Immunoarchitecture of Distinct Reticular Fibroblastic Domains in the White Pulp of Mouse Spleen

Abstract

The development of peripheral lymphoid tissues requires a series of cognate interactions between hemopoietic and stromal cell populations, including reticular fibroblasts, which form the mesenchymal scaffolding of distinct tissue compartments. Here we describe the formation of different fibroblastic domains in the mouse spleen white pulp by using two new rat monoclonal antibodies (MAbs). In the white pulp, MAb IBL-10 labels both T- and B-cell zone reticular elements at various intensities. The IBL-10^hi subset was found primarily at the edge between the peripheral part of the PALS and follicles, and the IBL-10^lo compartment was distributed evenly within the white pulp. The IBL-10^hi subset appeared during the first 2 postnatal weeks and was absent in SCID mice. The white pulp fibroblast subset identified with MAb IBL-11 had a different tissue distribution and kinetics of ontogeny, with an appearance overwhelmingly restricted to the PALS and a narrow rim at the edge of the follicular border area toward the marginal zone. The appearance of IBL-11–positive reticular cells was delayed compared with that of the IBL-10^lo–positive subset. The formation was independent of the influence of antigen receptor–bearing lymphocytes, as evidenced by the presence of IBL-11–positive fibroblasts in SCID mice. By transferring various lymphocyte subsets into SCID mice, partial compartmentalization of the white pulp fibroblasts could be induced, indicating that these mesenchymal fibroblast precursors retain their ability to differentiate upon encountering mature T- or B-cells.

Keywords

spleen white pulp stroma fibroblast heterogeneity SCID mouse

A general feature of the secondary lymphoid organs is their highly compartmentalized tissue structure, which correlates with their ability to mount effective immune responses. The ordered arrangement of their lymphohemopoietic and mesenchymal constituents is the result of successive cognate interactions between immature lymphoid cells and stromal precursors during tissue development, mediated by adhesion molecules, various members of the LT/TNF family and their receptors, and lymphotropic chemokines, respectively (Fu and Chaplin, 1999; Mebius 2003). These pathways must remain continuously functional both to maintain the compartmentalized structure of peripheral lymphoid tissues and to permit their eventual transformation during immune responses, including the generation of germinal centers (Mackay et al. 1997).

The spleen, as the single largest peripheral lymphoid organ in both humans and rodents, has a number of distinguishing features compared with other tissues of the immune system. The main parts of the spleen are the white pulp and the red pulp, with distinct and complementary functions. The boundary between the two regions is the marginal zone, containing specialized marginal sinus-lining cells surrounded by a fibroblastic reticular network, marginal zone macro-phage subsets, and a separate B-cell pool (Kraal 1992; Martin and Kearney 2002; Karlsson et al. 2003). In the white pulp, the lymphocyte composition of distinct T- and B-cell–rich compartments is similar to that in lymph nodes (Nolte et al. 2000). In the white pulp, the T-cells encircle the central arteriole, thus forming the periarteriolar lymphoid sheath (PALS), to which spheroid accumulations of B-cells and follicular dendritic cells as follicles are connected. The red pulp may serve as an auxillary myelopoietic compartment and functions as an exit route for the majority of recirculating lymphoid cells. In addition, it efficiently phagocytoses blood-borne antigens and decaying erythrocytes via its macrophages, and hosts clusters of plasma cells (MacLennan et al. 2003).

The movement of recirculating lymphoid cells and the flow of soluble compounds are greatly influenced by the mesenchymal scaffolding of the spleen, composed of reticular cells and a broad array of extracellular matrix components associated with these cells (Liakka and Autio-Harmainen 1992; Ocklind et al. 1993). This guiding function may also include the adsorption and selective display of homing chemokines produced by regional (follicular or T-cell zone–associated) stromal cells as directional cues (Nolte et al. 2003). In this complex arrangement, the fibroblastic reticular cells create a meshwork that may connect distant parts of the splenic lymphoid tissue and may thus contribute to the formation of a continuous route for migrating cells and soluble molecules as well, similarly to lymph nodes (Gretz et al. 1996). Furthermore, the possible role of reticular cells in influencing specific immune responses has been indicated by the trapping and possible retention of immune complexes coated via complement by ill-defined perivascular reticular cells (Taylor et al. 2002). Moreover, the addition of fibroblast cells in culture could promote the formation of plasma cells after in vitro activation of human B-cells, in which system the extent of this supportive effect was dependent on the tissue origin of fibroblastic cells (Skibinski et al. 1998).

Although many of the molecular mechanisms that contribute to the lymphocyte migration and positioning remain obscure, many aspects of the splenic fibroblastic reticular cells have remained enigmatic. A limited number of studies indicate their regional heterogeneity with regard to their phenotypic markers (Yoshida et al. 1991,1993). However, data concerning their developmental and functional specifics are sparse (ten Dam et al. 2003). In murine studies aimed at the splenic reticular architecture, the fibroblasts have been identified with the rat monoclonal antibody ER-TR7 (Van Vliet et al. 1986) against an undefined marker that, although providing valuable information on the general fibroblastic arrangement of the spleen and other lymphoid organs, does not reveal the regional complexity of the reticular meshwork. Therefore, any new reagent that can identify subset-restricted elements of the reticular fibroblastic compartments may be of potential use in providing more accurate data on the dynamics and architecture of the lymphoid-mesenchymal domains of the spleen.

Here we report the production of two new rat monoclonal antibodies (MAbs) directed against different elements of the splenic fibroblast network. Using these antibodies, we also analyze the postnatal appearance of distinct fibroblast subsets and their spatial relationship in both normal and SCID mice, and also in chimeric SCID mice reconstituted with mature lymphoid cells. Our data indicate that the reticular scaffolding in various parts of the splenic white pulp is composed of phenotypically different fibroblasts whose physiological tissue arrangement in normal mice requires the presence of lymphocytes.

Materials and Methods

Mice

Inbred BALB/c (H-2^d), C.B-17/Icr scid/scid (SCID, H-2^d), and C57Bl/6 (B6, H-2^b) mice were purchased from Charles River Laboratories (Wilmington, MA) and maintained at the University's SPF animal housing facility. (BALB/c × B6) F1 (H-2^d/b) mice were used as lymphocyte donors for SCID reconstitution experiments. SCID mice 4–6 weeks old were injected IV via the lateral tail vein with 2 × 106 cells from F1 mice (either mixed or purified T- or B-cells, respectively). After reconstitution, the SCID mice were provided with Ciprobay in their drinking water. For experiments on irradiation effects, 4–6-week-old BALB/c mice were irradiated using a single open field with 6 MeV photons of a Philips Sli linear accelerator at 10 Gy calculated at midline level of the mice; source-skin distance was 100 cm. For dose buildup compensation, an 18-mm bolus was used. Dose rate was at low setting. All procedures involving live animals were conducted in accordance with the guidelines set out by the Ethics Committee on Animal Experimentation of the University of Pécs.

Hybridoma Production

Splenic white pulp fragments from BALB/c mice that had been immunized IP with washed splenic red blood cells (SRBCs) were collected as described previously (Nolte et al. 2000). After isolation, the loose cells were separated from the fragments with repeated sedimentation at 1 × g in DMEM tissue culture medium, followed by injection into 8-week-old Wistar rats (white pulp material equivalent to five spleens on three occasions, 2 weeks apart each). The spleen cells from immunized Wistar rats were fused with Sp-2/0 Ag 14 myeloma-derived cells using polyethylene glycol MW 4000 and processed as previously described (Köhler and Milstein 1975). The hybridoma supernatants were tested for specific antibody production using immunohistochemistry (Balázs et al. 1999); the positive wells were cloned by repeated limiting dilution.

Antibodies and Other Reagents

Anti-endothelium and pan-reticular rat MAb IBL-7/22, DaB1 mouse alloantibody (IgG1 isotype) against MHC Class I (H-2K^b allotype), FITC-labeled rat anti-mouse CD45 (clone IBL-5/25), and anti-mouse MHC Class II (IBL-5/22) MAbs were produced in our lab (Balázs et al. 1998,1999, 2001). Antibodies against murine MadCAM-1 (clone MECA-89), VCAM-1 (clone 429), CR1/2: CD35/21 (clone 7G6), and B220/CD45RA (clone RA3–6B2) were purchased from BD Pharmingen (San Diego, CA). The anti-mouse CD3 MAb clone KT3 was obtained from Serotec (Crawley Down, UK). For detection of unlabeled primary antibodies, PE-conjugated rat anti-mouse IgG1, PE-conjugated goat anti-rat IgG (BD Pharmingen), or FITC-labeled mouse anti-rat IgG (Zymed; South San Francisco, CA), or biotinylated mouse monoclonal anti-rat K-chain (clone MRK-1; BD Pharmingen) antibodies were used, in conjunction with streptavidin-PE (BD Pharmingen) or extravidin-alkaline phosphatase (Sigma-Aldrich; Budapest, Hungary). The immunohistochemical (IHC) detection of fluorescein-labeled anti-B220 (BD Pharmingen) was performed using HRP-conjugated sheep anti-fluorescein IgG Fab (Roche; Mannheim, Germany).

Purification of T- and B-cells by MACS

Spleen cells (4 × 10⁷) from (BALB/c × B6) F1 mice were divided into two aliquots and incubated with FITC-conjugated anti-B220 MAb or anti-MHC class II MAb on ice. After washing, the cells were incubated with anti-FITC beads (Miltenyi Biotec; Bergisch Gladbach, Germany), followed by separation using VarioMACS equipment. For B-cell enrichment, the B220-positive cells were collected using an LS column; for T-cell purification, the MHC class II-positive cells were depleted with an LD column. After washing, 5 × 10⁶ donor cells (B-, T-, and mixed) were injected IV into three different groups of SCID recipients.

Immunohistochemistry and Immunofluorescence

The single-label IHC staining of cryostat sections from various tissues with rat hybridoma supernatants containing IBL-10 and IBL-11 antibodies followed by biotin-amplified alkaline phosphatase detection was performed as described, using NBT/BCIP as chromogen substrate in the presence of 1 mg/ml levamisol (Balázs et al. 2001). For double, immunohistochemistry the visualization of stromal components by alkaline phosphatase-based labeling was followed by a short treatment of slides with dinitro-phenyl-hydrazine (1 mg/ml in PBS) to block endogeneous peroxidase activity. After rinsing, the slides were treated with normal rat IgG at 20 μg/ml to saturate the remaining binding sites of previously applied anti-rat Ig reagent, then incubated with FITC-conjugated anti-B220 MAb. The reaction was developed with HRP-conjugated sheep anti-FITC Fab fragments, using amino-ethyl carbasol as chromogen and 0.01% H₂O₂ as substrate in a 0.1 M Na-acetate buffer, pH 5.2.

For immunofluorescence, the sections were incubated with unlabeled primary antibodies followed by FITC-conjugated anti-rat IgG. For dual labeling, the sections were then blocked with 20% normal rat serum in PBS, followed by incubation with the second biotinylated MAb. After rinsing, the sections were incubated with streptavidin–PE conjugate and viewed under an Olympus BX61 fluorescent microscope. The acquisition of digital pictures with a CCD camera and the morphometric image analysis were performed using the analySIS software.

Flow Cytometry

The reconstituted SCID mice were decapitated, and their blood was collected in tubes containing Na-heparin. Their spleens were removed and snap frozen for IHC evaluation. The undiluted blood was incubated with a cocktail of FITC-conjugated rat MAb against CD45, CD3, or anti-B220 antibodies and DaB1 MAb against H-2K^b on ice in the presence of 2.4G2 MAb (BD Pharmingen) to block FcγRII-mediated binding of immunoglobulins. After incubation, the cells were washed with PBS containing 0.1% BSA and Na-azide. The DaB1 (mouse IgG₁) alloantibody was revealed using PE-conjugated rat anti-mouse IgG₁. After the lysis of erythrocytes, the cells were analyzed with a Becton-Dickinson FACSCalibur using the CellQuest software. The fluorescence distribution was determined from 10,000 electronic events gated on forward- and side-scatter parameters.

Results

Complex Organization of White Pulp Reticular Meshwork in the Murine Spleen Delineated by Subset-specific Antibodies IBL-10 and IBL-11

The initial screening of hybridoma supernatants revealed a number of clones producing antibodies against various cellular components of both the marginal zone and central white pulp. The choice of white pulp fragments as immunogen from SRBC-primed mice proved rather effective because all of the positive clones derived from the fusion had a restricted reactivity against various components of those parts of the tissue. Because our aim was to produce antibodies capable of identifying distinct stromal elements with tissue distribution restricted to the white pulp, the clones producing antibodies with diffuse or nonselective reactivity were excluded from further studies.

With the exception of intense reaction against the trabeculae by the IBL-10 MAb, there was no specific staining of red pulp reticular components (fibers or cells) by either the IBL-10 or the IBL-11 MAb. In the white pulp, the IBL-10 antibody labeled the adventitia of central arterioles and also some reticular cells in both the follicles and the PALS. The boundary toward the marginal zone was only faintly decorated, whereas in the border region between the T and B zones, a more-pronounced reaction could often be observed. Some more intensely labeled cells could also be detected in the deeper regions of follicles (Figure 1). Induction of the germinal center by IP administration of SRBCs did not elicit the redistribution of IBL-10-reactive cells in any region of the spleen (not shown).

In addition to their mostly lymphocytic composition, the follicles also contain a number of nonlymphoid cells, including follicular dendritic cells (FDCs), whose occurrence is restricted to the follicles. The follicular appearance of strongly IBL-10-positive reticular cells prompted us to investigate their possible colocalization with FDCs using dual immunofluorescence with IBL-10 MAb and an anti-CD21/35 antibody to identify FDCs. We found that apart from a few incidental juxtapositioned cells, no clear relationship could be observed between the distribution of these two cell types (Figures 2A and 2B).

Figure 1

Splenic reactivity of IBL-10 (A,D) and IBL-11 (B,E) MAbs in serial sections from young adult mouse spleen. For comparison and tissue orientation, a pan-endothelium and reticular fibroblast staining (IBL-7/22, C) is shown; arrow in C indicates the position of central arteriole. Fo, follicles; PALS, periarteriolar lymphoid sheath. In (B) and (E), B cells are labeled red; the IBL-10 (D) and IBL-11 (E) reactive fibroblast subsets are stained gray blue/black. Bar = 100 μm.

Figure 2

Spatial relationship between various stromal elements and the fibroblast subsets identified with IBL-10 and IBL-11 MAbs. (A,B) Insets show IBL-10-positive fibroblasts (green) and FDCs (red anti-CR1.2) in the follicular area of the spleen. (C,D) Show IBL-11 fibroblasts (green) and MAd-CAM-1 (red) positive marginal sinus lining cells. (E–G) Show the relationship between IBL-10 and IBL-11 subsets, with the two smaller insets corresponding to the approximate area from the central part of the PALS and the peripheral segment of the follicles, respectively (indicated by rectangles).

Unlike the generally even distribution of IBL-10–positive reticular cells and fibers between follicles and PALS, IBL-11 MAb had a strong preference for T-cell zone reactivity (Figure 1). In addition, it also delineated an almost continuous rim within the follicles facing the white pulp proximal aspect of the marginal zone. The IBL-11-positive cells did not expand beyond the white pulp distal layer of marginal sinus, as indicated by their position relative to the sinus lining cells expressing MADCAM-1 only (Figures 2C and 2D). Similarly to IBL-10, the IBL-11 MAb had no detectable reactivity against any lymphohemopoietic cell type isolated from spleen, lymph nodes, bone marrow, and thymus (not shown).

The IHC reactions indicated that whereas the follicular and marginal zone reactivities of IBL-10 and IBL-11 MAbs were clearly different, in the PALS they may be related. To confirm this interpretation, a dual immunofluorescent staining was performed. We found that the inner PALS that surrounds the central artery containing IBL-10-positive adventitial cells appears to be dominated by IBL-11 single positive cells, while the outer PALS contains an IBL-10/11–coexpressing compartment. In the follicles, the IBL-10-positive cells seldom coexpress IBL-11, whose population is restricted to the cells adjacent to the marginal zone (Figures 2E and 2G).

Postnatal Development of the White Pulp Reticular Compartments Associated with Lymphocyte Colonization

Previous findings on the postnatal phase of spleen development have suggested the sequential reorganization of several stromal elements (including endothelial cells and FDC precursors) coupled with lymphocyte compartmentalization during the first 2 weeks after birth (Balázs et al. 2001; Balogh et al. 2001). We tested whether the fibroblastic cells identifiable with IBL-10 and IBL-11 MAbs also change their tissue distribution during the establishment of adult-type architecture.

Immediately after birth (age <D1) there were only IBL-10-positive reticular cells adjacent to the central artery, which at that stage was surrounded primarily by B-cells (identified as B220-positive lymphocytes). IBL-11 reactivity at this period could not be detected. A few days later (D3–5), a more-pronounced reactivity projecting toward the peripheral parts of the developing white pulp could be noted. By D7, a faint IBL-11 staining was observed, primarily located at the white pulp region distal to the central artery. At this stage, some focal accumulations of the B-cells could be noted, whereas the original ring-like cluster of B220-positive cells was dislodged from the immediate vicinity of the central artery by the gradual accumulation of T-cells. The border region between the T-cell-rich area and these premature follicles was sometimes found to contain the IBL-10^hi reticular cells mentioned previously. At ~D10, the IBL-11-positive fibroblasts showed a tendency for follicular paucity in appearance, which was accompanied by the formation of a loose rim composed of IBL-11–positive cells at the inner aspect of the MAdCAM-1–positive marginal sinus, similar to the adult-type distribution (Figure 3). The above sequence of events is summarized in Table 1.

Influence of Lymphocytes on the Establishment of Fibroblast Domains

It has been well established that the lymphoid cells play a major role in the functional diversification and compartmentalization of radiation-resistant stromal constituents during the formation of secondary lymphoid tissues (Fu and Chaplin 1999; Mebius 2003). Therefore, we investigated whether the reticular cells identified with IBL-10 and IBL-11 MAbs are also radiation resistant. We found that exposure to the supralethal dose of 10 Gy did not eliminate these cells. However, the removal of sensitive lymphoid cells induced the collapse of established networks and led to the bundling of both IBL-10 and IBL-11 elements (Figures 4A and 4B; for comparison with normal age-matched control, see Figure 1).

Because these mice had already established their fibroblastic domains before the irradiation, we next tested for the effect of absence of antigen receptor-bearing lymphocytes on the appearance of IBL-10-and IBL-11-positive components in SCID mice. It has previously been observed that the adoptive transfer of mature lymphoid cells into SCID recipients can induce the local differentiation of FDCs (Kapasi et al. 1993; Yoshida et al. 1995). We found that in control SCID mice, there was only a faint staining of fibroblasts recognized by IBL-10 MAb around the central artery. The IBL-10 compartment typically present at the border area between the follicles and PALS could not be observed. However, the red pulp trabecular staining with IBL-10 was preserved, as in normal mice. In contrast, IBL-11 MAb identified some strongly reactive cells grouped in a tight cluster at the periphery of the rudimentary white pulp area (Figures 4C and 4D). This indicated that the appearance of IBL-10^hi follicular fibroblasts requires the presence of mature lymphocytes. In contrast, the formation of the IBL-11-positive subset is independent of both T- and B-cells, even though its tissue arrangement is influenced by the lymphocyte compartments.

Next, we compared the ability of lymphocytes (either whole or purified T- and B-cells) to induce the rearrangement of fibroblast subsets reactive with IBL-10 and IBL-11 MAbs. The efficiency of cell transfer was evaluated by flow cytometric detection of donor H-2K^b alloantigen expression by the transplanted (B6 × BALB/c) F1 lymphocyte subsets among the total blood leukocytes. In contrast to the overwhelming majority of myeloid cells (granulocytes and monocytes), which appeared to be host derived, >90% of circulating lymphoid cells with lineage-restricted (B220 and CD3) differentiation markers expressed the donor-associated H-2K^b alloantigen in mice transplanted with pooled lymphocytes or with purified T- or B-cells, at a similar degree of donor cell load (Figure 5). We found that 1 week after the cell transfer, the tissue distribution of IBL-10– and IBL-11–positive fibroblast subsets was similar to that in normal mice, with some minor differences (Figure 6). We could observe the appearance of IBL-10^hi cells in the white pulp in all three groups of transplanted mice, and also the loosening of bundled IBL-11 cells at the peripheral parts of the white pulp, although the typical T-cell zone-associated reactivity of IBL-11–positive fibroblasts could not be established. We also found that this partial remodeling of IBL-11–positive reticular cells also occurred in recipients of F1 B-cells. The transfer of T-cells into SCID mice could also restore the presence of IBL-10^hi cells, which are typically restricted to the follicular area in normal mice. In that respect, there was no difference between the abilities of mixed or purified T- or B-cells to induce formation of IBL-10^hi–positive fibroblasts and rearrangement of IBL-11. However, the zonal enrichment of the IBL-10^hi cells at the periphery of follicles, as observed in normal mice, did not occur in T-cell–transplanted SCID recipients because these cells were distributed evenly in the white pulp. Moreover, the tendency of IBL-11 cells to be arranged around the marginal zone aspect of follicles and within the PALS (thus leading to the paucity of these cells in the central part of the follicle) was also diminished, because these fibroblasts were distributed throughout the partially restored white pulp in these mice without any indication of gradient formation (Table 2). Permitting the tissue remodeling in these reconstituted SCID mice to continue for 2 more weeks could not correct this abnormal pattern of IBL-10^hi and IBL-11 cell distribution (not shown).

Figure 3

Postnatal development of white pulp fibroblast compartments. Tissue samples from newborn (nb) (age <12 hr, A), D3 (B), D5 (C,D), and D7 (E,F) pups. Sections were labeled with IBL-10 MAb (A,B,C,E,G), IBL-11 MAb (D,F,H), anti-CD45RA/B220 (I), and anti-CD3 (J). The age of samples is indicated at the lower right corner of insets. Serial section from D10 samples (G-J) indicate the last stage before the final adult-type tissue architecture illustrated in Figure 1. Asterisks label the position of the central arteriole. Arrowheads in B,C,E,G point to trabecular reactivity of IBL-10 MAb in the red pulp. Arrows in (D) label the first appearance of IBL-11 dim cells. Bar in (J) = 100 μm for all panels.

Table 1

Sequence of tissue rearrangment processes of lymphocytic and fibroblast compartments in the postnatal spleen

	Marker/cell type
Time	IBL-10	IBL-11	CD3 (T-cells)	B220 (B-cells)
NB	Trabecular and periarteriolar staining	Not detectable	Small clusters around the artery	Small groups of cells in the red pulp and along the central artery
D3	More-pronounced trabecular reaction, clusters of cells radiating from the central artery	Not detectable	As above	Thicker (8–15) layers of cells around the artery
D5	Reticular staining across the white pulp area	Faint expression at the peripheral regions of the white pulp without marginal zone–associated reactivity	3–8 layers of cells immediately around the central artery	Cells distanced from the central artery
D7	As above	More-intense reactivity in the white pulp area	As above	Uneven focal accumulation of B-cells along the peripheral part of cells surrounding the artery
D10	Appearance of clusters containing more-intensely stained cells in the central area of developing white pulp	Demarcation of IBL-11–positive cells around the marginal zone, paucity of staining at areas corresponding to the follicles	10–30 cell layers filling up the periarteriolar region	As above

NB, newborn.

Figure 4

Effect of irradiation (A,B) and the SCID mutation (C,D) on the architecture of white pulp fibroblast subsets. Mice were irradiated, spleens were removed 48 hr later, and the serial sections were reacted with IBL-10 (A) or IBL-11 (B) MAb. (C) Arrow indicates a red pulp trabecule intensely stained with IBL-10 MAb. (D) Arrowhead points to a cluster of bundled IBL-11-positive fibroblasts. (C,D) Rp denotes red pulp. Asterisks identify central arteriole. Bar in (D) = 100 μm for panels A–D. The demonstrated reactions are a representative sample of results obtained on three separate occasions using groups of five mice.

Figure 5

Lymphoid chimerism in SCID mice after adoptive transfer of mixed or purified T and B cells from F1 donor mice 1 week after the cell transfer. (A) The various myeloid (granulocyte, R1 and monocyte, R2) donor/recipient chimerisms, according to their expression of donor H-2K^b alloantigen correlated with their CD45 display. Cells in R3 correspond to the lymphoid cells. Numbers indicate the frequency of cells in the given quadrants. (B) The results of mixed or purified T- and B-cell repopulation by F1 cells in SCID mice using the antibody combinations (indicated at the x and y axes, respectively). Numbers indicate the frequency of cells in the given quadrants. Data are representative of three independent experiments performed on three to five mice.

Figure 6

Reconstruction of SCID fibroblast compartments by the adoptive transfer of various donor cells (indicated at the top of columns). (A,C) Labeled with anti-B220 MAb. (B,D) CD3 expression. (E–G) labeled with IBL-10 MAb. (H–J) IBL-11 staining. Representative results obtained in three independent occasions using groups of five mice.

Discussion

The aim of our present study was to generate new rat MAbs capable of identifying distinct components of the reticular fibroblastic scaffolding of the white pulp of murine spleen and to define their ontogeny during the early postnatal period. These cells have been rather poorly defined in terms of cell surface markers, which has also prompted us to compare the reactivities of these new MAbs to those developed earlier (Table 3).

Table 2

Effects of adoptive cell transfer on the fibroblast domain organization in SCID mice

Donor cells	Appearance of IBL-10^hi cells	Accumulation of IBL-10^hi cells at the follicle/PALS border zone	Redistribution of bundled IBL-11–positive cells from the periphery of PALS	Segregation of IBL-11–positive cells in the white pulp a
Mixed T/B- cells	+	+	+	+/−
T-cells	+	−	+	−
B-cells	+	+	+	+/−
None	−	−	−	−

^aEstablishment of a distinct rim of IBL-10–positive fibroblasts around the marginal zone and/or their preferential enrichment in the PALS.

Taking the ER-TR7 MAb as a pan-fibroblast marker (expressed by both red and white pulp fibroblasts), it is noteworthy that the B-cell area probably contains fewer fibroblasts than the PALS, in which the fibroblasts form a dense network. This compartmentalized distribution is controlled by LTβ/LIGHT, because in their absence, the normally uneven pattern becomes diffuse despite the presence of mature lymphoid cells in normal numbers (Ettinger et al. 1996). Therefore, the successful segregation of fibroblasts between T- and B-cell zones is not simply dependent on appropriate lymphocyte numbers (low in SCID or RAG^–/–) but is also dependent on the lymphocytes' inducer capability, which is severely impaired in the absence of LTβR engagement. It is also important that no alterations of the red pulp reticular arrangements detectable with ER-TR7 have been reported in LT deficiency (Tumanov et al. 2003), indicating that the formation of reticular fibroblasts in the red pulp is regulated in a different fashion compared with the white pulp, presumably acting on different cells. In that respect, it is important to emphasize that the splenic reactivities of both IBL-10 and IBL-11 MAbs are almost completely restricted to the white pulp area, although to its different compartments. IBL-10 was found to react against the white pulp with a slight follicular dominance (particularly at the follicle–PALS border), whereas IBL-11 had a clearly overwhelming reaction in PALS, with a follicular reactivity restricted to its perimeter, corresponding to the follicular aspect of the marginal zone. Other antibodies with a similarly skewed labeling pattern include the BP-3 clone detecting a CD38-related epitope shared by immature lymphocytes and stromal cells, and WP-1 MAb, both preferentially reacting with stromal cells in splenic follicles (McNagny et al. 1991; Yoshida et al. 1991). WP-1 strongly labels follicles in normal mice, whereas in SCID mice, it recognizes reticular cells in rudimental “follicular” areas adjoining the PALS (Yoshida et al. 1993). In addition to the intense follicular reaction in normal mice, there is also a sparse reaction in the PALS with WP-1 MAb, which appears to be missing in the SCID spleen. The BP-3 antigen has been reported to be present in both neonatal and SCID mice (McNagny et al. 1991), further confirming that it differs from both IBL-10 (inasmuch as it is very much reduced in SCID) and IBL-11 (inasmuch as it is absent in neonates). It was also reported that these BP-3–positive stromal components also include FDCs (Yoshida et al. 1995), marginal sinus lining cells, and other unidentified stromal cells (Ngo et al. 1999). According to our findings, however, neither IBL-10 nor IBL-11 had any significant coexpression with CD21/35 displayed by FDCs. Moreover, the follicular reactivity of WP-1 MAb appears to be much more extensive compared with either IBL-10 or IBL-11 MAb, suggesting that the WP-1 MAb may, similar to BP-3, also identify other follicular stromal cells, in addition to the fibroblasts (Yoshida et al. 1991).

Table 3

Main characteristics of monoclonal antibodies used for identifying reticular fibroblast in the murine spleen a

	Regional reactivities
Clone/Marker	T-cell area	B-cell area	Red pulp	Presence in SCID	Reference
IBL-7/22	++	+	+	+	Balázs et al. 1999, and unpublished
ER-TR7	++	+	+	NR	Van Vliet et al. 1986
WP-1	+	++	−	+	Yoshida et al. 1991,1993
BP-3	+	++	−	+	McNagny et al. 1991
RPSC-2	+	−	++	+	Yoshida et al. 1991,1993
gp38	++	−	−	+	Farr et al. 1992
IBL-10	+	+	−	+/−	Present study
IBL-11	+	+	−	+	Present study

^aSymbols indicate the relative regional intensity of expression. NR, not reported.

With regard to the fibroblast components of the marginal zone area, it is interesting to note the close spatial relationship between the MAdCAM-1 reactivity of the sinus lining cells (Kraal et al. 1995) and the IBL-11 reactivity of adjacent fibroblasts. In humans, with no marginal sinus, the equivalent region, termed the perifollicular zone, contains MAdCAM-1-positive fibroblasts, which exhibit a number of unusual features, including the expression of smooth muscle α-actin, smooth muscle myosin, cytokeratin 18, and VCAM-1 and Thy-1 antigens, and which are thought to be involved in regulating T-cell migratory events (Steiniger et al. 2001). Although the MAdCAM-1 in unstimulated mice is primarily restricted to the sinus lining cells, its expression can also be observed at an extravascular location on FDCs in germinal centers, where the display of MAdCAM-1 antigen appears to delineate a continuous compartment originating from the marginal sinus (Balogh et al. 2002). We found that this upregulation of MAdCAM-1 by germinal center stromal cells in SRBC-immunized mice is not accompanied by the appearance of IBL-10 or IBL-11 epitopes on the MAdCAM-1-positive follicular cells in GCs, which suggests that the expression of these two markers is unrelated to the adaptation events of resting stromal elements to an altered lymphocytic microenvironment during GC formation.

The ability of transferred lymphoid cells to induce the appearance or redistribution of certain microenvironmental cells in immunodeficient SCID recipients has already been reported for FDCs and for reticular cells identifiable with WP-1 and RPSC-2 MAbs, respectively (Kapasi et al. 1993; Yoshida et al. 1993). The preference for a certain lymphocyte subset presenting inducer molecules, such as LTα/β2 or TNFα, appears to be lymphoid tissue specific because the lack of expression by B-cells has different degrees of inhibition of FDC differentiation in the spleen compared with that in lymph node or Peyer's patches (Tumanov et al. 2003). In our studies on the spleen, the transfer of various lymphoid cells at the same number could effectively induce the rearrangement of IBL-11–positive components, while it also initiated the appearance of IBL-10^hi cells in the follicular region. Our data indicate that the IBL-11-positive cells first appear at the peripheral part of the white pulp in both normal newborn and young adult SCID mice, but their subsequent rearrangement is probably related to the presence of mature lymphocytes and thus is impaired in SCID. It remains to be determined whether the remodeling of IBL-11–positive cells from their original bundled distribution is due to their migration or simply to their separation caused by the lymphocyte colonization. It was interesting to note, however, that this remodeling also occurred in the B-cell–reconstituted animals, despite the preferential T-zone expression of IBL-11–positive reticular cells in normal mice. Similarly, the IBL-10^hi cells in B-cell-transplanted SCID mice also appeared at the same density as in mice reconstituted with T-cells, arguing against a special T- or B-cell requirement for the establishment of this subset, in contrast to the primarily B-cell–dependent differentiation of FDCs (Wang et al. 2001). The failure to form an IBL-10^hi fibroblast margin at the border between T- and B-cell zones in T-cell–transplanted SCID recipients is probably related to the lack of B-cells. Therefore, we postulate that the B-cells (although effective inducers of their differentiation) probably restrict the infiltration of these cells into the deeper parts of the follicle, which was absent in these animals. Alternatively, the B-cells' role might be to direct the appearance of these (or similar cells) in the T-cell zone (Ngo et al. 2001). It appears, therefore, that although the major events can overwhelmingly be corrected by the transfer of B-cells alone into young adult SCID mice, it is likely that the establishment of normal structure requires the availability of both subsets at an earlier period. To define precisely the entity of other stromal reticular cells influenced by lymphoid cells and their ontogenic relationship with each other, a systemic approach to delineate these various reticular cells with a battery of available monoclonal antibodies appears desirable.

In summary, our new monoclonal antibodies detect two reticular fibroblast–associated antigens whose tissue expression is restricted to two different compartments of splenic white pulp. Their appearances have different ontogenic features and requirements for antigen receptor– bearing lymphocytes but they apparently display no particular lymphocyte preference for their induction and proper tissue distribution. Further analyses employing these new monoclonal antibodies in mutant mice with LT/TNF deficiency–related developmental abnormalities that affect the stromal architecture of peripheral lymphoid organs may further highlight the complex organization of the reticular fibroblast cells.

Footnotes

Acknowledgements

Supported by ETT grant No. 592/2003 from the Ministry of Health, Social and Family Affairs, Hungary (PB).

We gratefully acknowledge the expert contribution of Ms Judit Melczer in maintaining the hybridoma cells, and Prof Béla Somogyi for access to the FACSCalibur flow cytometer at the Department of Biophysics. For rat MAbs IBL-10 and IBL-11, please contact PB.

References

Balázs

Grama

Balogh

(1999) Detection of phenotypic heterogeneity within the murine splenic vasculature using rat monoclonal antibodies IBL-7/1 and IBL-7/22. Hybridoma 18:177–182.

Balázs

Horváth

Balogh

(1998) Simple determination of donor/host origin and donor leukocyte subsets in rat-mouse chimeras. J Immunol Methods 218:117–121.

Balázs

Horváth

Grama

Balogh

(2001) Phenotypic identification and development of distinct microvascular compartments in the postnatal mouse spleen. Cell Immunol 212:126–137.

Balogh

Aydar

Tew

Szakal

(2001) Ontogeny of the follicular dendritic cell phenotype and function in the postnatal murine spleen. Cell Immunol 214:45–53.

Balogh

Aydar

Tew

Szakal

(2002) Appearance and phenotype of murine follicular dendritic cells expressing VCAM-1. Anat Rec 268:160–168.

Ettinger

Browning

Michie

van Ewijk

McDevitt

(1996) Disrupted splenic architecture, but normal lymph node development in mice expressing a soluble lymphotoxin-beta receptor-IgG1 fusion protein. Proc Natl Acad Sci USA 93:13102–13107.

Farr

Berry

Kim

Nelson

Welch

Aruffo

(1992) Characterization and cloning of a novel glycoprotein expressed by stromal cells in T-dependent areas of peripheral lymphoid tissues. J Exp Med 176:1477–1482.

Chaplin

(1999) Development and maturation of secondary lymphoid tissues. Annu Rev Immunol 17:399–433.

Gretz

Kaldjian

Anderson

Shaw

(1996) Sophisticated strategies for information encounter in the lymph node: the reticular network as a conduit of soluble information and a highway for cell traffic. J Immunol 157:495–499.

10.

Kapasi

Burton

Shultz

Tew

Szakal

(1993) Induction of functional follicular dendritic cell development in severe combined immunodeficiency mice. Influence of B and T cells. J Immunol 150:2648–2658.

11.

Karlsson

Guinamard

Bolland

Sankala

Steinman

Ravetch

(2003) Macrophages control the retention and trafficking of B lymphocytes in the splenic marginal zone. J Exp Med 198:333–340.

12.

Köhler

Milstein

(1975) Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256:495–497.

13.

Kraal

(1992) Cells in the marginal zone of the spleen. Int Rev Cytol 132:31–74.

14.

Kraal

Schornagel

Streeter

Holzmann

Butcher

(1995) Expression of the mucosal vascular addressin, MAd-CAM-1, on sinus-lining cells in the spleen. Am J Pathol 147:763–771.

15.

Liakka

Autio-Harmainen

(1992) Distribution of the extracellular matrix proteins tenascin, fibronectin, and vitronectin in fetal, infant, and adult human spleens. J Histochem Cytochem 40:1203–1210.

16.

Mackay

Majeau

Lawton

Hochman

Browning

(1997) Lymphotoxin but not tumor necrosis factor functions to maintain splenic architecture and humoral responsiveness in adult mice. Eur J Immunol 27:2033–2042.

17.

MacLennan

Toellner

Cunningham

Serre

Sze

Zuniga

Cook

(2003) Extrafollicular antibody responses. Immunol Rev 194:8–18.

18.

Martin

Kearney

(2002) Marginal-zone B cells. Nature Rev Immunol 2:323–335.

19.

McNagny

Bucy

Cooper

(1991) Reticular cells in peripheral lymphoid tissues express the phosphatidylinositol-linked BP-3 antigen. Eur J Immunol 21:509–515.

20.

Mebius

(2003) Organogenesis of lymphoid tissues. Nature Rev Immunol 4:292–303.

21.

Ngo

Cornall

Cyster

(2001) Splenic T zone development is B cell dependent. J Exp Med 194:1649–1660.

22.

Ngo

Korner

Gunn

Schmidt

Riminton

Cooper

Browning

(1999) Lymphotoxin alpha/beta and tumor necrosis factor are required for stromal cell expression of homing chemokines in B and T cell areas of the spleen. J Exp Med 189:403–412.

23.

Nolte

Belien

Schadee-Eestermans

Jansen

Unger

van Rooijen

Kraal

(2003) A conduit system distributes chemokines and small blood-borne molecules through the splenic white pulp. J Exp Med 198:505–512.

24.

Nolte

Hoen

van Stijn

Kraal

Mebius

(2000) Isolation of the intact white pulp. Quantitative and qualitative analysis of the cellular composition of the splenic compartments. Eur J Immunol 30:626–634.

25.

Ocklind

Talts

Fassler

Mattsson

Ekblom

(1993) Expression of tenascin in developing and adult mouse lymphoid organs. J Histochem Cytochem 41:1163–1169.

26.

Skibinski

Skibinska

Stewart

James

(1998) Enhancement of terminal B lymphocyte differentiation in vitro by fibro-blast-like stromal cells from human spleen. Eur J Immunol 28:3940–3948.

27.

Steiniger

Barth

Hellinger

(2001) The perifollicular and marginal zones of the human splenic white pulp: do fibroblasts guide lymphocyte immigration?. Am J Pathol 159:501–512.

28.

Taylor

Pickering

Kosco-Vilbois

Walport

Botto

Gordon

Martinez-Pomares

(2002) The follicular dendritic cell restricted epitope, FDC-M2, is complement C4; localization of immune complexes in mouse tissues. Eur J Immunol 32:1888–1896.

29.

ten Dam

Hafmans

Veerkamp

van Kuppevelt

(2003) Differential expression of heparan sulfate domains in rat spleen. J Histochem Cytochem 51:727–739.

30.

Tumanov

Grivennikov

Shakhov

Rybtsov

Koroleva

Takeda

Nedospasov

(2003) Dissecting the role of lymphotoxin in lymphoid organs by conditional targeting. Immunol Rev 195:106–116.

31.

Van Vliet

Melis

Foidart

Van Ewijk

(1986) Reticular fibroblasts in peripheral lymphoid organs identified by a monoclonal antibody. J Histochem Cytochem 34:883–890.

32.

Wang

Sun

(2001) Complementary effects of TNF and lymphotoxin on the formation of germinal center and follicular dendritic cells. J Immunol 166:330–337.

33.

Yoshida

Kaji

Takahashi

van den Berg

Dijkstra

(1995) Host origin of follicular dendritic cells induced in the spleen of SCID mice after transfer of allogeneic lymphocytes. Immunology 84:117–126.

34.

Yoshida

Matsuura

Tamahashi

Takahashi

(1993) Development of antigenic heterogeneity in the splenic meshwork of severe combined immunodeficient (SCID) mice after reconstitution with T and B lymphocytes. Cell Tissue Res 272:1–10.

35.

Yoshida

Tamahashi

Matsuura

Takahashi

Tachibana

(1991) Antigenic heterogeneity of the reticular meshwork in the white pulp of mouse spleen. Cell Tissue Res 266:223–229.