2010. Unravelling mononuclear phagocyte heterogeneity. Nat Rev Immunol 10:453–460. Helft J, Ginhoux F, Bogunovic M, Merad M. 2010. Origin and functional heterogeneity of non-lymphoid tissue dendritic cells in mice. Immunol Rev 234:55–75. Hopkinson-Woolley J, Hughes D, Gordon S, Martin P. 1994. Macrophage recruitment during limb development and wound healing in the embryonic and foetal mouse. J Cell Sci 107(Pt 5):1159– 1167. Kalff JC, Schwarz NT, Walgenbach KJ, Schraut WH, Bauer AJ. 1998. Leukocytes of the intestinal muscularis: their phenotype and isolation. J Leukoc Biol 63:683–691. Knowles CH, De Giorgio R, Kapur RP, Bruder E, Farrugia G, Geboes K, Gershon MD, Hutson J, Lindberg G, Martin JE, MeierRuge WA, Milla PJ, Smith VV, Vandervinden JM, Veress B, Wedel T. 2009. Gastrointestinal neuromuscular pathology: guidelines for histological techniques and reporting on behalf of the Gastro 2009 International Working Group. Acta Neuropathol 118:271–301. Larsen JO. 1998. Stereology of nerve cross sections. J Neurosci Methods 85:107–118. Marteau P, Pochart P, Dore J, Bera-Maillet C, Bernalier A, Corthier G. 2001. Comparative study of bacterial groups within the human cecal and fecal microbiota. Appl Environ Microbiol 67: 4939– 4942. 1565 McGarry MP, Stewart CC. 1991. Murine eosinophil granulocytes bind the murine macrophage-monocyte speciﬁc monoclonal antibody F4/80. J Leukoc Biol 50:471–478. Mikkelsen HB. 1995. Macrophages in the external muscle layers of mammalian intestines. Histol Histopathol 10:719–736. Mikkelsen HB. 2010. Interstitial cells of Cajal, macrophages and mast cells in the gut musculature: morphology, distribution, spatial and possible functional interactions. J Cell Mol Med 14:818– 832. Mikkelsen HB, Garbarsch C, Tranum-Jensen J, Thuneberg L. 2004. Macrophages in the small intestinal muscularis externa of embryos, newborn and adult germ-free mice. J Mol Histol 35:377–387. Mikkelsen HB, Larsen JO, Hadberg H. 2008. The macrophage system in the intestinal muscularis externa during inﬂammation: an immunohistochemical and quantitative study of osteopetrotic mice. Histochem Cell Biol 130:363–373. Mikkelsen HB, Mirsky R, Jessen KR, Thuneberg L. 1988. 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. Mitsuoka T. 2000. Signiﬁcance of dietary modulation of intestinal ﬂora and intestinal environment. Biosci Microﬂora 19:15–25. Morris L, Graham CF, Gordon S. 1991. Macrophages in haemopoietic and other tissues of the developing mouse detected by the monoclonal antibody F4/80. Development 112:517–526. Ozaki H, Kawai T, Shuttleworth CW, Won KJ, Suzuki T, Sato K, Horiguchi H, Hori M, Karaki H, Torihashi S, Ward SM, Sanders KM. 2004. Isolation and characterization of resident macrophages from the smooth muscle layers of murine small intestine. Neurogastroenterol Motil 16:39–51. Smith PD, Ochsenbauer-Jambor C, Smythies LE. 2005. Intestinal macrophages: unique effector cells of the innate immune system. Immunol Rev 206:149–159. Takahashi K, Naito M, Schultz LD. 1992. Differentiation of epidermal Langerhans cells in macrophage colony-stimulating-factor-deﬁcient mice homozygous for the osteopetrosis (op) mutation. J Invest Dermatol 99:46S–47S.