The heteromeric lymphotoxin αβ ligand (LT) binds to the LTβ receptor (LTβR) and provides an essential trigger for lymph node (LN) development. LTβR signaling is also critical for the emergence of pathological ectopic lymph node-like structures and the maintenance of an organized splenic white pulp. To better understand the role of LT in development, the expression patterns of LTβ and LTβR mRNA were examined by in situ hybridization in the developing mouse embryo. Images of LTβ ligand expression in developing peripheral LN in the E18.5 embryo revealed a relatively early phase structure and allowed for comparative staging with LN development in rat and humans. The LTβR is expressed from E16.5 onward in respiratory, salivary, bronchial, and gastric epithelium, which may be consistent with early communication events between lymphoid elements and epithelial specialization over emerging mucosal LN. Direct comparison of mouse fetal and adult tissues by FACS analysis confirmed the elevated expression of LTBR in some embryonic epithelial layers. Therefore, surface LTBR expression may be elevated during fetal development in some epithelial layers.

Formation of the immune organs is of considerable interest not only as a classical developmental problem, but because organized pathological ectopic lymph node-like structures develop in chronically inflamed sites (1, 2). The discovery that the lymphotoxin (LT)3 system plays key roles in both embryonic and pathological lymphoid organ formation as well as in the maintenance of lymphoid architecture suggests that common mechanistic elements are being utilized in all three settings (3). There are also parallels between the signals employed in inflammatory events and Peyer’s patch (PP) development (4). Thus, an analysis of ontological motifs is likely to provide novel insight into the control of architecture and cellular positioning in both the normal and inflamed adult organs.

LTα and LTβ form a heteromeric complex that binds uniquely to the LTβR and deletion of any of these genes leads to loss of lymph node (LN) development (3, 5, 6). LTβR activation is coupled to NFκB-inducing kinase and IκB kinase (IKK) α and hence to NFκB activation and indeed NFκB-inducing kinase gene inactivation resembles LTβR disruption in terms of LN status (7, 8). A number of other gene deletions have led to loss of LN or PP development including RANK and its ligand TNF-related activation-induced cytokine, CXCR5, CCL13, VLA4, IL-7 and its receptor, relB, Ikaros, Id2, IKKα, and RORγ (9, 10).

The stages of murine LN development have been poorly analyzed due to the difficulty of identifying LN in the mouse embryo. Although there are photographs of developing rat and human LN, an actual microscopic image of a fetal mouse peripheral or mesenteric LN has not been reported (11, 12). In contrast, considerable progress has been made in visualizing PP development by use of in situ whole-mount histological methods (4, 13). Several stages in PP development can be recognized in the mouse from the initial anlage formation and expression of VCAM, to the seeding with the IL-7R+, CD4+, CD3 organizer cell and, finally, filling with mature lymphocytes. In LT-deficient mice, the earliest stages of PP formation cannot be detected (14). Similarly, Kim et al. (9) have postulated five stages of lymphatic and LN development and, in this scheme, LN development commences at stage III shortly after formation of the lymphatics at E12–13. In stage III, mesenchymal connective tissue invaginates into the lumen of the lymphatics, giving rise to a bulbous structure that will form the cortical/paracortical regions of the LN. The invagination is joined to the surrounding tissue by a hilus or stalk that will form the medullary end of the LN. The luminal space is bounded by endothelial cells from the lymphatics and probably capillaries and thin connective bridges link the invagination to the surrounding endothelium (12). Fetal blood flow is present at this point, providing a conduit for lymphocyte trafficking. At stage IV, cellular content increases and the lumen and bridges collapse to form the subcapsular region. In the final stage V, capsule formation occurs and mature lymphocytes begin to populate the structure.

Characterization of the expression patterns of the ligands and receptors in the LT pathway and the involved cell types is crucial for a complete understanding of this process. Even though the analysis of the RNA or protein expression patterns of a particular gene is technically straightforward, the expression pattern for a new cytokine system is often one of the last facets to be fully characterized. To study the question of when and where the LTβ and LTβR genes are expressed during lymph node organogenesis, we used in situ hybridization (ISH) methods. This analysis of LTβ expression led to visualization of developing LN and thus allowed for a comparative staging of murine LN formation with that in rat and humans. Surface LTβR expression was elevated in some embryonic mucosal epithelial layers and expression then decreased in the adult, suggesting that surface LTβR display may be a fetal event in some epithelial layers.

The murine LTβ sense and antisense probes were prepared from pGEM-3Z vector containing a 420-bp fragment of the mouse LTβ cDNA covering part of exons 3 and 4 (15). A 700-bp NotI fragment from the 5′ end of the mouse LTβR cDNA (16) was ligated into pcDNA3 (Invitrogen, San Diego, CA), yielding AN003. LTβR sense and antisense probes were prepared from this vector. Radiolabeled cRNAs were synthesized by in vitro transcription in the presence of 12.5 μM 35S-labeled UTP (400 Ci/mmol; Amersham). 35S-Labeled probes were subsequently reduced to an average size of 50–100 nt by mild alkaline hydrolysis as previously described (17). For the synthesis of 35S-labeled probes, all of the above reactions were performed in 10 mM DTT. Total RNA was extracted from 20 selected tissues of 3-mo-old NMRI mice as described below. Briefly, 0.5 g of each selected tissue was homogenized with a Polytron in 0.1 M Tris-HCl (pH 7.4), 0.1 M 2-ME, and 4 M guanidium thiocyanate. After addition of solid CsCl (0.4 g/ml), the homogenate was layered onto 1 ml of a 5.7 M CsCl/0.2 M EDTA (pH 7.0) cushion and centrifuged at 35,000 rpm for 20 h at 20°C. The pelleted RNAs were dissolved in 300 μl of 10 mM Tris-HCl (pH 8.1), 5 mM EDTA, and 0.1% SDS, extracted twice with phenol/chloroform, ethanol precipitated, and resuspended in water. Five micrograms of total RNA from each tissue was subsequently denatured with glyoxal, electrophoresed in 1.2% agarose gels and transferred overnight onto Hybond nylon membranes (Hybond-N; Amersham). Prehybridizations, hybridizations, and posthybridization washes were conducted as described previously (18).

Embryos were timed and segments of uteri or, at later stages, whole NMRI embryos were dissected, embedded in Tissue-Tek (Miles, Elkhart, IN), and frozen down in precooled methylbutane as described elsewhere (18). Cryostat tissue sections (5 μm) were mounted on poly-l-lysine (Sigma, St. Louis, MO)-coated microscope slides, fixed in 4% glutaraldehyde in PBS for 1–5 min, rinsed in PBS, and stored in 70% ethanol at 4°C until analyzed. Fixed sections were rinsed in 2× SSC (1× SSC = 150 mM NaCl and 15 mM sodium citrate, pH 7.0), acetylated with 0.25% (v/v) acetic anhydride in 0.1 M triethanolamine (pH 8.0) at room temperature for 10 min, incubated in 0.1 M Tris-HCl (pH 7.0)/0.1 M glycine at room temperature for 30 min, and prehybridized in 2× SSC/50% formamide at 50°C for 15 min. 35S-Labeled cRNAs (1–3 × 106 cpm) were applied to each section in 20–70 μl of hybridization mixture (50% formamide, 2× SSC, 10 mM DTT, 1 mg/ml BSA (RNase free), 0.15 mg/ml rRNA, and 5% dextran sulfate) for 3 h at 50°C. Slides were subsequently washed for 20 min in 2× SSC, 50% formamide at 50°C, and 10 mM DTT, and in 0.2× SSC and 50% formamide at 50°C for 30 min each. Unhybridized transcripts were digested with 10 μg/ml RNase A at 37°C for 30 min. The slides were washed again in 2× SSC/50% formamide at 50°C for 15 min, four changes of 2× SSC at room temperature, dehydrated in graded ethanol, and air dried. Sections were either directly exposed for 2–3 wk at room temperature to x-ray films (SB5; Eastman Kodak, Rochester, NY) between intensifying screens or immersed in NTB-2 emulsion (Eastman Kodak), diluted 1:1 in deionized water, exposed for 3–4 wk at 4°C, developed in Kodak D-19 developer, fixed in 30% sodium thiosulfate, and counterstained in 1% methylene blue. Controls of specificity included the systematic use of sense cRNA probes in each experiment.

The entire intestinal mass was dissected from timed C57BL/6 embryos and any connected mesenteric tissue was removed. The intestines were gently passed through an 18-gauge syringe needle to crudely disrupt the gut integrity and then washed in HBSS with 2% FBS and 10 mM HEPES buffer. Gut fragments were suspended in the above HBSS buffer with 1 mM EDTA (HBSS/EDTA) and rocked for 30 min at 37°C. The fragments were removed and resuspended in the same buffer for a second extraction. The cell suspensions were passed through a 70-μm nylon filter and the resultant mixture of epithelial cells and early lymphocytes was used for FACS analysis. Adult large or small intestines were flushed, sliced longitudinally, and the washed fragments subjected to HBSS/EDTA extraction as defined above. Adult spleen cells were obtained by subjecting glass slide-mashed spleens to collagenase digestion for 30 min at room temperature followed by RBC lysis with Gey’s solution, further EDTA release of cells from residual stroma, and filtration through a nylon filter. Fetal hepatocytes were obtained by passage of the liver through a 21-gauge syringe needle. Murine C26 colon carcinoma cells were obtained from D. LePage (Biogen, Cambridge, MA) and the Lewis lung carcinoma was purchased from the American Type Culture Collection (Manassas, VA). Embryonic fibroblasts were a gift from M. Scott (Biogen), grown in DMEM with 10% FBS, and analyzed at passage 2. Cells were removed with EDTA/PBS. Cells were suspended in PBS containing 5% FBS, 5% each of normal mouse and rat serum, and 10 μg/ml Fc Block (BD PharMingen, San Diego, CA). Cells were stained sequentially with 10 μg/ml hamster anti-murine LTBR ACH6 (19) or a control mAb Ha4/8 followed by the PE-anti-hamster mixture (BD PharMingen). Adult spleen cell preparations were stained as above for LTBR followed by FITC-anti-CD11c (HL3), CyChrome-anti-TCR-β, and allophycocyanin-anti-CD11b (Mac-1; all from BD PharMingen). All fetal or adult epithelial cells were further stained at the end for 5 min with 10 μg/ml 7-amino-actinomycin D and analyzed immediately without paraformaldehyde fixation. Analysis gates excluded any 7-amino-actinomycin D-positive dead cells.

The macroscopic distribution of LTβ and LTβR mRNA was determined by ISH of the respective 35S-labeled cRNA probes to sagittal sections of whole embryos ranging from E5.5 to E18.5 followed by direct exposure to x-ray films. Using this technique, LTβ and LTβR RNA were detected in certain macroscopically identifiable organs from E16.5 onward (Fig. 1). No signal was detected with the sense LTβ or LTβR cRNA probes (Fig. 1 and data not shown). As shown in Fig. 1, and in accordance with previous reported results, embryonic LTβ gene expression is fairly restricted, being macroscopically confined to the developing skin, thymus, gut, and brain (20). Simultaneous analysis of LTβR mRNA revealed a wider distribution in embryos of the same age (Fig. 1). Significant levels of LTβR mRNA were detected macroscopically in the developing sinus, submaxillary gland, thymus, muscle, lung, stomach, and gut. The regional distribution of LTβ and LTβR mRNA at E16.5 was very similar to that that of the E18.5 embryo (data not shown). A previous Northern blot analysis of LTβR RNA from whole embryos showed expression from as early as E7 (21). The expression patterns of LT and LTβR are summarized in Table I.

FIGURE 1.

Macroscopic localization of LTβ and LTβR mRNA by ISH of 35S-labeled cRNA probes to tissue sections of an E16.5 mouse embryo. LTβ mRNA (LTβ Antisense) is detectable in the developing thymus (Th), gut (Gu), and skin (Sk). LTβR mRNA (LTβR Antisense) is detectable in the developing sinus (Si), submaxillary gland (Gl), thymus (Th), lung (Lu), stomach (St), gut (Gu), and muscle (Mu). No signal was detected in adjacent sections hybridized to the sense LTβ (Sense) or LTβR probe (data not shown). Photographs were taken after a 15-day exposure at room temperature. original magnification, ×2.8).

FIGURE 1.

Macroscopic localization of LTβ and LTβR mRNA by ISH of 35S-labeled cRNA probes to tissue sections of an E16.5 mouse embryo. LTβ mRNA (LTβ Antisense) is detectable in the developing thymus (Th), gut (Gu), and skin (Sk). LTβR mRNA (LTβR Antisense) is detectable in the developing sinus (Si), submaxillary gland (Gl), thymus (Th), lung (Lu), stomach (St), gut (Gu), and muscle (Mu). No signal was detected in adjacent sections hybridized to the sense LTβ (Sense) or LTβR probe (data not shown). Photographs were taken after a 15-day exposure at room temperature. original magnification, ×2.8).

Close modal
Table I.

Expression of LTβ ligand and receptor RNA during developmenta

Gestational Age
12.5-Day14.5-Day16.5-Day18.5-Day
LTβLTβRLTβLTβRLTβLTβRLTβLTβR
Epithelia         
Keratinocytes − − − − ++ − +++ − 
Choroid plexus − − − − − − 
Respiratory epithelium NS NS NS NS − ++ − ++ 
Olfactory epithelium NS NS NS NS − − 
Otic epithelium NS NS − − − − − − 
Retinal pigment epithelium NS NS NS NS − − − − 
Proliferative lens epithelium − − − − − − 
Salivary glands NS NS NS NS − ++ − ++ 
Bronchial epithelium NS NS − − − 
Stomach NS NS NS NS − ++ − ++ 
Intestine − − − ++ − ++ 
Kidney tubular epithelium − − − − − − 
Testes NS NS NS NS − − − − 
Biliary epithelium NS NS NS NS − − − − 
Pancreatic acini and ductules NS NS NS NS − − 
Liver ++ − ++ − 
Ventricular neuroepithelium − − − − − − − − 
Spleen NS NS NS NS NS NS NS NS 
LN         
Peripheral NS NS NS NS NS NS ++ − 
Intestinal NS NS NS NS ++ − +++ − 
Thymus NS NS NS NS ++ +++ 
Adrenal gland NS NS NS NS − − − − 
Neuronal tissue         
Brain − − − − − − 
Spinal cord − − − − − − 
Photoreceptor layers of the eye − − − − − − 
Bone NS NS NS NS − 
Endothelia NS NS − − − − − − 
Mesenchyme         
Kidney − − − − − − − − 
Lung − − − − − − − − 
Skeletal muscle − − − − − − 
Heart muscle (ventricle) − − − − − − − − 
Gestational Age
12.5-Day14.5-Day16.5-Day18.5-Day
LTβLTβRLTβLTβRLTβLTβRLTβLTβR
Epithelia         
Keratinocytes − − − − ++ − +++ − 
Choroid plexus − − − − − − 
Respiratory epithelium NS NS NS NS − ++ − ++ 
Olfactory epithelium NS NS NS NS − − 
Otic epithelium NS NS − − − − − − 
Retinal pigment epithelium NS NS NS NS − − − − 
Proliferative lens epithelium − − − − − − 
Salivary glands NS NS NS NS − ++ − ++ 
Bronchial epithelium NS NS − − − 
Stomach NS NS NS NS − ++ − ++ 
Intestine − − − ++ − ++ 
Kidney tubular epithelium − − − − − − 
Testes NS NS NS NS − − − − 
Biliary epithelium NS NS NS NS − − − − 
Pancreatic acini and ductules NS NS NS NS − − 
Liver ++ − ++ − 
Ventricular neuroepithelium − − − − − − − − 
Spleen NS NS NS NS NS NS NS NS 
LN         
Peripheral NS NS NS NS NS NS ++ − 
Intestinal NS NS NS NS ++ − +++ − 
Thymus NS NS NS NS ++ +++ 
Adrenal gland NS NS NS NS − − − − 
Neuronal tissue         
Brain − − − − − − 
Spinal cord − − − − − − 
Photoreceptor layers of the eye − − − − − − 
Bone NS NS NS NS − 
Endothelia NS NS − − − − − − 
Mesenchyme         
Kidney − − − − − − − − 
Lung − − − − − − − − 
Skeletal muscle − − − − − − 
Heart muscle (ventricle) − − − − − − − − 
a

Levels of expression were assessed semiquantitatively by ISH of 35S-labeled murine LTβ or LTβR cRNA probes to frozen cryostat sections of embryos. −, No mRNA detectable; +, weak labeling; ++, moderate labeling; +++, strong labeling; NS, not studied.

The cellular distribution of LTβ mRNA in embryos ranging from 5.5 to 18.5 days of development was determined by ISH of 35S-labeled cRNA probes to tissue sections followed by detection with an autoradiographic emulsion. Significant levels of mRNA were detectable at E18.5 in LN-like structures that appear above the submaxillary gland, above the thymus, in the abdomen at the level of the kidneys just anterior to the developing hip joint, and in the femoral region (Fig. 2). Tentatively, this labeling is assigned to the developing cervical, para-aortic (or mediastinal), lumbar/caudal (also sometimes called para-aortic), and popliteal nodes. Structures shown in Fig. 2 closely resemble images of rat popliteal LN at E20 (12). One can see in these images of developing peripheral LN that the primary LTβ-positive structure is an invagination into a lumenal space. This structure closely approximates the stage III described by Kim et al. (9). Thin connecting bridges can be seen between the invagination and the enveloping endothelium and these resemble the bridges described in rat and human embryonic LN (11, 12). These fetal LN at E18.5 appear to be at a state before stage IV where the lumen collapses to form the subcapsular sinus. Although these images are not definitive, in Fig. 2,A, one connecting bridge which does not appear to be the connecting hilus contains LTβ+ cells and in Fig. 2,C, there is the suggestion of a LTβ-positive structure. Interestingly, there was the appearance of LTβ+ cells in the lining of either a capillary or the lymphatic lumen (Fig. 2, A and C). A whole-mount image of CD4+ cells in the neonatal mesenteric LN showed scattered colonies of CD4+ cells with little suggestion of bridges and/or luminal spaces (9), and images of neonatal peripheral LN have a fairly mature appearance (22, 23). Therefore, the E18.5 images shown here appear to be earlier in the developmental path and the final development from roughly stage III to stage V must occur relatively quickly within the last 24–36 h of fetal development.

FIGURE 2.

Microscopic localization by ISH of LTβ mRNA in developing LN-like structures in a mouse embryo. Shown are potential LN in an E18.5 embryo in the cervical (A), paraortic (B), and femoral (C) regions and an aggregate of LTβ-positive cells in a region resembling a PP (D). L, Lymphatic or vascular lumen; H, potential hilus structures; and the arrows point to potential connecting bridges between the invaginating mesenchymal tissue and the lumen endothelial lining. Dark field (right) and light field (left) photographs are contrasted. Final magnification is ×50 for the LN and ×40 for the PP.

FIGURE 2.

Microscopic localization by ISH of LTβ mRNA in developing LN-like structures in a mouse embryo. Shown are potential LN in an E18.5 embryo in the cervical (A), paraortic (B), and femoral (C) regions and an aggregate of LTβ-positive cells in a region resembling a PP (D). L, Lymphatic or vascular lumen; H, potential hilus structures; and the arrows point to potential connecting bridges between the invaginating mesenchymal tissue and the lumen endothelial lining. Dark field (right) and light field (left) photographs are contrasted. Final magnification is ×50 for the LN and ×40 for the PP.

Close modal

LTβ mRNA was also observed at the same stage of development in lymphoid-like aggregates in the intestinal submucosa (Fig. 2 D). This structure strikingly resembles a typical PP surrounded by well-separated villi. Although we were not able to precisely differentiate between PP in the small intestine and potentially submucosal B cell-rich follicles in the large bowel or perhaps even emerging cryptopatches, the shape of the villi is consistent with the small bowel. Labeled LN were not observed in the E16.5 embryo, which may simply reflect a low cellularity and hence an inability to find them. In general, LTβ gene expression during murine embryogenesis appears to be associated with the development of certain lymphoid organs, as shown here for the fetal thymus, intestinal lymphoid aggregates, and abdominal LN. Within the 12 sagittal sections analyzed in this study, there were none sufficiently lateral to include the fetal spleen.

Foci of cells expressing LTβ were detected in the fetal liver from E12.5 (Fig. 3) and similar numbers of foci were found at E14.5 and E16.5, although the colonies were more diffuse at the later stages. Curiously, there was not a large number of foci, suggesting that these foci may represent emerging colonies of a particular cell lineage rather than the massive hemopoiesis that occurs at this stage. These cells are likely to be hemopoietic since lower levels of transcript were detectable in the liver at later stages of development, which would be consistent with an early wave of hemopoietic cells (Table I). Hybridization of adjacent sections with the sense cRNA probe only revealed low background levels of signal, indicating that the observed signal was specific (Fig. 3).

FIGURE 3.

Microscopic localization by ISH of LTβ mRNA in developing mouse tissues. Focal expression of LTβ mRNA in the liver of an E12.5 embryo using an antisense probe (Liver AS) and roughly adjacent sections from the same liver hybridized to the sense LTβ cRNA probe (Liver S). LTβ mRNA is detected within thymocytes of the thymic cortex and medulla and in keratinocytes of the epidermis (skin) at E18.5. Dark field (right) and light field (left) photographs are contrasted. Final magnification is ×100 for the liver and ×50 for thymus and skin.

FIGURE 3.

Microscopic localization by ISH of LTβ mRNA in developing mouse tissues. Focal expression of LTβ mRNA in the liver of an E12.5 embryo using an antisense probe (Liver AS) and roughly adjacent sections from the same liver hybridized to the sense LTβ cRNA probe (Liver S). LTβ mRNA is detected within thymocytes of the thymic cortex and medulla and in keratinocytes of the epidermis (skin) at E18.5. Dark field (right) and light field (left) photographs are contrasted. Final magnification is ×100 for the liver and ×50 for thymus and skin.

Close modal

The strongest levels of embryonic LTβ gene expression were observed in the thymus, skin, brain, and LN-like structures described above. In the thymus, thymocytes within both the cortical and medullary regions were strongly labeled from E16.5 onward (Fig. 3). In contrast, in the adult thymus, LTβ expression was only observed in the medulla region (20). High levels of LTβ mRNA were also detectable at E16.5 and E18.5 in the epidermis (Fig. 3). Background levels of staining are always somewhat higher along the outer edge of the embryo yet this signal was specific, as no signal was observed in the epidermis with the sense probe (data not shown). As most cells in the epithelial layer appear to be LTβ positive, it is likely that these cells are developing keratinocytes. Similar skin labeling was also reported previously (20). The expression of LTβ mRNA in the developing brain was noted earlier and its significance remains unclear (20).

Microscopic analysis of sections hybridized to the 35S-labeled LTβR cRNA probe showed detectable levels of LTβR gene expression from E12.5–14.5 and onward with detectable albeit weak levels of message being seen at the cellular level in the epithelium of the gut and lung in E14.5 embryos (Table I). LN-like structures could not be observed using this probe. In E16.5 and E18.5 embryos, the highest abundance of LTβR mRNA was detected in epithelial cells of the developing gut, and transcripts were present in epithelial cells of the villi (Fig. 4). No significant background expression was observed in adjacent sections hybridized to the sense LTβR cRNA probe (Fig. 4). Likewise, significant levels of LTβR mRNA were detected from E16.5 onward in the epithelium lining of the stomach (Fig. 4), in respiratory epithelial cells of the lung and sinuses (Fig. 4), and in acinar glandular cells of the developing submaxillary gland (data not shown). In the developing thymus, LTβR transcripts were detected in both cortical and medullary thymic regions, but at comparatively lower levels than observed for LTβ (Table I). Taken together, during late embryogenesis, LTβR gene expression appears to predominate in several epithelial layers (Table I).

FIGURE 4.

Microscopic localization of LTβR mRNA in the E18.5 mouse embryo. Expression is observed in the epithelium layer of the intestine (Gut AS) which was not observed when a sense RNA probe was used (Gut S), in the lining of the stomach and the respiratory sinus (Res. Sinus). Dark field and light field photographs are contrasted. Gut AS, Gut antisense; Gut S, gut sense. Magnification is ×100 in each case.

FIGURE 4.

Microscopic localization of LTβR mRNA in the E18.5 mouse embryo. Expression is observed in the epithelium layer of the intestine (Gut AS) which was not observed when a sense RNA probe was used (Gut S), in the lining of the stomach and the respiratory sinus (Res. Sinus). Dark field and light field photographs are contrasted. Gut AS, Gut antisense; Gut S, gut sense. Magnification is ×100 in each case.

Close modal

We wished to ascertain whether LTBR expression in the mucosal linings of the gut and airway was elevated during development. The ISH studies presented here did not compare adult and fetal expression levels directly and, moreover, we do not appear to have an anti-murine LTBR mAb that works well in immunohistochemistry. Therefore, FACS analysis was performed on EDTA-dissociated epithelial cells derived from adult and fetal intestines. At E16–17, the fetal intestine expressed LTBR while fetal hepatocytes at the same developmental stage lacked appreciable expression (Fig. 5). Adult large and small intestinal epithelial cells expressed very low levels of LTBR. At birth, LTBR expression in the intestinal epithelium had already begun to drop down to the low levels seen in the adult. Several other cells were assessed for comparison. Two cultured epithelial tumor lines, the C26 colon and Lewis lung carcinomas, showed the highest level of surface expression (Fig. 5) and the B16-F10 melanoma line was similar (data not shown). As noted previously, splenic B cells (Fig. 5) were receptor negative as were T cells (data not shown). CD11b+ splenic monocytes were receptor positive (Fig. 5) as were CD11b+CD11c+CD11bCD11c+ and CD11c+CD8a+ splenic dendritic cells (data not shown). The ISH analysis did not reveal general expression in the embryo that one would perhaps expect for global fibroblast expression, yet early passage murine embryonic fibroblasts were receptor positive. Since we find that most cultured primary nonlymphoid cells from either murine or human sources are LTBR positive, receptor expression may be simply up-regulated upon transfer into tissue culture. This analysis shows that the high level of RNA expression revealed by ISH was paralleled by higher surface protein expression as assessed by FACS analysis of uncultured primary cells. Thus, surface LTBR display appears to be down-regulated upon maturation of the intestinal epithelial layer and this event may occur in several mucosal epithelial layers.

FIGURE 5.

FACS analysis of LTβR expression on the surface of intestinal epithelial cells from adult large, small, fetal (E16–17), and neonatal intestine. For comparison, receptor expression is shown on fetal (E16–17) hepatocytes, CD11b+ splenic macrophages, splenic B cells, embryonic fibroblasts, and the C26 colon and Lewis lung carcinomas.

FIGURE 5.

FACS analysis of LTβR expression on the surface of intestinal epithelial cells from adult large, small, fetal (E16–17), and neonatal intestine. For comparison, receptor expression is shown on fetal (E16–17) hepatocytes, CD11b+ splenic macrophages, splenic B cells, embryonic fibroblasts, and the C26 colon and Lewis lung carcinomas.

Close modal

LTβ and LTβR gene expression was assessed in 20 adult mouse tissues using Northern blot analysis. As shown in Fig. 6, expression of LTβ is fairly restricted in adult tissues. Strong levels of expression are observed in the thymus and spleen and lower levels are detectable in the large intestine, small intestine, and lung. Several bands are detectable in the bone, but only a fraction appears to migrate at the expected size. This may be due to some RNA degradation in this sample. This pattern of LTβ expression in adult tissues is in accordance with that observed by others who have detected high levels of LTβ mRNA by RT-PCR and Northern blot analyses in the thymus and spleen (15, 20). Therefore, it appears that both during development and in the adult mouse, LTβ gene expression occurs predominately in the lymphoid tissues.

FIGURE 6.

Northern blot analysis of LTβ and LTβR gene expression in adult mouse tissues. Five micrograms of total RNA prepared from the indicated tissues was hybridized to the 32P-labeled murine LTβ and LTβR riboprobes. Equal loading and the integrity of RNAs were verified by methylene blue staining of 18S rRNA.

FIGURE 6.

Northern blot analysis of LTβ and LTβR gene expression in adult mouse tissues. Five micrograms of total RNA prepared from the indicated tissues was hybridized to the 32P-labeled murine LTβ and LTβR riboprobes. Equal loading and the integrity of RNAs were verified by methylene blue staining of 18S rRNA.

Close modal

Northern blot analysis of LTβR gene expression revealed a distinct pattern of expression compared with that of LTβ, as most of the analyzed tissues contain detectable amounts of LTβR transcripts (Fig. 6). Expression was observed in the lung, kidney, stomach, small intestine, large intestine, liver, and adrenal gland. Receptor expression was noted to occur in liver, lung, and kidney in previous studies (16, 21). Moderate to low levels of LTβR mRNA were detected in all other tissues analyzed except for the pancreas. Therefore, the distribution of the receptor is relatively ubiquitous, although beyond the myeloid and follicular dendritic lineages, the types of cells that express the receptor remain ill-defined.

Whole-mount immunohistochemical or ISH methods have been very useful in the identification of developing PP; however, neither peripheral nor mesenteric LN have been imaged in the mouse embryo. LTβ expression was observed using ISH at sites of both developing peripheral LN and PP. Although the appearance of LTβ-positive cells in these embryonic LN was not surprising, the ISH signal allowed one to find the early LN within the embryos and visualize their structure. These images of developing E18.5 peripheral murine LN resemble closely those from the rat E20 and human embryos (11, 12). Given that E18.5 in murine development is close to birth, these early stage peripheral murine LN appear to be emerging late relative to the rat. It is possible that LTβ expression within the LN anlagen occurs earlier than E18.5 and we were simply unable to find the LN. Alternatively, given the rather rudimentary structure observed at E18.5, rapid LN development and filling may occur only after E16.5 and for this reason the rudimentary LN are effectively invisible.

In contrast to the relatively late appearance of LTβ-positive structures, it has been shown using maternal transfer to deliver in utero an LT inhibitor that LTαβ:LTβR communication was critical between E12 and E15 for the development of the LN anlagen after stage II (24). Studies on PP development have led to a model whereby a hemopoietic “inducer” cell displays surface LT in response to IL-7 signaling. The LT-positive cell triggers LTβR-positive mesenchymal cells to become a VCAM+/ICAM+ “organizer” (4, 13, 25). The genetic lack of LT blocks PP development at the earliest detectable stage, i.e., formation of the VCAM/ICAM+ anlage at E16; however, it has not been formally proven that a hemopoietic cell delivers this LT signal at the early stages. Clearly, the Ikaros mouse reveals the need for hemopoietic cells in general and Ikaros null mice lack the critical IL-7R+CD4+/−CD3LT+ cell that populates the rudimentary gut and mesenteric LN (22, 25, 26). It is certainly reasonable that the LT signal is delivered by a hemopoietic cell since expression in the adult appears to be limited to lymphoid cells and specifically to activated T, B, and NK cells and a subset of resting B cells (10, 19, 27, 28, 29). Nonetheless, the possibility remains that stromal elements may express LT and LN development may be more comparable to other organs, e.g., the liver or the pancreas, where a dialogue occurs between endothelial and endodermal tissues early in the process (30). There may be expression of LTβ along the lining of the lumen surrounding the rudimentary LN. This expression could be simply an artifact and certainly this study is not definitive. Alternatively, the luminal lining LTβR-positive cells may be hemopoietic cells in transit into the LN via the hilus or perhaps LT expression by stromal/endothelial cells.

LT was found to be present on the surface of a unique IL-7R+CD4+CD3 oligolineage progenitor cell and this cell is a likely source of LTαβ during the population of rudimentary mesenteric LN and PP by cells of the hemopoietic lineage (22, 26). Abundant numbers of the CD4+CD3LTαβ+ cells are easily observed by FACS in the rudimentary mesenteric LN and by whole-mount histological methods in PP. Thus, LTαβ expression by this cell is likely to be a dominant, if not sole, contributor to the LTβ signal observed at E18.5 in this study. Whether this cell provides the critical ontological LT input both early and later during filling of the rudimentary LN has not been unequivocally established. Only low numbers of these cells are detectable at E14.5 in the blood and spleen, yet none are found in the fetal liver (26). Attempts to identify surface LTαβ-positive cells in the E11.5–15.5 fetal liver by FACS were unsuccessful and therefore the nature of the LTβ-positive cell foci reported here is not clear (J. L. Browning, unpublished observations). It is possible that LTβ mRNA is expressed in the absence of LTα and in some settings LTβ expression appears to be constitutive (31). LTα RNA expression was not examined in this study and even in the adult spleen where LTβ expression is readily detected, LTα expression is relatively low (15, 32). RT-PCR analysis of LTα expression using RNA from various embryo parts showed LTα expression from E11 to E13 in placenta, head, liver, and yolk sac/blood; however, yolk sac expression ceased by E12–13 (33).

Expression of LTβ was seen in the other immune organs. The E16.5–18.5 thymus was LTβ positive, roughly correlating with the first wave of thymocyte processing. Since the fetal liver does not contain the CD4+CD3 cell, this LTβ-positive cell must be derived directly within the thymus or perhaps the gut cryptopatch. LTβR is also expressed in the thymus at this time even though a role for the LT system in thymic development or function has not been revealed in genetically deleted mice (34). The embryonic expression of LTβ in the skin is intriguing given reports of nonlymphoid expression of LTα RNA or protein in inflamed epithelial layers such as in lichenoid skin, in keratinocytes from the middle ear with the hyperproliferating disease cholesteatoma, in skin following hair regression, and in melanocytes (35, 36, 37, 38). The expression of LTα or LTβ in nonlymphoid/myeloid cells at sites of inflammation has not been extensively studied. Since LTβR signals through IKKα and deletion of IKKα results in an epidermal defect, the skin expression of LTβ is potentially interesting (39). However, such a skin defect was not observed in LTβ- or LTβR-deficient mice and therefore the epidermal defect is not linked to the LT pathway. LTβ expression was observed in the developing brain and LT proteins were found in astrocytes and oligodendrocytes in the human brain (28). Since some aspects of ontogeny can be recapitulated in inflammatory processes, these observations may be linked and hence LT ligand expression may not be solely limited to the T, B, and NK lineages.

Staining for LTβR using FACS methods showed a low level of expression in various monocytic and follicular dendritic populations (Fig. 5) (19, 40, 41). In this light, the inability to detect LTβR RNA in the developing LN is possibly not surprising. Lacking a strong signal or the presence of only a few LTβR-positive cells, the LN cannot be visualized. Therefore, even a signal of intermediate intensity may be difficult to score and the inability to detect LTBR expression that must exist in the fetal LN represents a limitation of the approach.

LTβR is expressed from E16 onward in mucosal epithelium lining cells, especially in the intestine, bronchi, and potentially in several ductal epithelial systems, e.g., salivary gland. Expression on the thymic epithelium may account for the receptor expression observed in the thymus. Likewise, in the liver, LTβR is not expressed on the hepatocytes as ascertained by FACS analysis, suggesting that either stromal expression accounts for the ISH results or the surface appearance is modulated. Why this receptor is expressed in these various developing epithelial cell types in the apparent absence of ligand is not clear. The ISH method may limit the ability to detect low levels of the LTβ ligand or other ligands such as LIGHT that bind LTβR may be critical. It is enticing to speculate that the receptor serves in the differentiation process whereby these cells specialize to form either lymphoid anlagen or the layer covering mucosal lymphoid organs in both the gut and the bronchus. During PP development, a VCAM/ICAM-positive mesenchymal cell in the emerging anlage is believed to be a critical source of chemokine expression and this cell presumably expresses LTβR (13). In this study, we have shown that the bulk of the fetal intestinal epithelial cells at E16 are LTBR positive. At this stage the PP anlage is just beginning to develop and the numbers of VCAM/ICAM-positive cells are very few. Therefore, it is likely that if this VCAM/ICAM-negative LTβR-positive cell is recruited into anlage formation, a conversion must occur that would resemble an inflammatory process. Mature intestinal epithelial cells are known to be capable of initiating an inflammatory program and therefore this model has some precedent (42). Inhibition of LTβR signaling in an adult mouse decreases the number of M cells in the follicle-associated epithelium, suggesting that lymphoid cells communicate with the epithelium via this system (43). It is reasonable to assume that during development the LTβR may play a similar regulatory role. Direct comparison by FACS of fetal and adult intestines confirmed that LTβR expression is higher in the fetus and begins to decrease at birth. This observation is interesting since many human breast and colorectal carcinomas are LTβR bright by immunohistochemical staining while much of the surrounding nontransformed stromal tissue is receptor dull (J. Browning, V. Bailly, and S. Violette, unpublished data) and a similar observation was reported for lung carcinomas (44). Therefore, in some epithelial layers, surface LTβR is displayed during development and perhaps some carcinomas recapitulate ontology by reinstating the higher fetal level of surface expression.

We gratefully acknowledge the excellent technical assistance of G. Radlgruber, R. Zanone, Porsnri Lawton, and Apinya Ngam-ek, the photographical assistance of M. Pisteur, and helpful input and critical reading by Paul Rennert and Jennifer Gommerman. We thank Dr. Dajun Yang of the Lombardi Cancer Institute for his original observations on LTβR expression in breast carcinomas and Tom Crowell, Sheila Violette, Bill Yang, Veronique Bailly, and Linda Griffith for further immunohistological assessment.

1

This work was supported by grants from the Swiss National Science Foundation, the Sir Jules Thorn Charitable Overseas Trust, the Ernst and Lucie Schmidheiny Foundation, and the Ciba-Geigy-Jubiläums-Stiftung.

3

Abbreviations used in this paper: LT, lymphotoxin; LN lymph node, PP, Peyer’s patch; ISH, in situ hybridization; IKK, IκB kinase.

1
Hjelmstrom, P., J. Fjell, T. Nakagawa, R. Sacca, C. A. Cuff, N. H. Ruddle.
2000
. Lymphoid tissue homing chemokines are expressed in chronic inflammation.
Am. J. Pathol.
156
:
1133
2
Luther, S. A., T. Lopez, W. Bai, D. Hanahan, J. G. Cyster.
2000
. BLC expression in pancreatic islets causes B cell recruitment and lymphotoxin-dependent lymphoid neogenesis.
Immunity
12
:
471
3
Fu, Y. X., D. D. Chaplin.
1999
. Development and maturation of secondary lymphoid tissues.
Annu. Rev. Immunol.
17
:
399
4
Nishikawa, S. I., H. Hashi, K. Honda, S. Fraser, H. Yoshida, H. Nakano, S. Nishikawa, P. Rennert, K. Ikuta, M. Tamechika, et al
2000
. Inflammation, a prototype for organogenesis of the lymphopoietic/hematopoietic system.
Curr. Opin. Immunol.
12
:
342
5
Futterer, A., K. Mink, A. Luz, M. H. Kosco-Vilbois, K. Pfeffer.
1998
. The lymphotoxin beta receptor controls organogenesis and affinity maturation in peripheral lymphoid tissues.
Immunity
9
:
59
6
Ware, C. F., T. L. VanArsdale, P. D. Crowe, J. L. Browning.
1995
. The ligands and receptors of the lymphotoxin system.
Curr. Top. Microbiol. Immunol.
198
:
175
7
Shinkura, R., K. Kitada, F. Matsuda, K. Tashiro, K. Ikuta, M. Suzuki, K. Kogishi, T. Serikawa, T. Honjo.
1999
. Alymphoplasia is caused by a point mutation in the mouse gene encoding NF-κB-inducing kinase.
Nat Genet
22
:
74
8
Matsushima, A., T. Kaisho, P. D. Rennert, H. Nakano, K. Kurosawa, D. Uchida, K. Takeda, S. Akira, M. Matsumoto.
2001
. Essential role of nuclear factor (NF)-κB-inducing kinase and inhibitor of κB (IκB) kinase α in NF-κB activation through lymphotoxin β receptor, but not through tumor necrosis factor receptor I.
J. Exp. Med.
193
:
631
9
Kim, D., R. E. Mebius, J. D. MacMicking, S. Jung, T. Cupedo, Y. Castellanos, J. Rho, B. R. Wong, R. Josien, N. Kim, P. D. Rennert, Y. Choi.
2000
. Regulation of peripheral lymph node genesis by the tumor necrosis factor family member TRANCE.
J. Exp. Med.
192
:
1467
10
Ansel, K. M., V. N. Ngo, P. L. Hyman, S. A. Luther, R. Forster, J. D. Sedgwick, J. L. Browning, M. Lipp, J. G. Cyster.
2000
. A chemokine-driven positive feedback loop organizes lymphoid follicles.
Nature
406
:
309
11
Bailey, R. P., L. Weiss.
1975
. Ontogeny of human fetal lymph nodes.
Am. J. Anat.
142
:
15
12
Eikelenboom, P., J. J. Nassy, J. Post, J. C. Versteeg, H. L. Langevoort.
1978
. The histogenesis of lymph nodes in rat and rabbit.
Anat. Rec.
190
:
201
13
Honda, K., H. Nakano, H. Yoshida, S. Nishikawa, P. Rennert, K. Ikuta, M. Tamechika, K. Yamaguchi, T. Fukumoto, T. Chiba, S. I. Nishikawa.
2001
. Molecular basis for hematopoietic/mesenchymal interaction during initiation of Peyer’s patch organogenesis.
J. Exp. Med.
193
:
621
14
Yoshida, H., K. Honda, R. Shinkura, S. Adachi, S. Nishikawa, K. Maki, K. Ikuta, S. I. Nishikawa.
1999
. IL-7 receptor α+CD3 cells in the embryonic intestine induces the organizing center of Peyer’s patches.
Int. Immunol.
11
:
643
15
Lawton, P., J. Nelson, R. Tizard, J. L. Browning.
1995
. Characterization of the mouse lymphotoxin-β gene.
J. Immunol.
154
:
239
16
Force, W. R., B. N. Walter, C. Hession, R. Tizard, C. A. Kozak, J. L. Browning, C. F. Ware.
1995
. Mouse lymphotoxin-β receptor: molecular genetics, ligand binding, and expression.
J. Immunol.
155
:
5280
17
Sappino, A. P., J. Huarte, J. D. Vassalli, D. Belin.
1991
. Sites of synthesis of urokinase and tissue-type plasminogen activators in the murine kidney.
J. Clin. Invest.
87
:
962
18
French, L. E., M. Hahne, I. Viard, G. Radlgruber, R. Zanone, K. Becker, C. Muller, J. Tschopp.
1996
. Fas and Fas ligand in embryos and adult mice: ligand expression in several immune-privileged tissues and coexpression in adult tissues characterized by apoptotic cell turnover.
J. Cell Biol.
133
:
335
19
Browning, J. L., I. D. Sizing, P. Lawton, P. R. Bourdon, P. D. Rennert, G. R. Majeau, C. M. Ambrose, C. Hession, K. Miatkowski, D. A. Griffiths, et al
1997
. Characterization of lymphotoxin-αβ complexes on the surface of mouse lymphocytes.
J. Immunol.
159
:
3288
20
Pokholok, D. K., I. G. Maroulakou, D. V. Kuprash, M. B. Alimzhanov, S. V. Kozlov, T. I. Novobrantseva, R. L. Turetskaya, J. E. Green, S. A. Nedospasov.
1995
. Cloning and expression analysis of the murine lymphotoxin β gene.
Proc. Natl. Acad. Sci. USA
92
:
674
21
Nakamura, T., K. Tashiro, M. Nazarea, T. Nakano, S. Sasayama, T. Honjo.
1995
. The murine lymphotoxin-β receptor cDNA: isolation by the signal sequence trap and chromosomal mapping.
Genomics
30
:
312
22
Mebius, R. E., I. L. Schadee-Eestermans, I. L. Weissman.
1998
. MAdCAM-1 dependent colonization of developing lymph nodes involves a unique subset of CD4+CD3 hematolymphoid cells.
Cell Adhes. Commun.
6
:
97
23
Mebius, R. E., P. R. Streeter, S. Michie, E. C. Butcher, I. L. Weissman.
1996
. A developmental switch in lymphocyte homing receptor and endothelial vascular addressin expression regulates lymphocyte homing and permits CD4+CD3 cells to colonize lymph nodes.
Proc. Natl. Acad. Sci. USA
93
:
11019
24
Rennert, P. D., J. L. Browning, R. Mebius, F. Mackay, P. S. Hochman.
1996
. Surface lymphotoxin α/β complex is required for the development of peripheral lymphoid organs.
J. Exp. Med.
184
:
1999
25
Yoshida, H., H. Kawamoto, S. M. Santee, H. Hashi, K. Honda, S. Nishikawa, C. F. Ware, Y. Katsura, S. I. Nishikawa, S. Schweyer, H. J. Radzun.
2001
. Expression of α4β7 integrin defines a distinct pathway of lymphoid progenitors committed to T cells, fetal intestinal lymphotoxin producer, NK, and dendritic cells.
J. Immunol.
167
:
2511
26
Mebius, R. E., P. Rennert, I. L. Weissman.
1997
. Developing lymph nodes collect CD4+CD3LTβ+ cells that can differentiate to APC, NK cells, and follicular cells but not T or B cells.
Immunity
7
:
493
27
Ware, C. F., P. D. Crowe, M. H. Grayson, M. J. Androlewicz, J. L. Browning.
1992
. Expression of surface lymphotoxin and tumor necrosis factor on activated T, B, and natural killer cells.
J. Immunol.
149
:
3881
28
Cannella, B., I. D. Sizing, C. D. Benjamin, J. L. Browning, C. S. Raine.
1997
. Abs to lymphotoxin α (LTα) and LTβ recognize different glial cell types in the central nervous system.
J. Neuroimmunol.
78
:
172
29
Gramaglia, I., D. N. Mauri, K. T. Miner, C. F. Ware, M. Croft.
1999
. Lymphotoxin αβ is expressed on recently activated naive and Th1-like CD4 cells but is down-regulated by IL-4 during Th2 differentiation.
J. Immunol.
162
:
1333
30
Bahary, N., L. I. Zon.
2001
. Development. Endothelium: chicken soup for the endoderm.
Science
294
:
530
31
Millet, I., N. H. Ruddle.
1994
. Differential regulation of lymphotoxin (LT), lymphotoxin-β (LT-β), and TNF-α in murine T cell clones activated through the TCR.
J. Immunol.
152
:
4336
32
Zinetti, M., S. Agyekum, T. Evans, J. Polak, J. Cohen.
1998
. The role of lipopolysaccharide, pro-inflammatory cytokines and bacterial superantigens in the transcriptional regulation of lymphotoxin α and β in mouse splenocytes.
Cytokine
10
:
940
33
Kohchi, C., K. Noguchi, Y. Tanabe, D. Mizuno, G. Soma.
1994
. Constitutive expression of TNF-α and -β genes in mouse embryo: roles of cytokines as regulator and effector on development.
Int. J. Biochem.
26
:
111
34
De Togni, P., J. Goellner, N. H. Ruddle, P. R. Streeter, A. Fick, S. Mariathasan, S. C. Smith, R. Carlson, L. P. Shornick, J. Strauss-Schoenberger, et al
1994
. Abnormal development of peripheral lymphoid organs in mice deficient in lymphotoxin.
Science
264
:
703
35
Middel, P., U. Lippert, K. M. Hummel, H. P. Bertsch, M. Artuc, S. Schweyer, H. J. Radzun.
2000
. Expression of lymphotoxin-α by keratinocytes: a further mediator for the lichenoid reaction.
Pathobiology
68
:
291
36
Seiberg, M., J. Marthinuss, K. S. Stenn.
1995
. Changes in expression of apoptosis-associated genes in skin mark early catagen.
J. Invest. Dermatol.
104
:
78
37
Melani, C., G. Parmiani, M. P. Colombo.
1993
. Lymphotoxin gene expression by melanocytes and melanoma cell lines and persistence of unspliced mRNA.
FEBS Lett.
335
:
114
38
Yan, S. D., C. C. Huang.
1991
. Lymphotoxin in human middle ear cholesteatoma.
Laryngoscope
101
:
411
39
Hu, Y., V. Baud, T. Oga, K. I. Kim, K. Yoshida, M. Karin.
2001
. IKKα controls formation of the epidermis independently of NF-κB.
Nature
410
:
710
40
Murphy, M., B. N. Walter, L. Pike-Nobile, N. A. Fanger, P. M. Guyre, J. L. Browning, C. F. Ware, L. B. Epstein.
1998
. Expression of the lymphotoxin beta receptor on follicular stromal cells in human lymphoid tissues.
Cell Death Differ.
5
:
497
41
Husson, H., S. M. Lugli, P. Ghia, A. Cardoso, A. Roth, K. Brohmi, E. G. Carideo, Y. S. Choi, J. Browning, A. S. Freedman.
2000
. Functional effects of TNF and lymphotoxin α1β2 on FDC-like cells.
Cell Immunol.
203
:
134
42
Dwinell, M. B., N. Lugering, L. Eckmann, M. F. Kagnoff.
2001
. Regulated production of interferon-inducible T-cell chemoattractants by human intestinal epithelial cells.
Gastroenterology
120
:
49
43
Debard, N., F. Sierro, J. Browning, J.-P. Kraehenbuhl.
2001
. The role of nature lymphocytes and lymphotoxin on the development of the follicle-associated epithelium and M cells in mouse Peyer’s patches.
Gastroenterology
120
:
1173
44
Boussaud, V., P. Soler, J. Moreau, R. G. Goodwin, A. J. Hance.
1998
. Expression of three members of the TNF-R family of receptors (4-1BB, lymphotoxin-β receptor, and Fas) in human lung.
Eur. Respir. J.
12
:
926