WO2007010259A1 - T cell memory assay - Google Patents

Abstract

The invention relates to assays for identifying potential immunosuppressive or immunostimulating compounds, to methods of studying immunodeficiency diseases in patients and to isolated cells from humans.

Description

T cell memory assay

The invention relates to assays for identifying potential immunosuppressive or immunostimulating compounds, to methods of studying immunodeficiency diseases in patients and to isolated cells from humans.

The success of the mammalian adaptive immune system depends on its capacity to make and sustain populations of memory B and T cells, and on their global distribution through an organized network of lymph nodes and mucosal associated lymphoid tissue (MALT). The development of lymphoid tissues in embryonic life is dependent on a specific population of cells, inducer cells, which colonizes lymph node and MALT

anlagen (1). Lymphotoxin beta receptor (LTβR)-signals from inducer cells to the

stromal cells they associate with switch on stromal cell expression of the chemokines CXCL13, CCL19 and CCL21 which recruit B, T and dendritic cells (DCs)₅ forming local aggregations or nodes of lymphoid tissue (2).

In addition to enabling the development of lymphoid tissue, inducer cells have been identified as responsible for the organization within the neonatal lymph node of B and T

cell areas (3). This is dependent on signals through both the LTβR and the closely

related tumour necrosis factor receptor (TNFR)-I (4). The genes for these receptor relatives are tightly linked on chromosome 12 in humans and chromosome 6 in mice (www.ensembl.org), implying that they arose by local gene duplication prior to

speciation of man and mouse. The mechanism by which TNFRl- and LTβR-signals

organize local B:T segregation is analogous to the LTβR-dependent development of lymph nodes: these signals switch on the expression of chemokine genes in discrete stromal cell populations within the lymph node (5). The expression of CCR7-ligands clusters dendritic cells (DCs) and T cells to form the T cell area (6); the expression of CXCR5-ligand brings B cells together to form follicles (7).

The expression of the T zone chemokines in lymph nodes is normal in RAG^{" "}mice, although the expression of the B zone chemokine, CXCL13, is reduced to -20%, and normal expression depends on B cells (Ngo et al., 2001). Like the LTa^{" "} lymphocyte transfer experiments, these data suggest that there is a non-B non-T cell capable of inducing normal (T zone) and partial (B zone) TNFRl and LTβR-dependent chemokine expression. Although in the spleen B cells are required for the normal expression of both B and T zone chemokines (Ngo et al., 2001), the fact that LTa^{" "} splenocytes segregate normally in the spleens of RAG^"''^" mice suggests strongly that a non-B non-T cell can provide stroma with the essential TNFRl- and LTβR-signals that are needed to induce both T zone and B zone chemokines.

Although there is excellent evidence that inducer cells are responsible for both the development and organization of neonatal lymphoid tissue, they are thought to disappear in adult life (8). Recently, the inventors described a distinctive CD4⁺CD3^"

CDl lc^'B220^"IL-7Rα⁺ (CD4⁺CD3^") accessory cell in the lymphoid tissues of adult mice

with a phenotype closely resembling that of inducer cells (9). They have shown that these cells, by virtue of their expression of two other genetically linked TNF family members, OX40-ligand (OX40L) and CD30-ligand (CD30L), are crucial for generating and sustaining the T cells that provide help for the development of high affinity memory antibody responses (9, 10). The inventors speculated that they were the adult equivalent of inducer cells and investigated this possibility by examining the expression in both

cell populations of mRNA for TNFRl -ligands and LTβR-ligands, together with a more

comprehensive set of immunity-related genes. Their analysis reveals that the genetic fingerprint of adult CD4⁺CD3^" cells, excepting only their expression of OX40L and CD30L, is highly correlated with the fingerprint of both embryonic (El 5) and neonatal inducer cells, and clearly different from those of dendritic cells and plasmacytoid dendritic cells (pDCs). Both adult and juvenile CD4⁺CD3^' cells express mRNA for the

splice variant of the retinoic acid orphan receptor gamma (RORγt) (lower in adult

populations) that is required for the development of lymph nodes and gut associated lymphoid tissues (11-13). Finally, as described below and elsewhere (14), they have identified signals that elicit an adult CD4⁺CD3^"phenotype from inducer cells. Therefore the inventors propose that adult CD4⁺CD3^" populations represent the mature adult phenotype of inducer cells.

Furthermore the inventors have identified that CD4+ CD3- cells express both CD4 and CXCR4 in mice and humans. This is an important finding as it indicates that these cells are likely to be targets for HIV. Acquired Immunodeficiency syndrome (AIDS) is caused by Human Immunodeficiency Virus (HIV). HIV is a retrovirus having a single stranded RNA genome. Related diseases in other animals include SIVl in chimpanzees and SIV2 in sooty mangabees. Viruses have a particular target cells in the animal or plants that they infect. The host cells for HIV are those carrying CD4 molecules including macrophages and CD4 T- lymphocytes (CD4 T-cells). HIV uses proteins on its surface called gpl20 and gp41 to attach to the CD4 molecule on the cells.

There are also a number of co-receptors on cells that HIV binds to. The most common co-receptors are CCR5 and CXCR4. CXCR4 is also known as fusin (Feng Y et al Science (1996) vol 272, pages 872-877). This is particularly used by syncytium- inducing (SI) strains of HIV. CXCR4 or SI strains tend to emerge in the body during the course of HIV infection.

One group of researchers have studied the way in which HIV slowly diversifies in the human body (Mullins J., 38^th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Diego (1998) abstract S-IOO). The study highlighted the need for treatments that interfere with the destruction of CD4 T-cells with the CXCR4 co- receptor. There is also some evidence that CXCR4 expression in macrophages is unregulated by some bacterial cell wall products, thus allowing infection by some strains of HIV ( Moriuchi M et al. J. Clin. Invest (1998), vol 102 (8) pages 1540-50). The inventors' data link CD4⁺CD3^" cells with memory antibody responses and maintenance of the integrity of lymphoid architecture, both of which are lost as HIV infection progresses (42, 44, 45). CD4⁺CD3^" cells in mice express CD4 and the chemokine receptor CXCR4, which are co-receptors for viral entry into human cells (46). Although HIV⁺ individuals are infected initially with the CCR5-tropic variant, which mainly targets CD4⁺ T-cell populations in the gut (47), mutation of the virus to the CXCR4-tropic form is associated with depletion of T cells in secondary lymphoid tissue and a poor prognosis (47).

High levels of HIV are known to be trapped on follicular dendritic cells in B-cell follicles (48), where CD4⁺CD3^" cells could be infected directly. The histopathology of HIV-infected lymph nodes is also consistent with the immunopathology of CD4⁺CD3^" cells. Lymph nodes from HIV-infected individuals show hyperplastic GC formation initially, but then the GCs involute and the follicular architecture disappears (49). If CD4⁺CD3^" cells became targets during the course of HIV infection, not only would it impair the capacity to make and maintain neutralizing antibody to the virus and other pathogens, but also, by destroying the cells that establish and maintain lymphoid architecture, undermine immune responses to the large number of normally nonpathogenic infections associated with AIDS.

How might CD4⁺CD3^" cells be destroyed? There are two possibilities: destruction by the virus or destruction by the host cytolytic CD8⁺ T-cell response against infected cells. There is some evidence for the latter in that invasion of B-cell follicles by CD8⁺ T cells is associated with follicular destruction (41). In humans, however, long-term survival with untreated HIV correlates with strong CD4⁺ and CDS⁺ T-cell responses against virally infected cells. This seems to argue against CD4 CD3^" cell destruction by CD8⁺ T cells being relevant to AIDS pathogenesis. However, the opposite is true in the natural non-human primate hosts. Chimpanzees (SIVl) (50) and Sooty Mangabeys (SIV2) (51) show high levels of viraemia without progressive CD4⁺ T-cell deficiency or AIDS, and this is associated with preservation of the follicular architecture in chimpanzees (41). This is not because of resistant mutations within the chimpanzee CXCR4 gene, which is identical to humans (52), but rather due to evidence of a hypoplastic CD8⁺ T-cell response against the virus (41,53), possibly allowing the survival of a CD4⁺CD3^" cell population.

As HIV maintains latency in memory CD4⁺ T cells and is therefore impossible to eradicate (54), an ignorant CD8⁺ T-cell response may have a selective survival advantage in the long-run, provided that the virus is not inherently lethal. By preserving CD4⁺CD3^"cell function, the likelihood of making and maintaining neutralizing antibody responses is increased, which limits the spread of virus, at the same time preserving immune responses to normally non-pathogenic organisms.

Why then does this not seem to be the case for long-term human survivors of HIV? Although there is excellent evidence that SIV/HIV mutates to evade recognition by individual MHC molecules (55), the large number of MHC class I molecules in humans may make it difficult for the virus to evade the CD8⁺ T-cell response entirely. Anything but an all out assault on virus-infected cells, which apparently is achieved by HIV survivors through the strength of their CD8⁺ T-cell response, allows persistent viraemia, the infection of CD4⁺CD3^" cells, and the subsequent attrition of the lymphoid architecture that the inventors believe leads to AIDS.

The inventors have identified the importance of CD4⁺CD3^' cells in organizing the development and perhaps maintenance of secondary lymphoid organs, as well as in the support of adaptive memory antibody responses. Expression of CD4 and CXCR4 by these cells in mice indicates that they may be targets for HIV in humans, and that their destruction by the host CD8+ T-cell response could account for many of the features associated with progressive disease. Pinpointing their identity in humans confirms this. The result is relevant to the control of the HIV pandemic. The inventors predict that in individuals that make suboptimal responses — those that progress to AIDS — strategies that evoke CD8⁺ T-cell responses (56) may aggravate disease, consequent on CD4⁺CD3^" cell destruction, whereas vaccine strategies designed to evoke neutralizing antibody responses (57) are likely to preserve CD4⁺CD3^" cells and lymphoid architecture and to improve prognosis.

The recognition that CD4+CD3- cells express the T cell survival ligand, OX40L is also important. The inventors have unexpectedly found that TLlA, through DR3, rapidly upregulates OX40L. This allows compounds that block expression of OX40L to be assayed. The blocking of OX40L leads to a specific loss of T cell memory. Hence this assay allows the identification of compounds having the potential to be T cell specific immunosuppressive drugs.

In neonatal mice, lymphoid tissue inducer cells (LTi), which elicit the development of lymph nodes and gut associated lymphoid tissue (Mebius, 2003), have been associated with the initial B:T segregation that occurs in the neonatal lymph node (Cupedo et al., 2004). LTi share a similar phenotype to a CD4⁺CD3^" cell present in adult mice (Lane et al., 2005) that the inventors have previously implicated in T cell memory for antibody responses (Gaspal et al., 2005; Kim et al., 2003). The inventors have extended previous observations demonstrating a role for a non-B non T cell in B:T segregation (Fu et al., 1997), and identified CD4 CD3^" cells as key orchestrators of lymphostromal organisation in secondary lymphoid tissues. They show that adult CD4⁺CD3^" cells express high levels of mRNA for LTa, LTβ, TNFa and LIGHT, which are the ligands for TNFRl and the LTβR. Levels of expression are comparable to those expressed in embryonic and neonatal LTi and at least an order of magnitude greater than in CDl Ic⁺ dendritic cells (DCs) or plasmacytoid DCs (pDCs). Furthermore, using adoptive cell transfer experiments, they demonstrate that the expression of these genes is functional: CD4⁺CD3^" cells and particularly LTi, derived from embryonic spleen, but not lymphocytes, pDCs and DCs, are able to restore a significant degree of B:T segregation in the spleens of LTα^"Λ mice. Using confocal microscopy, they demonstrate that this CD4⁺CD3^" cell is closely associated with vascular cell adhesion molecule- 1 positive (VCAM-I⁺) follicular dendritic cells (FDCs) in B cell areas and also with the VCAM-I⁺ stromal population within the T zone and that injection of CD4⁺CD3^" cells into LT α^"A mice restores VCAM-I expression to the emerging T zone. Furthermore, CD4 CD3^" cells are also found adjacent to high endothelial venules (HEV) so are well placed to provide the homeostatic LTβR-dependent maintenance of these blood vessels specialized for lymphocyte transport (Browning et al., 2005).

Although LTi share the phenotype of adult CD4⁺CD3^" cells (CD4⁺CD3^"CD1 lc^"B220^" IL-7Rα⁺) (Kim et al., 2003), and both function in B:T segregation, there is no direct evidence that they share a precursor-product relationship; indeed the inventors have previously reported that LTi lack expression of the T cell survival ligands, OX40-ligand (CD252,TNFSF4) and CD30-ligand (CD153, TNFSF8), but have speculated that this helps explain why T cell priming in neonatal life leads to tolerance rather than immunity (Kim et al., 2005). Here the inventors provide 3 pieces of evidence linking LTi and adult CD4⁺CD3^" cells. First, they show strong correlation between them by a quantitative analysis of their gene expression. They quantified the expression of a set of immunity-related genes that includes chemokine receptors, TNF family members and survival proteins. The expression patterns within subsets of T cells, of B cells and of DCs correlated closely with group identity; when subjected to the same analysis, adult CD4⁺CD3^' cells and LTi were closely correlated with one another but distinct from lymphocyte, NK cell, pDC and DC populations. Second, they show in vitro induction on LTi of the two TNF ligands linked with T cell survival, OXC40-ligand (L) and CD30L. Finally, they demonstrate in vivo that fetal LTi acquire key characteristic features of adult CD4⁺CD3^' cells, namely OX40L and CD30L, following transfer into an adult recipient.

A first aspect of the invention provides:

An assay for identifying a potential immunosuppressive or an immunostimulating compound comprising the steps of providing i) TLlA or a fragment or a derivative thereof and ii) a DR3 receptor or a fragment or a derivative thereof, and determining the affect of the compound on the binding of TLlA to the DR3 receptor.

The immunosuppressive or an immunostimulating response assayed is thought to be T cell specific because TLlA has the specific effect in CD4+CD3- cells of increasing OX40L production. OX40L, as discussed above is the T cell survival ligand.

TLlA is also known as TNFSFl 5 or TNF superfamily ligand (15).

DR3 (death receptor 3) is also known as Apo-3, LARD, TRAMP, TR3 and TNFRSF12 (TNF superfamily receptor 12)- see for example Marsters S.A. et al Curr. Biol., (1996), VoI 6, pages 1669-76. This receptor is also known to mediate activation of NfkappaB and to induce apoptosis in some cells. The receptor may be presented in vivo or in vitro. The receptor may, for example in a simplified assay form be immobilised on a suitable support, such as a microtitre plate by techniques well known in the art for the immobilisation of proteins. The binding of TLl A to the DR3 receptor may then be monitored by providing TLl A in a suitable medium, such as a buffered solution. Alternatively TLlA may be immobilised and the binding of the DR3 receptor to the TLlA may be detected.

The detection of the binding of one or other component, and the effect of the presence of the compound on the binding of the two components may be monitored in a number of ways. For example, one or other component may be labelled with a radiolabel or a coloured label. The labelling of proteins with radio and other labels is well known in the art. The amount of DR3 and TLlA bound together may be determined by measuring the amount of label retained after non-bound label has been washed away.

Alternatively, for example, the amount of bound and/or non-bound TLlA and DR3 may be determined via immunological techniques well known in the art, such as ELISA, sandwich assays, competition assays etc. Antibodies against TLlA and DR3 are commercially available. TLlA may be obtained from, for example, R&D Systems Inc, Minneapolis, MN, USA or Abeam Ltd. Anti-DR3 antibodies may be obtained, for example, from Affinity Bioreagents Inc, or Abeam Ltd. Anti-TLl A and/or anti-DR3A antibodies may be used in the assay. The effect of the compound on TLl A and DR3 binding may be compared to a control sample with, for example, no compound added.

Preferably the DR3 is provided on the surface of a CD4+CD3- cell.The effect of the compound on the binding of the TLl A to the DR3 receptors on the cell or the effect on signal transduction induced expression of OX40L may be determined.

The invention also provides:

An assay for identifying a potential immunosuppressive or an immunostimulating compound comprising the steps of:

a) providing a cell expressing a DR3 receptor or a fragment or a derivative thereof,

b) stimulating the cell with TLl A or a fragment or a derivative thereof

c) contacting the cell with the compound, and

d) determining the effect of the compound on OX40L, or a fragment or a derivative thereof, expression.

This aspect of the invention is more specific because it detects the effect of the compound on OX40L, the T cell survival ligand, expression. OX40L (OX40 ligand) is also known as CD134L (CD134 ligand)- see for example Akiba, H. et al, "Identification of rat OX40 ligand by molecular cloning", Biochem, Biophys., Res. Commun. (1998) 251, pages 131-6.

OX40L may be monitored by techniques known in the art.

TLlA, DR3 or OX40L may be mammalian versions of the proteins, preferably obtained from mouse, rat or humans. They may be the native protein to the cell, where used, or alternatively heterologous proteins introduced by e.g. introducing in a suitable expression vector know in the art

The term "or a derivative thereof " means that the protein preferably still has one or more of the functions of the wild type , TLlA, DR3 or OX40L. It is well known in the art that one or more conservative substitutions of amino acids may be made without altering the function of the protein.

Typical substitutions are among Ala, VaI, Leu and lie; among Ser and Thr; among the acidic residues Asp and GIu; among Asn and GIn; and among the basic residues Ly s and Arg; or aromatic residues Phe and Tyr. One or more amino acids may be substituted, deleted or added. Preferably, less than 10, less than 5, less than 2 amino acids are substituted, deleted or added in any combination.

Preferably, the TLlA derivative or fragment has the ability to bind and activate wild type DR3 receptor. Preferably, the DR3 receptor fragment or derivative has the ability to be bound by native TLlA and activate OX40L production in cells, such as CD4+CD3- cells or more preferably in DR3 transfected cell lines. . The OX40L derivatives or fragments preferably have the ability to bind OX40 receptors and preferably activate the receptors.

For example the OX40L may be monitored by immunological techniques. Anti-OX40L antibodies are known in the art, for example from eBioscience Inc, San Diego, California.

The cell may be a CD4+CD3- cell or more preferably in DR3 transfected cell lines. . Alternatively a cell may be transformed with a DNA sequence encoding DR3 with suitable coding signals to cause the DR3 to be expressed in the cell. Methods of expressing genes in cells are well known in the art. The advantage of such a system is that, instead of having to isolate CD4+CD3- cells or more preferably in DR3 transfected cell lines. , an easily obtainable or grown cell line may be used to study the DR3 receptor and the induction of OX40L in the cell.

Alternatively the effect of the compound on OX40L gene expression may be determined. Microarrays for monitoring protein expression are well known in the art. Other techniques, such as quantitative PCR may be used. Quantitative PCR is described in, or example, WO97/46714 and EP1288314A, incorporated herein by reference. Preferably the compound is added to the cell before the cell is stimulated with the TLlA.

Typical concentrations of TLlA are between 0.1 and lOOOng/ml, especially up to 500ng/ml, 300ng/ml, 200ng/ml, lOOng/ml, 50ng/ml or 20ng/ml. Especially preferred concentrations are lng/ml to 100ng/ml.

Preferably the effect monitored is a decrease in binding or a decrease in OX40L expression. This indicates that the compound is likely to have immunosuppressive effects. The ability to identify e.g. T cell specific immunosuppressive compounds has great potential for example for the development of compounds to reduce transplantation rejection. Additionally this may be used to identify compounds with the potential to treat autoimmune diseases such as systemic lupus erythematosis, systemic vasculitis mediated by anti-neutrophil cytoplasmic antibodies. Other diseases that may be treated, or have symptoms treated, with such compounds include diabetes and rheumatoid arthritis as the appearance of autoantibodies heralds the onset of disease by a number of years.

Preferably the cell is used in vitro. The cell is preferably mammalian, such as a mouse, ape, or a human cell.

The cell preferably has the following phenotype: CD4+CD3-CD1 lc-IL7receptor+, ckit+, common cytokine receptor

gamma chain positive, TRANCE+, RANK+, TNFRSF25+, CXCR5+, CXCR4+

A further aspect of the invention provides an isolated human cell having the phenotype:

CD4+CD3-CDl lc-IL7receptor+, ckit+, common cytokine receptor

gamma chain positive, TRANCE+, RANK+, TNFRSF25+, CXCR5+, CXCR4+

The isolation of such cells has not been carried out before. This is the first time that they have been recognized. The potential importance in AIDS research is discussed above.

Hence a further aspect of the invention provides:

A method of studying immunodeficiency in a patient comprising obtaining a sample of tissue containing CD4+CD3- cells and determining the amount of the cells in the sample. This allows an indication of the likely effect on T cell survival in the patient to be determined and followed. The invention also provides a method for identifying human CD4⁺CD3^" cells in human tissues using reagents that identify DR3.Preferably the immunodeficiency is AIDS. The method preferably provides an indication of the progression of the immunodeficiency by presence or lack of the cells.

It may be used to assist in determining, for example, whether a patient is likely to develop a reduced immune system because of HIV infection, and whether or when to supply drugs to assist in the suppression of e.g. HIV. This allows better control of the drugs given, which often have undesirable side effects.

As discussed above the cells are ideal targets for compounds, such as vaccines, to improve their survival, for example in the presence of HIV. The invention therefore provides a method of identifying a compound which assists in the survival of CD4+CD3- cells, comprising contacting CD4CD3- cells with the compound and determining the effect of the compound on survival of the cells

Preferably the cell monitored is:

CD4+CD3-CDl lc-IL7receptor+, ckit+, common cytokine receptor

gamma chain positive, TRANCE+, RANK+, TNFRSF25+, CXCR5+, CXCR4+

Preferably the cell is from a mammal, such as an ape, human or mouse.

The presence of the cell may determined using techniques known in the art for the isolation and detection of specific cells types by immunohistochemistry. For avoidance of any doubt the following abbreviations have been used herein:

CD4⁺CD3^" cells, CD4⁺CD3^"CDllc^"B220TL-7Rα⁺ cells. Such cells have been further

characterised to identify additional phenotypic components; DR3, death receptor 3; HVEM, herpes virus entry mediator; GITR, glucocorticoid-induced TNFR family- related gene; LIGHT, LT-related inducible ligand that competes for glycoprotein D binding to herpes virus entry mediator on T cells; TRANCE, TNF-related activation- induced cytokine; RANK, receptor activator of nuclear factor kappa B; GFP, Green fluorescent protein; mYa, million years ago.

The invention will now be described by way of example only with reference to the following figures:

Figure 1. Correlation between gene expressions in CD4⁺CD3^"

and other accessory cell populations.

Adult cell populations were isolated from RAG^"7" spleens. El 5 embryonic and day 2 neonatal cells were from normal BALB/c mice

a. CD8OD1 Ic⁺ versus CD8⁺CD1 Ic⁺ dendritic cells (DC), b. Adult CD4⁺CD3^" cells versus CD8⁺CD1 Ic⁺DCs. c. Adult CD4⁺CD3^" cells versus plasmacytoid dendritic cells. d. Adult versus El 5 CD4⁺CD3^" cells, e. Adult versus neonatal (day 2) CD4⁺CD3^' cells. f. Neonatal CD4⁺CD3^' cells from RAG^"Λ spleen versus RAG^'Λ lymph node.

These results are representative of at least two separate experiments.

Figure 2. Genes differentially expressed in CD4⁺CD3^" cells.

a. El 5 and adult CD4⁺CD3^" cells.

b. adult CD4⁺CD3^" cells cultured with or without TLlA.

c. El 5 CD4⁺CD3^" cells cultured with or without TLlA.

d. Correlation of gene expression in OX40L⁺ versus OX40L^" FACS sorted El 5 CD4⁺CD3^" cells after treatment with TLlA.

These results are representative of at least two separate experiments.

Figure 3. Effect of TL1 A on TNF ligand protein expression on CD4⁺CD3^" populations in vitro.

a. Adult CD4⁺CD3^" cells cultured with/withoutlOO ng/ml TLlA for 2 days.

b. Neonatal inducer cells (day 1) cultured with/without 100 ng/ml TLlA for 2 days.

c. El 5 inducer cells cultured with/without 100 ng/ml TLlA for 2 days.

Shaded histograms show control staining with biotinylated rat antibodies. This result is representative of four separate experiments. Figure 4. Effect of TNFR1 and LTβR signals on B:T segregation and CD4⁺CD3- cells.

a. Low power images from wild type mice (WT), LTa^{" "} mice, or mice injected with

lOOμg LTβR-Ig 4days previously. CD3 is shown in red and IgM in green. Yellow

shows co-localization of B and T cells. Squares outline insets shown at higher magnification in (b). Scale bars 100 μm.

b. Magnifications of insets from a-c respectively, demonstrating the location of CD4⁺CD3^" cells (red), IgM⁺ cells (grey), CDl Ic⁺ and CD3⁺CD4^" T cells (green), and CD3⁺CD4⁺ T and CD4⁺CD1 Ic⁺ DCs (yellow). Scale bars 20 μm.

Figure 5. Expression of CXCR4 by neonatal and adult CD4⁺CD3^" cells.

Shaded histograms show control staining with biotinylated rat antibodies.

Figure 6.

Digitally processed image of an immunostained section from mouse spleen to show the location of CD4⁺CD3^"CD1 Ic^" cells (red) in relation to CDl Ic⁺ DCs (green) and IgM⁺ B cell areas (gray). CD4⁺CD3^" cells are mainly located at the B:T interface and in B-cell follicle adjacent to follicular T cells whereas DCs are located in T cell area and red pulp. Plasma cells are located in the red pulp that surrounds the white pulp areas that contain lymphocytes. T=T cell area, B=B follicle, MZ=marginal zone. DC=dendritic cell. Figure 7.

Antigen-specific T cells are first primed on DCs in T cell area. T cells that upregulate CXCR5 and migrate into B follicles. The white arrow indicates CD4 T cell migration from T cell area to B follicle. In the follicle, follicular T cells (T_f) drive the proliferation of B cells that differentiate into centroblasts undergoing somatic mutation within immunoglobulin variable genes. Centroblasts proliferate rapidly and differentiate into centrocytes that compete for antigen fragments trapped on follicular dendritic cells (FDCs). B-cell mutants bearing high-affinity receptors take up antigen and present peptide fragments to T_f, which then provide selective CD40L-dependent rescue signals. When antigen is scarce, T_f are sustained by OX40 and CD30 signals from CD4⁺CD3^" cells, which express OX40L and CD30L constitutively so that that they can continue to select B cells. Centrocytes that are positively selected leave the GC and differentiate into memory B cells or mature into long-lived plasma cells.

Inset white shows a T cell and a DC (T:DC) interaction in T cell area (TZ). Inset red shows interaction between a T_f and a CD4 CD3^" cell within a B cell GC and the left shows this at high magnification.

Figure 8.

Re-activation mechanism of memory T cells (T_m) by CD4⁺CD3^" cells. Antigen-specific T_m that co-express CXCR5⁺CCR7⁺, transmigrate through high endothelial venules (HEVs) in lymph nodes via their expression of ligands for peripheral lymph node addressin (PNAd). IL-7 produced by putative stromal cells or follicular dendritic cells (FDCs) within lymphoid tissue upregulates expression of OX40 on T_m. T_m then interact with interface CD4⁺CD3^" cells, which express OX40L constitutively at the B:T border. The interaction of CD4 CD3^" cells and T_m cells provide survival signals to T_m through OX40 by upregulation of BCL-2 and BCL-XL. Inset shows high magnification of this interaction.

Figure 9

Summarises the identification of CD4+CD3- cells in humans

Fig 9a shows human CD4+CD3- cells in dark grey interacting with T cells (light grey)

Fig 9b shows the cells of the invention (see box- identified by e.g. FACS) are IL7R+cKit+CD4+ identified in human lymph node and spleen.

Fig 9c shows that the CD4+cKit+ cells also express Thy-1 and CXCR5 like the mouse cells

Fig 9d shows three populations (C-KIT++ (upper box), C-KIT+ (middle box) and cKIT- (lower box)) of cells from human total LN which were sorted, and then their mRNA expression was analysed.

Fig 9e shows the ckit+ cells that are of interest express the main genes of interest- LTalpha+, TNFalpha÷, LTbeta, Light

Fig 9f shows the ckit+ cells that are of interest express the main genes of interest- Rank, Trance and DR3 although OX40L is still present

Fig 9g shows the expression of the expected genes Fig 9h shows a comparison of human LN cells with mouse CD4+3- cells. 91 genes were compared. The correlation coefficients between mouse CD4+3- cells and human LN are given. A good correlation is observed between mouse and human cells

Figure 10

Summarises the correlation between gene expression of different cell types.

Figure 11

Summarises the correlation coefficients between gene expressions of the different cell types of Figure 10.

Figure 12

Regulation of B:T segregation and evidence that CD4⁺CD3^' cells are associated with VCAM-I⁺ stromal cells.

(A) Low power confocal images of spleen sections. CD3 is shown in red and IgM in green. Yellow shows co-localization of B and T cells.

(a) wild type mice (WT), (b) LTα^"Λ mice, (c) organization of the spleen of LTα^";" mice

10 days after transfer of splenocytes, (d) organization of LTα^"A splenocytes 10 days

after transfer into RAG^"Λ recipients, (e) organization of the spleen of LTα^";" mice 10

days after transfer of CDl Ic enriched DCs and pDCs, (f) organization of the spleen of

LTα^"Λ mice 10 days after transfer of CD4⁺CD3^" cells.

Scale bar, 100 μm. Results representative of at least two separate experiments. (B) Confocal images of spleen sections showing (a) wild type (WT) and (b) LTα^"Λ

mice. Dotted yellow area identifies B follicle in WT mice. To show association of VCAM-I⁺ cells with CD4⁺CD3^" cells confocal images from T cell deficient mice were analyzed at (c) low power, (d) high power of B cell area, and (e) high power of T cell area from (c).

Scale bar, 50 μm for low power, 20 μm for high power.

Figure 13

Role of LTi and CD4⁺CD3^" cells in B:T segregation

(A) mRNA Expression of TNFRl- and LTβR- ligands on LTi, CD4⁺CD3^'and other accessory cells.

These results are representative of at least four separate experiments.

(B) (a) White pulp area, (b) B cell area and (c) T cell area in LTa^"7" mice that were

adoptively transferred with splenocytes, CDl Ic⁺DCs and pDCs, CD4⁺CD3^" cells or El 5 LTi compared with normal white pulp areas.

Bar shows mean values with high and low values calculated from 10 different white pulp areas. Statistical differences were calculated using a non-parametric Mann- Whitney test. Results representative of at least 2 experiments. Figure 14

CD4⁺ populations in spleen and CD4⁺CD3^' cells associated with PNAd^* HEVs in lymph node

(A) CD4⁺ populations isolated from T cell deficient mice and characterized for expression of CDl Ic, B220 and OX40L and CD30L. Shaded histograms show control staining.

(B) Confocal images of (a) T cell deficient mice stained with B220 and CDl Ic in green, and CD4 in red, (b) normal mice stained with CD3, B220 and CDl Ic in green and CD4 in red at low power, and (c) lymph node from a normal mouse stained with CD3, B220 and CDl Ic in green, CD4 in red, and PNAd in blue (scale bar, 100 μm) and high power (scale bar, 20 μm).

Figure 15

Correlation between gene expressions in CD4⁺CD3^' and other cells.

Adult cell populations of CD4⁺CD3^" cells, pDCs, and DCs were isolated from RAG^"7" spleens. El 5 embryonic, day 2 neonatal LTi, follicular B, marginal zone B, and NK cells were from normal mice. ThI and Th2 cells were differentiated in vitro for 6 days.

The level of expression of individual mRNAs is expressed as a % of β-actin. X and Y

axes show relative levels ofmRNA expression of the two cell types being correlated. CC=correlation coefficient.

These results are representative of on average five separate experiments. Figure 16

Effects of TL1A signals on LTi and CD4^*CD3^" cells

(A) Genes differentially expressed in LTi and CD4⁺CD3^" cells.

(a) El 5 LTi and adult CD4⁺CD3^" cells.

(b) Adult CD4⁺CD3^" cells cultured with or without 100 ng/ml TLlA for 2 days.

(c) El 5 LTi cultured with or without 100 ng/ml TLlA for 2 days.

(d) Correlation of gene expression in OX40L⁺ versus OX40L^" FACS sorted E 15 LTi after treatment with TLlA for 2 days.

These results are representative of at least two separate experiments.

(B) Effect of TLl A on TNF ligand protein expression on LTi and CD4⁺CD3^" cells in vitro.

(a) Adult splenic CD4⁺CD3^" cells cultured with/withoutl 00 ng/ml TLl A for 2 days.

(b) Adult lymph node CD4⁺CD3^" cells cultured with/withoutl 00 ng/ml TLlA for 2 days.

(c) Neonatal LTi (day 1 -2) cultured with/without 100 ng/ml TL 1 A for 2 days.

(d) E15 LTi cultured with/without 100 ng/ml TLlA for 2 days.

(e) Neonatal LTi (day 1 -2) cultured with/without 100 ng/ml TL 1 A and/or 100 ng/ml IL-7 for 6 days.

Shaded histograms show control staining with biotinylated rat antibodies. This result is representative of four separate experiments. (C) In vivo upregulation of OX40L and CD30L on LTi (CD45.2) after transfer into an adult mouse (CD45.1).

Shaded histograms show control staining with biotinylated rat antibodies. This result is representative of two separate experiments.

Figure 17

Expression of BcI family members and chemokine/chemokine receptors and the ratio of adult CD4^*CD3^" cell number in RAGT^/m and B cell sufficient mice.

(A) Relative mRNA expression of Bcl-2, Bcl-xL, CCR7 and its ligand (CCL 19), CXCR5 and its ligand (CXCLl 3), and CXCR3. This result is representative of three separate experiments.

(B) Surface CXCR4 expression on neonatal LTi and adult CD4⁺CD3^" cells. Shaded histograms show control staining with biotinylated rat antibodies.

(C) Ratio of CD4⁺CD3^" cell numbers purified from a mixture of one RAG^"7" (CD45.2) and one B cell sufficient mouse (CD45.1).

Materials and Methods

Mice

All experiments were performed in accordance with the UK laws and with the approval

of the local ethics committee. Normal, RAGl deficient (RAGl^"7") and LTa^"7" mice were

bred and maintained in our animal facility. Neonatal lymph node or spleen CD4⁺CD3^" cells and spleen LTi were isolated from 1-2 day old normal BALB/c litters or RAGl^"7" mice. Spleens from BALB/c El 5 embryos were used to isolate El 5 CD4⁺CD3^" cells and spleen LTi.

For experiments to test the effects of acute blockade of LTβR-signals, 100 μg/mouse

LTβR-Ig (a kind gift of Dr Jeff Browning, Biogen) or control-Ig was injected into wild

type C57BL/6 mice i.p. 4 days prior to sacrifice.

E15 spleen organ culture

Embryos from normal pregnant BALB/c mice of gestation day 15 were obtained and the spleens removed. The spleens were then placed in culture medium with 100 ng/ml IL-7 (PeproTech EC). The spleens were mouth pipetted with a fine glass pipette onto a 0.8

μm sterile nucleopore filter (Millipore) on a sterile arti wrap sponge. The Petri dish was

then cultured in a contained humid environment in a 10% CO incubator, and left for 5 days. On day 5, cultured spleens were teased apart with fine forceps, and 100 ng/ml of recombinant mouse TLlA (R&D Systems) was added into the medium for a further 2 days of culture. Cell suspensions were then stained for flow cytometry analysis or for MoFIo cell sorting.

Preparation of neonatal inducer cells, adult CD4^*CD3^' cells, pDCs and DCs

Cell suspensions for isolation of CD4⁺CD3^" cells, DCs and pDCs were made from the spleens of adult RAG^"Λmice, as described previously (9, 14). Neonatal LTi were isolated from either BALB/c or C57B1/6 mice that were 1 or 2 days old. Briefly, CDl Ic⁺ cells were positively enriched by using CDl Ic coated magnetic beads (Miltenyi Biotec Ltd.), and then FACS sorted into CD8⁺ and CD8^" populations. CD4⁺ cells were enriched from CDl Ic⁺ depleted populations using CD4 coated magnetic beads, and the resulting CD4⁺ enriched populations sorted into CD4⁺CD3^"B22O^"CD1 Ic^" (CD4⁺CD3^") and CD4⁺CD3^"B220⁺CDl lc^low(pDC) populations. CD45^" stromal cells were FACS sorted from BALB/c. For Figure 14A, CD4 enriched populations were prepared without CDl Ic depletion from T cell deficient mice. Sorted cells were

suspended in 0.5 ml RLT buffer (QIAGEN) with 5 μl β-mercaptoenthanol (Sigma) and stored at -80°C prior to use.

With respect to Figures 10 to 17: for the preparation of El 5 LTi, embryos from normal pregnant mice of gestation day 15 were obtained and the spleens removed. The spleens were placed in culture medium with 100 ng/ml IL-7 (PeproTech EC) on a 0.8 μm sterile nucleopore filter (Millipore on a sterile arti wrap sponge. The Petri dish was then cultured in a contained humid environment in a 10% CO incubator for 5 days. On day 5, cultured spleens were teased apart with fine forceps and CD4 cells enriched as above.

With respect to Figures 10 to 17: follicular B (CD21^lowCD23⁺IgM^intermediate) cells, marginal zone B (CD21^hlghCD23^"IgM⁺) cells, and NK cells from normal mice were sorted to make cDNA. ThI and Th2 cells were prepared under ThI conditions (10 ng/ml IL- 12 and 10 μg/ml anti-IL-4) and Th2 conditions (10 ng/ml IL-4 and 10 μg/ml anti-IL- 12) for 6 days in vitro culture.

Stimulation of E15 or neonatal LTi or CD4⁺CD3^' cells for Figures 10 to 17:

Prepared cells were cultured with a wide range (0.1-100 ng/ml) of recombinant mouse TKlA (R&D) Systems) for 2 or 6 days of culture and then stained for flow cytometry analysis or for MoFIo cell sorting. Adoptive cell transfer for Figures 10 to 17:

3x10⁷ splenocytes from either LTa^"7" or normal mice were transferred i.p. into RAG^"7" hosts. 10 days post transfer the spleens of the injected mice were taken and stained for confocal microscope analysis.

1.6x10* CD4⁺CD3^" cells or IJxIO⁵LTi or IxIO⁶ CDl Ic⁺DCs and pDCs from RAG^"7" mice or 4x10⁶ splenocytes from normal mice were transferred i.v. into LTa^"7" recipient mice. 10 days post transfer the spleens of the injected mice were taken and stained fro confocal microscope analysis.

1x10 LTi cells from CD45.2 El 5 spleens cultured with IL-7 for 7 days were transferred i.p. into CD45.1 T- and NK-cell deficient mice. The 5 days post transfer the spleens of the transferred mice was taken and CDl lc-depleted CD4-enriched cells were overnight cultured and stained for flow cytometry analysis.

Ce// culture and FACS staining of TL1A treated populations

CD4-enriched populations were cultured alone or with 100 ng/ml IL-7 (PeproTech EC) and/or recombinant mouse TLlA (R&D Systems) over a wide range (0.1-100 ng/ml) for 2 days.

CD4⁺CD3^" cells were identified by excluding CD3⁺CD1 lc⁺B220⁺ cells with FITC labelled anti-CD3₅ anti-CD 1 Ic and anti-B220 mAbs (BD Biosciences). They were positively identified with anti-CD4 PE, and then stained with biotinylated mAbs against OX40L, CD30L and CXCR4 (BD Biosciences) or TRANCE (R&D systems) in conjunction with streptavidin CyChrome (BD Biosciences) as the second-step staining reagent.

RNA and cDNA preparation

Total RNA was prepared from FACS-sorted cell subsets using RNeasy^® Micro Kit according to the manufacturer's instructions (QIAGEN). The RNA was eluted at the final step with RNase-free water.

For generation of cDNA from RNA by reverse transcription the following method was

used. 1 μl of oligo-dT_12-18 primers (500 μg/ml) (Amersham Pharmacia Biotech) was

added to each 10 μl RNA sample, which was then denatured for 10 min at 70°C

followed by a quick 1 min chill on ice. 9 μl of the reverse transcription reaction mix

containing 0.5 μl H₂O, 4 μl 5X first strand buffer (GIBCO Invitrogen Co.), 2 μl 0.1 M

DTT (GIBCO Invitrogen Co.), 1 μl 10 mM dNTP (Promega), 0.5 μl RNase inhibitor

(RNAguard, Pharmacia) and 1 μl avian reverse transcriptase (M-MLV, Life

Technologies) was then added to each sample. Samples were then heated for 1 hour at

42⁰C followed by 10 min at 90°C. Finally cDNA samples were diluted to 100 μl with

RNase-free water.

TAQman low density array analysis

TAQman primer sets are designed to work with an efficiency approaching 100%, enabling the quantitative comparison of mRNA expression for different genes not only

within a cell type, but also between cells of different lineages. Housekeeping genes (β-

actin in these experiments) were used to correct for total mRNA, generating a scale from

0 to 10⁶ (the level to which the β-actin signal was corrected in all mRNA samples). TAQman low density real time PCR arrays (Applied Biosystems) were designed with a 96-gene format. The arrays contained TAQman primers and probes for all available TNF/TNFR family members and a selection of T cell cytokines and cytokine receptors, chemokines and their receptors, TLR molecules, costimulatory molecules, DC markers, transcription factors, and house keeping genes. A list of all of the genes measured is as follows:

chemokines (CCL19, CXCL12, CXCL13), chemokine receptors (CCR7, CXCR3,

CXCR5), cytokines (IL-lα, IL-I β, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-9, IL-IO, IL-

12p35, IL-12p40, IL-13, IL-15, IL-18, TSLP, IFN-αl, IFN-β, IFN-γ, TGF-βl), cytokine

receptors (IL-2Rα, IL-2Rβ, IL-2Rγ, IL-4Rα, IL-7Rα, IL-I ORa, IL-lORβ, IL-12Rβl,

IL-12Rβ2, IFN-γRl, IFN-γR2), costimulatory molecules (CD80, CD86, CTLA4, ICOS,

ICOSL)₅ DC marker (DC-SIGN, cathepsinS, Integrin alpha x), house keeping (CD4, β-

actin, 18S), MHC class (β-2m, CD74), TLR (MyD-88, TLR2, TLR3, TLR4, TLR5,

TLR7, TLR9), TNF family (LTa, TNFa, LTβ, OX40L, CD40L, FASL, CD70, CD30L,

4-1BBL, TRANCE, TWEAK, APRIL, BAFF, LIGHT, TLl), TNFR family (TNFRl,

TNFR2, LTβR, OX40, CD40, FAS, CD30, 4-1BB, RANK, TWEAKR, BAFFR,

HVEM, GITR, DR3), transcription factor (Bcl-2, Bcl-6, Bcl-xL, RORγ, GATA3, foxP3,

T-bet), others (perform, granzymeB).

cDNA was mixed with TAQman Universal PCR Master Mix (Applied Biosystems). This was added to the TAQman Low Density array, and PCR was performed in a 7900HT Fast Real-Time PCR System (Applied Biosystems) according to manufacturer's recommendations. The relative signal per cell was quantified by setting a threshold within the logarithmic phase of the PCR and determining the cycle number at which the fluorescence signal reached the threshold (C_t). The C_t for the target gene was subtracted from the Q for β- actin. The relative amount was calculated as 2 x 10 .

Conventional PCR analysis

Signals for RORγt, LTa and β-actin were analyzed by conventional PCR. β-actin was used as the housekeeping gene for sample normalization, prior to amplifying the target genes of interest. Reactions were conducted in a Peltier Thermal Cycler PTC-200 (MJ Research, Genetic Research Instrumentation) using ABgene PCR master mix. The first step of the reaction involved activation of Thermoprime DNA polymerase at 94°C for 5 min, then cycles of 94°C for 30 sec, annealing of primers at 55°C for 40 sec and extension of primed template at 72°C for 1 min.

PCR products were analyzed by ethidium bromide gel electrophoresis and identified by fragment size using Syngene Gel Documentation Gene Tools software.

The primer sequences were: β-actin : Forward (5'-ATC TAC GAG GGC TAT GCT

CTC C-3'), Reverse (5'-CTT TGA TGT CAC GCA CGA TTT CC-3'); RORγt :

Forward (5'-ACC TCC ACT GCC AGC TGT GTG CTG TC-3'), Reverse (5'-CAA

GTT CAG GAT GCC TGG TTT CCT C-3'); LTa : Forward (5'-CTC CAT CCT GAC

CGT TGT TT-3'), Reverse (5'-TAG ACC CAC AAA AAC CCT GC-3'). lmmunohistology for Confocal microscope

Spleen sections from normal, LTα^"Λ mice and mice sacrificed 4 days after i.p. injection

with either control-Ig or LTβR-Ig (lOOμg/ml) were stained as described previously (9).

Comparison of lymphocyte architecture in spleens from wildtype and variously treated or gene modified mice was performed using the Zeiss LSM software on confocal micrographs taken from sections stained for IgM and CD3 as described previously (Kim et al, 2003), in conjunction with staining with fluoresceinated anti- VCAM-I antibody (Becton Dickinson) at a previously optimized dilution. The intense staining of VCAM- I⁺ cells in splenic red pulp contrasted with less intense and patchier appearance of VCAM-I staining in white pulp areas. The inventors therefore used the limits of the VCAM- l^hl staining as a lymphocyte-independent indicator of the extent of the white pulp areas. Intuitively if B cells are stained with one fluorochrome and T cells with another, when there is increased B:T segregation then there will be less overlapping of colours. This is what the following objective algorithm was designed to test.

Regions of white pulp delineated by red pulp VCAM-I expression were extracted and the area (μm²) and total pixels determined by the Zeiss confocal software were recorded. Within the extracted region we enumerated the pixels registering intensity for IgM (B cells) and CD3 (T cells). Where B cells and T cells are within 0.16 μm (the pixel dimension using the x25 objective) of one another, the pixels are recorded as double positive. The inventors quantified double positivity by multiplying the IgM and CD3 matrices of the micrographs together to produce a separate array (IgM⁺CD3⁺) that we used as a measure of the degree of contact between B and T cells. Numbers of singly positive pixels (IgM or CD3) were determined by subtracting the numbers of IgM⁺CDS⁺ pixels from those in the respective original arrays. Pixel number (for each magnification on the confocal microscope there will be a fixed relationship between pixel number and area in μm²) was then used to provide an estimate of the areas within each white pulp region taken up by B cell or T cell membrane, or both (IgM⁺, CD3⁺ and IgM⁺CD3⁺ respectively). Spleen sections were examined systematically for all identifiable areas of white pulp; routinely, 10 different areas were photographed per spleen. Following statistical evaluation, mean values with ranges for each treatment were selected for display purposes.

Results

Inducer cells and adult CD4^*CD3^' cells display similar genetic fingerprints which readily distinguish them from other accessory cell types

Although relatively few genes are cell-specific, the inventors reasoned that the level of mRNA expression for a set of genes would be correlated in cells of related lineage, establishing a genetic fingerprint. To test this, we designed TAQman arrays for a panel of immunity-related genes (see materials and methods). Comparison of genes expressed within the two major subsets of DCs in mice, CD8⁺CD1 Ic⁺ and CD8^"CD1 Ic⁺ DCs, revealed a strong correlation coefficient (CC = 0.94) (Fig. Ia), suggesting that this presumption was valid. Comparison of gene expression between adult CD4⁺CD3^" cells and either CD8⁺ DCs or pDCs showed much weaker correlation coefficients (CC = 0.68 and 0.59, respectively) (Figs. Ib and Ic). In contrast, after OX40L and CD30L gene expression was excluded from the analysis, the gene expression in adult CD4⁺CD3^" cells was strongly correlated with embryonic (E15) spleen CD4⁺CD3^" (CC=0.86) (Fig. Id), and with neonatal spleen CD4⁺CD3^" populations (CC=0.90) (Fig. Ie); neonatal lymph node and neonatal spleen CD4⁺CD3^" cells were also strongly correlated (CC=O.88) (Fig. If).

The details of the distinctive gene profile established for the CD4⁺CD3^" cell type are tabulated: Table 1 shows TNF and TNF-receptor (TNF/TNFR) family members and Table 2 non-TNF/TNFR family members. Expression of mRNA in El 5 inducer cells was similar to neonatal inducer cells (data not shown). The analysis focused on genes

expressed at high levels (mRNA expressed at >0.2% of the β-actin signal). To simplify

analysis, gene expression has been categorized into 4 groups relative to expression of β-

actin: +++ > 10%, ++ 1-10%, + 0.2-1%, - <0.2%.

RORγt mRNA is expressed in adult CD4^*CD3^~ populations and DR3 signals effect a transition from a neonatal to an adult phenotype

RORγt is expressed in neonatal inducer cells (11) and is detected in both El 5 and adult

CD4⁺CD3^" cells (Fig. 2a). The levels of mRNA for RORγt were correlated with the

expression of the LTβR-ligands, LTβ and LIGHT, as has been reported for thymocytes

(15).

As the TNFR, DR3 is expressed in mRNA from both juvenile and adult CD4⁺CD3^' cells (Table 1), we tested the effects of adding the DR3-ligand, TLlA, to all populations of CD4⁺CD3^" cells. TLlA was added at 100 ng/ml for 2 days culture, but similar results were obtained from 1 ng/ml (data not shown) (Figs. 2 and 3). In adult CD4⁺CD3^" cell

populations, TLIA downregulated mRNA expression for RORγt and LTβR-ligands and

upregulated the expression of OX40L and TRANCE (Fig. 2b). Addition of TLlA to

El 5 CD4⁺CD3^" cells also downregulated RORγt expression with upregulation of TRANCE but particularly OX40L (Fig. 2c). The effects on mRNA were reflected by changes in protein expression at the cell surface on adult (Fig. 3a), neonatal (Fig. 3b) and El 5 (Fig. 3c) CD4⁺CD3^' cells. In all 3 groups, DR3-signals upregulated OX40L and TRANCE expression, but not CD30L expression. The inventors have previously reported that IL-7 signals upregulate CD30L on neonatal cells, but show here that El 5 spleens cultured with IL-7 (see experimental procedures) did not express CD30L (Fig. 3 c), which clearly need some other signal before they can respond to IL-7.

Only 60% of El 5 populations expressed high levels of OX40L after the addition of TLlA (Fig. 3c), making it possible that there were two precursors within the population defined by their expression of CD4, and absence of CD l ie, B220 and CD3 : one an inducer cell, the other the precursor of the OX40L⁺ CD4⁺CD3^" adult phenotype cell. To examine this possibility we compared the genetic fingerprint of DR3 signaled cells that were OX40L⁺ or OX40L^". With the exception of OX40L, the gene profiles were highly correlated (CC=0.95) (Fig. 2d).

Expression of TNF/TNFR family members

There are currently ~17 identified TNF and ~30 TNFR family members (www.gene.ucl.ac.uk/nomenclature/genefamily/tnftop.html). The gene array focused on TNF/TNFR family members linked with survival (Table 1). TNF family members associated with B cell survival and activation (APRIL, BAFF and CD40L) do not appear in the gene profiles of any of the CD4⁺CD3^" cell populations.

Of the 15 TNF family members on the array, 8 were expressed in mRNA from adult CD4⁺CD3^" cells. OX40L and CD30L are selectively expressed on adult cells and we have linked them with T cell survival for B cell help (10); otherwise embryonic and neonatal populations expressed a similar pattern. A striking feature of this pattern is the

presence of the ligands for TNFRl (TNFα and LTa) and LTβR (LTa and LTβ, and

LIGHT), all of which are linked with the segregation of B:T areas in lymphoid tissue

(4); the LTβR-ligands with TRANCE are linked with lymph node organogenesis (16).

Neonatal and adult CD4⁺CD3^" cells coordinately expressed 7 of the 14 TNFR family members, five of them strongly (++). The TNFR family members can also be grouped

into those involved in lymph node development and B:T segregation (1, 4) (LTβR,

TNFRl and RANK) and those linked with T cell activation (HVEM, TNFR2, 4-1BB, and DR3) (17). These four T cell-associated TNFR family members come from a gene cluster of seven TNFR family members on human chromosome 1 and mouse chromosome 4. Neither neonatal nor adult CD4⁺CD3^" cells express CD30, OX40 or GITR.

Expression of non-TNF/TNFR family genes

Table 2 summarizes the expression of non-TNF/TNFR family genes expressed on CD4⁺CD3^" cell populations. All of them were expressed at comparable levels in adult and embryonic/neonatal populations. There are three gene groups of particular interest: the chemokine receptors, the survival genes, and the GC-specific genes. CD4⁺CD3^" cells express both of the chemokine receptors, CXCR5 and CCR7, but not the pDC related receptor, CXCR3 (18). The inventors looked for but did not find mRNA for the ligands of CXCR5 (CXCLl 3), and CCR7 (CCL21 and CCLl 9), which occur in stromal populations (6, 7).

The survival genes, Bcl-2 and Bcl-xL, appear at high levels in all CD4⁺CD3^" populations. This helps explain the inventors' observation that CD4⁺CD3^" cells survive in culture for at least a week (unpublished observations), and exhibit slow turnover in vivo as assessed by uptake of thymidine analogues (13).

Finally, CD4⁺CD3^' cells express 2 genes associated with GCs: the transcription factor, GAT A-3, is required for Th2 cell development (19); the GC-specific transcriptional repressor, Bcl-6, controls GC formation.

Continued TNFR1 and LTβR signals maintain B:T segregation and the position of CD4⁺CD3^' cells in adult lymphoid tissue

TNFα expression is essential for B:T segregation but not lymph node formation (20).

Although B cells can express TNFα (4), purified B cells expressed insignificant

amounts of mRNA for TNFα compared with neonatal and adult CD4⁺CD3^" cells (Table

1). TNFRl and LTβR signals both contribute to B:T segregation (21), but most of the

effects of TNFRl signals lie upstream of LTβR signals (5). Therefore injection of

LTβR-Ig fusion proteins into adult mice tests the effects not only of LTβR signals, but also of the related component of TNFRl signals, on lymphoid organization (Fig. 4a).

Injection of LTβR-Ig was associated with rapid loss of the B:T interface, and diffusion

of T cells into B cell areas. The architecture was not as disturbed as it is in LTa^{" "} mice,

confirming that there are some effects of TNFRl signals independent of LTβR-

expression(21). The effects on the location of CD4⁺CD3^" cells were also clear (Fig. 4b). In normal mice, CD4⁺CD3^' cells are located at the B:T interface and within B follicles.

In both LTα^"Λ mice and mice injected with LTβR-Ig, CD4⁺CD3^" cells were still found

colocalized with B cells, but there was no longer evidence of alignment where the B:T border had been. CD4^*CD3^~ cells express the chemokine receptor CXCR4

The inventors have linked CD4⁺CD3^" cells with lymphoid organization and T cell memory for antibody responses (9, 10), both of which are characteristically lost during HIV infection. Neonatal and adult CD4⁺CD3^" cells expressed the HIV chemokine receptor, CXCR4, (Fig. 5). The ligand for CXCR4, CXCLl 2, is expressed within GCs and co-operates with CXCLl 3 in GC organization (22). This is likely to be the mechanism responsible for recruiting CD4⁺CD3^" cells into GCs (9).

Discussion

The inventors have demonstrated that embryonic inducer cell populations, responsible for the organogenesis of lymphoid tissue, share with adult CD4⁺CD3^" cells a distinctive genetic fingerprint that is quite different from DCs and pDCs. The difference between inducer and adult CD4 CD3^" cells is the expression of the two TNF family members, a method of identifying a compound which assists in the survival of CD4+CD3- cells, comprising contacting CD4CD3- cells with the compound and determining the effect of the compound on survival of the cells

OX40L and CD30L (14). Previously, they demonstrated that IL-7 induces CD30L expression on neonatal inducer cell populations in vitro (14). Here, they have shown that a signal from another TNF family member, TLlA, through DR3 expressed on both adult and juvenile populations, rapidly upregulates the expression of the T cell survival ligand, OX40L. This provides direct evidence that inducer cells found in the embryo require only two signals to elicit an adult CD4⁺CD3^" cell phenotype.

In neonatal life, inducer cells are found clustered close to blood vessels in spleen (13), a

potential rich source of the DR3-ligand, TLlA (23). TNFα expressed by neonatal inducer cell populations has the potential to induce TLlA (23) and also IL-7 (24) expression on endothelium (23), providing a signaling loop leading to their own maturation to OX40L- and CD30L-expressing adult phenotype CD4⁺CD3^' cells.

B:T segregation, a feature common to avian and mammalian immune systems but not lower vertebrate systems (25), is the platform for the development of both high affinity class switched antibodies and memory antibody responses. Neither of these functions

develops in LTa^{" "} mice, which fail to segregate B and T cells (4). These signals are

required constitutively; as we demonstrate here that acute blockade of LTβR-ligands by

a fusion protein, LTβR-Ig, disrupts B:T segregation in adult mice, as others have found

(26). Experimental evidence suggests that a non-lymphoid cell is responsible: LTα^-/"

lymphocytes segregate normally following adoptive transfer into irradiated wild type mice and produce class switched IgG antibodies following immunization, whereas normal lymphocytes are unable to segregate or respond following transfer into

irradiated LTα^"Λ recipients (27). The constitutive expression of the ligands for TNFRl

(TNFα and LTa) and LTβR (LTa and LTβ, and LIGHT) on adult CD4⁺CD3^" cells

makes them well placed to perform this function.

Primary antibody responses, normal in LTα^";"mice that lack B:T segregation (27),

require only transient T:B interactions to initiate extrafollicular growth of primary plasmablasts (28). In contrast, the process of affinity maturation within GCs requires iterative cycles of B cell proliferation, mutation and selection by CXCR5⁺ T cells (29, 30). B:T segregation, by excluding T cells of irrelevant specificity from the B follicle, provides a local environment into which locally primed CXCR5⁺ T cells migrate and select specific B cells efficiently. B:T segregation also mediates the establishment of the B:T interface that is critical for

memory antibody responses (28), and is maintained by continued LTβR-dependent

signals (5). During memory antibody responses, activated B cells are recruited by their co-expression of CCR7 and CXCR5 (31) to the B:T interface, where they interact with memory T cells (29), and where CD4⁺CD3^" cells are implicated in providing OX40L

and CD30L dependent survival signals to memory T cells (10). In LTα^"Λmice, the

chemokine cues that colocalize memory B (31) and T (32) cells at the B:T interface and enable the memory response are missing; consequently, memory T and B cells

transferred into LTa^{" "}recipients fail to generate a memory response (33). Likewise,

when we disrupted the B:T interface by LTβR-Ig injection, alignment of CD4⁺CD3^"

cells was lost.

The data are consistent with the hypothesis that it is the expression of LTβR- and

TNFRl -ligands by CD4⁺CD3^" cells that provides the segregated B:T infrastructure required for the development of GCs and immunoglobulin class switching. The additional expression of the TNF family members, CD30L and OX40L (9), supports the survival of both the CXCR5⁺ T cells fostering GC development in the B follicles and the T cells providing help for memory antibody responses (10). The profound defect in memory observed in mice that lack B:T segregation suggests that the OX40L and

CD30L T cell support function is likely to have evolved after the TNFRl- and LTβR-

ligand mediated B:T segregation. This proposed evolutionary sequence of events fits well with the observed developmental expression of TNF family members on

CD4⁺CD3^" cells. Whereas TNFRl- and LTβR-ligands are expressed from embryonic stages onwards, OX40L and CD30L only appear postnatally (14), and are regulated differently: DR3 signals upregulate OX40L and downregulate LTβR-ligands. The inventors conclude that CD4⁺CD3^" cells are the adult equivalent of inducer cells are at odds with observations that inducer cells are missing in adult lymphoid tissue (8).

Furthermore, RORγ deficient mice lack inducer cells (11-13) but demonstrate B:T segregation in the spleen (11, 12), a function we have attributed to adult CD4⁺CD3^"

cells. However, histological analysis of spleen tissue from RORγt deficient animals (kindly provided by Dan Littman, New York University School of Medicine) does demonstrate CD4⁺CD3^" cells in B follicles (Gaspal, McConnell and Lane, unpublished

observations). Their interpretation of these data is that RORγt expression is not

essential for the survival and function of CD4⁺CD3^' cells in the spleen, but is required for their differentiation into inducer cells that initiate lymph node organogenesis. The

failure to detect RORγt-GFP expressing inducer cells after birth in lymphoid tissue (13)

is attributable to the DR3 -mediated downregulation of RORγt gene expression in the

mature adult CD4⁺CD3^" cells that we report here, mirroring the situation in T cells,

which express RORγt mRNA as thymocytes (34) but which attenuate its expression as they mature (35). A definitive answer to this question will require isolation and

characterization of CD4⁺CD3^" cells from adult and neonatal RORγt-deficient mice.

In the inventors' view, the most logical interpretation of the data is that the LTβR-

dependent inducer cell function in lymph node organogenesis evolved from a cell

responsible for B:T segregation (LTβR- and TNFRl -dependent) and memory antibody

responses (OX40- and CD30-dependent) in spleen. A comparison of the genomes and immune systems of birds and mammals, which share an ancestor some 300 mYa (36), supports this sequence. Birds demonstrate B:T segregation in the spleen, make GCs and memory antibody responses (25), but lack lymph nodes, present in even the most

primitive mammals (37), an immune system that resembles the phenotype of RORγt deficient mice (11, 12). While they have an orthologue of TNFRl they appear to lack

the LTβR (www.ensembl.org/Gallus_gallus) essential for lymph node development.

This suggests a duplication at the TNFRl locus early in mammalian evolution

facilitated the LTβR-dependent development of a network of organized lymphoid

structures, conferring a selective advantage for mammals able to redistribute memory cells systemically (38), ensuring memory responses were global rather than remaining local to the site of immunization.

The inventors also demonstrate that the adult CD4⁺CD3^" cells, like the LTi with which they share a common phenotype (Kim et al., 2003; Lane et al., 2005), express high

levels of a second set of TNF family genes, LTa, LTβ and TNFα, that are linked with the organized B:T segregation observed in lymphoid tissues (Fu and Chaplin, 1999). In vivo, adult CD4⁺CD3^" cells are tightly linked to VCAM-I⁺ stromal cells in both T and B cell areas. The cells were also found adjacent to DCs and blood vessels in the spleen

and HEVs in lymph node, so are well situated to provide TNFRl- and LTβR-signals to

these cells. Furthermore, injection of either adult CD4⁺CD3^' cells or LTi into mice

deficient in LTa restored a significant degree of B:T segregation, whereas injection of

splenocytes (including B and T cells) or CDl Ic⁺ DC and pDC populations failed to do so.

At first glance these functions, B:T segregation and GC T cell survival, seem unrelated, but both are in fact integral to the development of high affinity antibody responses, which depend on the efficient iterative selection of GC B cells by CXCR5⁺ follicular T cells. By excluding CCR7⁺ T cells of irrelevant specificity but allowing locally primed CXCR5⁺ T cells into the B follicle, B:T segregation enables the process of B cell selection. Crucially, because CD4⁺CD3^" cells are attached directly to both FDCs and the T cells that select GC B cells (Gaspal et al., 2005; Kim et al., 2003), they effectively tether the selecting T cells to the FDC₅ forming the microenvironment for the B cell selection essential for affinity maturation.

The conclusion that CD4⁺CD3^" cells are important for organizing lymphoid tissues is supported by observations that lymph node expression of CCLl 9 and CCL21 is not dependent on B or T cells, unlike the expression of CXCL13 (Ngo et al., 2001) and the

observations that DCs depend on LTβR-signals (Kabashima et al., 2005; Wu et al.,

1999). However, the situation in spleen is less clear cut. Two independent studies demonstrate the importance B cells (Ngo et al., 2001; Tumanov et al., 2002). The latter

study in particular indicated the role of other cells as selective deletion of LTβ on B

cells only partially recapitulated the phenotype of the LTβ -deficient mouse. Furthermore, in the study by Ngo et al., careful dissection of the role of B cells demonstrated their requirement in neonatal life, and injection of normal B cells into RAG^"A mice failed to restore CCL21 expression in the spleen. The inventors hypothesize that this "developmental" effect of B cells might be attributable to the effect of B cells on CD4⁺CD3^' cell numbers which are increased ~6-fold in B cell sufficient mice compared with RAG^V" mice.

Although selective transgenic re-expression of LTa on B cells in LTa^";" mice restores

B:T segregation (Ngo et al., 2001), the inventors believe that, the architecture is abnormal in these mutant mice. The distribution of gp38, a marker for T zone stroma expressing CCR7-ligands, normally found throughout the T zone (Farr et al., 1992), is restricted to the B:T interface (Ngo et al., 2001). As CD4⁺CD3^" cells are aligned at this

B:T interface, this raises the alternative possibility that the LTa signal normally

provided by the CD4⁺CD3^" cells to T zone stroma is provided in trans to stromal cells at the B:T interface by B cells, restoring expression of T zone chemokines (hence B:T segregation) in the gp38 expressing cells at the B:T interface.

A recent publication has indicated an additional role for LTβR-signals in the

homeostasis of CD4⁺ myeloid DCs (Kabashima et al., 2005). Although this study

suggested that B cells provided the critical LTβR-signal for their survival, the inventors

have demonstrated that CD4⁺CD3^" cells cluster with CD4⁺ DCs both in vivo and ex vivo.

Not only are CD4⁺CD3^' cells able to provide LTβR-signals to these cells, but also

because CD4⁺CD3^" cells constitutively express TRANCE (Kim et al., 2003), they can signal survival through the ligand for TRANCE, RANK, expressed on DCs (Josien et al., 1999). TRAF6, a key target for RANK-signaling, is essential for the survival of CD4⁺ DCs (Kobayashi et al., 2003).

A second recent study demonstrated that the homeostatic persistence of high endothelial

venules (HEVs) was dependent on chronic LTβR-signals (Browning et al., 2005).

CD4⁺CD3^" cells are found immediately adjacent to HEVs and are therefore well placed

to provide the LTβR-signals required for their maintenance.

This observation provides a critical link with the cell responsible for lymph node

organogenesis as both share a role in the LTβR-dependent induction and maintenance of

HEVs (Mebius, 2003). LTi interact with VCAM-I⁺ stromal cells in the developing

lymph node to provide the LTβR-dependent signals required to upregulate the

expression of both CCR7- and CXCR5-ligands (Mebius, 2003), which are both essential for recruiting T and B cells to the neonatal lymph node (OhI et al., 2003). Furthermore,

neonatal LTi have also been implicated in the LTβ-dependent B:T segregation in neonatal lymph node (Cupedo et al., 2004). All of these functions are clearly analogous to those that the inventors have described for adult CD4⁺CD3^" cells.

Although adult CD4⁺CD3^" cells and LTi share a similar genotype compared with other cell types, and also share expression of similar set of protein markers at the cell surface, they clearly differ in their expression of the T cell survival proteins, CD30L and OX40L, which may help explain why T cell priming in the neonate results in tolerance rather than autoimmunity (Kim et al., 2005). CD30L expression can be induced on neonatal LTi in vitro with IL-7, and the expression of CD30L in vivo also appears to be IL-7 dependent (Kim et al., 2005). The inventors have now shown that a signal from another TNF family member, TLlA, through DR3 expressed on both adult CD4⁺CD3^" and LTi, rapidly upregulates the expression of the T cell survival ligand, OX40L on LTi and induces further upregulation of OX40L on adult CD4⁺CD3^" cells, hi neonatal life, LTi are found clustered close to blood vessels in spleen (Eberl et al., 2004), a potential rich source of the DR3-ligand, TLlA (Migone et al., 2002). TNFα expressed by

neonatal LTi has the potential to induce TLlA (Migone et al., 2002) and also IL-7 (Kroncke et al., 1996) expression on endothelium, providing a signaling loop leading to their own maturation to OX40L and CD30L expressing adult phenotype CD4⁺CD3^" cells. This conclusion is further supported by the observation that LTi upregulate expression of both OX40L and CD30L to levels comparable to that found on adult cells following adoptive transfer of LTi into adult mice in vivo. Although definitive proof that LTi give rise to adult CD4⁺CD3^" cells awaits cell fate mapping experiments (Eberl and Littman, 2004) the data provided here are suggestive of this sequence of events.

The inventors suggest the following model for the role of B cells and CD4⁺CD3^" cells in adult lymphoid tissue organization and function. In the T zone, CD4⁺CD3^" cells, by virtue of their constitutive antigen-independent expression of an array of TNF ligands, provide survival and differentiation signals to T zone stroma, memory T cells, CD4⁺ DCs and, in lymph node, to HEVs, so maintaining the integrity of the T zone for responses to new and previously encountered pathogens. In the B follicles, the inventors think their role is different. The expression of B zone chemokines depends on B cells activating follicular stromal cells. Furthermore, survival signals to the TNF receptor expressed on B cells, BAFF, are derived from stroma (Mackay and Browning, 2002). CD4⁺CD3^" cells express neither the ligands for BAFF nor the other key TNF receptor involved in B cell selection within GCs, CD40 (Kawabe et al., 1994). However, within the context of the GCs that develop within the B follicle, CD4⁺CD3^"

cells provide the critical link between FDCs (TNFRl- and LTβR-signals) and the

follicular T cells (CD30 and OX40) that select GC B cells by providing transient CD40- rescue signals, (Vinuesa et al., 2005). They therefore form the environment within which GC B cells are selected and without which affinity maturation of antibody responses fails. Seen in the context of selective rescue of GC B cells, it is not surprising that CD4⁺CD3^" cells lack constitutive expression of TNF receptors linked with B cell survival.

The functional description of the CD4⁺CD3^" accessory cell disclosed here is pertinent to the progression of HIV infection in humans to AIDS. The inventors believe that the expression by CD4⁺CD3^" cells of CXCR4⁺, which is associated with the development of normal GCs (Allen et al., 2004), is the link between the emergence of CXCR4 tropic variants of HIV and progressive disease (Nishimura et al., 2005; Nishimura et al., 2004; Scarlatti et al., 1997). HIV localizes on FDCs (Heath et al., 1995) with which CD4⁺CD3^" cells are directly associated and therefore potentially susceptible to infection. Destruction of CD4⁺CD3^" cells, either by CXCR4 tropic HIV variants or, more likely, by the host CD8 immune response that invades B follicles (Koopman et al., 1999), would account for many of the histological abnormalities and functional defects in immune function observed in humans following HIV infection. The signature of progressive disease: loss of the discrete B follicle structure (follicular fragmentation and lysis), disruption of the FDC network (Burke et al., 1994) and loss of the functional capacity to make de novo high affinity antibody responses (Ochs et al., 1988); is predicted by the loss of CD4⁺CD3^" cells. The consequent impairment of the capacity to mount neutralizing antibodies to the mutant CXCR4 tropic virions that emerge during the course of infection would help to explain the failure to control viral replication, with rising viral titers anticipating disease progression.

Although deletion of CD4⁺CD3^" cells in B follicles accounts for the abnormalities in the B cell immune system, it does not explain the susceptibility to opportunistic infections, a feature of CD4 T cell deficiency. The development of AIDS is heralded by a steep decline in T cell numbers and it has been suggested that this is due to direct T cell infection (Nishimura et al., 2005; Nishimura et al., 2004). However, there is a strong correlation between the benign nature of the disease observed in natural simian hosts and the preservation of the infrastructure of lymph nodes (Hirsch, 2004; Koopman et al., 1999). The inventors believe that the decline in T cell numbers in humans might be due at least in part to depletion of CD4⁺CD3^" cells in the T zone. This results not only in a deficit in the OX40- and CD30-signals that maintain memory T cell survival in that location but also the consequent decreased expression of CCR7-ligands by T zone stroma leading to failure to recruit DCs and T cells into secondary lymphoid organs. This view is supported by the observed decline in expression of these chemokines accompanying the progression to AIDS in a simian model (Choi et al., 2003). Finally, loss of these cells in the T zone predicts loss of HEVs, a feature which postdates lymphocyte loss from lymph nodes (Csanaky et al., 199_.1).

In summary, the inventors have demonstrated that adult CD4⁺CD3^' cells express a

highly distinctive set of immunity-related genes, including RORγt, which is shared with

the inducer cells essential for lymph node organogenesis. Furthermore, we have identified signals that directly induce the expression of the T cell survival ligands, OX40L and CD30L, transforming the inducer cell to the adult CD4⁺CD3^" phenotype. An imrnunopathology of CD4 CD3^" cells could be relevant to the immunodeficiency associated with HIV infection. Murine CD4⁺CD3^" cells express CXCR4 in addition to CXCR5 and CCR7. The human equivalent (39) is therefore a potential target for HIV, which is trapped on follicular dendritic cells (40) in close proximity to CD4⁺CD3^" cells. Destruction of CD4⁺CD3^" cells, either by the virus or by a host CD8 immune response (41), would account for both the defect in antibody production (42) and the loss of lymphoid architecture (43) characteristic of disease progression to AIDS.

All of the major abnormalities in lymph node architecture in HIV infection can be explained by loss of a human CD4⁺CD3^"CXCR4⁺cell that functions in an equivalent way to that which the inventors describe for the murine cell. The inventors have observed (unpublished observations) as have others (Grouard et al., 1996) a CD4⁺CD3^" cell in the B follicle and GCs of human lymphoid tissue that interacts with GC T cells, but it, unlike the murine CD4⁺CD3^" cell, expresses the integrin, CDl Ic. There is also some evidence that the human LTi is CD4⁺CD1 Ic⁺CD3^" (Spencer et al., 1986). With regards to the pathogenesis of HIV the inventors suggest that these cells are depleted in human and animal models of AIDS, and whether they are preserved in the tissues of natural hosts that harbor the virus but fail to develop either the abnormalities in lymphoid tissue architecture or immunodeficiency associated with pathogenic infection.

Additional data:

CD4⁺CD3^"cell interactions with primed CD4⁺ T cells.

Digital confocal microscopy of spleen sections enabled the identification of interactions between T cells in GCs and a CD4⁺CD3^" cell population with dendritic morphology but lacking the mouse dendritic-cell (DC) marker CDl Ic 59. CD4⁺CD3^" cells could be found not only in the B-cell follicles and GCs₅ where they were interacting with follicular T cells, but they were also aligned at the B:T interface where the T-cell zone and B-cell follicle adjoin, which is a site of T:B collaboration in both primary and secondary antibody responses 60,61 (Figure 6). Priming of marked transgenic CD4⁺ T cells occurred first on the conventional DC population in the T-cell zone, following which some of the CD4 T cells migrated to the outer T-cell zone and into B-cell follicles, where a substantial fraction was found to interact with CD4⁺CD3^" cells 59.

Due to technical difficulties in isolating CD4⁺CD3^" cells from mice with an intact CD4⁺ T-cell repertoire, these cells were first isolated from T-cell-deficient mice. Two CD4⁺ populations could be distinguished from CDl Ic⁺ bright DCs. One expressed B220 (CDl Ic low) and was identified as the plasmacytoid DC population 62; the other was CD4⁺CD3^"CD1 lc-β220TL-7R⁺. Humans also have a distinct CD4⁺CD3^"cell population within GCs (63). Although the human cells express CDl Ic, they express CD4 and their location is the same, so we regard them as equivalent to the mouse CD4⁺CD3^" population.

Which signals do primed T cells receive from CD4⁺CD3^" cells?

Because of their close association with primed T cells, both inside B-cell follicles (follicular T cells) (Figure 7) and at the interface between the B- and T-cell areas (newly primed and recirculating memory T cells) (Figure 8), it seemed plausible that CD4⁺CD3^" cells provide co-stimulatory signals to T cells.

Unlike conventional DCs, which express and can be activated through CD40, CD4⁺CD3^" cells lack significant CD40 expression (59). Furthermore, their expression of the conventional DC-associated costimulatory molecules such as CD80 and CD86 is low (59). However, two T-cell costimulatory molecules are expressed at high levels by CD4⁺CD3^" cells: OX40-ligand (OX40L; TNFSF4) and CD30-ligand (CD30L; TNFSF8). These are both members of the TNF family, whose receptors (OX40 and CD30) are expressed by primed but not naϊve T cells. In the context of the GC environment, OX40L can also be expressed by activated B cells 65. Uniquely, however, the expression of both OX40L and CD30L by CD4⁺CD3^" cells is high and constitutive, is independent of antigen activation and is not modified by cytokines, particularly IL-4 (65). Also, it has been shown from the histology that there is membrane contact between OX40^hi CD4⁺ T cells and OX40L⁺ CD4⁺CD3^" cells that would enable delivery of an OX40 signal to the primed T cells (59). Function of OX40 and CD30 signals delivered by CD4⁺CD3^~

cells.

OX40 and CD30 (the receptors for OX40L and CD30L expressed by CD4⁺CD3^" cells) are genetically linked in a TNF-receptor (TNFR) cluster containing 7 members, located on mouse chromosome 4 and human chromosome 1. Like many members of the TNFR family, they share common signaling pathways: both bind to members of the TNF- related adaptor factor (TRAF) family (TRAFl, TRAF2, TRAF3 and TRAF5) (66,67), and OX40 signals have been shown to upregulate expression of anti-apoptotic Bcl-2- family members that promote survival (68). Given that activated T cells can express both receptors, and that CD4⁺CD3^' cells express both ligands, it is expected that there should be partial redundancy between OX40 and CD30 signals, as shown by the results from single- and double-deficient mice.

Th2-cell survival in vitro.

CD4⁺ T cells activated in vitro upregulate expression of both OX40 and CD30, but their expression is particularly marked on Th2-differentiated cells (cultured in the presence of IL-4) 59. Although many studies have linked OX40 and CD30 signals with preferential Th2-cell development in vitro, CD4⁺ T cells deficient in OX40 and/or CD30 can still proliferate and differentiate into Th2 cells 59, 69. The most important difference that these signals make is to the survival of Th2, but not ThI cells. Co-culture of normal Th2-differentiated cells with CD4⁺CD3^" cells shows independent additive effects of OX40 and CD30 signals on Th2-cell survival (69).

In keeping with the survival effects of OX40 and CD30 signals in vitro, in vivo analysis of the immune response of mice deficient in both OX40 and CD30 showed independent and additive effects of these genes on two aspects of T-cell help for B cells (69): affinity maturation and memory T-cell help for secondary antibody responses.

Affinity maturation supported by follicular T cells.

The survival of the follicular T cells that orchestrate affinity maturation of GC B cells depends on additive signals through CD30 and OX40 (69). Transgenic T cells deficient in both CD30 and OX40 signals initially proliferated normally, but like their in vitro counterparts, they failed to survive. As a consequence, although T- cell help for primary antibody responses was normal, and GC development was initiated, GCs involuted after 7 days. GC failure led to impaired affinity maturation of antibody responses. OX40- and CD30-deficient CD4⁺ T cells in vivo had restricted survival in B- cell follicles, indicating that OX40 and CD30 have overlapping roles in the maintenance of follicular T cells. These data support the in vitro observation that the role of OX40 and CD30 signals from CD4⁺CD3^" cells, which express OX40L and CD30L, is to keep follicular T cells alive.

It is not difficult to see why follicular T-cell survival is important for the affinity maturation of antibodies in GCs. As affinity maturation progresses, the antigen driving the reaction becomes scarce (antigen is removed by phagocytes and masked by secreted antibodies). It is precisely at this time that follicular T cells are needed to select rare B- cell mutants with high-affinity receptors. Rapidly proliferating GC B centroblasts exit the cell cycle and differentiate into centrocytes that compete for antigen fragments trapped on follicular dendritic cells. Successful B-cell mutants bearing high-affinity receptors are able to take up antigen and successfully present peptide fragments to follicular T cells, which then provide selective CD40L-dependent rescue signals. The constitutive expression of OX40L and CD30L by CD4⁺CD3^" cells, irrespective of antigen concentration, equips them to provide antigen-independent survival signals to follicular T cells so that they can continue to select B cells (Figure 7).

T cell help for memory antibody responses.

Memory antibody responses in OX40 and CD30 double-deficient mice are markedly reduced (69). By contrast, defects in memory antibody responses in mice deficient in either OX40 or CD30 were sufficiently mild to be ignored by many investigators who studied OX40 70-73 or CD30 (74, 75) single-deficient mice.

Memory T cells that provide help for B-cell responses recirculate through lymphoid organs (76) ensuring that the production of memory antibody responses does not depend on the site of initial immunization. The effects of OX40 and CD30 deficiency on memory antibody responses indicated that CD4⁺CD3^" cells might specifically influence the survival of these memory T cells, as well as the follicular T cells involved in affinity maturation. Evidence from mouse studies indicates that recirculating memory T cells do not arise from the CXCR5-expressing follicular T-cell population responsible for affinity maturation in GCs, which does not recirculate (77). A subset of memory T cells that home to lymph nodes express both CXCR5 and CCR7 (78). Co-expression by B cells of CCR7 (the ligand for which produced by stromal cells in the T-cell zone (79)) and CXCR5 (the ligand for which is produced by stromal cells in the B-cell areas (80)) results in the alignment of B cells at the B:T interface (81), and this seems also to be the case for the CXCR5⁺CCR7⁺ memory T cells in mice (Reinhold Forster, personal communication) . In addition to being located in B-cell follicles adjacent to follicular T cells, CD4⁺CD3^' cells are also found at the B:T interface. Although mouse reagents are not available to check for surface CCR7 expression, there is evidence that CCR7 mRNA is expressed by the CD4⁺CD3^" cells (Kim, M. and Lane, P., unpublished observations). So, similar to their B- and T-cell counterparts, there are probably two subsets of CD4⁺CD3^" cells; one CXCR5⁺ subset localized in B-cell follicles that interacts with CXCR5⁺ follicular T cells, and the other subset at the B:T interface that co-expresses CXCR5 and CCR7 and interacts with CXCR5⁺CCR7⁺ recirculating memory T cells.

Recirculating memory T cells do not normally express either OX40 or CD30, but can be induced to re-express OX40 (69) when exposed to the common gamma-chain signaling cytokine, IL-7, which has been implicated in the maintenance of memory CD4⁺ T cells (82-84). We think that memory T cells, co-expressing CXCR5 and CCR7, upregulate OX40 in response to IL-7 signals, perhaps from stromal cells or follicular dendritic cells (85), as they migrate through secondary lymphoid organs, allowing them to receive OX40-dependent survival signals from CD4 CD3^" cells each time they passed through lymphoid tissue (Figure 8). T-cell-dependent memory B-cell responses are therefore also promoted by CD4⁺CD3^" cell-mediated survival of memory T-cell responses.

Neonatal CD4⁺CD3^" cells do not express OX40L or CD30L.

In contrast to adult mice, CD4⁺CD3^" cells isolated from neonates lack expression of the molecules associated with T-cell memory, OX40L and CD30L, indicating that the expression of these molecules is developmentally regulated (86). Their absence is specific: the expression of TNF -family ligands associated with the development of lymph nodes is otherwise comparable in adult and neonatal CD4⁺CD3^" cell populations (59). This may help to explain the observations of Medawar, who reported 50 years ago that immunization of neonatal rodents resulted in tolerance rather than immunity (87). The inventors predict that T-cell help for B-cell responses would effectively be aborted in neonatal rodents as a consequence of the absence of T-cell survival signals through OX40 and CD30 from CD4⁺CD3^" cells. This does not render the neonate immunodeficient because of protection from maternal antibodies at this time. By the time of weaning, expression levels of OX40L and CD30L on CD4⁺CD3^" cells are normal (86) and mice become immunocompetent to respond.

Are CD4⁺CD3^" cells related to the cells that induce lymph

nodes?

Detailed phenotyping of adult CD4⁺CD3^" cells revealed that they are similar to a CD4⁺ population first described by Mebius to colonize neonatal lymph nodes (88) but also found in neonatal spleen (89) in mice. The name "inducer" was later coined for these cells because of their essential role in the induction of lymph nodes, Peyer's patches (58) and, more recently, isolated lymphoid follicles in the submucosa (90).

The splice variant of the retinoic orphan receptor gamma (RORγt) is expressed by

inducer cells and required for their function in the induction of lymph nodes (91, 92).

Using a green fluorescent protein (GFP) marker to identify RORγt-expressing cells

specifically, Eberl and Littman (90) reported a phenotype for these cells similar to the CD4⁺CD3^" cells that we have identified in adults (59). Evidence from adult (59) and neonatal (88-90, 93) mice indicates that the shared phenotype between these CD4⁺CD3^" populations is: RORγt⁺, CD45⁺, CD4⁺, LTa⁺, LTβ⁺, TRANCE⁺, c-kit⁺, IL7-Ra⁺, IL2-

Ra⁺, common cytokine receptor gamma chain⁺, CXCR5⁺, CCR7⁺, α4β7⁺, Thyl.2⁺,

CDl Ib^", CDl Ic^", B220^", NKl .1^", CD8^" and CD3^". We have found that RORγt-specific

mRNA is expressed by adult CD4⁺CD3^" cell populations, albeit at lower levels than in neonates (Kim, M. and Lane, P., unpublished observations).

The expression of TNF -family members linked with the development and organization

of lymphoid tissue (58) — lymphotoxin-α (LT-α; TNFSFl), LT-β (TNFSF3) and

TRANCE (TNFSFl 1) — is also common to both adult and neonatal CD4⁺CD3^" cells. These data are consistent with adult CD4 CD3^" cells being related to neonatal inducer

cells. A difficulty with this argument is that RORγt-regulated GFP-expressing inducer

cells are reported to be absent from adult secondary lymphoid organs (Dan Littman,

personal communication). This may be because RORγt expression is attenuated in adult

CD4⁺CD3^" cells as it is in T cells, which express RORγt as double-positive thymocytes

(94), but which lose expression as they mature (94).

Immune memory and high affinity antibody responses

antedated lymph node development

Assuming that adult CD4⁺CD3^" cells and neonatal inducer cells are related, what was the original evolutionary function of cells of this phenotype? Birds and mammals evolved from a common reptilian ancestor ~300 million years ago (95). In terms of their immune responses, both form GCs when immunized, make high-affinity class- switched antibodies and have B- and T-cell memory. However, birds lack lymph nodes, unlike the most primitive mammals, the monotremes (96). This indicates that the genetic and cellular mechanisms for maintaining CD4⁺ memory T cells to support antibody production were established before birds and mammals went their separate evolutionary ways.

The key advantage of mammalian immunity is that the major investment made during the GC reaction in terms of B-cell affinity maturation is not lost; the memory response is systemic, not local to the site of initial vaccination, because memory lymphocytes recirculate (76). Once local immune memory was established, one could envision a selective survival advantage for animals able to redistribute the memory response globally. Lymph nodes provide the infrastructure for a systemic response; the blood vessels and lymphatics supply the conduit.

Genetic paleontology supports this as an evolutionary sequence. The chicken genome published in silico (ensembl.org/Gallus_gallus/) contains orthologues of many genes that are essential for GC development (such as CD28 and CD40), and also orthologues of CD30 and OX40, which we have identified as important for T-cell memory. In

contrast to these genes, the master gene for lymph-node development, LTβR, seems to

be missing from the chicken genome. Therefore, a cell type to control lymph-node development (the neonatal inducer cell) may have arisen simply through the evolution of a new function for an old cell type: the adult CD4⁺CD3^' cell that regulates the development and maintenance of memory antibody responses.

Identification of CD4+CD3- cells in humans

The identification of the cells in humans is summarised in Figure 9. The cells were identified in suspensions of cell suspensions of human lymph nodes of donors from the liver transplant program where relatives had informed consent.

Analysis of cell populations allowed the inventors to identify a CD4+IL7Ralpha+cKIt+CXCR5+thyl+ cell that closely resembles the murine CD4+CD3- cell that had previously been identified in mice. Use gene microarrays to fingerprint these cells, we found that they expressed a very similar pattern of genes to the mouse population i.e. TRANCE+, Rank+, DR3 (TNFRSF25)+, CXCR4+ amongst other genes. For this reason they are thought to represent the human equivalent of the cell.

Evidence that TNFR1- and LTβR-signals from a non-B non-T cell plays a critical role In B:T segregation

Many groups have demonstrated that continuous LTβR-signals are required to maintain B:T segregation and B follicles (Mackay et al., 1997; Ngo et al., 1999). The inventors

confirmed these findings by acutely blocking LTβR-signals (data not shown). Injection

of LTβR-Ig was associated with rapid loss of discrete B follicles and VCAM-I⁺ FDCs

as reported by others (Mackay and Browning, 1998). LTα^"A mice, which also have low

levels of TNF α (Alexopoulou et al., 1998) and are therefore deficient in both TNFRl-

and LTβR-signals, show an even greater degree of disorganisation of lymphocytes

(Figure 12Ab), confirming that these two signalling pathways have independent effects on lymphoid organization (Koni et al., 1997; Kuprash et al., 2002).

Although lymphocytes, particularly B cells, express both TNFRl and LTβR-ligands

(Figure 13A) (Fu and Chaplin, 1999), injection of normal splenocytes into LTα^"Λ mice failed to restore a B:T segregated architecture (Figure 12Ac) as reported previously (Fu et al., 1997). RAG^"7" mice lack B and T cells but other cell types are present. To test whether the environment within RAG^"'^" mice was capable of segregating B and T cells,

splenocytes from LTa^"7" mice were transferred into RAG^"7" mice. 10 days after cell

transfer, the reconstituted spleens were examined for lymphocyte architecture. Whether

reconstituted with LTa^"7" (Figure 12Ad) or normal (data not shown) lymphocytes, RAG^"

^7" spleens demonstrated clear B:T segregation by comparison with spleens from LTa^"7"

mice (Figure 12Ad).

Adult CD4^*CD3^~ and embryonic/neonatal LTi but notpDCs or DCs express high levels of LTa, LT β and TNFα

To investigate which non-B non-T cell in adult tissue expressed LTa, LTβ and TNFα,

the inventors purified CD8⁺CD1 lc⁺B220^" and CD8^"CD1 lc⁺B220^" DCs, pDCs (CD4⁺CD1 lc^IowB220⁺) and the CD4⁺CD3^" accessory cells that the inventors have recently characterized (CD4⁺CD3^'CD1 lc^"B220^") from RAG^"7" mice. TaqMan quantitative PCR probes are designed to work with 100% efficiency, allowing comparison of the levels of expression of different genes after correction for total mRNA expression. Results are shown in Figure 13 A. Adult CD4⁺CD3^" cells expressed high

levels of LTa by conventional semiquantitative PCR (Table 1), TNFα, LTβ and LIGHT

(also a LTβR-ligand), and levels were comparable to those expressed in either El 5 or

neonatal LTi. In contrast, the expression of these molecules on pDC, NK₁ or DC subpopulations was at least an order of magnitude less. Table 1 shows TNF and TNF-receptor (TNF/TNFR) family members and Table 2 non- TNF/TNFR family members. Expression ofmRNA in El 5 inducer cells was similar to neonatal inducer cells (data not shown). The analysis focused on genes expressed at

high levels (mRNA expressed at >0.2% of the β-actin signal). To simplify analysis,

gene expression has been categorized into 4 groups relative to expression of β-actin:

+++ > 10%, ++ 1-10%, + 0.2-1%, - <0.2%.

Adult CD4⁺CD3^' cells but not splenocytes are capable of organizing lymphocytes in LJd^1" mice

The strong expression of both TNFRl- and LTβR-ligands on both LTi and adult

CD4⁺CD3^' cells suggested that they might be capable of organizing LTα^"A lymphocytes.

To test this directly, CD4⁺CD3^" cells from RAG^"A mice were transferred intravenously

into LTa^"7" recipient mice. 10 days post transfer there was clear evidence of B:T '

segregation in LTα^"Λ mice (Figure 12Af). In contrast, mice that received CDl Ic

enriched fractions (CDl Ic⁺ DCs and CDl lc^low pDCs as controls) or splenocytes showed little evidence of B:T segregation (Figure 12Ae).

As outlined in the methods section, for each experiment, 10 areas of splenic white pulp

from reconstituted LTα^";"mice were analyzed to identify the degree of B:T segregation.

Compared with control LTα^"Λ spleen sections, white pulp areas containing B and T

lymphocytes were significantly larger in mice reconstituted with CD4⁺CD3^" cells (p=0.005) and E15 LTi (p=0.001) but not CDlIc⁺ populations (p=0.07) or splenocytes (p=0.28) (Figure 13Ba). This was due to significantly increased T cell free B cell areas for CD4⁺CD3^" cells (p=0.005) and LTi (p=0.002) but not CDl Ic⁺ cells (p=0.6) or splenocytes (p=0.28) (Figure 13Bb). The B cell free T cell areas were also significantly bigger in the spleens transferred with CD4⁺CD3^" cells (ρ=0.008) and LTi (ρ=0.004) and also CDl Ic⁺ cells (p=0.04) but not splenocytes (ρ=0.39) (Figure 13Bc).

CD4^*CD3~ cells are closely associated with VCAM-1^* stromal cells in B follicles and also in the T zone

In the developing embryo, LTi interact with stromal cells in lymph node anlagen to upregulate the expression of the chemokines that recruit lymphocytes to form lymph nodes (Mebius, 2003), and there is evidence that LTi are responsible for B:T segregation in the neonatal lymph node (Cupedo et al., 2004). Effective delivery of

LTβR-signals to these stromal cells involves interactions between the LTi cell integrin,

(X₄P₁, and its ligand, VCAM-I, which is expressed on stroma (Finke et al., 2002). The inventors reasoned that adult CD4⁺CD3^" cells would function in a similar way. To examine their relationship with VCAM-I⁺ stromal cells the inventors first identified VCAM-I⁺ populations in normal adult mice. The red pulp of the spleen exhibits strong staining for VCAM-I cells, but there is also generally less intense staining in white pulp areas, with discrete staining in both B follicles (including FDC populations) and T

cell areas (Figures 12Ba). Comparable to normal mice, LTα^"Λ mice show strong

staining for VCAM-I in the red pulp of the spleen, but VCAM-I staining is largely missing from white pulp areas, which although lymphocyte rich, show no segregation of

B and T cells (Figure 12Bb). In mice injected with LTβR-Ig, where LTβR- but not

TNFRl- signals are blocked, VCAM-I expression is maintained in both T zone and red pulp areas but there is selective loss of VCAM-I expression within B follicles consistent with rapid loss OfVCAM-I⁺ expressing FDCs (data not shown) (Mackay and Browning, 1998).

To examine the relationship between CD4 CD3^" cells and the VCAM-I⁺ stromal cells the inventors examined sections of splenic tissue from T and NK cell deficient mice (Hollander et al., 1995), where the absence of CD4⁺CD3⁺ T cells in the DC rich areas makes it straightforward to identify other CD4⁺ cells (Figure 12Bc). As reported previously (Kim et al., 2003), the inventors found CD4⁺CD3^" cells in B follicles, and these were closely associated with VCAM-I⁺ cells (Figures 12Bc (low magnification) and 12Bd (high magnification)). However, in the T and NK cell deficient mice the inventors were also able to identify a similar association between a CD4⁺CD3^' population and the local VCAM-I⁺ population amongst the CDl Ic⁺ DCs in the area populated by T cells in normal mice (Figures 12Bc (low magnification) and 12Be (high magnification)) .

To identify further the T zone-associated CD4⁺CD3^" cell population, the inventors positively selected CD4⁺ cells from T cell deficient mice (Hollander et al., 1995). Four populations could be identified. The first is the CD4⁺CD3^" cell population which the inventors have characterized previously (Kim et al., 2003), that lacks expression of B220 and CDl Ic but which expresses high levels of OX40L and CD30L (Figure 14Aa). The second is the pDC population that expresses B220 and low levels of CDl Ic but lacks expression of OX40L and CD30L (Figure 14Ab) and the third is CD4^lowCDl Ic⁺ myeloid DC that expresses low levels of OX40L but not CD30L (Figure 14Ad). Whereas the first three populations are homogeneous with respect to size, the fourth population (CD4⁺CDl lc⁺) is heterogeneous, having two populations (Figure 14Ac). The smaller population closely resembles classical myeloid DC phenotype cells, whereas the larger population looks like clusters of CD4⁺CD3^" cells and myeloid DCs (its phenotype is mixture of the phenotypes of these two cell types). Although the clusters expressed slightly higher levels of B220, the inventors think this reflects the increased fluorescence of the large clusters.

To positively identify CD4⁺CD3^" cells in T cell areas, the inventors stained sections from mice deficient in T and NK cells (Hollander et al., 1995; Wang et al,, 1996) and excluded CD3, CDl Ic and B220 using FITC conjugated antibodies (green) and counterstained with CD4 (red) (Figure 14Ba). Although some CD4⁺ cells expressed either CDl Ic or B220 (yellow), there are also many red cells that lack expression of either B220 or CDl Ic indicating that CD4⁺CD3^" cells are located in the T zone. The location of CD4⁺CD3^" cells in the T cell areas was not an artefact of T cell deficient mice, as careful analysis of normal mouse spleen revealed CD4 CD3^" cells within the T cell areas (Figure 14Bb). These data indicate that CD4⁺CD3^" cells associate with stroma in both B and T cell areas and are therefore well positioned to provide the TNFRl- and

LTβR- signals that induce chemokine expression. Furthermore, like LTi which are

found clustered around blood vessels in neonatal spleen (Eberl et al., 2004), CD4⁺CD3^" cells are also found around central arterioles in the spleen (Figures 14Ba and 14Bb) and in lymph node are found associated with PNAd⁺ high endothelial venules (HEVs) (Figure 14Bc).

LTi and adult CD4^*CD3^' cells display similar genetic fingerprints which readily distinguish them from other cell types

The data above suggested that not only could CD4⁺CD3^" cells supply the same function in B:T segregation within the adult lymphoid tissue as the LTi in the neonatal lymph node, but that neonatal LTi could also organize the spleens of adult UTa ^1' mice.

Compatible with the view that they are related cells, the two cell types share a common

phenotype that includes CD4⁺CD3^"CD1 lc^"B220^"IL-7Rα⁺, common cytokine receptor γ-

chain (γc)⁺ ₅ CD45⁺, Thyl⁺, TRANCE⁺, RANK⁺, and MHC class II^low (Lane et al.,

2005). Although relatively few genes are cell-specific, the inventors reasoned that the level oϊmRNA expression for a set of genes would be correlated in cells of related lineage, allowing them to be identified by their genetic fingerprint. To test this, the inventors designed TaqMan arrays for a panel of immunity-related genes (see materials and methods). Comparison of genes expressed within the two major subsets of DCs in mice, CD8⁺CD1 Ic⁺ and CD8^"CD1 Ic⁺ DCs (correlation coefficient (CC) = 0.94), ThI and Th2 differentiated T cells (CC = 0.86), and marginal zone and follicular B cells (CC = 0.95), revealed a strong correlation coefficient for the related cells (Figure 15), but there was little correlation between cells of different types (see Figures 10 and 11), lending validity to the use of the fingerprinting method. Comparison of gene expression between adult CD4⁺CD3^" cells and either CD8⁺ DCs (CC = 0.68), pDCs (CC = 0.59), NK (CC = 0.66), Th2 (CC = 0.63), ThI (CC = 0.76), follicular B (CC = 0.70), and marginal zone B cells (CC = 0.65) showed much weaker correlation coefficients. In contrast, after OX40L and CD30L gene expression was excluded from the analysis, the gene expression in adult CD4⁺CD3^" cells was strongly correlated with embryonic (El 5) spleen LTi (CC = 0.86), and with neonatal spleen LTi (CC = 0.90); neonatal lymph node and neonatal spleen LTi were also strongly correlated (CC = 0.88). Although there was some evidence that adult CD4⁺CD3^" cells were correlated with CD8^" DCs (CC = 0.81), the inventors think this reflects contamination of myeloid DCs by CD4⁺CD3^" cells with which they are clustered in vitro and in vivo (Figure 15). The details of the distinctive gene profile of TNF and TNF-receptor (TNF/TNFR) family members established for the CD4⁺CD3^" cell type are tabulated (Table 1). Expression of mRNA in El 5 LTi was similar to neonatal LTi (data not shown). The inventors' analysis focused on genes expressed at high levels (mRNA expressed at

>0.2% of the β-actin signal). To simplify analysis, gene expression has been

categorized into four groups relative to expression of β-actin: (1) +++ >10%, (2) ++ 1-

10%, (3) + 0.2-1%, (4) - <0.2%.

RORγt mRNA is expressed in adult CD4⁺CD3T cells and DR3 signals u p regulate OX40L expression on LTi

Because of the similar genetic fingerprint of LTi and adult CD4⁺CD3^" cells the inventors looked for mRNA expression of the splice variant of the retinoic acid orphan receptor,

RORγt (Sun et al, 2000), a gene essential for the function of LTi in lymph node

development. Both adult CD4⁺CD3^" cells and embryonic/neonatal LTi expressed mRNA

for RORγt but levels were 4-fold higher in embryonic LTi than in adult CD4⁺CD3^" cells

(Figure 16Aa) and both expressed both TNFRl- and LTβR- ligands (Figures 13 A,

16Aa, and Table 1).

The key difference between LTi and adult CD4⁺CD3^' cells is that the former lack expression of the T cell survival TNF ligands, OX40L and CD30L (Figure 16Aa) (Kim et al., 2005). The inventors have previously found that IL-7 signals upregulate CD30L expression on neonatal LTi, but these signals had no effect on OX40L expression. The TaqMan low density arrays had demonstrated mRNA for the TNFR family member, DR3 (TNFRSF25), on both LTi and adult CD4⁺CD3^" cells (Table 1), so the inventors tested the effects of the recombinant TNF ligand, TLlA (TNFSF15), on embryonic/neonatal LTi and CD4⁺CD3^" cells. TLlA was added at 100 ng/ml for 2 days in culture (Figures 16Ab and 16Ac), and similar results were obtained with TLlA added at 1 ng/ml (data not shown). In adult CD4⁺CD3^" cells, TLlA downregulated mRNA

expression for RORγt and upregulated the expression of OX40L and TRANCE (Figure

16Ab).

Addition of TLlA to El 5 LTi also downregulated RORγt expression with upregulation

of both TRANCE and particularly OX40L (Figure 16Ac). The effects on mRNA were reflected by changes in protein expression at the cell surface on adult splenic CD4⁺CD3^" cells (Figure 16Ba), adult lymph node CD4⁺CD3^" cells (Figure 16Bb), neonatal LTi (Figure 16Bc) and El 5 LTi (Figure 5Bd). In all three groups, 48 hour DR3-signals upregulated OX40L and TRANCE expression but had little effect on CD30L expression. Co-stimulation of neonatal LTi with TLlA for 6 days upregulated OX40L, TRANCE and also CD30L, whereas IL-7 alone upregulated CD30L and TRANCE but not OX40L (Figure 16Be). Together IL-7 and TLlA showed additive effects.

However, the fact that mice deficient in γc or IL-7 signals have normal levels of OX40L

but not CD30L and TRANCE show an important role for IL-7 signals in CD30L expression (Kim et al, 2005), and therefore the effects of TLlA in the 6 day experiments may be indirectly mediated through IL-7.

Because only 60% of El 5 LTi populations expressed high levels of OX40L after the addition of TLl A (Figure 5Bd), it was possible that there were two precursors within the population defined by their expression of CD4, and absence of CDl Ic, B220 and CD3: one an LTi cell, the other the precursor of the OX40L⁺ CD4⁺CD3^" adult phenotype cell. To examine this possibility the inventors compared the genetic fingerprint of DR3 signaled cells that were OX40L⁺ or OX40L^". With the exception of OX40L, the gene profiles were highly correlated (CC = 0.95) (Figure 16Ad), suggesting a single population of cells. The inventors conclude that the reason that not all LTi upregulate OX40L is technical and related to the diffusion of TLlA into the embryonic spleen fragments. In pilot experiments, when TLlA was added to whole embryonic spleens, OX40L induction was only seen on a small fraction of the LTi; this fraction increased substantially to the levels reported when the cultured El 5 spleens were teased apart (see materials and methods).

Fetal LTi upregulate expression of both OX40L and CD30L after transfer into adult recipients

To test directly whether LTi were capable of upregulating OX40L and CD30L in vivo, LTi were prepared from CD45.2 El 5 spleens, and transferred into an adult CD45.1 recipient that lacked T cells (isolation of CD4⁺CD3^" populations from T cell sufficient mice is technically difficult). 5 days later, CD4⁺ cells were enriched from the spleen, and stained with the allotype marker to identify transferred LTi, and OX40L and CD30L (Figure 16C). Transferred LTi were clearly identifiable in adoptive recipients, and whilst they were negative for CD30L and OX40L prior to cell transfer, they showed expression levels of OX40L and CD30L comparable to host adult CD4⁺CD3^" cells, indicating that fetal LTi acquire hallmarks of adult CD4⁺CD3^" cells in vivo.

Adult CD4^*CD3^' cells and LTi share expression ofchemokine receptors and survival proteins and express CXCR4 and the numbers of these cells in the spleen are augmented by B cells

The survival genes, Bcl-2 and Bcl-xL, are expressed at high levels in both adult CD4⁺CD3^" cells and LTi (Figure 17A). This helps explain our observation that CD4⁺CD3^" cells survive in culture for at least a week (unpublished observations), and correlates with the slow turnover observed in vivo for LTi as assessed by expression of the division marker, Ki67 (Eberl et al., 2004).

CD4⁺CD3^" cell location in B and T cell areas fits with their expression of both T zone (CCR7) and B follicle (CXCR5) chemokine receptors but they do not express the pDC related receptor, CXCR3 (Cella et al., 1999). Again this expression pattern was shared with LTi. The inventors looked for but did not find mRNA for the ligands of CCR7 (CCLl 9), and CXCR5 (CXCLl 3), which occur in stromal populations (Figure 6A) (Gunn et al., 1998; Luther et al., 2000). By staining with monoclonal antibodies, the inventors also identified expression of CXCR4 on both LTi and adult CD4⁺CD3^" cells (Figure 17B).

Previously, the inventors have reported that the yield of CD4⁺CD3^" cells from B cell sufficient mice was greater than from RAG^";"mice (-1.5:1) (Kim et al., 2005). However, this does not take into account the relative efficiency with which these cells are isolated in these different mouse strains. To control for this the inventors made a cell preparation from 2 spleens (the first from a CD45.2 RAG^";" mouse and the second from a CD45.1 B cell sufficient mouse deficient in T cells (Hollander et al., 1995)). Isolation of CD4⁺CD3^" cells from this mixture revealed a 6: 1 excess of cells derived from the B cell sufficient mouse indicating a clear effect of B cells on splenic CD4⁺CD3^" numbers (Figure 17C). Further support that all of the major abnormalities in lymph node architecture in HIV infection can be explained by loss of a human CD4⁺CD3^"CXCR4⁺ cell that functions in an equivalent way to that which are described for the murine cell (below).

The inventors have observed (unpublished observations) as have others (Grouard et al., 1996) a CD4+CD3-cell in the B follicle and GCs of human lymphoid tissue that interacts with GC T cells, but it, unlike the murine CD4+CD3- cell, expresses the integrin, CDl Ic. There is also some evidence that the human LTi is CD4+CD1 lc+CD3- (Spencer et al., 1986). The inventors suggest that, with regard to the pathogenesis of HIV, these cells are depleted in human and animal models of AIDS, but are preserved in the tissues of natural hosts that harbour the virus but fail to develop either the abnormalities in lymphoid tissue architecture or immunodeficiency associated with pathogenic infection.

Table 1. Expression of TNF/TNFR family members

New TNF Chr CD4⁺CD3^" CD4⁺CD3- B

Group TNF pDCs DCs nomenclature m/h Adult Neonatal cells

I LTa TNFSFl 17/6 ++ ++ _ _ n.d.

TNFa TNFSF2 17/6 ++ ++ — — —

LTβ TNFSF3 17/6 -H- +++ + — ++

LIGHT TNFSF14 17/19 + + — _ —

TRANCE TNFSFI l 14/13 + + - - -

II FASL TNFSF6 1/1

GITRL TNFSF 18 1/1 +* n.d. _* _* n.d.

OX40L TNFSF4 1/1 +++ — . — + _

CD30L TNFSF8 4/9 ++ — _ — _

TLlA TNFSF 15 4/9 — . — — —

CD27L TNFSF7 17/19 — — — — —

4-1BBL TNFSF9 17/19 _ — + + —

CD40L TNFSF5 x/x - - - - -

III TWEAK TNFSF12 11/17 + + _ + _

APRIL TNFSF13 11/17 _ — _ _ —

BAFF TNFSF13B 8/13 - - - - -

New TNFR Chr CD4⁺CD3- CD4⁺CD3 B

Group TNFR pDCs DCs nomenclature m/h Adult Neonatal cells

I RANK TNFRSFI lA 1/18 ++ ++ -H-

TNFRl TNFRSFlA 6/12 + ++ -H- + —

LTβR TNFRSF3 6/12 + + - - -

II CD27** TNFRSF7 6/12 n.d.

CD40 TNFRSF5 2/20 +/- +/- _ ++ -H-

OX40 TNFRSF4 4/1 _ — — + +/-

CD30 TNFRSF8 4/1 _ — — — . —

GITR TNFRSF 18 4/1 _ _ — — H-

TNFR2 TNFRSFlB 4/1 -H- ++ + -H- +/-

4-1BB TNFRSF9 4/1 + + _ + —

HVEM TNFRSF14 4/1 ++ -H- -H- + —

DR3 TNFRSF25 4/1 -H- -H- - - -

III TWEAKR TNFRSF12 17/16 _ _

BAFFR TNFRSF13C 15/22 - - - - ++

+++ >10% β-actin signal, -H- 1 - 10%, + 0.2 - 1%, - <0.2% n.d.; not determined

*; detected by GITR-Ig protein

**; detected by flow cytometry Table 2. Expression of non-TNF/TNFR family immune genes

Chr CD4⁺CD3- CD4⁺CD3^"

Group Gene pDCs DCs m/h Adult Neonatal

Chemokine CCR7 11/17 ++ + + +++ receptors CXCR5 9/11 ++ ++ — —

CXCR3 x/x - - ++ -

Bd-2 Bcl-2 1/18 -H- ++ _ _ family Bcl-xL 2/1 ++ -H- + +

Transcription Bcl-6 16/3 -H- ++ factors GATA-3 2/10 ++ ++ _ _.

T-Bet 11/17 - - - -

TLR/IL1R Myd-88 9/3 ++ ++ ++ +

Common IL2Rγ X/X +++ +++ ++ + γ chain IL2Rα 2/10 ++ ++ - + cytokine IL4Rα 7/16 + + ++ ++ receptors IL7Rα 15/5 ++ +++ ++ +

Other ILl ORa 9/11 ++ + cytokine ILlORβ 16/21 + + -H- + receptors IL12Rβl 8/19 + ++ - -

IL12Rβ2 6/1 - - - -

IFNγRl 10/6 ++ + - -

IFNγR2 16/21 ++ ++ ++ ++

Costimulation CD80 16/3 ++ molecules CD86 16/3 — — — ++

CTLA-4 1/2 _ . — — —

ICOS 1/2 -H- ++ _ . _

ICOSL 10/21 + +/- - +

+++ >10% β-actin signal, ++ 1 - 10%, + 0.2 - 1%, - <0.2% References

1. Mebius, R.E. 2003. Organogenesis of lymphoid tissues. Nat Rev Immunol 3:292-303.

2. Luther, S.A., K.M. Ansel, and J. G. Cyster. 2003. Overlapping roles of CXCL13, interleukin 7 receptor alpha, and CCR7 ligands in lymph node development. J Exp Med 197:1191-1198.

3. Cupedo, T., F.E. Lund, V.N. Ngo, T.D. Randall, W. Jansen, M.J. Greuter, R. de Waal-Malefyt, G. Kraal, J.G. Cyster, and R.E. Mebius. 2004. Initiation of cellular organization in lymph nodes is regulated by non-B cell-derived signals and is not dependent on CXC chemokine ligand 13. J Immunol 173:4889-4896.

4. Fu, Y.X., and D. D. Chaplin. 1999. Development and maturation of secondary lymphoid tissues. Annu Rev Immunol 17:399-433.

5. Ngo, V.N., H. Korner, M. D. Gunn, K.N. Schmidt, D.S. Riminton, M. D. Cooper, J. L. Browning, J. D. Sedgwick, and J.G. Cyster. 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.

6. Luther, SA, H. L. Tang, P.L Hyman, A.G. Farr, and J.G. Cyster. 2000. Coexpression of the chemokines ELC and SLC by T zone stromal cells and deletion of the ELC gene in the plt/plt mouse. Proc Natl Acad Sci U S A 97:12694-12699.

7. Gunn, M.D., V.N. Ngo, K.M. Ansel, E.H. Ekland, J.G. Cyster, and LT. Williams. 1998. A B-homing chemokine made in lymphoid follicles activates Burkitt's lymphoma type receptor-1. Nature 391 :799-802.

8. Eberl, G. 2005. Opinion: Inducible lymphoid tissues in the adult gut: recapitulation of a fetal developmental pathway? Nat Rev Immunol 5:413-420.

9. Kim, M.Y., F.M. Gaspal, H. E. Wiggett, F.M. McConnell, A. Gulbranson- Judge, C. Raykundalia, LS. Walker, M. D. Goodall, and P.J. Lane. 2003. CD4(+)CD3(-) Accessory Cells Costimulate Primed CD4 T Cells through OX40 and CD30 at Sites Where T Cells Collaborate with B Cells. Immunity 18:643-654.

10. Gaspal, F.M. C, M. Kim, F.M. McConnell, C. Raykundalia, V. Bekiaris, and P.J. L. Lane. 2005. Mice deficient in OX40 and CD30 signals lack memory antibody responses because of deficient CD4 T cell memory. J. Immunol. 174:3891-3896.

11. Sun, Z., D. Unutmaz, Y.R. Zou, M.J. Sunshine, A. Pierani, S. Brenner- Morton, R.E. Mebius, and D. R. Littman. 2000. Requirement for RORgamma in thymocyte survival and lymphoid organ development. Science 288:2369-2373.

12. Kurebayashi, S., E. Ueda, M. Sakaue, D. D. Patel, A. Medvedev, F. Zhang, and A.M. Jetten. 2000. Retinoid-related orphan receptor gamma (RORgamma) is essential for lymphoid organogenesis and controls apoptosis during thymopoiesis. Proc Natl Acad Sci U S A 97:10132- 10137.

13. Eberl, G., S. Marmon, M.J. Sunshine, P.D. Rennert, Y. Choi, and D.R. Littman. 2004. An essential function for the nuclear receptor RORgamma(t) in the generation of fetal lymphoid tissue inducer cells. Nat Immunol 5:64-73.

14. Kim, M. -Y., G. Anderson, I.-L Martensson, L. Erlandsson, W. ArIt, A. White, and P. J. L. Lane. 2005. OX40-ligand and CD30-ligand are expressed on adult but not neonatal CD4+CD3- inducer cells: evidence that IL7 signals regulate CD30-ligand but not OX40-ligand expression. J. lmmunol.:\r\ press.

15. Silva-Santos, B., D.J. Pennington, and A.C. Hayday. 2005. Lymphotoxin- mediated regulation of gammadelta cell differentiation by alphabeta T cell progenitors. Science 307:925-928.

16. 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, and Y. Choi. 2000. Regulation of peripheral lymph node genesis by the tumor necrosis factor family member TRANCE. J Exp Med 192:1467-1478.

17. Croft, M. 2003. Co-stimulatory members of the TNFR family: keys to effective T-cell immunity? Nat Rev Immunol 3:609-620.

18. CeIIa, M., D. Jarrossay, F. Facchetti, O. Alebardi, H. Nakajima, A. Lanzavecchia, and M. Colonna. 1999. Plasmacytoid monocytes migrate to inflamed lymph nodes and produce large amounts of type I interferon. Nat Med 5:919-923.

19. Zheng, W., and R.A. Flavell. 1997. The transcription factor GATA-3 is necessary and sufficient for Th2 cytokine gene expression in CD4 T cells. Cell 89:587-596.

20. Pasparakis, M., L. Alexopoulou, V. Episkopou, and G. Kollias. 1996. Immune and inflammatory responses in TNF alpha-deficient mice: a critical requirement for TNF alpha in the formation of primary B cell follicles, follicular dendritic cell networks and germinal centers, and in the maturation of the humoral immune response. J Exp Med 184:1397-1411.

21. Koni, P.A., R. Sacca, P. Lawton, J. L. Browning, N. H. Ruddle, and R.A. Flavell. 1997. Distinct roles in lymphoid organogenesis for lymphotoxins alpha and beta revealed in lymphotoxin beta-deficient mice. Immunity 6:491-500.

22. Allen, CD., K.M. Ansel, C. Low, R. Lesley, H. Tamamura, N. Fujii, and J. G. Cyster. 2004. Germinal center dark and light zone organization is mediated by CXCR4 and CXCR5. Nat Immunol 5:943-952.

23. Migone, T.S., J. Zhang, X. Luo, L. Zhuang, C. Chen, B. Hu, J. S. Hong, J.W. Perry, S.F. Chen, J.X. Zhou, Y.H. Cho, S. Ullrich, P. Kanakaraj, J. Carrell, E. Boyd, H. S. Olsen, G. Hu, L. Pukac, D. Liu, J. Ni, S. Kim, R. Gentz, P. Feng, PA Moore, S.M. Ruben, and P. Wei. 2002. TL1A is a TNF-like ligand for DR3 and TR6/DcR3 and functions as a T cell costimulator. Immunity 16:479-492. 24. Kroncke, R., H. Loppnow, H. D. Flad, and J. Gerdes. 1996. Human follicular dendritic cells and vascular cells produce interleukin-7: a potential role for interleukin-7 in the germinal center reaction. Eur J Immunol 26:2541-2544.

25. Zapata, A., and CT. Ameimiya. 2000. Phylogeny of lower vertebrates and their immunological structures. In Origin and evolution of the vertebrate immune system. L. du Pasquier and G.W. Litman, editors. Springer, Berlin. 67-110.

26. Mackay, F., G. R. Majeau, P. Lawton, P. S. Hochman, and J. L. 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.

27. Fu, Y.X., H. Molina, M. Matsumoto, G. Huang, J. Min, and D. D. Chaplin. 1997. Lymphotoxin-alpha (LTalpha) supports development of splenic follicular structure that is required for IgG responses. J Exp Med 185:2111-2120.

28. Liu, Y.J., J. Zhang, P.J. Lane, E.Y. Chan, and I. C. MacLennan. 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.

29. Breitfeld, D., L OhI, E. Kremmer, J. Ellwart, F. Sallusto, M. Lipp, and R. Forster. 2000. Follicular B helper T cells express CXC chemokine receptor 5, localize to B cell follicles, and support immunoglobulin production. J Exp Med 192:1545-1552.

30. Schaerli, P., K. Willimann, A.B. Lang, M. Lipp, P. Loetscher, and B. Moser. 2000. CXC chemokine receptor 5 expression defines follicular homing T cells with B cell helper function. J Exp Med 192:1553-1562.

31. Reif, K., E.H. Ekland, L OhI, H. Nakano, M. Lipp, R. Forster, and J.G. Cyster. 2002. Balanced responsiveness to chemoattractants from adjacent zones determines B-cell position. Nature 416:94-99.

32. Ansel, K.M., V.N. Ngo, P.L. Hyman, S.A. Luther, R. Forster, J. D. Sedgwick, J. L. Browning, M. Lipp, and J.G. Cyster. 2000. A chemokine- driven positive feedback loop organizes lymphoid follicles. Nature 406:309-314.

33. Fu, Y.X., G. Huang, Y. Wang, and D.D. Chaplin. 2000. Lymphotoxin- alpha-dependent spleen microenvironment supports the generation of memory B cells and is required for their subsequent antigen-induced activation. J Immunol 164:2508-2514.

34. He, Y.W., M.L. Deftos, E.W. Ojala, and M.J. Bevan. 1998. RORgamma t, a novel isoform of an orphan receptor, negatively regulates Fas ligand expression and IL-2 production in T cells. Immunity 9:797-806.

35. He, Y.W., C. Beers, M.L. Deftos, E.W. Ojala, K.A. Forbush, and M.J. Bevan. 2000. Down-regulation of the orphan nuclear receptor ROR gamma t is essential for T lymphocyte maturation. J Immunol 164:5668- 5674. 36. Kumar, S., and S. B. Hedges. 1998. A molecular timescale for vertebrate evolution. Nature 392:917-920.

37. Connolly, J. H., P.J. Canfield, S.J. McClure, and RJ. Whittington. 1999. Histological and immunohistological investigation of lymphoid tissue in the platypus (Ornithorhynchus anatinus). J Anat 195 ( Pt 2): 161 -171.

38. Gowans, J. L., and J.W. Uhr. 1966. The carriage of immunological memory by small lymphocytes in the rat. J Exp Med 124:1017-1030.

39. Grouard, G., I. Durand, L Filgueira, J. Banchereau, and Y.J. Liu. 1996. Dendritic cells capable of stimulating T-cells in germinal-centers. Nature 384:364-367.

40. Heath, S.L, J.G. Tew, A.K. Szakal, and G. F. Burton. 1995. Follicular dendritic cells and human immunodeficiency virus infectivity. Nature 377:740-744.

41. Koopman, G., A.G. Haaksma, J. ten Velden, CE. Hack, and J. L. Heeney. 1999. The relative resistance of HIV type 1 -infected chimpanzees to AIDS correlates with the maintenance of follicular architecture and the absence of infiltration by CD8+ cytotoxic T lymphocytes. AIDS Res Hum Retroviruses 15:365-373.

42. Ochs, H. D., A.K. Junker, A.C. Collier, F.S. Virant, H. H. Handsfield, and RJ. Wedgwood. 1988. Abnormal antibody responses in patients with persistent generalized lymphadenopathy. J Clin Immunol 8:57-63.

43. Burke, A.P., D. Anderson, P. Mannan, J. L. Ribas, Y.H. Liang, J. Smialek, and R. Virmani. 1994. Systemic lymphadenopathic histology in human immunodeficiency virus-1 -seropositive drug addicts without apparent acquired immunodeficiency syndrome. Hum Pathol 25:248-256.

44. Janoff, E. N., Hardy, W. D., Smith, P. D. & Wahl, S. M. Humoral recall responses in HIV infection. Levels, specificity, and affinity of antigen-specific IgG. J. Immunol. 147, 2130-2135 (1991).

45. Mori, S., Takami, T., Nakamine, H., Miyayama, H. & Nakamura, S. Involution of lymph node histiocytes in AIDS. Actα. Pathol. Jpn. 39, 496-502 (1989).

46. Fauci, A. S. HIV and AIDS: 20 years of science. Nat. Med. 9, 839-843 (2003).

47. Brenchley, J. M. et al. CD4+ T cell depletion during all stages of HIV disease occurs predominantly in the gastrointestinal tract. J. Exp. Med. 200, 749-759 (2004).

48. Heath, S. L., Tew, J. G., Szakal, A. K. & Burton, G. F. Follicular dendritic cells and human immunodeficiency virus infectivity. Nature 311, 140-144 (1995).

49. Burke, A. P. et al. Systemic lymphadenopathic histology in human immunodeficiency virus-1 -seropositive drug addicts without apparent acquired immunodeficiency syndrome. Hum. Pathol. 25, 248-256 (1994).

50. Hahn, B. H., Shaw, G. M., De Cock, K. M. & Sharp, P. M. AIDS as a zoonosis: scientific and public health implications. Science 287, 607-614 (2000).

51. Stebbing, J., Gazzard, B. & Douek, D. C. Where does HIV live? N. Engl. J. Med. 350, 1872-1880 (2004). 52. Pretet, J. L., Zerbib, A. C, Girard, M., Guillet, J. G. & Butor, C. Chimpanzee CXCR4 and CCR5 act as coreceptors for HIV type 1. AIDS. Res. Hum. Retroviruses 13, 1583-1587 (1997).

53. Silvestri, G. et al. Nonpathogenic SIV infection of sooty mangabeys is characterized by limited bystander immunopathology despite chronic high-level viremia. Immunity 18, 441-452 (2003).

54. Siliciano, J. D. et al. Long-term follow-up studies confirm the stability of the latent reservoir for HIV-I in resting CD4+ T cells. Nat. Med. 9, 727-728 (2003).

55. Price, D. A. et al. T cell receptor recognition motifs govern immune escape patterns in acute SIV infection. Immunity 21, 793-803 (2004).

56. McMichael, A. & Hanke, T. The quest for an AIDS vaccine: is the CD8+ T-cell approach feasible? Nat. Rev. Immunol. 2, 283-291 (2002).

57. Burton, D. R. et al. HIV vaccine design and the neutralizing antibody problem. Nat. Immunol. 5, 233-236 (2004).

58. Mebius, R. E. Organogenesis of lymphoid tissues. Nat. Rev. Immunol. 3, 292- 303 (2003).

59. Kim, M. Y. et al. CD4(+)CD3(-) Accessory Cells Costimulate Primed CD4 T Cells through OX40 and CD30 at Sites Where T Cells Collaborate with B Cells. Immunity 18, 643-654 (2003).

60. Liu, Y. J., Zhang, J., Lane, P. J., Chan, E. Y. & MacLennan, I. C. 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 (1991).

61. Garside, P. et al. Visualization of specific B and T lymphocyte interactions in the lymph node. Science 281, 96-99 (1998).

62. Nakano, H., Yanagita, M. & Gunn, M. D. CDl lc(+)B220(+)Gr-l(+) cells in mouse lymph nodes and spleen display characteristics of plasmacytoid dendritic cells. J. Exp. Med. 194, 1171-1178. (2001).

63. Grouard, G., Durand, L, Filgueira, L., Banchereau, J. & Liu, Y. J. Dendritic cells capable of stimulating T-cells in germinal-centers. Nature 384, 364-367 (1996).

64. Linton, P. J. et al. Costimulation via OX40L expressed by B cells is sufficient to determine the extent of primary CD4 cell expansion and Th2 cytokine secretion in vivo. J. Exp. Med. 197, 875-883 (2003).

65. Kim, M. Y. et al. OX40 Signals during Priming on Dendritic Cells Inhibit CD4 T Cell Proliferation: IL-4 Switches off OX40 Signals Enabling Rapid Proliferation of Th2 Effectors. J. Immunol. 174, 1433-1437 (2005).

66. Aggarwal, B. B. Signalling pathways of the TNF superfamily: a double-edged sword. Nat. Rev. Immunol. 3, 745-756 (2003).

67. Croft, M. Co-stimulatory members of the TNFR family: keys to effective T-cell immunity? Nat. Rev. Immunol. 3, 609-620 (2003).

68. Rogers, P. R., Song, J., Gramaglia, L, Killeen, N. & Croft, M. OX40 promotes Bcl-xL and Bcl-2 expression and is essential for long- term survival of CD4 T cells. Immunity 15, 445-455. (2001). 69. Gaspal, F. M. C. et al. Mice deficient in OX40 and CD30 signals lack memory antibody responses because of deficient CD4 T cell memory. J Immunol. 174, 3891-3896 (2005).

70. Chen, A. I. et al. Ox40-ligand has a critical costimulatory role in dendritic cell:T cell interactions. Immunity 11, 689-698 (1999).

71. Kopf, M. et al. OX40-deficient mice are defective in Th cell proliferation but are competent in generating B cell and CTL Responses after virus infection. Immunity 11, 699-708 (1999).

72. Pippig, S. O. et al. Robust B cell immunity but impaired T cell proliferation in the absence of CD134 (OX40). J Immunol. 163, 6520-6529 (1999).

73. Murata, K. et al. Impairment of Antigen-presenting Cell Function in Mice Lacking Expression of OX40 Ligand. J Exp. Med. 191, 365-374 (2000).

74. Amakawa, R. et al. Impaired negative selection of T cells in Hodgkin's disease antigen CD30-deficient mice. Cell 84, 551-562 (1996).

75. Texido, G. et al. Somatic hypermutation occurs in B cells of terminal deoxynucleotidyl transferase-, CD23-, interleukin-4-, IgD- and CD30-deficient mouse mutants. Eur. J. Immunol. 26, 1966-1969 (1996).

76. Gowans, J. L. & Uhr, J. W. The carriage of immunological memory by small lymphocytes in the rat. J. Exp. Med. 124, 1017-1030 (1966).

77. Ansel, K. M., McHeyzer- Williams, L. J., Ngo, V. N., McHeyzer- Williams, M. G. & Cyster, J. G. In vivo-activated CD4 T cells upregulate CXC chemokine receptor 5 and reprogram their response to lymphoid chemokines. J. Exp. Med. 190, 1123-1134 (1999).

78. Breitfeld, D. et al. Follicular B helper T cells express CXC chemokine receptor 5, localize to B cell follicles, and support immunoglobulin production. J. Exp. Med. 192, 1545-1552 (2000).

79. Luther, S. A., Tang, H. L., Hyman, P. L., Farr, A. G. & Cyster, J. G. Coexpression of the chemokines ELC and SLC by T zone stromal cells and deletion of the ELC gene in the plt/plt mouse. Proc. Natl. Acad. Sci. USA 97, 12694-12699 (2000).

80. Gunn, M. D. et al. A B-homing chemokine made in lymphoid follicles activates Burkitt's lymphoma type receptor- 1. Nature 391, 799-802 (1998).

81. Reif, K. et al. Balanced responsiveness to chemoattractants from adj acent zones determines B-cell position. Nature 416, 94-99 (2002).

82. Li, J., Huston, G. & Swain, S. L. IL-7 promotes the transition of CD4 effectors to persistent memory cells. J. Exp. Med. 198, 1807-1815 (2003).

83. Kondrack, R. M. et al. Interleukin 7 regulates the survival and generation of memory CD4 cells. J. Exp. Med. 198, 1797-1806 (2003).

84. Seddon, B., Tomlinson, P. & Zamoyska, R. Interleukin 7 and T cell receptor signals regulate homeostasis of CD4 memory cells. Nat. Immunol. 4, 680-686 (2003).

85. Kroncke, R., Loppnow, H., Flad, H. D. & Gerdes, J. Human follicular dendritic cells and vascular cells produce interleukin-7: a potential role for interleukin-7 in the germinal center reaction. Eur J Immunol 26, 2541-2544 (1996). 86. Kim, M.-Y. et al. OX40-ligand and CD30-ligand are expressed on adult but not neonatal CD4+CD3- inducer cells: evidence that IL7 signals regulate CD30- ligand but not OX40-ligand expression. J Immunol, In press (2005).

87. Billingham, R. E., Brent, L. & Medawar, P. B. Activity acquired tolerance of foreign cells. Nature 172, 603-606 (1953).

88. Mebius, R. E., Rennert, P. & Weissman, I. L. Developing lymph nodes collect CD4+CD3- LTbeta÷ cells that can differentiate to APC, NK cells, and follicular cells but not T or B cells. Immunity 7, 493-504 (1997).

89. Eberl, G. et al. An essential function for the nuclear receptor RORgamma(t) in the generation of fetal lymphoid tissue inducer cells. Nat. Immunol. 5, 64-73 (2004).

90. Eberl, G. & Littman, D. R. Thymic origin of intestinal alphabeta T Cells Revealed by Fate Mapping of RORgammat+ Cells. Science 305, 248-251 (2004).

91. Sun, Z. et al. Requirement for RORgamma in thymocyte survival and lymphoid organ development. Science 288, 2369-2373. (2000).

92. Kurebayashi, S. et al. Retinoid-related orphan receptor gamma (RORgamma) is essential for lymphoid organogenesis and controls apoptosis during thymopoiesis. Proc. Natl. Acad. ScL USA 97, 10132-10137 (2000).

93. OhI, L. et al. Cooperating mechanisms of CXCR5 and CCR7 in development and organization of secondary lymphoid organs. J Exp. Med. 197, 1199-1204 (2003).

94. He, Y. W. et al. Down-regulation of the orphan nuclear receptor ROR gamma t is essential for T lymphocyte maturation. J. Immunol. 164, 5668-5674 (2000).

95. Kumar, S. & Hedges, S. B. A molecular timescale for vertebrate evolution. Nature 392, 917-920 (1998).

96. Connolly, J. H., Canfield, P. J., McClure, S. J. & Whittington, R. J. Histological and immunohistological investigation of lymphoid tissue in the platypus (Ornithorhynchus anatinus). J. Anat. 195 ( Pt 2), 161-171 (1999).

Alexopoulou, L., Pasparakis, M., and Kollias, G. (1998). Complementation of lymphotoxin alpha knockout mice with tumor necrosis factor-expressing transgenes rectifies defective splenic structure and function. J Exp Med 188, 745-754.

Browning, J. L., Allaire, N., Ngam-Ek, A., Notidis, E., Hunt, J., Perrin, S., and Fava, R. A. (2005). Lymphotoxin-beta Receptor Signaling Is Required for the Homeostatic Control of HEV Differentiation and Function. Immunity 23, 539-550.

Choi, Y. K., Fallert, B. A., Murphey-Corb, M. A., and Reinhart, T. A. (2003). Simian immunodeficiency virus dramatically alters expression of homeostatic chemokines and dendritic cell markers during infection in vivo. Blood 101, 1684-1691. Csanaky, G., Pap, T., Kalasz, V., and Kelenyi, G. (1991). Blood vessels in human immunodeficiency virus-related lymphoadenopathy: high endothelial venules and lymphocyte migration. Apmis 99, 640-644.

Eberl, G., and Littman, D. R. (2004). Thymic origin of intestinal alphabeta T Cells Revealed by Fate Mapping of RORgammat+ Cells. Science 305, 248-251. Farr, A. G., Berry, M. L., Kim, A., Nelson, A. J., Welch, M. P., and Aruffo, A. (1992). Characterization and cloning of a novel glycoprotein expressed by stromal cells in Iⁿdependent areas of peripheral lymphoid tissues. J Exp Med 176, 1477-1482.

Finke, D., Acha-Orbea, H., Mattis, H., Lipp, M., and Kraehenbuhl, J. P. (2002). CD4+CD3- Cells Induce Peyer's Patch Development: Role of 41 Integrin Activation by CXCR5. Immunity 17, 363-373.

Hirsch, V. M. (2004). What can natural infection of African monkeys with simian immunodeficiency virus tell us about the pathogenesis of AIDS? AIDS Rev 6, 40-53.

Hollander, G. A., Wang, B., Nichogiannopoulou, A., Platenburg, P. P., van Ewijk, W., Burakoff, S. J., Gutierrez-Ramos, J. C, and Terhorst, C. (1995). Developmental control point in induction of thymic cortex regulated by a subpopulation of prothymocytes. Nature 373, 350-353.

Josien, R., Wong, B. R., Li, H. L., Steinman, R. M., and Choi, Y. (1999). TRANCE, a TNF family member, is differentially expressed on T cell subsets and induces cytokine production in dendritic cells. J Immunol 162, 2562-2568.

Kabashima, K., Banks, T. A., Ansel, K. M., Lu, T. T., Ware, C. F., and Cyster, J. G. (2005). Intrinsic lymphotoxin-beta receptor requirement for homeostasis of lymphoid tissue dendritic cells. Immunity 22, 439-450.

Kawabe, T., Naka, T., Yoshida, K., Suematsu, S., Yoshida, N., Kishimoto, T., and Kikutani, H. (1994). The immune response in CD40-deficient mice: impaired immunoglobulin class switching and germinal center formation. Immunity 1, 167-178.

Kim, M.-Y., Anderson, G., Martensson, L-L., Erlandsson, L., ArIt, W., White, A., and Lane, P. J. L. (2005). OX40-ligand and CD30-ligand are expressed on adult but not neonatal CD4+CD3- inducer cells: evidence that IL7 signals regulate CD30-ligand but not OX40-ligand expression. J Immunol 174, 6686-6691.

Kobayashi, T., Walsh, P. T., Walsh, M. C, Speirs, K. M., Chiffoleau, E., King, C. G., Hancock, W. W., Caamano, J. H., Hunter, C. A., Scott, P., et al (2003). TRAF6 is a critical factor for dendritic cell maturation and development. Immunity 19, 353-363.

Kuprash, D. V., Alimzhanov, M. B., Tumanov, A. V., Grivennikov, S. L, Shakhov, A. N., Drutskaya, L. N., Marino, M. W., Turetskaya, R. L., Anderson, A. O., Rajewsky, K., et al. (2002). Redundancy in tumor necrosis factor (TNF) and lymphotoxin (LT) signaling in vivo: mice with inactivation of the entire TNF/LT locus versus single- knockout mice. MoI Cell Biol 22, 8626-8634.

Lane, P. J. L., Gaspal, M. C, and Kim, M.-Y. (2005). Two sides of a cellular coin: CD4+CD3- cells orchestrate memory antibody responses and lymph node organisation. Nature Rev Immunol 5, 655-660.

Mackay, F., and Browning, J. L. (1998). Turning off follicular dendritic cells [letter]. Nature 395, 26-27.

Mackay, F., and Browning, J. L. (2002). BAFF: a fundamental survival factor for B cells. Nat Rev Immunol 2, 465-475.

Ngo, V. N., Cornall, R. J., and Cyster, J. G. (2001). Splenic T zone development is B cell dependent. J Exp Med 194, 1649-1660.

Nishimura, Y., Brown, C. R., Mattapallil, J. J., Igarashi, T., Buckler- White, A., Lafont, B. A., Hirsch, V. M., Roederer, M., and Martin, M. A. (2005). Resting naive CD4+ T cells are massively infected and eliminated by X4-tropic simian-human immunodeficiency viruses in macaques. Proc Natl Acad Sci U S A 102, 8000-8005.

Nishimura, Y., Igarashi, T., Donau, O. K., Buckler- White, A., Buckler, C₅ Lafont, B. A., Goeken, R. M., Goldstein, S., Hirsch, V. M., and Martin, M. A. (2004). Highly pathogenic SHIVs and SIVs target different CD4+ T cell subsets in rhesus monkeys, explaining their divergent clinical courses. Proc Natl Acad Sci U S A 101, 12324- 12329.

OhI, L., Henning, G., Krautwald, S., Lipp, M., Hardtke, S., Bernhardt, G., Pabst, O., and Forster, R. (2003). Cooperating mechanisms of CXCR5 and CCR7 in development and organization of secondary lymphoid organs. J Exp Med 197, 1199-1204.

Scarlatti, G., Tresoldi, E., Bjorndal, A., Fredriksson, R., Colognesi, C, Deng, H. K., Malnati, M. S., Plebani, A., Siccardi, A. G., Littman, D. R., et al. (1997). In vivo evolution of HIV-I co-receptor usage and sensitivity to chemokine-mediated suppression. Nat Med 3, 1259-1265.

Spencer, J., MacDonald, T. T., Finn, T., and Isaacson, P. G. (1986). The development of gut associated lymphoid tissue in the terminal ileum of fetal human intestine. Clin Exp Immunol 64, 536-543.

Tumanov, A., Kuprash, D., Lagarkova, M., Grivennikov, S., Abe, K., Shakhov, A., Drutskaya, L., Stewart, C, Chervonsky, A., and Nedospasov, S. (2002). Distinct role of surface lymphotoxin expressed by B cells in the organization of secondary lymphoid tissues. Immunity 17, 239-250.

Vinuesa, C. G., Tangye, S. G., Moser, B., and Mackay, C. R. (2005). Follicular B helper T cells in antibody responses and autoimmunity. Nat Rev Immunol 5, 853-865.

Wang, B., Hollander, G. A., Nichogiannopoulou, A., Simpson, S. J., Orange, J. S., Gutierrez-Ramos, J. C, Burakoff, S. J., Biron, C. A., and Terhorst, C. (1996). Natural killer cell development is blocked in the context of aberrant T lymphocyte ontogeny. Int Immunol 8, 939-949.

Wu, Q., Wang, Y., Wang, J., Hedgeman, E. O., Browning, J. L., and Fu, Y. X. (1999). The requirement of membrane lymphotoxin for the presence of dendritic cells in lymphoid tissues. J Exp Med 190, 629-638.

Claims

Claims

1) An assay for identifying a potential immunosuppressive or an immunostimulating compound comprising the steps of providing i) TLl A or a fragment or a derivative thereof and ii) a DR3 receptor or a fragment or a derivative thereof, and determining the affect of the compound on the binding of TLl A to the DR3 receptor.

2) An assay according to claim 1 wherein the DR3 is provided on the surface of a CD4+CD3- cell.

3) An assay according to claim 1 or claim 2, wherein the affect of the compound is to decrease the binding of TLlA

4) An assay according to any preceding claim which is a competitive binding assay.

5) An assay according to any preceding claim, wherein labelled ILIA is provided and the amount of labelled TLlA bound to the DR3 receptor, or not bound to the DR3, receptor is determined.

6) An assay according to any preceding claim, wherein the assay is an immunoassay comprising the use of an anti-TLAl A antibody to detect bound or unbound TLAlA. 7) An assay for identifying a potential immunosuppressive or an immunostimulating compound comprising the steps of:

a) providing a cell expressing a DR3 receptor or a fragment or a derivative thereof,

b) stimulating the cell with TLl A or a fragment or a derivative thereof

c) contacting the cell with the compound, and

d) determining the effect of the compound on OX40L, or a fragment or a derivative thereof, expression.

8) Method according to claim 7, wherein the cell is a CD4+CD3- cell or a cell transfected with a nucleic acid molecule encoding DR3.

9) An assay according to claim 7 or claim 8, wherein the cell is contacted with the compound before the cell is stimulated with TLlA.

10) An assay according to claims 7, 8, or 9 wherein a decrease in the amount of OX40L indicates the compound to have potential immunosuppressive activity.

11) An assay according to any one of claims 7-10, wherein the amount of OX40L produced is determined by means of an immunoassay, microarray assay, or quantitative PCR . 12) An assay according to any preceding claim, wherein the CD4+CD3- cell is a mouse cell or a human cell.

13) An assay according to claim 12, wherein the cell is in vitro.

14) An assay according to any one of claims 2-13, wherein the cell has the phenotype:

CD4+CD3-CDl lc-IL7receptor+, ckit+, common cytokine receptor gamma chain positive, TRANCE+, RANK+, TNFRSF25+, CXCR5+, CXCR4+

15) Isolated human cells expressing the phenotype:

CD4+CD3-CDl lc-IL7receptor+, ckit÷, common cytokine receptor gamma chain positive, TRANCE+, RANK+, TNFRSF25+, CXCR5+, CXCR4+

16) A method of studying immunodeficiency in a patient comprising obtaining a sample of tissue containing CD4+CD3- cells and determining the amount of the cells in the sample.

17) A method according to claim 16, wherein the immunodeficiency is Acquired Immunodeficiency Syndrome (AIDS). 18) A method according to any one of claims 16 or 17, wherein the method provides an indication of the progression of the immunodeficiency.

19) A method according to any one of claims 16-18, wherein the method provides an indication of the likelihood of developing AIDS further.

20) Method of identifying a compound which assists in the survival of CD4+CD3- cells, comprising contacting CD4CD3- cells with the compound and determining the effect of the compound on survival of the cells

21) Method according to claim 20 wherein the cells are infected with HIV.