Medullary thymic epithelial cells (mTEC) play an important and unique role in central tolerance, expressing tissue-restricted Ags (TRA) which delete thymocytes autoreactive to peripheral organs. Since deficiencies in this cell type or activity can lead to devastating autoimmune diseases, it is important to understand the factors which regulate mTEC differentiation and function. Lymphotoxin (LT) ligands and the LTβR have been recently shown to be important regulators of mTEC biology; however, the precise role of this pathway in the thymus is not clear. In this study, we have investigated the impact of this signaling pathway in greater detail, focusing not only on mTEC but also on other thymic stromal cell subsets. LTβR expression was found in all TEC subsets, but the highest levels were detected in MTS-15+ thymic fibroblasts. Rather than directing the expression of the autoimmune regulator Aire in mTEC, we found LTβR signals were important for TRA expression in a distinct population of mTEC characterized by low levels of MHC class II (mTEClow), as well as maintenance of MTS-15+ fibroblasts. In addition, thymic stromal cell subsets from LT-deficient mice exhibit defects in chemokine production similar to that found in peripheral lymphoid organs of Lta−/− and Ltbr−/− mice. Thus, we propose a broader role for LTα1β2-LTβR signaling in the maintenance of the thymic microenvironments, specifically by regulating TRA and chemokine expression in mTEClow for efficient induction of central tolerance.

Thymic stromal cells (TSC)7 provide chemokines, cytokines, and cell surface molecules essential for the differentiation of thymocytes. The stroma is composed of thymic epithelial cells (TEC), dendritic cells (DC), fibroblasts, macrophages, and endothelial cells that provide distinct signals governing thymopoiesis (1). Medullary TEC play a unique role in central tolerance, ectopically expressing a vast range of tissue-restricted Ag (TRA) that mediate deletion of thymocytes which pose a danger to peripheral organs and tissues (2). About one-third of TRA expression by mTEC is controlled by the transcriptional regulator Aire and it has been proposed that other “Aire-like” factors exist which promote expression of Aire-independent TRA (3, 4). The importance of TRA expression to T cell central tolerance is exemplified by the disease autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy (APECED). This multiorgan autoimmune disease is caused by deficiency of the transcriptional regulator AIRE in humans and Aire in mouse models of the disease (5). Importantly, loss of a single TRA gene, even in the presence of functional Aire, is sufficient to break central tolerance and cause targeted autoimmunity (6).

TRA expression and Ag presentation are regulated during mTEC differentiation. These functions are reduced in mTEC expressing low levels of CD80 and MHC class II (MHCII) molecules (mTEClow) compared with the more mature, Aire-expressing CD80highMHCIIhighmTEC subset (mTEChigh) (4, 7, 8). Given the essential role of mTEC in central tolerance, it is important to understand factors that govern their differentiation.

The TNF-related cytokines, lymphotoxin (LT) α and LTβ, and the LTβR have been recently shown to influence mTEC differentiation (9, 10). LTα can form a soluble homotrimer, LTα3, which binds to TNFR1 and TNFR2 as well as the herpesvirus entry mediator. However, it is the membrane-bound heterotrimer LTα1β2 and LT-related inducible ligand that competes for glycoprotein D binding to herpesvirus entry mediator on T cells (LIGHT), which signal exclusively through the LTβR (11). LTβR signals activate NF-κB-dependent production of chemokines and cytokines critically involved in the organogenesis and development of peripheral lymphoid organs (12). In lymph nodes and Peyer’s patches, the LTα1β2-LTβR axis regulates cross-talk between the hemopoietic and stromal compartment that is necessary for their development and maintenance of proper architecture (13).

Both Lta−/− and Ltbr−/− mice exhibit autoimmunity characterized by lymphocytic infiltration of organs and production of autoantibodies to several tissues (9, 10). Importantly, autoimmunity (albeit mild) was transferred upon engraftment of Ltbr−/− thymic stroma into thymectomized recipients, indicating a level of stromal-dependent defect in central tolerance (14). The precise role of LTβR signaling in the thymus, however, remains unclear.

Early studies using Lta−/− and Ltbr−/− mice indicated signaling through LTβR directly regulated Aire expression and Aire-dependent TRA transcription in mTEC (10). In contrast, a similar study of Ltbr−/− mice indicated no change in mTEC-dependent Aire or TRA transcription on a per cell basis, but perturbed mTEC organization and differentiation (9). In support of this finding, more recent reports demonstrated that the expression and function of Aire in mTEC was unaffected in the absence of LTβR signaling (15, 16). Thus, the emerging consensus is that LTβR signaling is required for proper organization and differentiation of mTEC; however, the mechanism by which LTβR regulates these effects remains undefined.

Previous studies predominantly focused on the influence of LT signals on the mTEChighAire+ subset of TSC. Given the importance of interplay among thymic cell types for negative selection, we sought to clarify and extend these data, analyzing the influence of the LT pathway on this and other TSC subsets. We find that although LTβR signaling was not required for differentiation of Aire-expressing mTEChigh, it was important for expression of Aire-independent TRA and Ulex europeaus agglutinin 1 (UEA-1) binding in mTEClow, as well as the maintenance of normal numbers of MTS-15+ fibroblasts. Importantly, TSC populations of Lta−/− and Ltb−/− mice exhibit a previously unreported deficiency in chemokine production, which was restored upon treatment with LTβR agonist Abs. These data refine and extend the role of the LTβR pathway in the thymus to the maintenance of certain TSC subsets and regulation of key molecules involved in TSC organization and function.

Lta−/− and Ltb−/− mice were generated and maintained on a C57BL/6 background as previously described (17, 18). Mice were housed at the Department of Biochemistry Animal Facility (Monash University, Clayton, Victoria, Australia) according to institutional guidelines. All mice used in this study were between 4 and 6 mo of age, unless otherwise stated.

Mice were injected i.p. with 100 μg of LTβR-agonist Ab (4H8) or an isotype control in sterile PBS. Mice were either injected once or daily for 3 consecutive days. Thymus and spleen were harvested 8 h or 3 days following the first injection.

The following Abs used in this study were purchased from BD Pharmingen, unless otherwise stated: MTS-10 (19) and MTS-15 (20) were grown in our laboratory, anti-epithelial cell adhesion molecule (EpCAM, clone G8.8a; a gift from Dr. A. Farr, Department of Biological Structure, University of Washington, Seattle, WA), anti-Aire (clone 5H12-2), anti-LTβR (clone 4H8), IgG2a isotype control (clone RTK2758; BioLegend), pan-cytokeratin (DakoCytomation), anti-Ly51 (clone 6C3), anti-I-A/I-E (clone M5/114.15.2), anti-CD80 (clone 16-10A1), anti-CD86 (clone GL1), anti-CD40 (clone 3/23), anti-H2-Db (clone KH95), anti-CD45 (clone 30-F11), anti-CD31 (clone MEC13.3), anti-CD8α (clone 53-6.7), anti-CD4 (clone GK1.5), anti-TCRβ (clone H57-597), and anti-Ki67 (clone B56). The lectin Ulex europaeus agglutinin 1 (UEA-1) was purchased from Vector Laboratories. Secondary reagents were streptavidin-allophycocyanin (BD Pharmingen), streptavidin-Alexa Fluor 488, anti-rat Ig Alexa Fluor 488, anti-rat Ig Alexa Fluor 568, anti-rabbit Ig Alexa Fluor 647 (all from Molecular Probes), and anti-rat IgG2cFITC (Southern Biotechnology Associates).

Thymic tissue was dissected and cleaned to remove excess fat and connective tissue. Thymic lobes were suspended in Tissue-Tek OCT compound (Sakura Finetek) and snap frozen in a liquid nitrogen and isopentane slurry. Sections of 8–10 μm were cut on a Leica CM1850 cryostat and left to air dry at 4°C for 30 min. Primary Abs were added to sections and slides were incubated at room temperature for 15 min, then washed in PBS for 5 min. Secondary Abs were then applied to slides, which were incubated for 15 min, washed, and then mounted with a coverslip using fluorescent mounting medium (DakoCytomation). Images were acquired on a Bio-Rad MRC 1024 confocal microscope and analyzed using LaserSharp2000 software (Bio-Rad).

Individual thymic digestion.

Digestion of individual thymi was performed as previously described in the study of Gray et al. (21). Briefly, thymi were digested in collagenase/DNase, then collagenase/Dispase/DNase (Roche) and resulting suspensions were passed through 100-μm mesh to remove debris. Excluding the first thymocyte wash, all fractions from each thymus were pooled and counted using a Z2 Coulter Counter (Beckman Coulter). For flow cytometric analysis, 5 × 106 cells from each individual thymus were stained with appropriate Abs.

Pooled thymic digestion.

Enzymatic digestion was performed as previously described (21) on pools of 6–10 thymi per group. Final digestions were incubated with anti-CD45 microbeads (Miltenyi Biotec) according to the manufacturer’s protocols. This suspension was run on an AutoMACS (Miltenyi Biotec) to deplete CD45+ cells. CD45+ cells were rerun using the same program to recover any remaining unlabeled cells. Depleted cells (CD45) were pooled and recovered by centrifugation.

Immunofluorescent staining was performed as previously reported (20). Sample data from 1 × 104 CD45 cells were acquired on a FACSCalibur (BD Biosciences) using up to four fluorescent channels and analyzed using CellQuest software (BD Biosciences). Statistical analysis was performed with SPSS version 15.0 software using the Mann-Whitney U test.

TSC depleted of CD45+ cells were stained with appropriate immunofluorescent markers in FACS/EDTA (5 mM EDTA in PBS with 2% FCS and 0.02% NaN3) and sorted on a FACSVantage cell sorter (BD Biosciences) at no more than 3 × 103 cells per second. Samples were collected in 30% (v/v) FCS in RPMI 1640, recovered by centrifugation, counted, and analyzed for purity. Populations were sorted to >95% purity.

Total RNA was isolated from sorted TSC or whole tissues using TRI reagent (Molecular Research Center) and 1-bromo-3-chloropropane phase separation reagent (Molecular Research Center) according to the manufacturer’s instructions. RNA was reverse transcribed using Superscript III (Invitrogen Life Technologies) and oligo(dT) oligonucleotides (Invitrogen Life Technologies) according to the manufacturer’s protocol.

Quantitative PCR was performed on a Corbett Rotor-Gene 3000 (Corbett Research) in 10-μl reactions using SYBR Green Supermix (Invitrogen Life Technologies) with appropriate primers (200 nM). After initial holds for 2 min at 50°C, then 10 min at 95°C, PCR was performed with 40 cycles of 95°C for 15 s and 60°C for 60 s. Target transcript levels relative to those of GAPDH were determined using the 2−ΔCt method. Analysis of relative gene expression was performed using real-time quantitative PCR and the 2−ΔCt method (22). Refer to Table I for primer sequences used.

Table I.

Primer sequences

GeneForward (5′–3′)Reverse (5′–3′)
Lta GCTTGGCACCCCTCCTGTC GATGCCATGGGTCAAGTGCT 
Ltb CCAGCTGCGGATTCTACACCA AGCCCTTGCCCACTCATCC 
Ltbr CCAGATGTGAGATCCAGGGC GACCAGCGACAGCAGGATG 
Tnfsf14 (LIGHT) CAGGCCCCTACAGACAACAC ACTCGTCTCCCACAAGGAACT 
Aire GGTTCTGTTGGACTCTGCCCTG TGTGCCACGACGGAGGTGAG 
Csna TTTGCTATGCCCAGACTTCA TTTCCTCACTGCTGCTATGC 
Csng TCTGGCAAAGCACGAAATAAAGG AGTTGTTTGGAAGAACACGCTA 
Ins2 GACCCACAAGTGGCACAA ATCTACAATGCCACGCTTCTG 
Spt1 GTGTTGCTTGGTGTTTCCAC GCAGAATCAGCAGTTCCAGA 
Csnb GGCACAGGTTGTTCAGGCTT AAGGAAGGGTGCTACTTGCTG 
Csnk ATTCTGGCATTAACTCTGCCC AAAGATGGCCTGTAGTGGTAGTA 
Gad1 QIAGEN validated primers for quantitative RT-PCR (Cat.No. QT00163527) 
Fabp9 QIAGEN validated primers for quantitative RT-PCR (Cat.No. QT00115913) 
Col2 AGAACAGCATCGCCTACCTG CTTGCCCCACTTACCAGTGT 
Crp QIAGEN validated primers for quantitative RT-PCR (Cat.No. QT00255444) 
Tgn QIAGEN validated primers for quantitative RT-PCR (Cat.No. QT00116592) 
K14 QIAGEN validated primers for quantitative RT-PCR (Cat.No. QT00114422) 
Ccl19 GCTAATGATGCGGAAGACTG ACTCACATCGACTCTCTAGG 
Ccl21 GCAGTGATGGAGGGGGTCAG CGGGGTGAGAACAGGATTGC 
Ccl17 CTGCTCGTTCTGGGGAC TGTTTGTCTTTGGGGTCTGC 
Ccl22 GGTCCCTATGGTGCCAATG TTATCAAAACAACGCCAGGC 
Ccl25 TGAAACTGTGGCTTTTTGCC GTCAAGATTCTCATCGCCCTC 
Cxcl12 GCTCTGCATCAGTGACGGTA TGTCTGTTGTTGTTCTTCAGC 
Cxcl13 GCACAGCAACGCTGCTTCT TCTTTGAACCATTTGGCAGC 
Il6 TGTATGAACAACGATGATGCACTT ACTCTGGCTTTGTCTTTCTTGTTATC 
Tnfa CATCTTCTCAAAATTCGAGTGACAAGCC TGGGAGTAGACAAGGTACAACCCATC 
Relb QIAGEN validated primers for quantitative RT-PCR (Cat.No. QT00172578) 
Fgf7 GCGCAAATGGATACTGACACG GGGCTGGAACAGTTCACACT 
Fgf10 GACCAAGAATGAAGACTGTCCG TACAGTCTTCAGTGAGGATACC 
Gapdh ACCATGTAGTTGAGGTCAATGAAGG GGTGAAGGTCGGTGTGAACG 
GeneForward (5′–3′)Reverse (5′–3′)
Lta GCTTGGCACCCCTCCTGTC GATGCCATGGGTCAAGTGCT 
Ltb CCAGCTGCGGATTCTACACCA AGCCCTTGCCCACTCATCC 
Ltbr CCAGATGTGAGATCCAGGGC GACCAGCGACAGCAGGATG 
Tnfsf14 (LIGHT) CAGGCCCCTACAGACAACAC ACTCGTCTCCCACAAGGAACT 
Aire GGTTCTGTTGGACTCTGCCCTG TGTGCCACGACGGAGGTGAG 
Csna TTTGCTATGCCCAGACTTCA TTTCCTCACTGCTGCTATGC 
Csng TCTGGCAAAGCACGAAATAAAGG AGTTGTTTGGAAGAACACGCTA 
Ins2 GACCCACAAGTGGCACAA ATCTACAATGCCACGCTTCTG 
Spt1 GTGTTGCTTGGTGTTTCCAC GCAGAATCAGCAGTTCCAGA 
Csnb GGCACAGGTTGTTCAGGCTT AAGGAAGGGTGCTACTTGCTG 
Csnk ATTCTGGCATTAACTCTGCCC AAAGATGGCCTGTAGTGGTAGTA 
Gad1 QIAGEN validated primers for quantitative RT-PCR (Cat.No. QT00163527) 
Fabp9 QIAGEN validated primers for quantitative RT-PCR (Cat.No. QT00115913) 
Col2 AGAACAGCATCGCCTACCTG CTTGCCCCACTTACCAGTGT 
Crp QIAGEN validated primers for quantitative RT-PCR (Cat.No. QT00255444) 
Tgn QIAGEN validated primers for quantitative RT-PCR (Cat.No. QT00116592) 
K14 QIAGEN validated primers for quantitative RT-PCR (Cat.No. QT00114422) 
Ccl19 GCTAATGATGCGGAAGACTG ACTCACATCGACTCTCTAGG 
Ccl21 GCAGTGATGGAGGGGGTCAG CGGGGTGAGAACAGGATTGC 
Ccl17 CTGCTCGTTCTGGGGAC TGTTTGTCTTTGGGGTCTGC 
Ccl22 GGTCCCTATGGTGCCAATG TTATCAAAACAACGCCAGGC 
Ccl25 TGAAACTGTGGCTTTTTGCC GTCAAGATTCTCATCGCCCTC 
Cxcl12 GCTCTGCATCAGTGACGGTA TGTCTGTTGTTGTTCTTCAGC 
Cxcl13 GCACAGCAACGCTGCTTCT TCTTTGAACCATTTGGCAGC 
Il6 TGTATGAACAACGATGATGCACTT ACTCTGGCTTTGTCTTTCTTGTTATC 
Tnfa CATCTTCTCAAAATTCGAGTGACAAGCC TGGGAGTAGACAAGGTACAACCCATC 
Relb QIAGEN validated primers for quantitative RT-PCR (Cat.No. QT00172578) 
Fgf7 GCGCAAATGGATACTGACACG GGGCTGGAACAGTTCACACT 
Fgf10 GACCAAGAATGAAGACTGTCCG TACAGTCTTCAGTGAGGATACC 
Gapdh ACCATGTAGTTGAGGTCAATGAAGG GGTGAAGGTCGGTGTGAACG 

In peripheral lymphoid organs, LTβR is expressed exclusively by stromal cells and transduces signals from both LTα1β2 and LIGHT produced by hemopoietic cells (23). Although a previous study showed thymic expression of LTβR by in situ hybridization, the stromal defects reported in LT-deficient mice prompted more precise analysis (24). Analysis of thymic sections stained with Abs to LTβR revealed low level expression by reticular stromal cells throughout the thymus (Fig. 1,A). Flow cytometric analysis revealed LTβR expression on CD45MHCII+ TEC and higher levels on CD45MHCII nonepithelial cells (non-TEC), as indicated by the increased ratio of LTβR to isotype median expression levels (Fig. 1 B).

FIGURE 1.

Expression of LTβR and its ligands on thymic cell populations. A, Thymic sections from C57BL/6 mice stained with anti-LTβR or isotype control Abs (green). B, Flow cytometric analysis of LTβR expression (solid line) compared with isotype (dashed line), on CD45 stroma gated on MHCIIhigh (hi), low (lo), or negative (−) cells, as quantified by the numerical ratio of LTβR to isotype control median expression levels for each gated population. C, PCR of LTα, LTβ, LIGHT, and LTβR expression in purified thymic cell populations (see Table II), relative to whole TSC expression, standardized to 1. Mean and SE were determined from two to three experiments for each population.

FIGURE 1.

Expression of LTβR and its ligands on thymic cell populations. A, Thymic sections from C57BL/6 mice stained with anti-LTβR or isotype control Abs (green). B, Flow cytometric analysis of LTβR expression (solid line) compared with isotype (dashed line), on CD45 stroma gated on MHCIIhigh (hi), low (lo), or negative (−) cells, as quantified by the numerical ratio of LTβR to isotype control median expression levels for each gated population. C, PCR of LTα, LTβ, LIGHT, and LTβR expression in purified thymic cell populations (see Table II), relative to whole TSC expression, standardized to 1. Mean and SE were determined from two to three experiments for each population.

Close modal

To more precisely analyze the distribution of LT signaling components, thymic stromal and thymocyte subsets were FACS purified using defined markers (see Table II) and transcription of LTβR, LTα, LTβ, and LIGHT was assessed by real-time PCR. Fig. 1 C illustrates that significant levels of LTβR transcription were found in all TEC subsets, with slightly higher levels in cortical TEC (cTEC) and mTEClow. The high levels of LTβR transcription by CD45MHCII non-TEC, which has been shown to consist predominantly of fibroblasts and endothelium (20), supported the flow cytometry data and were predominantly derived from MTS-15+ fibroblasts. LTβR transcription was low in DC and undetectable in thymocytes. Consistent with a previous study (9), LT transcripts were found predominantly on CD4+ and CD8+ single-positive (SP) thymocytes. In addition, we found high expression levels of LTα and the majority of transcript for LIGHT in stromal subsets, suggesting that LTβR ligands are also derived from TSC subsets, not just mature thymocytes. Together, these data extend the expression profile of LTβR to all major TEC subsets and fibroblasts and reveal distinct thymocyte and stromal-derived ligands.

Table II.

Phenotype of thymic and spleen cell populations

Cell PopulationPhenotype
Whole TSC CD45 
TEC CD45EpCAM+MHC II+ 
Non-TEC CD45MHCII 
cTEC CD45MHCII+UEA1 
MHCIIhighmTEChigh CD45 MHCIIhi UEA1+ 
MHCIIlowmTEClow CD45MHCIIlowUEA1+ 
Thymic fibroblasts (MTS-15) CD45 MTS-15+ 
Thymic endothelium (CD31) CD45CD31+ 
Thymic DC (ThyDC) CD11c+MHCII+ 
Splenic DC (SplDC) CD11c+MHCII+ 
TN thymocytes CD3 CD4CD8 
DP thymocytes CD4+CD8+ 
CD4+ SP thymocytes (4 SP) CD3+CD4+CD8 
CD8+ SP thymocytes (8 SP) CD3+CD4CD8+ 
Cell PopulationPhenotype
Whole TSC CD45 
TEC CD45EpCAM+MHC II+ 
Non-TEC CD45MHCII 
cTEC CD45MHCII+UEA1 
MHCIIhighmTEChigh CD45 MHCIIhi UEA1+ 
MHCIIlowmTEClow CD45MHCIIlowUEA1+ 
Thymic fibroblasts (MTS-15) CD45 MTS-15+ 
Thymic endothelium (CD31) CD45CD31+ 
Thymic DC (ThyDC) CD11c+MHCII+ 
Splenic DC (SplDC) CD11c+MHCII+ 
TN thymocytes CD3 CD4CD8 
DP thymocytes CD4+CD8+ 
CD4+ SP thymocytes (4 SP) CD3+CD4+CD8 
CD8+ SP thymocytes (8 SP) CD3+CD4CD8+ 

The broad expression of LTβR and its ligands extends on previous studies and prompted a re-examination of the consequence of LT deficiency on specific TSC subsets. The architecture and composition of thymi from adult Lta−/− and Ltb−/− mice were compared with age-matched controls by immunohistochemistry (Fig. 2). The lectin UEA-1 recognizes a carbohydrate epitope highly expressed on a subset of mTEC by histology and with which Aire is associated (25). It should be noted that although histological observations indicated that UEA-1 bound only a minor fraction of mTEC (26), flow cytometric studies reveal that UEA-1 binds to essentially all mTEC in normal young mice, perhaps due to the heightened sensitivity of flow cytometry, epitope masking by histology, or differential surface and intracellular expression (Ref. 20 and present study).

FIGURE 2.

Reduced UEA-1 but not Aire expression in the thymic medulla of Lta−/− and Ltb−/− mice. Thymic sections from Lta−/−, Ltb−/−, and control mice were stained for medullary epithelial markers UEA-1 (A; green), Aire (B; red), and merge of UEA-1/Aire and pan epithelial marker keratin (C; blue), including higher magnification (second row). Images are representative of three to four experiments.

FIGURE 2.

Reduced UEA-1 but not Aire expression in the thymic medulla of Lta−/− and Ltb−/− mice. Thymic sections from Lta−/−, Ltb−/−, and control mice were stained for medullary epithelial markers UEA-1 (A; green), Aire (B; red), and merge of UEA-1/Aire and pan epithelial marker keratin (C; blue), including higher magnification (second row). Images are representative of three to four experiments.

Close modal

By histology, the medullae of Lta−/− and Ltb−/− mice exhibited similar size but had notably reduced UEA-1 staining (Fig. 2,A). Previous histological analysis has revealed a loss of Aire expression in thymi from Ltbr−/− and Lta−/− mice, without significant alterations in the differentiation or distribution of UEA-1+ medullary epithelium (10). In contrast, Boehm et al. (9) demonstrated that signaling through the LTβR was needed for both optimal numbers and organization of UEA-1+ mTEC, a finding supported recently by Venanzi et al. (16). Although the thymic medullary defects observed in mice deficient for LTα and LTβ ligands appear to be generally less severe than in mice deficient for the LTβR (9, 16), a loss of homogenous UEA-1 distribution has been observed within the medullary regions of Ltb−/− mice (9). In line with recent reports (15, 16), we found that Aire protein expression was readily detectable on thymic sections of both Lta−/− and Ltb−/− mice (Fig. 2,B) and, upon higher magnification, appeared to be of similar intensity and localization to controls (Fig. 2 C, second panel). Therefore, the disruption of UEA-1 distribution was not accompanied by loss of Aire expression.

Flow cytometry was used to determine the precise effects of LTα or LTβ deficiency on mTEC phenotype, TSC numbers, and Aire expression. The total thymic cellularity of LTα- and LTβ-deficient mice was similar to controls and thymocyte subsets; triple negative (TN), double positive (DP), and CD4 and CD8 SP (four SP and eight SP, respectively) were not different (data not shown). Within the CD45 TSC compartment, surface staining with the epithelial cell adhesion molecule, EpCAM, defines TEC while binding of UEA-1 and Ly-51 distinguishes mTEC and cTEC, respectively (20). In control mice, approximately two-thirds of EpCAM+ TEC bound high levels of the lectin UEA-1 (UEA-1high) (Fig. 3,A). Analysis of both Lta−/− and Ltb−/− TEC revealed a significant drop in the proportion of UEA-1high cells compared with wild-type (wt) controls (EpCAM+UEA-1high mean ± SD; wt 64.1% ± 7.92, Lta−/− 31.7% ± 6.81, Ltb−/− 25.3% ± 6.20; p < 0.001), with an increased proportion of TEC-expressing low levels of UEA-1 (EpCAM+UEA-1low mean ± SD; wt 35.3% ± 7.28, Lta−/− 67.3% ± 6.82, Ltb−/− 74.75 ± 6.21; p < 0.001; Fig. 3 A).

FIGURE 3.

Flow cytometric analysis of TSC subsets in Lta−/− and Ltb−/− mice. A, UEA-1 expression on CD45, EpCAM+ TEC with regions gating UEA-1high and UEA-1low subsets. B, UEA-1 expression on CD45MHCII+ TEC with regions gating MHChighUEA-1high and MHClowUEA-1high TEC subsets. C, Expression of Ly-51 and UEA-1 positively identify cTEC and mTEC populations, respectively, on CD45EpCAM+ TEC. D, UEA-1 and Aire expression on CD45EpCAM+ TEC with regions gating UEA1highAire+ and UEA1highAire populations. E, Regions show MTS-15+PDGFRα+ and MTS15PDGFRα+ thymic fibroblast populations, gated on CD45 TSC. Dot plots are representative of 6–10 individual thymic digests. F, Enumeration of TSC populations, defined in Table II, in Lta−/−, Ltb−/− and control mice. Mean and SE were generated from two experiments each using five individual thymus digestions per group. ∗, p < 0.05.

FIGURE 3.

Flow cytometric analysis of TSC subsets in Lta−/− and Ltb−/− mice. A, UEA-1 expression on CD45, EpCAM+ TEC with regions gating UEA-1high and UEA-1low subsets. B, UEA-1 expression on CD45MHCII+ TEC with regions gating MHChighUEA-1high and MHClowUEA-1high TEC subsets. C, Expression of Ly-51 and UEA-1 positively identify cTEC and mTEC populations, respectively, on CD45EpCAM+ TEC. D, UEA-1 and Aire expression on CD45EpCAM+ TEC with regions gating UEA1highAire+ and UEA1highAire populations. E, Regions show MTS-15+PDGFRα+ and MTS15PDGFRα+ thymic fibroblast populations, gated on CD45 TSC. Dot plots are representative of 6–10 individual thymic digests. F, Enumeration of TSC populations, defined in Table II, in Lta−/−, Ltb−/− and control mice. Mean and SE were generated from two experiments each using five individual thymus digestions per group. ∗, p < 0.05.

Close modal

To determine whether the reduction in UEA-1 staining reflected loss of this mTEC subset or a change in phenotype, other mTEC markers were analyzed. Differential expression of MHCII molecules highlights further heterogeneity within mTEC and correlates with differentiation (8). Strikingly, the reduction in UEA-1 binding observed in both Lta−/− and Ltb−/− mice was found to occur specifically within the mTEClow, not mTEChigh subset (Fig. 3,B). Despite reduced UEA-1 binding, expression of MHC class I molecules and the costimulatory molecules CD80, CD86, and CD40 were unaltered on both mTEChigh and mTEClow subsets (data not shown). Costaining of anti-Ly-51 and UEA-1 to define cTEC and mTEC compartments showed they were present in equivalent proportions and number in wt, Lta−/− and Ltb−/− mice (Fig. 3, C and F), despite the drop in UEA-1 intensity within the mTEC population of LT-deficient mice (Fig. 3,C). Consistent with a recent report (16), Aire expression was also normal in Lta−/− and Ltb−/− mTEC in terms of staining intensity and cell number (Fig. 3, D and F). In addition, Aire+ mTEC in Lta−/− and Ltb−/− mice remained UEA-1high, consistent with the reduction of UEA-1 staining specifically in the AiremTEClow subset observed by FACS (Fig. 3,D). Although some Aire+ cells appeared to be UEA-1 by histology in wt and LT−/− thymic sections, this is likely to reflect in situ epitope masking, differences in the subcellular localization between Aire and UEA-1 and/or the sensitivity of FACS and immunohistology. The high expression of LTβR by non-TEC prompted a detailed analysis of this subset in Lta−/− and Ltb−/− mice. Thymic fibroblasts can be distinguished by expression of the α variant of the platelet-derived growth factor receptor α (PDGFRα) (27) and a subset therein by the Forssman glycolipid recognized by the MTS-15 Ab (28). Interestingly, CD45MTS-15+ thymic fibroblasts were dramatically reduced in both proportion and number in Lta−/− and Ltb−/− mice (Fig. 3, E and F), while CD45PDGFRα+MTS15 cells were not obviously affected. This suggests that, in accordance with the high levels of LTβR transcript found in MTS-15+ cells, LTβR signals are critical for expansion and/or maintenance of this population. CD31+ endothelium, which is closely associated with surrounding MTS-15+ regions, did not show this dependence (data not shown). In summary, both LTα- and LTβ-deficient mice exhibited a more complex phenotype than previously appreciated, characterized by reduced levels of UEA-1 binding specifically on AiremTEClow cells and severe loss of MTS-15+ fibroblasts.

In view of the broader TSC defects observed in LT-deficient mice, we undertook an assessment of TRA expression in all major TEC subsets, not just the mTEChigh explored in previous studies (9, 16). We first established the relative expression of a panel of Aire-dependent and Aire-independent TRA as defined previously (3, 4, 14) within TEC subsets of wt mice. Consistent with previous reports (4), mTEChigh cells from control mice were the predominant source of Aire and the Aire-dependent TRA transcripts casein α (Csna), casein γ (Csng), insulin 2 (Ins2), and salivary protein 1 (Sp1) (Fig. 4,B). The Aire-independent TRA transcripts casein β (Csnb), casein κ (Csnk), glutamic acid decarboxylase 1 (GAD1), fatty acid-binding protein 9 (Fabp9; note, partial Aire dependency reported) (3), and thyroglobulin (Tgn) were also highly expressed in mTEChigh cells, while collagen type 2 (Col2) and C-reactive protein (CRP) were higher in the mTEClow subset. Relative TRA expression in cTEC was very low. Transcription of K14, a marker for mTEC subsets by histology (26), was almost 3-fold higher in mTEClow than mTEChigh (Fig. 4 B), indicating marked differential expression of this molecule between these subsets.

FIGURE 4.

TRA expression in Lta−/− and Ltb−/− TEC subsets. A, Dot plots gated on CD45MHCII+ TEC showing regions used for sorting of mTEChigh, mTEClow, and cTEC populations for control and Lta−/− mice. B, PCR analysis of: Aire, casein α (Csnα), casein γ (Csng), insulin 2 (Ins2), salivary protein 1 (Sp1), casein β (Csn β), casein κ (Csnk), glutamic acid decarboxylase 1 (GAD1), fatty acid-binding protein 9 (Fabp9), type 2 collagen (Col2), C-reactive protein (CRP), thyroglobulin (Tgn), and keratin 14 (K14) in control TEC subsets, relative to highest expression level, standardized to 1. C–F, PCR analysis of TRA expression in Lta−/− mTEChigh (C), Lta−/− mTEClow (D), Ltb−/− mTEChigh (E), and Ltb−/− mTEClow (F). Fold change in transcript levels are shown relative to age-matched controls, standardized to 1 (dashed line). Means and SE generated were from three to four experiments for each population.

FIGURE 4.

TRA expression in Lta−/− and Ltb−/− TEC subsets. A, Dot plots gated on CD45MHCII+ TEC showing regions used for sorting of mTEChigh, mTEClow, and cTEC populations for control and Lta−/− mice. B, PCR analysis of: Aire, casein α (Csnα), casein γ (Csng), insulin 2 (Ins2), salivary protein 1 (Sp1), casein β (Csn β), casein κ (Csnk), glutamic acid decarboxylase 1 (GAD1), fatty acid-binding protein 9 (Fabp9), type 2 collagen (Col2), C-reactive protein (CRP), thyroglobulin (Tgn), and keratin 14 (K14) in control TEC subsets, relative to highest expression level, standardized to 1. C–F, PCR analysis of TRA expression in Lta−/− mTEChigh (C), Lta−/− mTEClow (D), Ltb−/− mTEChigh (E), and Ltb−/− mTEClow (F). Fold change in transcript levels are shown relative to age-matched controls, standardized to 1 (dashed line). Means and SE generated were from three to four experiments for each population.

Close modal

Consistent with the flow cytometric analyses, similar levels of Aire transcript were found in mTEChigh from wt, Lta−/− (Fig. 4,C), and Ltb−/− (Fig. 4,E) mice. Interestingly there was slightly decreased expression of Aire-dependent TRA in mTEChigh from Lta−/− and Ltb−/− mice, with transcripts between 40 and 60% of normal levels (Fig. 4, C and E). A similar trend was observed for some Aire-independent TRA in Lta−/−and Ltb−/− mTEChigh. It was the mTEClow population of Lta−/−and Ltb−/− mice, however, that was most severely compromised in its ability to produce TRA transcripts compared with wt levels; Csnb, Csnk, and Fabp9 were >2-fold reduced, while Col2 and CRP were reduced ∼7- and 33-fold in Lta−/− mice and 5-fold and 13-fold in Ltb−/− mice, respectively (Fig. 4, D and F). Despite these differences, K14 transcription was normal in Lta−/− mTEClow.

Thus, extensive analysis of mTEC subsets revealed that LTα and LTβ signals were not required for the expression of Aire, but were critical for normal expression of Aire-independent TRA within the AiremTEClow population.

In secondary lymphoid tissues, LTα1β2- LTβR signaling regulates the expression of the lymphoid-organizing chemokines CCL19/ELC (EBL-1 ligand chemokine), CCL21/SLC (secondary lymphoid tissue chemokine), CXCL12/SDF1 (stromal cell-derived factor 1), and CXCR13/BLC (B lymphocyte chemokine) by lymphoid stromal cells (11). In light of the thymic architecture defects observed in LT-deficient mice, we hypothesized that the LTβR pathway may similarly regulate chemokine expression and organization of the thymic stroma. The chemokines CCL19 and CCL21 are important for proper organization and development of the thymic medulla (29). Fig. 5 shows that mTEChigh and mTEClow subsets from Lta−/− mice had reduced CCL19 transcripts compared with wt controls (2- and 7-fold decreases, respectively), while expression of CCL21 and other CC chemokines was normal. Other changes included an approximate 3-fold reduction in the LTβ transcript in both Lta−/− mTEChigh and mTEClow subsets. Analysis of Ltb−/−mTEChigh and mTEClow subsets recapitulated the findings in Lta−/− TEC subsets, with a reduction in expression of CCL19 but not other CC chemokines (Fig. 5, E and F). Interestingly, Lta−/− mice demonstrated an 5- to 6-fold increase in the IL-6 transcript, an indicator of thymic injury in other models (28), in cTEC, mTEClow, and non-TEC subsets (Fig. 5, B–D). This was not found in Ltb−/− TSC subsets, suggesting that disruption of alternate LTα signaling pathways, such as LTα3-TNFR as opposed to LTα1β2- LTβR, underlies this defect.

FIGURE 5.

Chemokine and cytokine expression in Lta−/− and Ltb−/− TSC populations. PCR analysis of chemokine and cytokine transcription in Lta−/− mTEChigh (A), Lta−/− cTEC (B), Lta−/− mTEClow (C), Lta−/− non-TEC (D), Ltb−/− mTEChigh (E), and Ltb−/− mTEClow (F). Transcript levels are shown relative to age-matched wt coda standardized to 1 (dashed line). Means and SE were generated from three to four different experiments for each populations.

FIGURE 5.

Chemokine and cytokine expression in Lta−/− and Ltb−/− TSC populations. PCR analysis of chemokine and cytokine transcription in Lta−/− mTEChigh (A), Lta−/− cTEC (B), Lta−/− mTEClow (C), Lta−/− non-TEC (D), Ltb−/− mTEChigh (E), and Ltb−/− mTEClow (F). Transcript levels are shown relative to age-matched wt coda standardized to 1 (dashed line). Means and SE were generated from three to four different experiments for each populations.

Close modal

To confirm the defects in LT-deficient mTEC were primarily due to lack of LTβR signaling, Lta−/− mice were injected with an Ab agonist for the LTβR (LTβR-ag) or an isotype control Ab. Medullary TEC subsets were analyzed 8 h after LTβR-ag treatment for normalization of phenotype and TRA and chemokine expression. The activity of LTβR-ag was confirmed by analysis of whole spleen, revealing greater than 2-fold increases in several chemokines as previously reported (12) (Fig. 6,A). Fig. 6,B demonstrates that transcription of all genes analyzed in Lta−/− mTEChigh cells remained unchanged 8 h after LTβR-ag treatment, including Aire, Aire-dependent, and -independent TRA and chemokines. In contrast, the Lta−/− mTEClow subset demonstrated increased expression of most genes normally depressed, including 4-fold up-regulation of CCL19 and increased transcription of Aire-independent TRA: Csnb, Csnk, Fabp9, Col2, and CRP (Fig. 6,C). CCL21, which was not depressed in TSC from Lta−/− or Ltb−/− mice, was also increased 2-fold after LTβR-ag treatment in mTEClow from Lta−/− mice. Interestingly, the increased gene transcription in Lta−/− mTEClow following LTβR ligation was accompanied by partial restoration of UEA-1high binding levels within this subset compared with Lta−/− mice treated with isotype control Ab (isotype-Ig). Further LTβR-ag treatment (three times daily injections) increased the percentage of UEAhigh cells specifically within the Lta−/− mTEClow population, while UEA-1 intensity on the Lta−/− mTEChigh population remained unchanged (Fig. 6 D and data not shown). Together, these data show that the compromised mTEClow compartment in Lta−/− mice is a primary defect that can be restored by LTβR stimulation.

FIGURE 6.

Direct stimulation of LTβR restores mTEClow defects in Lta−/− mice. Lta−/− mice were injected with a LTβR agonist (LTβR-ag) and changes in the spleen and thymic stroma were assessed 8 h later. Whole wt spleen (A), Lta−/− mTEChigh (B), and Lta−/− mTEClow (C) subsets with fold changes for each are shown relative to Lta−/− mice injected with isotype control Ab, standardized to 1 (dashed line). Means and SE generated from two experiments. D, Flow cytometric analysis of UEA-1 binding in Lta−/− mTEClow showing the percentage of UEA1highmTEClow cells after a one-time LTβR injection (8 h later) or three-time daily LTβR-ag injections compared with injection with isotype control (isotype-Ig).

FIGURE 6.

Direct stimulation of LTβR restores mTEClow defects in Lta−/− mice. Lta−/− mice were injected with a LTβR agonist (LTβR-ag) and changes in the spleen and thymic stroma were assessed 8 h later. Whole wt spleen (A), Lta−/− mTEChigh (B), and Lta−/− mTEClow (C) subsets with fold changes for each are shown relative to Lta−/− mice injected with isotype control Ab, standardized to 1 (dashed line). Means and SE generated from two experiments. D, Flow cytometric analysis of UEA-1 binding in Lta−/− mTEClow showing the percentage of UEA1highmTEClow cells after a one-time LTβR injection (8 h later) or three-time daily LTβR-ag injections compared with injection with isotype control (isotype-Ig).

Close modal

The thymic medulla is critical for the comprehensive induction of central tolerance (30). Autoimmunity observed with deficiencies in the CCR7-CCR7L chemokine (31) and LTβR pathways (9, 14) have emphasized this importance; however, in the case of the latter, the underlying reasons for defective central tolerance are not clear. In this study, we demonstrate that rather than specifically effecting the Aire+ mTEC population, the LT pathway is necessary for the normal activity and organization of mTEClow and MTS-15+ fibroblast subsets. Importantly, LTβR stimulation regulates the expression of chemokines critical for the migration of thymocytes to medullary regions and many Aire-independent TRA expressed by mTEClow.

Medullary TEC govern the maturation and negative selection of thymocytes and are composed of phenotypically distinct subsets specialized to perform their unique function. Our data indicate that all mTEC in normal mice show high levels of UEA-1 staining by flow cytometry, but the reduced labeling observed in Lta−/− or Ltb−/− mice was due to a specific decrease within the mTEClow, not mTEChigh subset. Recent reports have shown that mTEClow can give rise to mTEChigh (8, 15); however, despite the defects we observed in Lta−/− and Ltb−/− mTEClow, their differentiation into the mTEChighAire+-expressing subset was not apparently impeded. In addition, there was no numerical deficiency in any TEC population including the Aire-expressing mTEChigh subset. Two separate reports, however, have shown a loss of mTEChigh in the Ltbr−/− mouse model (9, 16). Thus, it seems that LTβR signals contribute to maintenance of optimal mTEC numbers, but ligands other than LTα- or LTβ-containing molecules can provide sufficient stimuli for this function. For example, LIGHT expression by TSC observed in this study may fulfill this role.

Conversely, lack of LTα and LTβ caused a mild reduction in some Aire-dependent and -independent TRA in mTEChigh and a severe decrease in TRA expression in mTEClow. The TRA phenotype we observed in mTEChigh contrasts with two earlier studies that found similar TRA transcripts in mTEChigh from Ltbr−/− and Lta−/− mice using gene microarray (16) or in mTEChigh from Ltbr−/− mice using end-point PCR (9). This discrepancy may reflect the greater sensitivity of quantitative PCR used in our study and/or differences in animal age (4–8 wk compared with 4–6 mo in our study). Importantly, we show that the impairment of TRA production was much greater in the mTEClow subset, particularly for CRP and Col2, which were markedly decreased. This is noteworthy because mTEClow are a major source of CRP and Col2 in normal mice, and reduced central tolerance to Col2 has been revealed by an increased susceptibility to collagen-induced arthritis (14). This suggests that the mTEClow subset may play a more significant role in the negative selection of autoreactive thymocytes then previously appreciated (7), be it via direct contact with thymocytes or the provision of Ag to DC, which process and cross-present TRA peptides to delete self-reactive thymocytes (25, 32). In this way, LTα- or LTβ-mediated TRA expression in mTEClow might buffer negative selection and indeed be essential for tolerance in certain models.

Our data reveal a function for LTα and LTβ signals in regulating the expression of the tissue-organizing chemokine CCL19 in the thymus, in a fashion similar to that observed in peripheral lymphoid stroma. Reduced expression of both CCL19 and CCL21 in the spleen of Lta−/− and Ltb−/− mice caused disruption of the splenic microenvironment, including loss of T cell zone compartmentalization (11, 17). Thus, a model has been proposed where the LTα1β2-LTβR pathway regulates expression of CCL19, CCL21, and CXCL13 chemokines by lymphoid stromal cells, which in turn regulate LT expression on lymphocytes, establishing cytokine circuits leading to organization and homeostasis of peripheral lymphoid tissues (11). A similar mechanism may contribute to the formation of the thymic medulla, whereby expression of LTβR ligands by maturing thymocytes induces CCL19 and CCL21 expression by mTEC. Expression of CCL19 and CCL21 has been shown to be critical for the migration of positively selected thymocytes to the medulla and, in turn, cross-talk-mediated induction of mTEC (29). LTβR signals may establish or reinforce this pathway to a degree critical for central tolerance. In this context, it should be noted that LTβR-deficient mice exhibit greater disruption of the thymic medulla than Lta−/− or Ltb−/− mice (9, 16), again pointing to an important role for other LTβR ligands.

During review of this manuscript, a similar finding was published by the Fu group (33) demonstrating a loss of CCL19 and CCL21 in the thymus of LTβR-deficient mice which effected thymocyte differentiation in a TCR-transgenic model. Our data refine this study by pinpointing deficiencies and LTβR-dependent increases in CCL19 and CCL21 to the mTEClow subset of both LTα- and LTβ-deficient mice. The relative importance of CCL19 production by TSC subtypes and the quantitative threshold below which this impinges on thymocyte selection are interesting new questions that should be addressed in future studies.

Administration of LTβR-ag Abs in vivo affected increased TRA and chemokine expression only in the mTEClow subset within 8 h of treatment and was paralleled by increased UEA-1 binding, specifically within this subset. This rapid response indicates a primary and ongoing requirement for LTβR for the phenotype and function of mTEClow and that the less severe mTEChigh defects may be secondary, derived perhaps from their differentiation from impaired mTEClow precursors or a lack of LTβR-dependent signals from mTEClow or fibroblasts.

The broader analysis of the LT pathway in TSC also revealed a surprisingly high dependence on LTα or LTβ in MTS-15+PDGFRα+ thymic fibroblasts. MTS-15+ fibroblasts are found throughout the microenvironment of both thymus and spleen, predominantly associated with CD31+ blood vessels (28). In the thymus, MTS-15+ fibroblasts have been shown to provide growth factors such as fibroblast growth factor (FGF) 7 and FGF-10 critical for TEC proliferation (28, 34). The loss of these cells due to LT deficiency may indirectly affect mTEC phenotype through loss or reduction of important mesenchymal-epithelial interactions. MTS-15PDGFRα+ fibroblasts, however, were not effected by loss of LT signaling, and further study is needed to ascertain the roles of the these two phenotypically distinct thymic fibroblast populations in the thymic microenvironment.

It is becoming increasingly clear that many ligand-receptor pathways are involved in cross-talk-dependent differentiation and maintenance of mTEC, including but not limited to RANK (15), LTβR (9, 10), and CD40 (7) molecules on TSC. The dependence of mTEC on these signals is mediated through NF-κB signaling and, in general, severe medullary defects are found in mice deficient or mutant for downstream NF-κB members (35). However, despite significant overlap and redundancy, the overall NF-κB signal resulting from differential receptor activation is likely to vary slightly and thereby elicit distinctive cellular response patterns (36). By analyzing all of the major TSC subsets individually, the present study further delineates the direct and indirect effects of LTβR signals on specific TSC populations. This highlights the complexity of the thymic microenvironment and emphasizes the need to integrate the impact of such pathways on its various elements.

We thank Mark Malin for invaluable technical assistance and Darren Ellemor and Andrew Fryga for expert cell sorting.

Richard Boyd is Chief Scientific Officer of Norwood Immunology. Richard Boyd and Ann Chidgey receive consultancies from Norwood Immunlogy. Norwood Immunology provides research support for this program.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by grants from the Australian National Health and Medical Research Council and funding from Norwood Immunology Ltd. and the Australian Stem Cell Centre (to R.L.B.). D.H.D.G. was supported by a National Health and Medical Research Council C. J. Martin Overseas Training Fellowship. H.S.S. was supported by National Health and Medical Research Council Fellowships 171601 and 461204, National Health and Medical Research Council Program Grants 257501 and 264573, and Eurothymaide, 6th Framework Programme of the European Union.

7

Abbreviations used in this paper: TSC, thymic stromal cell; Aire, autoimmune regulator; DC, dendritic cell; TRA, tissue-restricted Ag; LT, lymphotoxin; TEC, thymic epithelial cell; cTEC, cortical TEC; mTEC, medullary TEC; MHCII, MHC class II; wt, wild type; UEA-1, Ulex europeaus agglutinin 1; EpCAM, epithelial cell adhesion molecule; PDGFRα, platelet-derived growth factor receptor α; FGF, fibroblast growth factor; LTBR-ag, LTBR agonist.

1
Anderson, G., W. E. Jenkinson, T. Jones, S. M. Parnell, F. A. M. Kinsella, A. J. White, J. E. Pongrac’z, S. W. Rossi, E. J. Jenkinson.
2006
. Establishment and functioning of intrathymic microenvironments.
Immunol. Rev.
209
:
10
-27.
2
Kyewski, B., L. Klein.
2006
. A central role for central tolerance.
Annu. Rev. Immunol.
24
:
571
-606.
3
Anderson, M. S., E. S. Venanzi, L. Klein, Z. Chen, S. P. Berzins, S. J. Turley, H. von Boehmer, R. Bronson, A. Dierich, C. Benoist, D. Mathis.
2002
. Projection of an immunological self shadow within the thymus by the Aire protein.
Science
298
:
1395
-1401.
4
Derbinski, J., J. Gabler, B. Brors, S. Tierling, S. Jonnakuty, M. Hergenhahn, L. Peltonen, J. Walter, B. Kyewski.
2005
. Promiscuous gene expression in thymic epithelial cells is regulated at multiple levels.
J. Exp. Med.
202
:
33
-45.
5
Mathis, D., C. Benoist.
2007
. A decade of AIRE.
Nat. Rev.
7
:
645
-650.
6
DeVoss, J., Y. Hou, K. Johannes, W. Lu, G. I. Liou, J. Rinn, H. Chang, R. R. Caspi, L. Fong, M. S. Anderson.
2006
. Spontaneous autoimmunity prevented by thymic expression of a single self-antigen.
J. Exp. Med.
203
:
2727
-2735.
7
Gray, D. H. D., N. Seach, T. Ueno, M. K. Milton, A. Liston, A. M. Lew, C. C. Goodnow, R. L. Boyd.
2006
. Developmental kinetics, turnover, and stimulatory capacity of thymic epithelial cells.
Blood
108
:
3777
-3785.
8
Gray, D., J. Abramson, C. Benoist, D. Mathis.
2007
. Proliferative arrest and rapid turnover of thymic epithelial cells expressing Aire.
J. Exp. Med.
204
:
2521
-2528.
9
Boehm, T., S. Scheu, K. Pfeffer, C. C. Bleul.
2003
. Thymic medullary epithelial cell differentiation, thymocyte emigration, and the control of autoimmunity require lympho-epithelial cross talk via LTβR.
J. Exp. Med.
198
:
757
-769.
10
Chin, R. K., J. C. Lo, O. Kim, S. E. Blink, P. A. Christiansen, P. Peterson, Y. Wang, C. Ware, Y. X. Fu.
2003
. Lymphotoxin pathway directs thymic Aire expression.
Nat. Immunol.
4
:
1121
-1127.
11
Schneider, K., K. G. Potter, C. F. Ware.
2004
. Lymphotoxin and LIGHT signaling pathways and target genes.
Immunol. Rev.
202
:
49
-66.
12
Dejardin, E., N. M. Droin, M. Delhase, E. Haas, Y. Cao, C. Makris, Z.-W. Li, M. Karin, C. F. Ware, D. R. Green.
2002
. The lymphotoxin-β receptor induces different patterns of gene expression via two NF-κB pathways.
Immunity
17
:
525
-535.
13
Tumanov, A. V., D. V. Kuprash, S. A. Nedospasov.
2003
. The role of lymphotoxin in development and maintenance of secondary lymphoid tissues.
Cytokine Growth Factor Rev.
14
:
275
-288.
14
Chin, R. K., M. Zhu, P. A. Christiansen, W. Liu, C. Ware, L. Peltonen, X. Zhang, L. Guo, S. Han, B. Zheng, Y.-X. Fu.
2006
. Lymphotoxin pathway-directed, autoimmune regulator-independent central tolerance to arthritogenic collagen.
J. Immunol.
177
:
290
-297.
15
Rossi, S. W., M. Y. Kim, A. Leibbrandt, S. M. Parnell, W. E. Jenkinson, S. H. Glanville, F. M. McConnell, H. S. Scott, J. M. Penninger, E. J. Jenkinson, et al
2007
. RANK signals from CD4+3 inducer cells regulate development of Aire-expressing epithelial cells in the thymic medulla.
J. Exp. Med.
204
:
1267
-1272.
16
Venanzi, E. S., D. H. D. Gray, C. Benoist, D. Mathis.
2007
. Lymphotoxin pathway and Aire influences on thymic medullary epithelial cells are unconnected.
J. Immunol.
179
:
5693
-5700.
17
Ngo, V. N., H. Korner, M. D. Gunn, K. N. Schmidt, D. Sean Riminton, M. D. Cooper, J. L. Browning, J. D. Sedgwick, J. G. Cyster.
1999
. Lymphotoxin αβ 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.
18
Riminton, D. Sean, H. Korner, D. H. Strickland, F. A. Lemckert, J. D. Pollard, J. D. Sedgwick.
1998
. Challenging cytokine redundancy: inflammatory cell movement and clinical course of experimental autoimmune encephalomyelitis are normal in lymphotoxin-deficient, but not tumor necrosis factor-deficient, mice.
J. Exp. Med.
187
:
1517
-1528.
19
Godfrey, D. I., D. J. Izon, C. L. Tucek, T. J. Wilson, R. L. Boyd.
1990
. The phenotypic heterogeneity of mouse thymic stromal cells.
Immunology
70
:
66
-74.
20
Gray, D. H. D., A. P. Chidgey, R. L. Boyd.
2002
. Analysis of thymic stromal cell populations using flow cytometry.
J. Immunol. Methods
260
:
15
-28.
21
Gray, D. H., A. L. Fletcher, M. Hammett, N. Seach, T. Ueno, L. F. Young, J. Barbuto, R. L. Boyd, A. P. Chidgey.
2008
. Unbiased analysis, enrichment, and purification of thymic stromal cells.
J. Immunol. Methods
329
:
56
-66.
22
Livak, K. J., T. D. Schmittgen.
2001
. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-ΔΔCT) method.
Methods
25
:
402
-408.
23
Gommerman, J. L., J. L. Browning.
2003
. Lymphotoxin/LIGHT, lymphoid microenvironments and autoimmune disease.
Nat. Rev. Immunol.
3
:
642
-655.
24
Browning, J. L., L. E. French.
2002
. Visualization of lymphotoxin-β and lymphotoxin-β receptor expression in mouse embryos.
J. Immunol.
168
:
5079
-5087.
25
Kyewski, B., J. Derbinski.
2004
. Self-representation in the thymus: an extended view.
Nat. Rev. Immunol.
4
:
688
-698.
26
Klug, D. B., C. Carter, E. Crouch, D. Roop, C. J. Conti, E. R. Richie.
1998
. Interdependence of cortical thymic epithelial cell differentiation and T-lineage commitment.
Proc. Natl. Acad. Sci. USA
95
:
11822
-11827.
27
Jenkinson, W. E., S. W. Rossi, S. M. Parnell, E. J. Jenkinson, G. Anderson.
2007
. PDGFRα-expressing mesenchyme regulates thymus growth and the availability of intrathymic niches.
Blood
109
:
954
-960.
28
Gray, D. H. D., D. Tull, T. Ueno, N. Seach, B. J. Classon, A. Chidgey, M. J. McConville, R. L. Boyd.
2007
. A unique thymic fibroblast population revealed by the monoclonal antibody MTS-15.
J. Immunol.
178
:
4956
-4965.
29
Ueno, T., F. Saito, D. H. D. Gray, S. Kuse, K. Hieshima, H. Nakano, T. Kakiuchi, M. Lipp, R. L. Boyd, Y. Takahama.
2004
. CCR7 Signals are essential for cortex-medulla migration of developing thymocytes.
J. Exp. Med.
200
:
493
-505.
30
Hogquist, K. A., T. A. Baldwin, S. C. Jameson.
2005
. Central tolerance: learning self-control in the thymus.
Nat. Rev. Immunol.
5
:
772
-782.
31
Kurobe, H., C. Liu, T. Ueno, F. Saito, I. Ohigashi, N. Seach, R. Arakaki, Y. Hayashi, T. Kitagawa, M. Lipp, et al
2006
. CCR7-dependent cortex-to-medulla migration of positively selected thymocytes is essential for establishing central tolerance.
Immunity
24
:
165
-177.
32
Gallegos, A. M., M. J. Bevan.
2004
. Central tolerance to tissue-specific antigens mediated by direct and indirect antigen presentation.
J. Exp. Med.
200
:
1039
-1049.
33
Zhu, M., R. K. Chin, A. V. Tumanov, X. Liu, Y.-X. Fu.
2007
. Lymphotoxin receptor is required for the migration and selection of autoreactive T cells in thymic medulla.
J. Immunol.
179
:
8069
-8075.
34
Jenkinson, W. E., E. J. Jenkinson, G. Anderson.
2003
. Differential requirement for mesenchyme in the proliferation and maturation of thymic epithelial progenitors.
J. Exp. Med.
198
:
325
-332.
35
Derbinski, J., B. Kyewski.
2005
. Linking signalling pathways, thymic stroma integrity, and autoimmunity.
Trends Immunol.
26
:
503
-506.
36
Yin, L., L. Wu, H. Wesche, C. D. Arthur, J. M. White, D. V. Goeddel, R. D. Schreiber.
2001
. Defective lymphotoxin-β receptor-induced NF-κB transcriptional activity in NIK-deficient mice.
Science
:
2162
-2165.