¹Regenerative Medicine Institute and Department of Physiology, National Centre for Biomedical Science Engineering, National University of Ireland, Galway, Ireland. E-mail: rod.ceredig@nuigalway.ie

Lymphoid organs are frequently classified as 'primary', 'secondary' or 'tertiary'. Primary lymphoid organs, namely bone marrow and thymus, are the generators of B and T cells, respectively. Secondary lymphoid organs are formed during development at distinct anatomical sites, whereas tertiary lymphoid organs are induced structures, frequently forming at sites of chronic inflammation. It is becoming increasingly evident that so-called lymphoid tissue inducer (LTi) cells play an important role in the organization of all lymphoid structures.¹ This initial classification of lymphoid organs also implied that primary lymphoid organs were exclusively producers of cells—that is, the cells would only leave and not re-enter. In the classic experiments of Gowans and Knight,² where lymphocyte recirculation patterns in the rat were characterized, cells were radio-labeled before being injected into adults. Results indicated that very few injected cells were ever found in the thymus. Subsequent experiments, however, indicated that there were some T cells re-entering the thymus.³ In contrast, the fact that the bone marrow contains mature, functional B cells that have re-entered from the periphery has been known and accepted for some time. However, renewed interest in T-cell re-entry into the thymus has recently been generated largely because of additions to the experimental armamentarium, including the use of reporter mice and more sophisticated labeling techniques. The series of articles in this special issue deal with the nature and possible functional consequences of cellular traffic, both lymphoid and myeloid, back to the 'primary' lymphoid organs.

Before briefly introducing these interesting and possibly provocative articles, some general points need to be made. The generation of new experimental tools allows a fresh approach to the old questions. Hopefully, rather than rediscovering the wheel, these will generate additional information. Use of cell-tracking probes⁴ has clearly revolutionized our ability to trace cell movements and quantify cell division, whereas advances in microscopy⁵ have allowed us to visualize cellular interactions, especially in the anatomically superficial secondary lymphoid organs. Nevertheless, the primary lymphoid organs, especially the bone marrow, are relatively inaccessible to real-time analysis, making the important cellular interactions that certainly take place therein difficult to visualize. Few experimentalists venture into such inaccessible regions.

The fact that lymphocytes migrated along distinct pathways was described a long time ago. However, one of the next steps is to try and fully understand how this occurs in order to eventually be able to manipulate it. It is naïve to imagine that we understand how lymphocyte behavior, in particular migration patterns, is controlled. We have not yet even fully characterized what the molecular receptor/ligand interactions are that control these movements. Cells cannot think, they have to be told what to do. More strikingly, we do not even know what all the cellular elements of the immune system are; this is particularly striking when one deals with the so-called 'monocyte', the supposed precursor of both tissue macrophages, dendritic cells and perhaps other myeloid elements.

The article by Pillai reviews their data on a relatively recently described peri-sinusoidal niche for recirculating B cells in the bone marrow. Focusing again on the bone marrow, Di Rosa describes the role of T cells therein, highlighting not only their memory status, but also how they may contribute to homeostasis of bone metabolism and hematopoiesis. The two articles by Varol and Proietto are devoted to myeloid cells: the first is related to the bone marrow, the second to the thymus. Myeloid cells can certainly be envisioned as transporters of exogenous materials, whose presence in lymphoid organs could certainly have functional repercussions.

As for the functional consequences of cellular migration back to the primary lymphoid organs, both T and B cells express a considerable panoply of antigens, and it is self-evident that there must be mutual tolerance of their respective lymphocyte-specific antigens. This mutual tolerance must, of course, include their respective T-cell receptor (TCR) or B-cell receptor (BCR) antigen-specific receptors, which are themselves neoantigens. Bogen and Weiss⁶ experimentally showed T-cell tolerance to immunoglobulins, and that B cells must also enter the thymus in order to achieve tolerance to other B-cell-specific antigens. The fact that the best mouse anti-mouse anti-TCR antibodies are generated in TCR KO mice also indicates that mature T cells must enter the bone marrow to induce B cell tolerance to TCR molecules.⁷ Migrating lymphocytes can also secrete soluble factors and therefore influence the function of primary lymphoid organs in 'trans'.

In the great thymus debate, the number of mature T cells re-entering are few and rarely found in the cortex, where the positive selection by epithelial cells supposedly takes place. Sprent and Surh paint a rather gloomy picture of the possible function of incoming cells. An important point of definition stems from this article, namely the use of the term 're-circulation'. This should be confined to what it actually means, namely cells going into the thymus and then coming out again; the latter has not yet been formally shown. Using parabiosis and reporter mice, we^{8, 9, 10} and others^{11, 12} have shown that T cells return to the thymus; however, the challenge raised by Pam Fink was to try and demonstrate a function for such cells. The recent demonstration by Kirberg et al.,¹⁰ using a rather novel experimental setting, that T cells entering the thymus may initiate positive selection to antigens that they alone possess, has been met with some skepticism. This group reviews this work and provides some additional data on this model. In their article, Hale and Fink comprehensively review this topic, including their work with RAG-GFP reporter mice, where with time, peripheral-type T cells do accumulate in the thymus. On a more optimistic note, they speculate on the possible functional consequences of T cells 'coming home' to the thymus. The possibility exists that incoming cells can transfer antigen perhaps to recipient epithelial elements. If this were the case, then incoming cells, whose numbers seem to be relatively constant in different experimental settings and which are known to persist in the medulla for a long time,¹³ could have a significant effect by influencing selection in 'trans'. The relatively recent notion that CD4⁺ thymocyte positive selection can be mediated either by epithelial cells or by lymphocytes themselves has been reviewed by Lee et al. Lymphocyte-mediated selection of T cells results in a subpopulation of T cells with unique properties. Species differences in MHC class II antigen expression make this model difficult to study in mice; however, an alteration in mouse class II expression by manipulation of the class II transactivator (CIITA) adds an experimental arm to this endeavor.

Rather than cell movement from primary lymphoid organs being unidirectional, cells can re-enter, bringing with them additional antigenic determinants picked up in the periphery; the latter possibility is much more likely for myeloid elements. To depict visually the possible functional consequences of such cellular re-entry, I have modified a cartoon created by Hiroshi Kawamoto for the thymus but which is also applicable to the bone marrow (Figure 1). Given that as a consequence of infections and other perturbations an organism's internal 'milieu' is not constant, lymphocyte selection must also adapt dynamically to such an altered scenario.

Figure 1.

In the left panel is the well-recognized representation of the thymus school created by Dr Hiroshi Kawamoto. Here are its depicted rare progenitor undergraduates entering the thymus school to be educated in the well-ordered morning selection classes. In panel b is a modification of this image, where graduates re-enter at a later time; they appear as more 'experienced' bearded brethren carrying gifts of dirty washing. Their entry may perturb, or modify, the well-organized selection classes. Panel a courtesy of Dr Hiroshi Kawamoto, panel b courtesy of the former University of Birmingham graphics department.

Full figure and legend (634K)