ARTICLE

The Three-dimensional Structure of Human Splenic White Pulp Compartments

Correspondence to: Birte Steiniger, Inst. of Anatomy and Cell Biology, Robert-Koch-Str. 6, D-35033 Marburg, Germany. E-mail: steinigb@mailer.uni-marburg.de

	Summary

Top Summary Introduction Materials and Methods Results Discussion Literature Cited

The precise arrangement of B- and T-lymphocytes in the different compartments of the human splenic white pulp is still largely unknown. We therefore performed a 3D reconstruction of 150 serial sections of a representative adult human spleen alternately stained for CD3 and CD20. The results indicate that the T-cell regions of human spleens may be interrupted by B-cell follicles. Therefore, there is no continuous periarteriolar lymphatic T-cell sheath (PALS) around white pulp arterioles. An arteriole may be surrounded by T-lymphocytes at one level, then run across a follicle without any T-cells around, and finally re-enter a T-cell region. T- and B-cell compartments are intricately interdigitated in the human splenic white pulp. CD4⁺ T-lymphocytes and the typical fibroblasts of the T-cell region may extend as a thin shell at the follicular surface within the marginal zone. On the other hand, IgD⁺⁺ B-cells continue from the follicular outer marginal zone along the surface of the T-cell region. Our findings indicate that the microanatomy of the splenic white pulp differs between humans and rodents. This may have consequences for the immigration of recirculating lymphocytes and for initial interactions among antigen-specific T- and B-lymphocytes. (J Histochem Cytochem 51:655–663, 2003)

Key Words: human spleen, white pulp, T-cell regions, follicles, marginal zone, CD3, CD20, CCL21

	Introduction

Top Summary Introduction Materials and Methods Results Discussion Literature Cited

IN CONTRAST TO mice and rats, the compartments of the splenic white pulp are ill-defined in humans. The location of human T-cell regions with respect to the so-called "central arterioles" is especially controversial. In addition, the arrangement of B-cell compartments, especially the location of the marginal zone (MZ), is insufficiently described in humans. Because MZ B-cells are believed to decisively contribute to the sepsis-protective effect of the spleen (Amlot and Hayes 1985 ), there is a need for better understanding of the composition of the human splenic white pulp.

The nomenclature for some parts of the splenic white pulp vasculature was coined in rats (Snook 1950 , Snook 1964 ) and the structure of the white pulp has been thoroughly described in this species (Veerman and van Ewijk 1975 ; Dijkstra et al. 1985a ; Van Rooijen et al. 1989 ; Kraal 1992 ). In rats, small branches of the splenic artery, i.e., the central arterioles, represent the innermost structures around which lymphocytes aggregate. These arterioles are concentrically surrounded by a T-cell-rich region, the periarteriolar lymphatic sheath (PALS). At regular intervals, accumulations of recirculating B-cells, the follicles, are attached to the surface of the PALS. PALS and follicles are covered by a broad MZ, which delimits both regions from the splenic red pulp. In rats it is still controversial whether the majority of lymphocytes in the MZ are memory-type or preactivated polyreactive B-cells (Liu et al. 1988 , Liu et al. 1991 ; Dammers et al. 2000 ). The MZ is divided from the PALS and follicles by a pale-staining borderline, which represents the blood-filled marginal sinus and the accompanying marginal metallophilic macrophages (Dijkstra et al. 1985b ). Recirculating B- and T-lymphocytes are presumed to enter the splenic white pulp via this marginal sinus and subsequently via the MZ (Nieuwenhuis and Ford 1976 ).

We have previously shown that the splenic white pulp of humans differs from that of rats with respect to several microanatomic (especially microvascular) features (Steiniger et al. 1997 , Steiniger et al. 2001 ; Steiniger and Barth 2000 ). Humans exhibit an additional region surrounding the follicles, the perifollicular zone, which harbors a part of the open splenic circulation and the so-called sheathed capillaries. In contrast to rats, a marginal sinus and marginal metallophilic macrophages do not occur. In humans, B-lymphocytes of the typical MZ phenotype (IgM⁺IgD^{- or ±}) predominantly surround the follicles and are almost absent from the surface of the T-cell regions. MZ phenotype B-cells form the broad inner part of the follicular MZ (iMZ), and there is an additional smaller outer follicular MZ (oMZ) composed of B-cells with a recirculating phenotype (IgM⁺IgD⁺⁺). These B-cells may continue at the surface of the T-cell regions. Interestingly, at the border between the follicular iMZ and oMZ, shell-like accumulations of CD4⁺ T-cells can be found. These T-cells closely accompany fibroblasts of a peculiar phenotype, which also form a meshwork in the entire T-cell region. The fibroblasts express smooth muscle -actin and myosin, MAdCAM-1, VCAM-1, and VAP-1, as well as thrombomodulin, cytokeratins 8 and 18, and Thy-1 (Steiniger et al. 2001 ). Therefore, follicles of the human splenic white pulp appear to be embedded in a kind of extension of the T-cell region into the MZ. This phenomenon has obviously led to the erroneous description of the human MZ as a T-cell compartment in previous publications (Hsu et al. 1983 ; Hsu 1985 ).

To more precisely define the T- and B-cell regions of the human splenic white pulp and their relationship to arterioles, we have performed a 3D reconstruction of 150 serial sections of a representative adult human spleen alternately stained for CD3 and CD20. The specimen was carefully selected from a panel of 73 human spleens primarily derived from traffic accident victims of different ages, which had been thoroughly characterized for the distribution of B- and T-lymphocytes, macrophages, and fibroblasts (Steiniger et al. 2001 ). Selection of the specimen is crucial, because it is well known that, in contrast to laboratory rodents, certain aspects of splenic microanatomy are individually variable in humans (Van Krieken et al. 1985 ; Van Krieken and te Velde 1988 ). This variability probably represents the state of the individual with respect to novel antigens reaching the blood. Therefore, the overall amount of splenic white pulp, the frequency of follicles, and the occurrence of secondary follicles may differ among human specimens. In addition to the individual clinical condition and the consequences of emergency treatment, age may also contribute to alterations of white pulp morphology. Full-blown secondary follicles are more likely to occur in children whereas young adults often exhibit secondary follicles with decaying germinal centers. In most older individuals, only primary follicles are encountered in the splenic white pulp.

The central question to be answered by our investigation concerned the relationship between white pulp arterioles and follicles. We had frequently observed rather large arterioles without any accompanying T-cells within follicles, and even within germinal centers of the human splenic white pulp. Therefore, we wished to clarify whether these intrafollicular vessels represented genuine "central arterioles" that had lost their T-cell sheaths or whether they should be regarded as side-branches of central arterioles. Together with our previous findings (Steiniger et al. 2001 ), our present results demonstrate that, in humans, the regions of T- or B-cell predominance are less continuous than in the white pulp of rodents. The T-cell regions are interrupted by follicles and most of the MZ is absent from the T-cell regions. In humans, central arterioles may lose their T-cell sheaths, cross a follicle, and regain a T-cell sheath afterwards. Neither the term "periarteriolar lymphatic sheath" nor "central arteriole" is really appropriate to describe the relation of T-cell-rich regions and arterioles in the human splenic white pulp.

	Materials and Methods

Top Summary Introduction Materials and Methods Results Discussion Literature Cited

Processing of Samples
A representative spleen was chosen from a pool of 73 organs that had been thoroughly investigated by immunohistology (Steiniger et al. 2001 ). The organ was obtained from a 21-year-old woman with blunt abdominal trauma, splenic rupture, and no additional clinical diagnoses. Samples of 1 cm³ were fixed in 4% formol/H₂O for 24 hr and embedded in paraffin. To alleviate orientation of the sections for 3D reconstruction, a corner of one embedded tissue block was cut to form a right angle. The paraffin was then removed and the specimen was re-embedded for sectioning.

Immunohistology
Demonstration of CD3 and CD20 in Serial Sections. A series of 150 sections was prepared on silanized slides and alternately stained for CD3 (polyclonal, DAKO; Hamburg, Germany, no. A 0452) and CD20 (L26, DAKO no. M 0755). In brief, after deparaffinization, formol pigment was removed with picric acid and the sections were preincubated with 0.05% protease type XIV (Sigma, Deisenhofen, Germany; no. P-5147) in TBS for 15 min at RT. After washing, sections were immersed in 0.15% H₂O₂ in PBS for 30 min at RT to block endogenous peroxidase. After an additional wash, anti-CD3 was applied to the sections at 1:80 and L26 at 1:150 in PBS with 1% BSA, 0.1%NaN₃ (PBS/BSA/NaN₃), and 3 µg/ml avidin for 60 min at RT. For detection of anti-CD3, a secondary biotinylated goat anti-rabbit IgG (Vector Labs no. BA-1000; via Alexis, Grünberg, Germany) was used at 1:200. For L26, a biotinylated goat anti-mouse IgG (Vector Labs no. PK-6102) was used at the same dilution. The secondary reagents were diluted in PBS with 5% normal rat serum and 20 µg/ml biotin and applied for 30 min at RT. Finally, avidin–biotinylated peroxidase complexes (ABC; Vectastain Elite ABC Kit, Vector Labs no. PK 6102) were prepared according to the manufacturer's recommendation and reacted with the sections for 30 min at RT. After washing in TBS, the peroxidase complexes were revealed by diaminobenzidine and the sections were lightly counterstained with Mayer's hemalum. The first sections of the series served as controls and were incubated with omission of the primary antibody. Normal serum of the respective species in place of primary antibodies did not produce background staining.

Demonstration of CCL21 (SLC) and Smooth Muscle -actin. Rabbit anti-human CCL21 (Exodus-2, SLC) affinity-purified antibodies were obtained from Preprotech (No. 500-P109) via TEBU (Frankfurt, Germany). The antibodies were diluted 1:1000 in PBS/BSA/NaN₃ and applied for 20 hr at 4C to paraffin sections that had been autoclaved for 20 min in H₂O/EDTA, pH 8.0, for antigen retrieval. MAb asm-1 against smooth muscle -actin was obtained from Progen (Heidelberg, Germany; no. 61001) and used at 1:200. Before incubation with this antibody, paraffin sections were autoclaved in citrate buffer, pH 6.0. Demonstration of bound antibody was as described above, using anti-rabbit or anti-mouse ABC. Omission of primary antibody and application of normal rabbit or mouse serum were used as controls.

Image Acquisition and 3D Reconstruction
The right angle cut at one corner of the sections was used for overall orientation. A representative area close to this landmark, containing three follicles, was chosen and photographed with a digital video camera using a x4 lens with an x1.25 optovar. The pictures were superimposed with the help of the Openlab program (Version 1.7.8; Improvision, Heidelberg, Germany) and the sectioned arterioles were used for orientation. By means of this program, the sections could only be moved in x- and y-directions without rotation. T- and B-cell regions and vessels were delineated and imported into two new files. These were used for 3D reconstruction with 3D-Doctor (Version 2.0 for demonstration; Able Software, Lexington, KY). Areas of interest were delineated and visualized by "complex surface rendering" and "full rendering" to view surface structures of compartments. For transparent pictures demonstrating branching of the arterioles inside the respective compartment, "surface-rendering" and "transparent object-setting" were chosen.

Four informative levels were defined and represented in a different color for direct comparison with the immunostained sections. For improved visualization, 3D images of the computer-generated reconstructions were drawn by hand.

	Results

Top Summary Introduction Materials and Methods Results Discussion Literature Cited

Arrangement of Arterioles in Relation to B- and T-cell Regions of the White Pulp
Two and a half follicles, designated I to III, were reconstructed from serial sections stained for CD20 using the L26 antibody (Fig 1A and Fig 1B). For direct comparison with the immunohistological staining results, sections are highlighted in yellow or green color at four different levels (Fig 1 and Fig 2). Surface rendering includes the most peripheral immunoreactive lymphocytes of the white pulp. Thus, Ia in Fig 1A and Fig 1B indicates a T-cell region covered by a thin continuous layer of L26-positive B-cells, which continue into the oMZ of the adjacent follicles. The lower part of this T-cell region is still observed as a hollow area at level 2 in Fig 1D and Fig 1F. A similar phenomenon occurs in the vicinity of follicle III (Fig 1D, Fig 1F, Fig 2D, Fig 2F, Fig 3G, and Fig 3H). L26 reacts with all B-cells, and therefore, germinal center, mantle zone, and MZ are not distinguished after immunostaining (Fig 3C and Fig 3E). However, these internal regions do show up in sections stained for CD3 because of the nuclear counterstaining in the CD3^- mantle zones and the CD3⁺ T-cells in the germinal centers (Fig 3D and Fig 3F).

View larger version (59K): [in this window] [in a new window]   Figure 1. Three-dimensional structure of B-cell regions and arterioles of the human splenic white pulp, visualized by MAb L26 against CD20. I, II and III indicate follicles. Ia is a T-cell region covered by a thin layer of B-cells. 1 and 2 represent two arterioles entering the specimen from above. Further branches of these vessels are numbered in hierarchical order. Levels 1 to 4 correspond to Fig 3A, Fig 3C, Fig 3E, and Fig 3G. (A,B) Surface view before (A) and after (B) rotation. (C,D) Transparent rendering before (C) and after (D) rotation. (E,F) Optimized drawings of surface (E) and transparent (F) rendering. Insets in B and C represent directions of rotation. Inset in C is a magnification of the branching point of arteriole 2.1. Final magnifications x20.

View larger version (54K): [in this window] [in a new window]   Figure 2. Three-dimensional structure of T-cell regions and arterioles of the human splenic white pulp, visualized by antibodies against the CD3 -chain. 1 and 2 represent two arterioles entering the specimen from above. Further branches of these vessels are numbered in hierarchical order. Levels 1 to 4 correspond to Fig 3B, Fig 3D, Fig 3F, and Fig 3H. (A,B) Surface view before (A) and after (B) rotation. (C,D) Transparent rendering before (C) and after (D) rotation. (E,F) Optimized drawings of surface (E) and transparent (F) rendering. Insets in A,B and D represent directions of rotation and/or a reduced version of Fig 1A, Fig 1B, or 1D for comparison to B-cell regions. Inset in C is a magnification of the branching point of arteriole 2.1. Final magnification x20.

View larger version (160K): [in this window] [in a new window]   Figure 3. Consecutive serial sections stained for B-cells (CD20) and T-cells (CD3) at the four selected levels indicated in Fig 1 and Fig 2. The numbering of arterioles and their branches is identical to Fig 1 and Fig 2. (A,C,E,G) distribution of CD20; (B,D,F,H) distribution of CD3; (A,B) level 1; (C,D) level 2; (E,F) level 3; (G,H) level 4. Within A–H the letters in white circles represent: A, follicle I; B, area of mixed T and B cellularity; C, follicle II; D, T-cells in the MZ of follicle I; E, T-cell region near follicle II; F, follicle III; G, T-cell region near follicle III. Final magnification x57.

Only arterioles in the white pulp, i.e., vessels with smooth muscle cells in their walls, were included for reconstruction. Two larger arterioles, named 1 and 2, and their branches are represented in blue color in Fig 1 and Fig 2. They form the innermost structures of the reconstructed white pulp area. The specimen was orientated such that these vessels enter from above and branch into finer arterioles towards the bottom of the reconstruction. Successive branches of the arterioles are designated in numerical order. The blood flow in the arterioles is from level 1 to level 4. To permit a better integration of 2D and 3D information and to prevent superpositioning of the vessels, the reconstructed part of the white pulp is also depicted after slight rotation. In the "normal" position, the lower margin of the immunostained areas in Fig 3 represents the anterior aspect of the white pulp in Fig 1A, Fig 1C, Fig 2A, and Fig 2C). From this position the specimen is rotated 15° to the right and 25° anteriorly, as indicated in Fig 1B, Fig 1D, Fig 2B, and Fig 2D.

Slight distortions of the single sections during the cutting process, the fact that only every second section was represented, and the inability of rotating the sections with the Openlab program led to an irregular outline of the vessels in Fig 1C, Fig 1D, Fig 2C, and Fig 2D. In addition, the four highlighted section levels were difficult to visualize after transparent rendering. To overcome these restrictions, an additional optimized version of the reconstruction was drawn by hand (Fig 1E, Fig 1F, Fig 2E, and Fig 2F).

The reconstruction revealed the following course of white pulp arterioles. The feeding arterioles 1 and 2 are connected by an anastomosis at level 1. Arteriole 1 gives off two branches within follicle I, designated 1.1 and 1.2 in Fig 1 Fig 2 Fig 3. Arteriole 2 splits into three branches, 2.1, 2.2, and 2.3, at the upper pole of follicle II. Arterioles 2.1 and 2.2 divide into three further branches at the border between white and red pulp, while arteriole 2.3 gives off branch 2.3.1 within follicle II and then continues in direction to follicle III. Therefore, there is no continuous single "central arteriole," but arterioles regularly branch into vessels of higher order, especially in white pulp follicles.

The location of the arterioles with respect to B- and T-cell regions is also evident from Fig 1 Fig 2 Fig 3. At levels 1 and 2 (Fig 1F, Fig 2F, and Fig 3A–3D), the vessels are covered by mixtures of B- and T-cells. Between both levels the vessels have, however, passed a T-cell-rich region, which is still partially visible at level 2 (Fig 1F, Fig 3C, and Fig 3D). Mixtures of T- and B-cells also occur around arterioles of higher order, especially at the surface of follicles, e.g., around arterioles 2.1, 2.1.1 and 2.1.2 (Fig 3C and Fig 3D). The course of arteriole 2.3 is most informative. This vessel is still located within the T-cell region at its origin above level 2. It then passes into a small region of mixed cellularity and enters follicle II, where it is accompanied only by B-cells, as observed at level 3 (Fig 1F, Fig 2A–2F, Fig 3E, and Fig 3F). At this level, arteriole 2.3 runs through the mantle zone of follicle II (Fig 3 F). It then leaves follicle II and again enters a T-cell region close to follicle III at level 4, where it occupies an eccentric position (Fig 2E, Fig 2F, Fig 3G, and Fig 3H). Because there is no other arteriole of this size associated with the white pulp at both levels 3 and 4, this finding indicates that "central arterioles" may run through a T-cell region, then leave this region, cross a follicle, and finally re-enter the next T-cell region. In other words, T-cell regions may be interrupted by follicles and then reappear again, so that a continuous PALS does not exist. In addition, most arterioles do not occupy a central position in the T-cell region or in the follicle.

Superficial Interdigitation of B- and T-cell Regions and Distribution of CCL21
The 3D reconstruction also demonstrated an intricate interdigitation of B- and T-cells at the surface of the respective regions. Previous investigations (Steiniger et al. 2001 ) had shown that, in many specimens, a thin row of B-cells, a part of which were IgD⁺⁺, extended from one follicular oMZ along the surface of a T-cell region to the next oMZ. This phenomenon is clearly visible in Fig 3G. In addition, CD4⁺ T-cells formed shell-like accumulations within the follicular MZ, which were most likely located at the border between iMZ and oMZ (Steiniger and Barth 2000 ; Steiniger et al. 2001 ). These MZ T-cells are also observed in Fig 2D, Fig 2F, Fig 3D, and Fig 3F). The density of the MZ T-cell shell is variable within one and the same follicle, but it also varies among different follicles in the same organ (Fig 3D and Fig 3F). At the poles of the follicles, these T-cells are difficult to visualize (Fig 3H). Obviously, tangential sectioning makes the cells appear more diffusely distributed. Interestingly, the CD4⁺ T-cells were associated with a special type of fibroblast expressing smooth muscle -actin in the MZ and also in the entire T-cell region (Steiniger et al. 2001 ). Further immunohistological characterization of these fibroblasts now indicates that they may be also positive for CCL21 (SLC) both in the MZ and in the T-cell region (Fig 4A and Fig 4B). However, the intensity of staining is variable in different organs. Surprisingly, the most strongly CCL21-positive structure in human spleens is the endothelium of the arterioles in T-cell regions (Fig 4A). In contrast, arteriolar endothelium in follicles and elsewhere in the spleen is always CCL21-negative.

View larger version (180K): [in this window] [in a new window]   Figure 4. Distribution of (A) CCL21 (SLC) and (B) smooth muscle -actin in human splenic white pulp fibroblasts. (A) Specimen from a 6-year-old girl after blunt abdominal trauma. Final magnifications: A x63; B x100.

	Discussion

Top Summary Introduction Materials and Methods Results Discussion Literature Cited

The precise localization of T- and B-cell regions in human spleens can be revealed only by 3D imaging. Because the diameter of a follicle is on the order of about 1 mm, there is presently no alternative to reconstruction of immunostained serial sections. Immunostaining for light microscopy gives superior morphological information and permanent specimens. This method, however, requires that primary 2D non-digital information must be transformed into digital data for further processing. Therefore, certain limitations cannot be avoided, which are discussed below.

Our reconstruction of two and a half human splenic follicles associated with two larger T-cell regions reveals fundamental differences to splenic white pulp morphology described in rodents: Human T-cell regions do not represent continuous PALS but instead are elliptical elongated structures, which are interrupted by follicles. However, a thin extension of the outer T-cell region, composed of typical fibroblasts and CD3⁺CD4⁺ T-cells, surrounds the follicles within the MZ. The fact that these fibroblasts probably contain CCL21, which is a chemokine with a preferential effect on naive CD4⁺ T-lymphocytes (Gunn et al. 1998 ), leads to the speculation that they might guide immigrating CD4⁺ T-cells into the white pulp after the T-cells have arrived via the open perifollicular circulation (Steiniger et al. 2001 ). Proof of this is, however, still lacking. Although the pattern we show for CCL21 in the T-cell areas corresponds to that found in mouse spleens by in situ hybridization (Gunn et al. 1998 ), it remains to be shown that CCL21 is endogenously produced. This is especially relevant with respect to the intense staining in arteriolar endothelium observed only in T-cell areas. In addition, the identity of the cells positive for smooth muscle -actin and CCL21 needs to be shown by double staining. However, this is difficult because of the different methods of antigen retrieval necessary for the two primary reagents used. Up to now, we have not found phenotypical differences between CD4⁺ MZ T-cells and the CD4⁺ T-cells in the T-cell areas.

The shape of the T-cell accumulations in the follicular MZ represented in Fig 2 is somewhat astonishing. The T-cells do not appear to form a closed shell around follicle I, but instead appear as ring-like structures visible only around the equator of the follicle. This phenomenon is most likely caused by omission of every second section and by tangential and even horizontal sectioning through the T-cell shell in the upper and lower parts of the follicle. At such low angles, band-like accumulations of T-cells can no longer be distinguished because the cells are increasingly dispersed in the plane of cutting. Rings of T-cells are also present in the MZ of follicle II above level 3, but their number is somewhat reduced in comparison to follicle I. Therefore, in Fig 2 MZ T-cells are only visualized where they are found most strongly accumulated. In the upper part of follicle III, MZ T-cells were almost indistinguishable from the red pulp. If follicle III is larger than follicles I and II, this may be also due to primarily tangential sectioning in the upper part of this follicle.

The interdigitation of T- and B-cells at the surface of the follicles and, respectively, of the T-cell regions may provide a means of alleviating cellular interactions among both cell types early after immigration into the white pulp. From previous investigations (Steiniger et al. 2000, Steiniger et al. 2001 ), we know that the B-lymphocytes located at the surface of the T-cell regions at least partially exhibit the IgD⁺⁺ phenotype of small recirculating B-cells. Therefore, they may correspond to the B-lymphocytes located in the outer PALS in rats and mice (Gulbranson-Judge and MacLennan 1996 ). The additional presence of IgD⁺⁺ B-cells in a special compartment of the follicular MZ has not been described in rodents and is obviously peculiar to humans.

Periarteriolar areas of mixed T–B cellularity are also unknown in rodents. We are convinced that this phenomenon is due to the simultaneous presence of T- and B-cells and not to T-cell crossreactivity of the CD20 reagent used. It has been described that CD20 may be present in T-cell neoplasias (Quintanilla-Martinez et al. 1994 ) and in a small number of normal peripheral blood T-cells (Hultin et al. 1993 ; Storie et al. 1995 ). Although we cannot exclude single double-positive T-cells, there is no indication that CD20 expression may occur in normal T-cells to such an extent that it could explain our findings. Areas of mixed cellularity were also found in the majority of other spleens investigated using different reagents, e.g., antibodies directed to CD79a or to IgD.

The reconstruction also reveals fundamental differences among rodents and humans with respect to the localization and branching pattern of white pulp arterioles. In humans, the same vessel may run through T-cell regions and then through follicles. Most arterioles branch dichotomously at the follicular surface, or even within a follicle. The arterioles do not occupy a central position in the T-cell areas or follicles. Therefore, the present nomenclature for splenic arterioles is not appropriate to describe the microanatomy of the human splenic white pulp.

Our findings can be generalized, although the extent of the white pulp is clearly variable in humans. Up to now, we have thoroughly investigated 73 spleens for most of the immunologically relevant cell types (Steiniger et al. 2001 ), describing the invariant immunohistological features. Therefore, we can be sure that the chosen organ is really representative of young adults. In addition, our results are in accordance with the majority of immunohistological studies published on human spleens thus far (Grogan et al. 1983 , Grogan et al. 1984 ; Tanaka et al. 1984 ; Timens and Poppema 1985 ; Van Krieken and te Velde 1986 ; Buckley et al. 1987 ; Timens et al. 1989 ; Buckley 1991 ). The histological intricacies of human B- and T-cell regions went, however, undetected in these publications.

The need to also perform drawings of the 3D computer-generated images indicates that the distortions caused by cutting and mounting of paraffin sections and the incongruencies introduced by omitting every second section could not be remedied by the versions of the computer programs applied for reconstruction. In addition, the fully transparent and 3D impression is perceived only if the image is freely rotated on the computer display. Therefore, we had to resort to hand-made drawings to integrate all information. To exclude misinterpretation of the data, we defined four levels in the reconstructed image, which can be directly compared to the original 2D immunostained sections shown in Fig 3.

Further research should be centered on elucidating the reasons for the peculiar arrangement of T- and B-cell regions in human spleens. In vitro studies of isolated fibroblasts, T-cells, and antigen-presenting cells may clarify how fibroblasts contribute to specific T-cell responses. Microdissection of immunostained B-cells from the surface of the T-cell region and from the follicular oMZ and iMZ may contribute to verifying these cells as recirculating B-cells and MZ B-memory cells, respectively.

	Acknowledgments

Immunostaining and 3D reconstruction were performed by L. Rüttinger as part of his M.D. thesis. We thank C. Fiebiger for precise observations and the talented artwork optimizing the 3D reconstructions. B. Herbst and A. Seiler provided excellent technical assistance. Thanks are due to Dr T. Hurek (Max-Planck Institute for Terrestrial Microbiology; Marburg, Germany) for support and discussions during data acquisition for 3D reconstruction and for the permission to use the Openlab program. Dr K. Troidl (Hochschulrechenzentrum of Marburg University) kindly provided assistance with scanning and processing of the photographs. Dr M. Bette and O. Stehling (Institute of Anatomy and Cell Biology, University of Marburg, Germany) also spent much time on adapting the digital images to a format suitable for publication.

Received for publication September 24, 2002; accepted December 18, 2002.

	Literature Cited

Top Summary Introduction Materials and Methods Results Discussion Literature Cited

Amlot PL, Hayes AE (1985) Impaired human antibody response to the thymus-independent antigen, DNP-Ficoll, after splenectomy. Lancet I: 1008–1011

Buckley PJ (1991) Phenotypic subpopulations of macrophages and dendritic cells in human spleen. Scan Microsc 5:147-158

Buckley PJ, Smith MR, Braverman MF, Dickson SA (1987) Human spleen contains phenotypic subsets of macrophages and dendritic cells that occupy discrete microanatomic locations. Am J Pathol 128:505-520[Abstract]

Dammers PM, Visser A, Popa ER, Nieuwenhuis P, Kroese FGM (2000) Most marginal zone B cells in rat express germline encoded Ig VH genes and are ligand selected. J Immunol 165:6156-6169[Abstract/Free Full Text]

Dijkstra CD, Döpp EA, Joling P, Kraal G (1985a) The heterogeneity of mononuclear phagocytes in lymphoid organs: distinct macrophage subpopulations in the rat recognized by monoclonal antibodies. Immunology 54:589-599[Medline]

Dijkstra CD, van Vliet E, Döpp EA, van der Lelij AA, Kraal G (1985b) Marginal zone macrophages identified by a monoclonal antibody: characterization of immuno- and enzyme-histochemical properties and functional capacities. Immunolgy 55:23-30[Medline]

Grogan TM, Jolley CS, Rangel CS (1983) Immunoarchitecture of the human spleen. Lymphology 16:72-82[Medline]

Grogan TM, Rangel CS, Richter LC, Wirt DP, Villar HV (1984) Further delineation of the immuno-architecture of the human spleen. Lymphology 17:61-68[Medline]

Gulbranson–Judge A, MacLennan ICM (1996) Sequential antigen-specific growth of T cells in the T zones and follicles in response to pigeon cytochrome c. Eur J Immunol 26:1830-1837[Medline]

Gunn MD, Tangemann K, Tam C, Cyster JG, Rosen SD, Williams LT (1998) A chemokine expressed in lymphoid high endothelial venules promotes the adhesion and chemotaxis of naive T lymphocytes. Proc Natl Acad Sci USA 95:258-263[Abstract/Free Full Text]

Hsu S-M (1985) Phenotypic expression of B lymphocytes. III. Marginal zone B cells in the spleen are characterized by the expression of Tac and alkaline phosphatase. J Immunol 135:123-130[Abstract]

Hsu S-M, Cossman J, Jaffe ES (1983) Lymphocyte subsets in normal human lymphoid tissues. Am J Clin Pathol 80:21-30[Medline]

Hultin LE, Hausner MA, Hultin PM, Giorgi JV (1993) CD20 (pan-B-cell) antigen is expressed at a low level on a subpopulation of human T lymphocytes. Cytometry 14:196-204[Medline]

Kraal G (1992) Cells in the marginal zone of the spleen. Int Rev Cytol 132:31-74[Medline]

Liu Y-J, Oldfield S, MacLennan ICM (1988) Memory B cells in T cell-dependent antibody responses colonize the splenic marginal zones. Eur J Immunol 18:355-362[Medline]

Liu Y-J, Zhang J, Lane PJL, Chan EY-T, MacLennan ICM (1991) Sites of specific B cell activation in primary and secondary responses to T cell-dependent and T cell-independent antigens. Eur J Immunol 21:2951-2962[Medline]

Nieuwenhuis P, Ford WL (1976) Comparative migration of B- and T-lymphocytes in the rat spleen and lymph nodes. Cell Immunol 23:254-267[Medline]

Quintanilla–Martinez L, Preffer F, Rubin D, Ferry JA, Harris NL (1994) CD20⁺ T-cell lymphoma. Neoplastic transformation of a normal T-cell subset. Am J Clin Pathol 102:483-489[Medline]

Snook T (1950) A comparative study of the vascular arrangements in mammalian spleens. Am J Anat 87:31-77

Snook T (1964) Studies on the perifollicular region of the rat's spleen. Anat Rec 1148:149-159

Steiniger B, Barth P (2000) Microanatomy and function of the spleen. Adv Anat Embryol 151:1-100

Steiniger B, Barth P, Herbst B, Hartnell A, Crocker PR (1997) The species-specific structure of microanatomical compartments in the human spleen: strongly sialoadhesin-positive macrophages occur in the perifollicular zone, but not in the marginal zone. Immunology 92:307-316[Medline]

Steiniger B, Barth P, Hellinger A (2001) The perifollicular and marginal zones of the human splenic white pulp: do fibroblasts guide lymphocyte immigration? Am J Pathol 159:501-512[Abstract/Free Full Text]

Storie I, Wilson GA, Granger V, Barnett D, Reilly JT (1995) Circulating CD20^dim T-lymphocytes increase with age: evidence for a memory cytotoxic phenotype. Clin Lab Haematol 17:323-328[Medline]

Tanaka H, Takasaki S, Sakata A, Muroya T, Suzuki T, Ishikawa E (1984) Lymphocyte subsets in the white pulp of human spleen in normal and diseased cases. Acta Pathol Jpn 34:251-270[Medline]

Timens W, Boes A, Poppema S (1989) Human marginal zone B cells are not an activated B cell subset: strong expression of CD21 as a putative mediator for rapid B cell activation. Eur J Immunol 19:2163-2166[Medline]

Timens W, Poppema S (1985) Lymphocyte compartments in human spleen: an immunohistologic study in normal spleens and noninvolved spleens in Hodgkin's disease. Am J Pathol 120:443-454[Abstract]

Van Krieken JHJM, te Velde J (1986) Immunohistology of the human spleen: an inventory of the localization of lymphocyte subpopulations. Histopathology 10:285-294[Medline]

Van Krieken JHJM, te Velde J (1988) Normal histology of the human spleen. Am J Surg Pathol 12:777-785[Medline]

Van Krieken JHJM, Te Velde J, Kleiverda K, Leenheers–Binnendijk L, Van de Velde CJH (1985) The human spleen; a histological study in splenectomy specimens embedded in methylmethacrylate. Histopathology 9:571-585[Medline]

Van Rooijen N, Claassen E, Kraal G, Dijkstra CD (1989) Cytological basis of immune functions of the spleen. Immunocytochemical characterization of lymphoid and non-lymphoid cells involved in the ‘in situ’ immune response. Prog Histochem Cytochem 19:1-69

Veerman AJP, van Ewijk W (1975) White pulp compartments in the spleen of rats and mice. A light and electron microscopic study of lymphoid and non-lymphoid cell types in T- and B-areas. Cell Tissue Res 156:417-441[Medline]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Steiniger, B. Articles by Barth, P. J. Search for Related Content PubMed PubMed Citation Articles by Steiniger, B. Articles by Barth, P. J. Social Bookmarking                      What's this?

Guidelines | Subscriptions | About | exPRESS - Current - Archive | Business Information | Contact