The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

The Journal of Immunology, 2008, 181, 400-407
Copyright © 2008 by The American Association of Immunologists, Inc.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Martins, V. C.
Right arrow Articles by Bleul, C. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Martins, V. C.
Right arrow Articles by Bleul, C. C.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*UniGene

Ltβr Signaling Does Not Regulate Aire-Dependent Transcripts in Medullary Thymic Epithelial Cells1

Vera C. Martins, Thomas Boehm and Conrad C. Bleul2

Max-Planck-Institute-for-Immunobiology, Department for Developmental Immunology, Freiburg, Germany


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Thymic medullary epithelial cells (mTECs) play a major role in central tolerance induction by expressing tissue-specific Ags (TSAs). The expression of a subset of TSAs in mTECs is under the control of Aire (autoimmune regulator). Humans defective for AIRE develop a syndrome characterized by autoimmune disease in several endocrine glands. Aire has been proposed to be regulated by lymphotoxin β receptor (Ltβr) signaling and there is evidence that, additionally, Aire-independent transcripts may be regulated by this pathway. Given the potential clinical importance of Aire regulation in mTECs for the control of autoimmunity, we investigated the relation between Ltβr signaling and TSA expression by whole genome transcriptome analysis. In this study, we show that the absence of Ltβr has no effect on the expression of Aire and Aire-dependent TSAs. Also, the lack of Ltβr signaling does not disturb regulatory T cells or the distribution of dendritic cells in the thymus. However, mTECs in Ltβr-deficient mice show an aberrant distribution within the thymic medulla with disruption of their three-dimensional architecture. This is predicted to impair the interaction between mTECs and thymocytes as shown by the reduced surface uptake of MHCII by mature thymocytes in Ltβr-deficient mice. We propose that the physiological medullary architecture ensures negative-selection by supporting lympho-epithelial interaction through a large epithelial cell surface distributed evenly across the medulla.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The adaptive immune system relies on a diverse repertoire of Ag receptors, each generated by a random process of somatic gene rearrangement. The randomness of the process, however, poses the potential problem of generating Ag receptors that are self-reactive. Cells expressing self-reactive receptors must be negatively selected or suppressed to preserve self-tolerance. The thymus is the lymphoid organ responsible for the generation of the diverse and self-tolerant repertoire of T cells. After positive selection in the thymus cortex, developing thymocytes reach the thymic medulla, interact with APCs, and are negatively selected if the TCR-peptide-MHC interaction occurs with very high avidity. High affinity interactions below a given threshold have been proposed to direct the cells to a regulatory T cell (Treg)3 fate (reviewed in Ref. 1).

Medullary thymic epithelial cells (mTECs) are a fundamental component of the thymic stroma. They play a unique role by expressing tissue specific Ags (TSAs) (reviewed in Ref. 2) and contribute to negative selection of self-reactive thymocytes through direct presentation and cross-presentation by dendritic cells (DCs) (3). The only known regulator for the promiscuous expression of TSAs in mTECs is Aire (autoimmune regulator) (4). Humans defective for AIRE develop autoimmune polyendocrine syndrome-type 1, also known as autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy. This is a syndrome characterized by autoimmune processes in parathyroid and adrenal glands associated with mucocutaneous candidiasis (5). Mice deficient for Aire show reduced expression of TSAs in mTECs and develop autoimmunity against several endocrine organs, resembling the phenotype of the human disease (4).

Treating autoimmune diseases that may be caused by suboptimal deletion of self-reactive thymocytes in the thymus, i.e., a defect in central tolerance induction, in man is a major challenge in the clinic. In respect to central tolerance induction, it has been proposed that signaling through the lymphotoxin β receptor (Ltβr) up-regulates the expression of Aire and Aire-dependent TSAs (6). If this hypothesis proved to be true, treatment with agonistic Abs specific for LTβR might be a clinically important approach that could lead to the mitigation of autoimmune diseases. However, our own experiments have shown no evidence for a direct regulation of Aire by Ltβr signaling (7).

Ltβr-deficient mice have an abnormal thymic medulla, produce autoantibodies against several organs (6, 7), and develop lymphocytic infiltrations (6, 8). Furthermore, these mice are also defective for the development of secondary lymphoid organs (8). Therefore, it is not clear whether the accumulation of lymphocytes in peripheral tissues results from an aberrant distribution caused by the secondary lymphoid organ defect or arises from impaired central tolerance induction in the thymus.

In this study, we address four open questions related to the importance of Ltβr signaling for the establishment of central tolerance in the thymus: first, is the autoimmune phenotype in Ltβr-deficient mice caused by a defect in central tolerance? Second, what transcriptional changes occur in mTECs in the absence of Ltβr signaling and are these similar to those seen in Aire-deficient mice? Third, what cellular component(s) is (are) associated with the breakdown of central tolerance in the Ltβr-deficient mice? Finally, is the interaction between lymphocytes and stroma altered in the absence of Ltβr signaling?


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Mice

C57BL/6, Ltβr-deficient (knockout, KO), and nude (C57BL/6 background; The Jackson Laboratory) mice were kept under specific pathogen-free conditions in the mouse facility of the Max Planck Institute for Immunobiology. Ltβr-deficient mice were backcrossed to C57BL/6 at least six generations and genotyped as described (8). Experimentation and animal care was in accordance with the guidelines of the Max-Planck-Institute for Immunobiology.

Autoantibody production by Western blotting

Western blotting was conducted according to standard protocols. In brief, whole organ extracts from liver, lung, stomach, and pancreas from Rag-1-deficient mice were prepared in 50 mM Tris/1% NP40/0.5% deoxylic acid/0.1% SDS/137.5 mM NaCl/10% Glycerin/1 mM EDTA lysis buffer supplemented with a mixture of proteinase inhibitors (complete, mini; Roche) and run in a 10% acrylamide gel. Following electrophoresis, proteins were blotted onto nitrocellulose membranes that were subsequently cut in strips. Each strip was incubated with diluted serum from one mouse and developed by incubation with goat anti-mouse IgG-HRP (Dako) followed by ECL detection (Amersham Biosciences).

Thymus transplants and analysis for lymphocytic infiltrates

Newborn Ltβr-deficient or wild-type (wt) thymi were transplanted under the kidney capsule of adult (5–9 wk of age) nude mice (C57BL/6 background) and hosts were sacrificed 14–17 wk later. Testis/ovary, salivary gland, stomach, liver, adrenal gland, and pancreas were dissected out and cryosectioned with every fifth or sixth section collected. Sections were fixed in 75% acetone/25% methanol and stained with H&E according to standard protocols.

Flow cytometry, cell sorting, and purification of mTECs

Thymi of five to six mice (4–6 wk old) were finely minced and stirred in RPMI 1640/20 mM HEPES/2% FCS medium for 10 min at room temperature, followed by three incubations of 10 min at 30°C in RPMI 1640/0.2 mg/ml collagenase IV/20 mM HEPES/2% FCS/25 µg/ml DNase and supernatant of all these fractions was discarded. Three incubations of 25 min at 30°C were performed in the same medium supplemented with dispase, followed by two incubations of 15 min at 37°C with 0.05% Trypsin/0.5 mM EDTA (pH 8.0)/0.3% BSA/25 µg/ml DNase. Cells from the latter five fractions were collected, stained, and sorted as CD45-negative, G8.8-positive, and CDR1-negative. The forward- and side-scatter gate was set to exclude thymocytes. For analysis and cell sorting of thymocytes and splenocytes, single-cell suspensions were prepared and stained with CD4-FITC (GK1.5) or -allophycocyanin (L3T4), CD8-PE-Cy7 or -FITC (53-6.7), CD25-PE (PC61.5), CD62L-allophycocyanin (MEL-14), CD69-biotin (H1.2F3), and MHCII-PE (AF6–120.1) purchased from eBioscience or BD Pharmingen and Foxp3-allophycocyanin (FJK-16s, staining kit) from eBioscience.

Gene expression analysis

A total of 10 µg total RNA from Ltβr-deficient and C57BL/6 mTECs was preamplified and biotinylated in two independent experiments. In brief, RNA was extracted using TRI reagent (Sigma-Aldrich) according to the manufacturers’ recommendations. RNA quantification and quality was measured in a Bioanalyzer machine (Agilent). Total mRNA was reverse-transcribed using an oligo(dT)-T7 promoter primer for the first strand synthesis. The second strand was synthesized and the cDNA was purified by phenol-chloroform extraction. GeneChip IVT labeling kit (Affymetrix) was used to synthesize the biotin-labeled cRNA, which was then purified and quantified. Biotin-labeled cRNA was hybridized to Affymetrix GeneChip Mouse Genome 430 2.0 Arrays for each genetic background according to the manufacturers’ protocol.

Comparison of the Ltβr-deficient/wt and Aire-deficient/wt transcriptomes and TSA identification

Raw data from Aire-deficient mTEC transcriptome analyses were downloaded from www.ncbi.nlm.nih.gov/geo. Raw data were loaded in GeneSpring software (Silicon Genetics) for normalization and Aire-deficient vs wt data were analyzed separately from Ltβr-deficient vs the corresponding wt data. The U74Av2 chip (Affymetrix) used for the Aire-deficient transcriptome analysis samples 12,000 genes that are represented within the 39,000 genes (Mouse Genome 430 2.0 chips; Affymetrix) sampled by the Ltβr-deficient transcriptome analysis. For the Ltβr-deficient/wt vs Aire-deficient/wt analysis, differences above a 2-fold change in expression levels were considered. Gene symbols were used to identify common up- or down-regulated transcripts in Aire-deficient and Ltβr-deficient mTECs. TSA status was attributed as follows: all probe set identifiers of transcripts with an at least 2-fold change in expression levels were loaded into Affymetrix Netaffx Analysis Center (www.affymetrix.com/analysis/netaffx/index.affx) and individually analyzed for their expression profile in the Unigene database (Unigene ID Build 167, 28 Oct 2007) using the direct link from the Affymetrix Netaffx Analysis Center. Genes were considered tissue specific if they were expressed in five or less tissues (from a total of 47) in the UniGene database. "Embryonic tissue" was only considered if this was the only tissue where the transcript was detected. Transcripts that had been removed from the Unigene database were not considered for analysis (52 transcripts among the decreases and 45 transcripts among the increases in LTβRKO; 9 transcripts among the decreases and 7 transcripts among the increases in AireKO).

Quantitative real-time PCR

Quantitative PCR (qPCR) was conducted with a LightCycler System (Roche) using a LightCycler-Fast-Start DNA Master SYBR Green I kit. Primers were designed to span intron-exon boundaries and yield 150–200 bp fragments. The specificity of products was assessed by melting curve analysis and gel electrophoresis. For quantitative analysis, the Fit Points Method of the Light Cycler Software version 3 was used. Results were normalized to expression levels of hypoxanthine-guanine phosphoribosyltransferase (Hprt) and the relative amount of each transcript was determined between wt and Ltβr-deficient mTECs. Primers used were: Hprt: 5'-ggttaagcagtacagcccca-3'; 5'-caagggcatatccaacaacaaact-3'; Aire: 5'-gtagcagcctaaagcctgtg-3'; 5'-acacggcacactcatcctcg-3'; Insulin2: 5'-ccaccagccctaagtgatcc-3'; 5'-tagagagcctccaccaggtg-3'; Salivary protein1: 5'-aaatgataacagtaccgggg-3'; 5'-tgcaaactcatccacgttgt-3'; Rgs13: 5'-tggagcacagtgatgagaac-3'; 5'-gatgatagcttcccgtgttg-3'; N cadherin: 5'-agatactgtggagcctgatg-3'; 5'-gttgtcagcagatttaaggc-3'; Keratin77: 5'-agcatcatcgatgccgtacg-3'; 5'-tctggatgttacggttgagc-3'; Keratin2–6a: 5'-gatcgaccacgttaagaagc-3'; 5'-atgtcctgtttggccttctg-3'; Collagen3 {alpha}1: 5'-cccagcaaaacaaaaaccac-3'; 5'-agcatttgacctgtattggg-3'; and Ccl11: 5'-tcacggtcacttccttcac-3'; 5'-ctttgcccaacctggtcttg-3'. Comparing array hybridization and qPCR experiments, a linear relationship exists between the fold changes (FC): for up-regulated genes FC(array) = 0.93 x FC(qPCR) + 3.25; for down-regulated genes: FC(array) = 0.15 FC(qPCR) + 1.75. This indicates that, in comparison to the qPCR results, the degree of down-regulation is underestimated by the array.

Immunohistochemistry

The 8 µm cryosections from Ltβr-deficient and wt thymus (4 wk old) were fixed (75% acetone/25% methanol) and incubated with Ab, followed by secondary Ab and/or streptavidin detection when required according to standard protocols. Abs were diluted in 1x PBS/0.5% BSA and slides were mounted with Fluoromount-G (Southern Biotechnology Associates). The following reagents and Abs were used: Ulex europaeus agglutinin1 (UEA-1; Vector Laboratories), Troma-1 (anti-cytokeratin 8; a gift from Dr. R. Kemler, Max Planck Institute for Immunobiology, Freiburg, Germany), anti-cytokeratin 5 (MK5; Covance), MHCII I-Ab/d (BD Biosciences), MIDC-8 (Serotec), anti-mouse Aire (kindly provided by Dr. H. Scott, Walter and Elizabeth Hall Institute, Melbourne, Australia), donkey anti-rat IgG Cy3 (Dianova), goat anti-rat IgG Alexa488, and streptavidin Alexa647 (Molecular Probes).

Proliferation/Treg assay

A total of 50,000 CD4+CD25– (effector T cells, Teff) Ltβr-deficient or wt splenocytes were sorted in triplicates and incubated for 72 h at 37°C in RPMI 1640/10% FCS/2 mM L-glutamine/2 mM HEPES/1 mM sodium pyruvate/2-ME in the presence of 1 µg/ml anti-CD3 (3C11) Ab and 50,000 T cell depleted and irradiated (300 rad) wt splenocytes. Where indicated, CD4+CD25+ (Treg) Ltβr-deficient or wt splenocytes were sorted into the same wells in the indicated proportions. Tritium (1.5µCi/ml) was added to the cultures for the last 6 h of incubation. Tritium incorporation was measured in a Trace 96 (Berthold) automatic filter counting system.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Ltβr signaling is required for central tolerance induction

Ltβr-deficient mice present with signs of autoimmunity, namely lymphocytic infiltrations in several organs (6, 8) and autoantibody production against liver, lung, pancreas, salivary gland, and stomach (6, 7). Autoantibodies in Ltβr-deficient mice are directed against several distinct tissue Ags as shown by Western blotting analysis using liver extracts (Fig. 1A). The patterns of autoantibody production were considerably diverse among individual mice, indicating that the absence of Ltβr signaling leads to a set of self-reactive lymphocytes that differs between individuals.


Figure 1
View larger version (39K):
[in this window]
[in a new window]

  		FIGURE 1. Autoimmunity in Ltβr-deficient mice results from a defect in central tolerance. A, Western blot analysis with Rag2KO whole liver extract stained with sera from Ltβr-deficient (KO, lanes 1–13) and wt (lanes 14–17) mice aged between 3 and 12 mo. B and C, Newborn Ltβr-deficient (LTβRKO) and wt thymi (one per mouse) were transplanted under the kidney capsule of nude (C57BL/6 background) mice, and recipients were analyzed for the presence of infiltrates 14–17 wk later by examination of H&E-stained sections. B, Each circle depicts the spectrum of disease in individual transplant recipients, and black regions represent the presence of mononuclear infiltrates in the corresponding organ. C, Representative H&E-stained sections from thymus transplant recipients showing extensive lymphoid infiltrates only in mice that received Ltβr-deficient thymi (KO, right), compared with recipients of wt thymi (left).

 
To understand the basis for the loss of immunologic tolerance in Ltβr-deficient mice, we first addressed whether this was due to a defect in peripheral or central tolerance induction mechanisms. Ltβr-deficient mice have a serious defect in the development of the secondary lymphoid organs (8). These mice lack lymph nodes and Peyer’s patches and have an abnormal organization of the spleen (8). Moreover, Ltβr-deficient mice show an aberrant structure in the three-dimensional arrangement of mTECs (7) that play an important role in central tolerance induction (2). Because both central and peripheral lymphoid organs are affected, organ infiltrates could be caused by a lack of central tolerance due to the defect in the thymus or be a consequence of aberrant distribution of lymphocytes in a mouse without lymph nodes. To address this problem, we transplanted wt or Ltβr-deficient newborn thymi (both C57BL/6 background) under the kidney capsule of C57BL/6 nude recipients and analyzed them 14–17 wk later for signs of autoimmunity. Ovary/testis, salivary gland, stomach, liver, adrenal gland, and pancreas were harvested, sectioned, stained, and analyzed for the presence of mononuclear infiltrations. Recipients receiving thymi from Ltβr-deficient donors developed lymphocytic infiltrates in liver, pancreas, and salivary gland (Fig. 1, B and C), while only one of the recipients from a wt thymus developed infiltrations in the liver (Fig. 1B). These findings indicate that central tolerance induction is impaired in Ltβr-deficient mice.

Ltβr signaling does not control Aire expression

In agreement with a central tolerance defect in Ltβr-deficient mice, previous reports suggested that Ltβr signaling positively regulated the expression of Aire and of both Aire-dependent as well as Aire-independent TSAs (6, 9). Furthermore, administration of an agonistic anti-Ltβr mAb was shown to increase TSA expression (6, 9). These findings opened up the clinically relevant possibility of treating autoimmune diseases by increasing TSA expression. To investigate the validity of this hypothesis in more detail we purified wt and Ltβr-deficient mTECs by cell sorting (Fig. 2A) and performed a whole transcriptome analysis. The microarray analysis for the wt vs Ltβr-deficient comparison covered a total of 39,000 transcripts, including the 12,000 transcripts surveyed in the earlier Aire-deficient transcriptome analysis (4). If Ltβr signaling positively regulated the expression of Aire and its target genes, one would expect overlap between the genes differentially regulated in Aire-deficient and Ltβr-deficient mTECs. However, this is not the case. Because of the use of different arrays, the datasets from Aire-deficient vs wt (4) and Ltβr-deficient vs wt (this paper) could not be directly compared. Rather, we generated lists of differentially regulated genes, determined the FC for each gene and then compared these lists. To avoid missing TSAs, which are expected to be expressed at very low levels, we did not filter the results with respect to expression levels. For Aire-deficient mTECs, 256 genes were down-regulated and 432 genes up-regulated at least 2-fold among 12,000 investigated transcripts (Supplemental Tables III and IV)4; using the same criteria, 553 down-regulated genes and 526 up-regulated genes were detected in the Ltβr-deficient mTEC comparisons among 39,000 investigated transcripts (Table I and Supplemental Tables I and II). Interestingly, only minimal overlap was found among these sets of genes (Table II); only 18 down-regulated and 6 up-regulated genes were found to be in common between the Aire-deficient and the Ltβr-deficient mTEC comparisons. Six of these down-regulated transcripts and one of the up-regulated were classified as tissue specific. Overall, the percentage of genes differentially expressed in Aire-deficient mTECs is twice the number in Ltβr-deficient mTECs, 5.7 vs 2.8% (p < 0.0001, {chi}2 test of association, Yates value 189.25); this difference is even more pronounced for genes deregulated at least 4-fold, 0.93 vs 0.19% (Table I). Thus, Aire deficiency has a much greater effect on transcriptional programs than Ltβr deficiency. This is also true for TSA gene expression, which dominates the Aire differences in the category of genes with at least 4-fold changes (Table I).


Figure 2
View larger version (31K):
[in this window]
[in a new window]

  		FIGURE 2. Signaling through Ltβr is not involved in the transcription of Aire and Aire-dependent TSAs in mTECs. A, Purification strategy used for sorting wt and Ltβr-deficient CD45-negative CDR1-negative G8.8-positive mTECs. The graphs show the results of a representative sort followed by reanalysis by the same instrument. Plots are gated for CD45-negative cells and set to exclude thymocytes by forward and side scatter. R3 contains 39.7% of all events in the left plot and 93.3% in the plot to the right. B, Quantitative RT-PCR analysis for the indicated transcripts on cDNA isolated from purified mTECs of Ltβr-deficient (ko) and wt mice as shown in A. Samples were normalized for the expression of Hprt. Diagrams show arbitrary units with the wt sample set to 1.

 

View this table:
[in this window]
[in a new window]

  		Table I. Tabulation of the number of differentially expressed transcripts for wt to KO transcriptome comparisonsa

 

View this table:
[in this window]
[in a new window]

  		Table II. Transcripts coregulated in Aire-deficient and LTβR-deficient mTECsa

 
Consistent with the previous results, no change was observed in the mRNA levels for Aire, Insulin2, and Spt1 in Ltβr-deficient mTECs as confirmed by qPCR (Fig. 2B). Contrary to a recent report (9), Collagen type 2 was not down-regulated in Ltβr-deficient mTECs in our analysis (data not shown and Supplemental Table II). Collectively, our analysis showed no evidence for the proposed regulation of Aire and Aire-dependent transcripts by Ltβr signaling because the transcriptome comparisons show only minimal overlap.

Ltβr-deficient mTECs have an architectural defect

Importantly, the transcriptome comparison of Ltβr-deficient to wt mTECs revealed reduced expression of several transcripts involved in epithelial cell polarization, membrane compartmentalization and cell-cell contact formation. Examples are Psd93 (discs large homologue 2, involved in the formation of the immunological synapse) (10), Advillin, Filaggrin, Keratins 10, 77, and 222, Myosin, N-cadherin (11), Neurexin I, and Integrin-{alpha}6 among others (data not shown and Fig. 2B). The differential expression of several transcripts detected by the array analysis was confirmed by qPCR (Fig. 2B).

Because a considerable number of differentially expressed transcripts pointed at a deregulation of structural components, we analyzed the medullary thymus architecture of Ltβr-deficient and wt mice by multicolor immunohistochemistry and confocal microscopy visualizing mTECs to an extent that was not possible previously (Fig. 3, A and B). The architectural defect in Ltβr-deficient mTECs affects the entire medullary epithelium (Fig. 3A). Instead of being spread throughout the medulla, mTECs clump and the three-dimensional network is disrupted, leaving large medullary areas devoid of mTECs (Fig. 3A). UEA-1+ and MHCIIhigh cells, which were shown to express high levels of TSAs (12), are particularly affected, and one can observe that they are mainly positioned at the periphery of the LTβRKO medulla instead of covering the whole area (Fig. 3, A and B). The general organization of the thymus into cortex and medulla is, however, preserved because the distribution of thymocytes is maintained, with CD4+CD8+ double-positive thymocytes localizing to the cortex and CD4+CD8– and CD4–CD8+ single-positive thymocytes to the medulla (Fig. 3C). Given the functional importance of Aire for central tolerance, we investigated its expression in the LTβRKO thymus and found isolated mTECs that expressed the Aire protein (Fig. 3, D and E). These cells are equally distributed throughout the wt medulla but are restricted to the periphery of medullary regions in the Ltβr-deficient thymus, leaving large areas devoid of Aire-expressing mTECs. Collectively, the data indicate that in the absence of Ltβr, there is a reduced density and disturbed distribution of mTECs that are known to express high levels of TSAs, suggesting that autoimmunity in these mice might arise as a result of impaired interaction with the TSA-expressing mTECs. Because DCs are also important players in the induction of central tolerance, we sought to investigate whether their distribution is affected by the absence of Ltβr signaling. However, the distribution of thymic DCs in the medullary thymic microenvironment appears to be unaffected in Ltβr-deficient mice (Fig. 3F).


Figure 3
View larger version (61K):
[in this window]
[in a new window]

  		FIGURE 3. There is an architectural defect in Ltβr-deficient mTECs whereas DCs are evenly distributed throughout the medulla. A–E, Confocal microscopy (x200 or x630 (for E) original magnification) of immunostained 4-wk-old thymus sections from LTβRKO (KO) and wt mice. A, Thymi were stained with Abs specific for keratin 8 (K8, in red) to identify cortical TEC. The mTEC were identified by staining with anti-keratin 5 (K5 in green). UEA-1-positive mTECs are represented in blue. B, Cortical TECs, in red, were stained as in A, UEA-1-positive cells are represented in green and MHCIIhigh cells are stained in blue. C, Thymocytes are stained for CD4 in green, CD8 in blue (double-positive thymocytes appear as cyan) and UEA-1-positive mTECs appear in red. The medulla is outlined by a white line. D, K5 staining of mTECs appears in red, UEA-1 staining in blue and staining with anti-Aire mAb is shown in green. E, Presents higher magnification images (x630) of the thymi shown in D. F, Sections were immunostained for mTECs using anti-K5 in green and DC by MIDC8, in red.

 
Ltβr does not influence the production and function of CD4+CD25+ Treg

The phenotype observed in Ltβr-deficient mTECs has similarities to the one in aly/aly mice (7, 13), which are defective for NF-{kappa}B-inducing kinase (14), a factor downstream of the Ltβr (15). The transplantation of aly/aly thymi into nude recipients also induces autoimmunity (16). Because aly/aly mice show an impaired production of CD4+CD25+ Treg (16), we sought to analyze whether this was also the case in the Ltβr-deficient mice. Treg are produced in the thymus and can also be generated in the periphery upon Ag stimulation. CD4+CD25+Foxp3+ cells are the best characterized Treg population and are known to have an important influence on immune responses (reviewed in Ref. 1).

We determined the percentage of CD4+CD25+Foxp3+ Treg in the thymus of Ltβr-deficient and wt mice by flow cytometry (Fig. 4A) and found no difference between them in terms of fraction among CD4+ cells and in absolute numbers (Fig. 4, A and B). Although no significant differences were detected concerning the numbers of Treg in the thymus, it was possible that the produced cells were not fully functional. Therefore, we tested their capacity to suppress the proliferation of Teff in vitro. In this assay, Treg activity was investigated by measuring the ability of increasing numbers of CD4+CD25+ Treg splenocytes to suppress the proliferation of the Teff. Both wt and Ltβr-deficient CD4+CD25– splenocytes proliferated to a similar extent (Fig. 4C). Addition of CD4+CD25+ splenocytes (known to contain Treg) to the CD4+CD25– effector cell cultures inhibited proliferation in a dose-dependent manner irrespective of whether the responding or the added cells were Ltβr-deficient or wt (Fig. 4C).


Figure 4
View larger version (40K):
[in this window]
[in a new window]

  		FIGURE 4. The Ltβr-deficient thymus produces normal numbers of functional CD4+CD25+Foxp3+ Treg. A, The gating strategy used to identify Treg in the thymus is shown. Thymocytes were first gated as CD4+CD8– cells (plots to the left) and the expression of CD25 and Foxp3 in gated CD4+ cells is shown in the plots to the right. Numbers indicate the percentage of cells falling into the respective gate or quadrant. B, Fraction in CD4+CD8– cells and absolute numbers of Treg per Ltβr-deficient (KO) and wt thymus. Each symbol represents a single mouse, and the mean is represented by a bar. C, Splenocytes from KO or wt mice were sorted as CD4+CD25– (Teff) and stimulated for 72 h with anti-CD3 and irradiated splenocytes depleted for T cells. CD4+CD25+ cells (Treg) splenocytes were sorted onto the indicated wells, in the given ratios. The graph represents the mean and SD of triplicate wells. The three last bars are controls: wt CD4+CD25+ stimulated only with anti-CD3 Ab (Teff without APC), Teff cultured without anti-CD3 (Teff without aCD3), and irradiated splenocytes depleted for T cells cultured with anti-CD3 (APC plus aCD3). Proliferation of the cells was measured as cpm. A representative experiment of three independent experiments is shown.

 
Ltβr-deficient thymocytes with the phenotype of recent thymic emigrants (RTE) interact less with the thymic stroma

Not only the reduced expression of TSAs in Aire-deficient mice but also the restricted positioning of Aire-expressing mTECs at the periphery of the medulla could lead to the reduced access of potentially autoreactive thymocytes to TSA-containing microenvironments, resulting in insufficient negative selection. The above results indicate that the mTEC network does not homogeneously cover the whole medulla and that the most affected mTECs are the ones expressing Aire and MHCII at high levels. We, therefore, hypothesized that if the available surface area of MHCIIhigh mTECs is reduced in Ltβr-deficient thymi, the frequency and duration of contacts with thymocytes might also be reduced. As a measure for contact frequency and duration we determined the amount of epithelial-thymocyte MHCII transfer. MHC molecules can be transferred from the APCs to lymphocytes upon MHC/TCR interaction (reviewed in Ref. 17). This transfer of MHC molecules also occurs in the thymus and developing thymocytes have been shown to acquire MHC molecules expressed by the stroma compartment (18). Indeed, it has been shown that the amount of MHC on the surface of thymocytes can be used as a tracing system for thymocyte-stromal cell interactions (18). The level of MHCII at the surface of thymocytes is far below the levels usually detected on DCs or epithelial cells (data not shown). However, in comparison to an isotype control, thymocytes could be shown to have surface MHCII, confirming that FACS analysis is sensitive enough to detect these low levels of MHCII (18) (Fig. 5B). Interestingly, we could not detect differences in the levels of MHCII on the surface of double-positive thymocytes between wt and Ltβr-deficient mice (Fig. 5A) (the mean fluorescence intensity in the KO is 94 ±5% of the wt). This is in accordance with the histological results that show no abnormalities in the cortical epithelium, where CD4+CD8+ double-positive thymocytes reside (Fig. 3, A and B). However, the thymocytes with the RTE phenotype have a lower amount of MHCII on their surface (mean fluorescence intensity in the KO is 66 ±12% of the wt) in the absence of Ltβr signaling (Fig. 5B), supporting our hypothesis that lympho-epithelial interactions depend on the proper mTEC architecture.


Figure 5
View larger version (26K):
[in this window]
[in a new window]

  		FIGURE 5. Ltβr-deficient RTEs interact less with the thymic stroma. The histogram plots show MHCII-specific staining on the surface of double-positive thymocytes (A) and thymocytes with the RTE phenotype (B). Ltβr-deficient (black line) and wt (gray line) thymocytes were stained, analyzed by FACS and compared with thymocytes stained with an isotype control (shaded histogram). Results of repeat experiments are represented by symbols in the graphs to the right. The mean fluorescence intensity of MHCII staining in Ltβr-deficient (KO) thymocytes for each experiment was normalized to that of wt thymocytes. Each symbol represents one experiment.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
Autoimmune diseases like rheumatoid arthritis, systemic lupus erythematodes, scleroderma, inflammatory bowel disease, and others cause significant morbidity and mortality. Although many of the pathomechanisms leading to autoimmune disease have been correlated with the breakdown of peripheral tolerance, recent years have seen a revival of the important role of central tolerance in the prevention of autoimmunity (2, 19). In this study, we present evidence to suggest that the autoimmunity in Ltβr-deficient mice is caused by the combined effects of dysregulation of Aire-independent TSAs and aberrations in medullary architecture. We show that in the absence of Ltβr, mTECs express reduced amounts of transcripts involved in epithelial cell polarization and appear to interact less with thymocytes while thymic DCs and Treg seem unaffected. Thus, the molecular mechanisms underlying autoimmune processes in mice lacking Ltβr appear to be fundamentally different from those in Aire and NF-{kappa}B-inducing kinase-deficient mice.

Central tolerance induction, i.e., negative selection of potentially self-reactive thymocytes in the thymus, is a coordinated process that depends on the interaction between thymocytes and two key stromal components in the thymus, namely thymic epithelium and bone marrow-derived DCs (2). For this interaction to be productive, self-Ags must be presented as peptide-MHC complexes that are accessible to thymocytes. Thymocytes are thought to be deleted if they are reactive to the presented self-Ags, or allowed to undergo further development if they recognize self-Ags with low affinity. Although it is conceptually easy to understand how tolerance to widely expressed Ags is maintained, this is more difficult in the case of Ags that are specific to organs like the brain or the eye. The expression of a given TSA is restricted to a very small number of mTECs (20), resulting in very low overall levels of expression. Still, it has been described that mTECs are competent mediators of negative selection of TSA-reactive thymocytes, although some uptake and cross-presentation by DCs also occurs (3). Because the expression of TSAs is so restricted, the proper presentation of these rare Ags is of critical importance because it must ensure that all thymocytes are exposed to all TSAs, each expressed by a minute number of mTECs, before exiting into the periphery. Thus, tolerance induction might fail either when TSAs are not presented by peptide-MHC complexes (4, 21) or when such complexes cannot be contacted by thymocytes.

Autoimmunity in Ltβr-deficient mice has been proposed to be caused by the lack of Ltβr signaling-mediated regulation of Aire expression (6). If this were true, significant overlap between the changes in the expression profiles of Aire- and Ltβr-deficient mTECs would be expected. We show here that this is not the case. In a non-exclusive scenario, Ltβr signaling might control a unique set of TSAs, independent of Aire, as proposed previously (9). Although this might be true for a small number of TSAs, it does not appear to be the sole mechanism, as it is the case for Aire. The absence of one single TSA was shown to induce organ-specific autoimmunity (21). However, this does not appear to be the case of the Ltβr-deficient mice, which develop a generalized autoimmune phenotype, where several organs are targeted and a polyclonal repertoire of self-reactive immunoglobulins is produced. The pattern of autoantibody production indicates that some common targets exist, while individual mice also have unique targets. This is consistent with a scenario where some self-Ags are missing and others are presented in a suboptimal fashion to developing thymocytes, resulting in a random element of autoantibody production.

Could the autoimmune phenotype observed in Ltβr-deficient mice be caused by structural abnormalities in the medulla? Several observations support this interpretation. First, we find that autoimmunity is transplantable and, thus, resides in the stromal compartment of the thymus. Second, the multicolor immunohistochemistry data show areas that clearly lack mTECs but are packed with lymphocytes; third, there is evidence that thymocytes take up reduced amounts of surface MHCII from thymic stroma cells, suggesting reduced lympho-epithelial interaction.

Taken together, our results show that Ltβr signaling in the thymic stroma is required for central tolerance induction in an Aire-independent manner. Instead, Ltβr signaling is required for mTECs to acquire their proper three-dimensional conformation, which optimizes the accessibility of self-peptide MHC to maturing thymocytes. Central tolerance induction in the thymus requires an ideal microenvironment where self-Ags are optimally displayed, promoting the interaction between the thymocytes and the thymic epithelium.

While this manuscript was under review, two other reports were published that support our conclusions. Venanzi et al. (22) use a whole-transcriptome approach where mTECs from Aire- and Ltβr-deficient were analyzed and only a minimal overlap was detected, showing that the role of Aire and Ltβr in mTECs is independent and does not involve lack of negative selection. As their transgenic approach (3) does not address mTEC-mediated deletion of CD4 cells, the result might be explained by the intact deletion mediated by DCs, which we observe to be unaffected in Ltβr-deficient mice. The second publication (23) detected a defect in negative selection when using a different transgenic approach (3); this system addresses CD8 T cells that are deleted by direct Ag presentation by mTECs.

In conclusion, a plausible scenario explaining the autoimmunity in Ltβr-deficient mice emerges. Loss of central tolerance is not caused by lack of Aire function but results from the combined effects of reduced expression of a unique set of TSAs and aberrations in stromal architecture reducing the interactions between thymocytes and cells capable of presenting self Ags.


   Acknowledgments
 
We thank Dr. R. Kemler (Max Planck Institute for Immunobiology, Freiburg, Germany) for providing the anti-cytokeratin 8 (Troma-1) Ab and Dr. Hamish Scott (Walter and Elizabeth Hall Institute, Melbourne, Australia) for providing the Aire Abs. We thank A. Wuerch and J. Wersing for FACS sorting and J. Kirberg for helpful discussions. We thank N. Plaster, I. Rode, L. Calderon, and J. Swann for discussion of the manuscript.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
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 V.C.M. was supported by a fellowship from Fundação para a Ciência e Tecnologia, Portugal. Back

2 Address correspondence and reprint requests to Dr. Conrad C. Bleul, Max-Planck-Institute-for-Immunobiology, Stuebeweg 51, Freiburg 79108, Germany. E-mail address: bleul{at}immunbio.mpg.de Back

3 Abbreviations used in this paper: Treg, regulatory T cell; TSA, tissue-specific Ag; mTEC, medullary thymic epithelial cell; Ltβr, lymphotoxin β receptor; DC, dendritic cell; wt, wild type; KO, knockout; Hprt, hypoxanthine-guanine phosphoribosyltransferase; qPCR, quantitative PCR; Teff, T effector cell; RTE, recent thymic emigrant. Back

4 The online version of this article contains supplemental material. Back

Received for publication August 17, 2007. Accepted for publication April 30, 2008.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Sakaguchi, S.. 2004. Naturally arising CD4+ regulatory t cells for immunologic self-tolerance and negative control of immune responses. Annu. Rev. Immunol. 22: 531-562. [Medline]
  2. Kyewski, B., L. Klein. 2006. A central role for central tolerance. Annu. Rev. Immunol. 24: 571-606. [Medline]
  3. 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. [Abstract/Free Full Text]
  4. 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. [Abstract/Free Full Text]
  5. Bjorses, P., J. Aaltonen, N. Horelli-Kuitunen, M. L. Yaspo, L. Peltonen. 1998. Gene defect behind APECED: a new clue to autoimmunity. Hum. Mol. Genet. 7: 1547-1553. [Abstract/Free Full Text]
  6. 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. [Medline]
  7. 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. [Abstract/Free Full Text]
  8. Futterer, A., K. Mink, A. Luz, M. H. Kosco-Vilbois, K. Pfeffer. 1998. The lymphotoxin β receptor controls organogenesis and affinity maturation in peripheral lymphoid tissues. Immunity 9: 59-70. [Medline]
  9. 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. [Abstract/Free Full Text]
  10. Ludford-Menting, M. J., J. Oliaro, F. Sacirbegovic, E. T. Cheah, N. Pedersen, S. J. Thomas, A. Pasam, R. Iazzolino, L. E. Dow, N. J. Waterhouse, et al 2005. A network of PDZ-containing proteins regulates T cell polarity and morphology during migration and immunological synapse formation. Immunity 22: 737-748. [Medline]
  11. Miyatani, S., K. Shimamura, M. Hatta, A. Nagafuchi, A. Nose, M. Matsunaga, K. Hatta, M. Takeichi. 1989. Neural cadherin: role in selective cell-cell adhesion. Science 245: 631-635. [Abstract/Free Full Text]
  12. 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. [Abstract/Free Full Text]
  13. Miyawaki, S., Y. Nakamura, H. Suzuka, M. Koba, R. Yasumizu, S. Ikehara, Y. Shibata. 1994. A new mutation, aly, that induces a generalized lack of lymph nodes accompanied by immunodeficiency in mice. Eur. J. Immunol. 24: 429-434. [Medline]
  14. Shinkura, R., K. Kitada, F. Matsuda, K. Tashiro, K. Ikuta, M. Suzuki, K. Kogishi, T. Serikawa, T. Honjo. 1999. Alymphoplasia is caused by a point mutation in the mouse gene encoding Nf-{kappa}b-inducing kinase. Nat. Genet. 22: 74-77. [Medline]
  15. Matsushima, A., T. Kaisho, P. D. Rennert, H. Nakano, K. Kurosawa, D. Uchida, K. Takeda, S. Akira, M. Matsumoto. 2001. Essential role of nuclear factor (NF)-{kappa}B-inducing kinase and inhibitor of {kappa}B (I{kappa}B) kinase {alpha} in NF-{kappa}B activation through lymphotoxin β receptor, but not through tumor necrosis factor receptor I. J. Exp. Med. 193: 631-636. [Abstract/Free Full Text]
  16. Kajiura, F., S. Sun, T. Nomura, K. Izumi, T. Ueno, Y. Bando, N. Kuroda, H. Han, Y. Li, A. Matsushima, et al 2004. NF-{kappa}B-inducing kinase establishes self-tolerance in a thymic stroma-dependent manner. J. Immunol. 172: 2067-2075. [Abstract/Free Full Text]
  17. Wetzel, S. A., D. C. Parker. 2006. MHC transfer from APC to T cells following antigen recognition. Crit. Rev. Immunol. 26: 1-21. [Medline]
  18. Merkenschlager, M.. 1996. Tracing interactions of thymocytes with individual stromal cell partners. Eur. J. Immunol. 26: 892-896. [Medline]
  19. Mathis, D., C. Benoist. 2004. Back to central tolerance. Immunity 20: 509-516. [Medline]
  20. Derbinski, J., A. Schulte, B. Kyewski, L. Klein. 2001. Promiscuous gene expression in medullary thymic epithelial cells mirrors the peripheral self. Nat. Immunol. 2: 1032-1039. [Medline]
  21. Devoss, J., Y. Hou, K. Johannes, W. Lu, G. I. Liou, J. Rinn, H. Chang, 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. [Abstract/Free Full Text]
  22. Venanzi, E. S., D. H. Gray, C. Benoist, D. Mathis. 2007. Lymphotoxin pathway and Aire influences on thymic medullary epithelial cells are unconnected. J. Immunol. 179: 5693-5700. [Abstract/Free Full Text]
  23. 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. [Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Martins, V. C.
Right arrow Articles by Bleul, C. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Martins, V. C.
Right arrow Articles by Bleul, C. C.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*UniGene

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS