TNF-like Ligand 1A (TL1A) Gene Knockout Leads to Ameliorated Collagen-Induced Arthritis in Mice: Implication of TL1A in Humoral Immune Responses

Xuehai Wang, Yan Hu, Tania Charpentier, Alain Lamarre, Shijie Qi, Jiangping Wu and Hongyu Luo
J Immunol December 1, 2013, 191 (11) 5420-5429; DOI: https://doi.org/10.4049/jimmunol.1301475

Abstract

TNF-like ligand 1A (TL1A), also known as TNFSF15, is a member of the TNF superfamily. Its known receptor is death receptor 3 (DR3). In humans, TL1A also binds to a secreted TNF family member called decoy receptor 3, which interferes with the interaction between TL1A and DR3. TL1A/DR3 signal has been implicated in several autoimmune diseases in animal models as well as in clinical conditions. We generated TL1A gene knockout (KO) mice to assess its role in collagen-induced arthritis (CIA), a mouse model of human rheumatoid arthritis. The KO mice were fertile and had no visible anomalies. Their lymphoid organ size and cellularity, T and B cell subpopulations, Th cell and regulatory T cell development in vivo and in vitro, and antiviral immune responses were comparable to those of wild-type mice. However, the KO mice presented ameliorated CIA in terms of clinical scores, disease incidence, and pathological scores. The KO mice had reduced titers of pathogenic anti-collagen Abs in the sera. No apparent defect was found in the function of follicular Th cells. We revealed that plasma cells but not B cells expressed high levels of DR3 and were direct targets of TL1A. In the presence of TL1A, they survived better and produced more pathogenic Ab. This study presented novel knowledge about the role of TL1A in humoral immune responses and its mechanism of action in CIA pathogenesis.

Introduction

Tumor necrosis factor–like ligand 1A (TL1A), also called TNFSF15, is a member of the TNF superfamily. It was identified during homology searches in an endothelial cell cDNA library. It is produced by the endothelial cells (1), monocytes, dendritic cells (DC), T cells, NKT cells, synovial fibroblasts and chondrocytes, either upon stimulation or in situ in inflammatory sites (2–6). Like other TNF superfamily members, TL1A can either be membrane-bound or soluble after being cleaved from the cell surface, but the cleavage happens on DC and not on T cells (1, 7). It was also discovered that the soluble form of TL1A was more potent in inducing IFN-γ production by T cells (8).

The receptor of TL1A death receptor 3 (DR3) is also called TNFRSF25, Apo3, WSL-1, TRAMP, or LARD (9). In addition to the full-length isoform, murine DR3 was reported to have two truncated variants: one secreted form displaying no transmembrane region and another lacking the fourth cysteine-rich domain in its extracellular sequence (10). A repertoire of all DR3-expressing cells has not yet been completed. DR3 is detectable on T cells, NK cells, NKT cells, osteocytes, renal tubule epithelial cells, and in vitro plasma cells, but primary B cells express very low levels of DR3 (3, 11–13). Previous studies of the immunological functions of TL1A and DR3 primarily focused on T cells, NK cells, and NKT cells (3, 12, 14–16). In addition to DR3, decoy receptor 3 also could bind to TL1A in humans and blocks the interaction between TL1A and DR3 (1). Mice do not have a human decoy receptor 3 ortholog.

Earlier in vitro studies have shown that TL1A promotes T cell proliferation, Th17 differentiation and the production of proinflammatory cytokines by T cells (9), indicating its potential role in inflammatory autoimmune diseases. The presence of TL1A in diseased tissues, fluids, and sera from patients of chronic colitis and rheumatoid arthritis (RA) corroborates the findings in animal models (2, 5, 17–19). In vivo studies of different animal models have provided direct evidence supporting the role of TL1A in autoimmune diseases. We have demonstrated that the administration of recombinant TL1A to mice with collagen-induced arthritis (CIA) aggravates the disease (5). Furthermore, overexpression of TL1A in transgenic mice results in spontaneous inflammatory bowel disease as well as increased number of Th cells (9, 20, 21). In contrast, DR3 gene knockout (KO) mice have reduced disease severity of allergic lung inflammation and Ag-induced arthritis and present less effective antiviral and antibacterial immune responses (15, 22–25). Blocking DR3–TL1A interaction by TL1A-neutralizing Ab would partially protect the KO mice from CIA, lung inflammation, and 2,4,6-trinitrobenzene sulfonic acid– and dextran sodium sulfate–induced colitis (3, 4, 9, 23). TL1A KO mice are less susceptible to experimental autoimmune encephalomyelitis (EAE) induction (22).

RA is a chronic polygenic autoimmune disease characterized by chronic systemic inflammation that may affect various tissues and organs. Joints and their surrounding tissues are most frequently affected. Often there is persistent synovitis and presence of autoantibodies leading to progressive destruction of cartilage and bone in the joints (26). The pathogenesis of RA is not fully understood with three primary outstanding questions to be fully answered: 1) How is the immune response initiated by genetic and/or environmental factors? 2) How does the immune response lead to local joint inflammation? 3) How does the inflammation cause bone destruction. There are some findings from patients and animal models that have provided some partial answers to these questions (27–36).

In the current study, we generated TL1A KO mice to study the role of TL1A in CIA pathogenesis. We found TL1A KO mice had significantly lower CIA clinical scores and incidence and lower serum anti-collagen Ab titers than their wild-type (WT) counterparts. Deletion of TL1A minimally affected T cell functions; however, TL1A could directly deliver survival signals to plasma cells as a way to promote pathogenic collagen-specific Ab production. Our findings revealed previously undocumented functions of TL1A in RA pathogenesis.

Materials and Methods

Generation of TL1A KO mice

A PCR fragment amplified from the TL1A cDNA was used as probe to isolate genomic bacterial artificial chromosome DNA clone 112H6 from the 129/sv mouse bacterial artificial chromosome genomic library RPCI-22. The targeting vector was constructed by recombination and routine cloning methods using an 11-kb TL1A genomic fragment from clone 112H6 as the starting material (Fig. 1A) (37). The 2.05-kb ScaI–XhoI genomic fragment containing exon 1 was replaced by a 1.1-kb Neo cassette from pMC1Neo Poly A (Fig. 1A). The final targeting fragment was excised from its cloning vector backbone by NotI digestion and electroporated into R1 embryonic stem (ES) cells for G418 (38). The targeted ES cell clones were injected into C57BL/6 blastocysts. Chimeric male mice were mated with C57BL/6 females to establish mutated TL1A allele germline transmission.

Southern blotting with a probe corresponding to the 5′ sequence outside the targeting region, as illustrated in Fig. 1A (hatched rectangle), was used to screen gene-targeted ES cells and eventually for the confirmation of the gene deletion in mouse tail DNA. The targeted allele showed an 11.7-kb StulI/StulI band and the WT allele, a 5.4-kb StulI/StulI band.

The TL1A KO mice were subsequently backcrossed to the DBA/1LacJ for eight generations for experimentation in this study. The littermate WT mice served as controls. In the beginning of the experiments, mice were on average 8–12 wk of age.

PCR was adopted for routine genotyping of the targeted allele(s). The following PCR conditions were applied: 4 min at 94°C, followed by 35 cycles of 30 s at 94°C, 30 s at 68°C, and 30 s at 72°C with final incubation at 72°C for 10 min. One forward primer and two reverse primers (forward primer, 5′-TCC ACA GCC AAC ATA GGC AAG GA-3′; reverse primer 1, 5′-GTG TGG CTT GCA ACA GGA AAT GGA-3′; and reverse primer 2, 5′-ACC TGC GTG CAA TCC ATC TTG TTC-3′) were included in the PCR, which amplified a 581-bp fragment from the targeted allele and a 335-bp fragment from the WT allele. All mice were housed under specific pathogen-free conditions and studied according to protocols approved by the Institutional Animal Protection Committees of the Centre hospitalier de l’Université de Montréal and Institut national de la recherche scientifique - Institut Armand-Frappier.

Reverse transcription-quantitative PCR

Murine TL1A and DR3 mRNA was measured by reverse transcription-quantitative PCR (RT-qPCR). Total RNA was isolated from cells or tissues with TRIzol reagent (Invitrogen, Carlsbad CA). For TL1A mRNA measurement, the forward primer (5′-TCA TTT CCC ATC CTC GCA GGA CTT-3′) and reverse primer (5′-TAA TTG TCA GGT GTG CTC TCG GCT-3′) were used to generate a 166-bp fragment. For measurement of mRNA of all three DR3 isoforms (the full-length one, the one with transmembrane domain deletion, and the one with deletion of the extracellular carbohydrate recognition domain), the forward primer (5′-AGA GGT ATG GCC CGT TTT G-3′) and reverse primer (5′-AAG TGG TTG TCT CTG GTC AAG-3′) were used to generate a 133-bp fragment. For mRNA of the full-length DR3 isoform, the forward primer (5′-TGC CTG GCT GGC TTC TAT ATA CGT G-3′) and reverse primer (5′-ACA GAC AGC AGT GCA AGC CTT A-3′) were used to generate a 92-bp fragment. The PCR conditions for both reactions were as follows: 2 min at 95°C followed by 45 cycles of 10 s at 95°C, 15 s at 56°C, and 20 s at 72°C. Samples were in triplicates. β-actin mRNA levels was measured as described in our previous publication and taken as internal controls (5). The data were expressed as signal ratios of either TL1A mRNA/β-acting mRNA or DR3 mRNA/β-actin mRNA.

Flow cytometry

Single-cell suspensions from the thymus, spleen, and lymph nodes (LN) were prepared and stained immediately or after culture for two-, three-, four-, or five-color flow cytometry. Mouse anti-mouse TL1A mAb (Tandys1a) was from eBioscience (San Diego, CA). Anti-programmed death-1 (PD-1) mAb; anti-GL7 mAb, anti-ICOS mAb, anti–L-selectin (CD40L) mAb, anti-CXCR5 mAb, and anti-DR3 mAb were from BD Biosciences (Mississauga, ON, Canada) and eBioscience. The rest of the mAbs used in flow cytometry are described in our previous publications (39).

Cell culture and in vitro Th cell differentiation

In general, total spleen or LN cells, T cell subpopulations, B cell subpopulations, or bone marrow cells were prepared, isolated with Miltenyi beads or sorted by flow cytometry as indicated, and then cultured in RPMI 1640 medium containing 10% FCS, 100 μg/ml streptomycin, 100 U/ml penicillin G, 1× nonessential amino acids, 1 μM sodium pyruvate, 2.5 μM 2-ME, and other supplements as indicated. In some cases, recombinant mouse TL1A (aa 76–252; R&D Systems, Minneapolis, MN) was added to the culture.

In vitro Th differentiation was conducted as follows. Naive CD4 T cells (CD4+CD62L+CD44low) were isolated from TL1A KO or WT mouse spleens with MagCellect Mouse Naive CD4+ T cell Isolation kits (R&D Systems). T cell–depleted TL1A KO or WT spleen cells were irradiated at 3000 rad and used as feeder cells. The naive CD4 cells (0.1 × 106/well) were mixed with the feeder cells (0.5 × 106/well) and cultured in 96-well plates in the presence of soluble anti-CD3ε mAb (clone 145-2C11, 2 μg/ml; BD Biosciences). Cultures were supplemented with recombinant mouse IL-12 (10 ng/ml; R&D Systems) and anti–IL-4 mAb (10 μg/ml; R&D Systems) for the Th1 condition; recombinant mouse IL-4 (20 ng/ml; R&D Systems), anti–IL-12 mAb (10 μg/ml; BD Biosciences), and anti–IFN-γ mAb (10 μg/ml; R&D Systems) for the Th2 condition; recombinant mouse IL-6 (20 ng/ml; R&D Systems), recombinant human TGF-β1 (5 ng/ml; R&D Systems), and anti–IL-4 and anti–IFN-γ mAb (10 μg/ml; R&D Systems) for the Th17 condition; recombinant human TGF-β1 (5 ng/ml; R&D Systems), anti–IL-4, and anti–IFN-γ mAb (10 μg/ml; R&D Systems) for the regulatory T (Treg) cell condition; and recombinant mouse IL-6 (20 ng/ml; R&D Systems), anti–IL-4, and anti–IFN-γ mAb (10 μg/ml) for the follicle T help (Tfh)–like cell condition, without any supplement for the Th0 condition.

Mouse CIA model

Eight- to 12-wk-old TL1A KO mice and their WT littermates in the DBA/1LacJ background were used in the experiments. Details of the induction of CIA in these mice were described previously (5). Briefly, mice were immunized with single intradermal injection at the base of the tail with 100 μg bovine type II collagen (BTIIC; Chondrex, Redmond, WA), which was emulsified in equal volumes of CFA (4 mg/ml Mycobacterium tuberculosis, strain H37Ra; Difco, Detroit, MI) on day 0. The mice were examined for their development and severity of arthritis from day 20 to 50. Disease severity was scored on a scale from 0 to 4 by visual inspection of each paw according to methodology of Brand et al. (40) with some modifications as follows: 0, normal paw; 1, erythema and mild swelling confined to two or more digits or the tarsal or the ankle joint; 2, erythema and mild swelling of any two regions of the metatarsal, tarsal, or ankle joint; 3, erythema and moderate swelling extending from the ankle to metatarsal joints; and 4, erythema and severe swelling encompassing the ankle, foot and digits, or ankylosis of the limb. The scores for each of four paws were added to give a final score, such that the maximal severity score was 16 for each mouse. The mice were considered as having arthritis, if their clinical score was equal to or above 1.

Histology

Mice immunized with BTIIC/CFA were sacrificed on day 50, and then, their paws including ankles were surgically removed and fixed in 10% buffered formalin. Following decalcification in 5% formic acid, the specimens were processed for paraffin embedding. Tissue sections (7 μm) that were stained with H&E and safranin O (for cartilage staining) were scored according to three parameters (i.e., lining hyperplasia, bone erosion destruction, and cell infiltration) in one-way blind fashion with the examiner not knowing the identity of the sections. Each parameter was scored on a 0–3 scale, as previously described, and all four paws (one section per paw) of each animal were scored. The overall score of an animal was the sum of the three parameters of the four paws (41).

Isolation of infiltrating cells of hematopoietic origin in paws

The mouse vascular system was perfused with 20 ml PBS under anesthesia. The paws were then harvested, skinned, and minced into small pieces, which was then digested with collagenase II (2 mg/ml; Chondrex, Redmont, WA) and dispase II (250 μg/ml; Roche Diagnostics, Indianapolis, IN) in 5 ml HBSS at 37°C for 40 min. The digested product was washed and passed through cell strainers (BD Biosciences, San Jose, CA) of 70 μm in pore size, consecutively. Cells in the passing-through liquid were enumerated and stained with anti-CD45.2, CD11c, CD11b, Thy1.2, and B220 mAbs and analyzed and counted with flow cytometry.

ELISA for the titers of anti-BTIIC Abs

ELISA was used to measure the titers of mouse serum anti-BTIIC Abs. Flat-bottom 96-well plates (Costar EIA/RIA plate number 3590, Fisher Scientific, Pittsburgh, PA) were coated with BTIIC in PBS (100 μl/well) overnight at 4°C. After blocking for 1 h at room temperature with 3% BSA in PBS, the plates were washed with PBS containing 0.05% Tween 20. Samples in duplicate were added to the wells (50 μl/well) and incubated overnight at 4°C. After extensive washes, 50 μl HRP-conjugated goat anti-mouse IgG1, IgG2a, and IgG2b antisera (1:5000 dilution; Southern Biotechnology Associates, Birmingham, AL) or rabbit anti-mouse IgG (1:5000 dilutions; GE Healthcare Life Sciences, Little Chalfont, Buckinghamshire, U.K.) were added to each well. Two hours later, the plates were washed, and tetramethylbenzidine (50 μl/well; BD Biosciences) was added, followed by 10-min incubation at room temperature in the dark. Finally, 2 N H2SO4 (50 μl/well) was added to the wells to stop the reaction, and optical densities were determined at 450 nm. The serum anti-BTIIC Ab titer of a WT with the highest level of anti-BTIIC Ab on day 28 after the BTIIC/CFA immunization was arbitrarily designated as a titer of one and used as a standard to determine the titers of all the serum samples.

ELISPOT for enumeration of anti-BTIIC Ab–producing cells

Multiscreen filter plate (Millipore, Billerica, MA) was coated with BTIIC as described for ELISA. The wells were blocked with culture medium (RPMI 1640 medium containing 10% FCS) for 2 h at 37°C and then washed. Cells from draining LN (dLN) (1 × 105 cells/100 μl/well) or bone marrow (5 × 105 cells/100 μl/well) in culture medium were added to the wells and incubated for 4 h at 37°C. The plates were then washed, and biotin–streptavidin-conjugated AffiniPure F(ab′)2 fragment of goat anti-mouse IgG Ab (Jackson ImmunoResearch Laboratories, West Grove, PA) was applied. After overnight incubation at 4°C, the plates were extensively washed. Alkaline–phosphatase-conjugated streptavidin (1:60; R&D Systems) was added to the wells and incubated for 2 h at room temperature. The spots were visualized by adding 5-bromo-4-chloro-3-indolyl phosphate/NBT substrate (R&D Systems). Pictures of each well were taken with an ELISPOT reader (C.T.L., Shaker Heights, OH), and spots were counted visually.

Results

Generation of TL1A KO mice and general features of their immune system

To understand the role of TL1A in immune responses, we generated TL1A KO mice. The targeting strategy was illustrated in Fig. 1A. Germ-line transmission of targeted TL1A was confirmed by Southern blotting using tail DNA (Fig. 1B), and PCR was used for routine genotyping of KO, WT, and heterozygous pups (Fig. 1C). TL1A deletion of KO mice at the mRNA level was verified by RT-qPCR as shown in Fig. 1D. The deletion of TL1A protein in T cells from KO mice was demonstrated in Fig. 1E, according to flow cyotmetry. The KO mice were fertile and had no visible anomaly. Lymphoid organs such as the thymus, spleens, and LN were of normal sizes and cellularity (data not shown). The CD4CD8 double-negative, CD4CD8 double-positive, CD4 single-positive, and CD8 single-positive subpopulations in the TL1A KO thymi were comparable to those in the WT thymi (Fig. 2A). The CD4 and CD8 T cell percentages (Fig. 2A) and B cell percentages (Fig. 2B) in the spleen and LN of TL1A KO mice were in the normal ranges. These results indicate that TL1A KO does not significantly affect T and B cell development.

FIGURE 1.
FIGURE 1.

Generation of TL1A KO mice. (A) Targeting strategy to generate TL1A KO mice. The shaded rectangle on the 5′ side of the mouse TL1A WT genomic sequence represents the sequence serving as a probe for genotyping by Southern blotting analysis. (B and C) Genotyping of TL1A mutant mice. Tail DNA was digested with StuI and analyzed by Southern blotting, with the 5′ probe whose location is indicated in (A). Arrowheads indicate 11.7-kb bands representing the WT allele and a 5.4-kb band representing the recombinant allele (B). PCR was used for routine genotyping with ear-lobe tissue DNA. The 335-bp bands representing the WT allele and the 581-bp bands representing the recombinant allele are indicated by arrowheads (C). (D) Absence of TL1A mRNA expression in TL1A KO T cells. Spleen T cells from WT and TL1A KO mice were stimulated with solid anti-CD3 (2 μg/ml) and anti-CD28 (2 μg/ml) for 24 h. Cell RNA was extracted and analyzed by RT-qPCR for TL1A expression. The results are expressed as ratios of TL1A versus β-actin signals with means ± SD indicated. (E) Absence of TL1A protein expression in KO T cells. WT and TL1A KO spleen T cells were stimulated as in (D) and stained with anti-CD4 mAb and anti-TL1A Ab and analyzed by two-color flow cytometry. The shaded lines represent the isotypic Ab control, and the thick lines represent TL1A staining of CD4+-gated T cells of WT (left panel) and KO (right panel) mice. Experiments in (D) and (E) were repeated at least twice and representative data are shown. HET, Heterozygous.

FIGURE 2.
FIGURE 2.

TL1A KO mice presented normal lymphocyte subpopulations. Experiments in this figure were repeated more than three times, and representative data are presented (A, B, D). (A) T cell populations in WT and TL1A KO LN, spleens, and thymus. Thymocytes, mesenteric LN cells, and splenocytes were analyzed by two-color flow cytometry for percentages of different T cell subpopulations. (B) B cell populations in the LN and spleen of WT and KO mice. Mesenteric LN cells and splenocytes were analyzed by two-color flow cytometry for percentages of CD19+/B220+ B cells. (C) Normal Th1, Th17, and Treg populations in TL1A KO mice ex vivo. WT or TL1A KO spleen cells were stimulated with PMA (50 ng/ml) and ionomycin (1 μg/ml) for 4 h in the presence of BD GolgiStop and analyzed by flow cytometry ex vivo for Th1, Th17, and Treg cell populations based on intracellular IFN-γ, IL-17, and Foxp3 staining. The bar graphs represent the summary of data from three to four experiments, with mean + SD indicated. (D) Normal differentiation of TL1A-KO CD4 cells in vitro. Naive TL1A-KO CD4 cells were cultured under conditions favoring Th1, Th2, Th17, and Treg cells, using WT or TL1A KO feeder cells, as indicated. Their intracellular cytokine or Foxp3 expression was quantified by three-color flow cytometry on day 3 for Th1, Th17, and Treg cells and on day 5 for Th2 cells after 4 h PMA (50 ng/ml) and ionomycin (1 μg/ml) stimulation in the presence of BD GolgiStop before the assay.

We examined Th1, Th17 and Treg cell populations in the spleen and LN of TL1A KO and WT mice ex vivo, but no apparent differences were found (Fig. 2C). Th2 cells in the KO and WT spleens were below the reliable detection level (data not shown). We next assessed whether a lack of TL1A affected Th1, Th2, Th17, and Treg cell development in vitro. We compared WT and KO T cell–depleted splenocytes in their capacity to support the development of the said T cell populations, using naive KO CD4 T cells as starting cells, which were used because they did not produce TL1A to confound result interpretation. The KO non-T splenocytes were capable of supporting the naive KO CD4 T cells differentiating into Th1, Th2, Th17, and Treg cells in a similar fashion as were their WT counterparts in supporting of naive KO CD4 T cells (Fig. 2D). This suggests that a lack of TL1A minimally affects Th1, Th2, Th17, and Treg development in vitro.

The antiviral immune responses of TL1A KO mice were evaluated in a lymphocytic choriomeningitis virus (LCMV) infection model. As illustrated in Supplemental Fig. 1A, the number of CD4 cells and CD8 cells in the spleens on day 8 postinfection presented no significant differences in WT and TL1A KO mice. The absolute numbers and relative percentages of LCMV-specific, tetramer-positive (gp33-41+ and np396-404+) CD8 cells in virus-infected mice were all increased in comparison with uninfected control C57BL/6 mice (data not shown), but there were no significant differences between TL1A KO and WT mice with regard to these parameters (Supplemental Fig. 1B, 1C). The absolute number and relative percentages of LCMV-specific TNF-α–producing CD4 (gp61-80+) and CD8 cells (gp33-41+) (Supplemental Fig. 1D, 1E), and LCMV-specific IFN-γ–producing CD4 (gp61-80+) and CD8 cells (gp33-41+) (Supplemental Fig. 1F, 1G) in TL1A KO mice were comparable to those in WT controls. This result indicates that TL1A is not essential in anti-LCMV immune responses.

TL1A KO mice presented ameliorated CIA

Our previous study demonstrated that administering recombinant TL1A aggravates CIA (5). Bull et al. (23) demonstrated that neutralizing Ab against TL1A partially blocks Ag-induced arthritis. We wondered whether a total lack of TL1A would affect RA pathogenesis in the CIA model. TL1A KO and WT mice in the DBA/1LacJ background were immunized with BTIIC/CFA to induce CIA. The dLN of the immunized mice were assessed on day 14 after BTIIC/CFA immunization, a midterm point between the starting of the immunization and a full-fledged CIA. The KO dLN were smaller in size (Fig. 3A, first panel) and lower in cellularity (second panel) compared with their WT counterparts. The absolute numbers of T cells and B cells in these KO dLN were significantly lower than those in the WT dLN (third and fourth panels). We assessed the size of the germinal centers (GC) and the percentage and absolute number of GC B cells in the dLN. As shown in Supplemental Fig. 2A, TL1A KO dLN showed no abnormality in term of the size of the GC. This is consistent with the fact that the percentages of GC B cells among the total dLN cells were comparable between WT and KO (Supplemental Fig. 2B). However, the absolute number of GC B cells in the dLN was significantly lower in KO than in WT mice (Supplemental Fig. 2C), because of a reduction of the cellularity of the KO dLN.

FIGURE 3.
FIGURE 3.

TL1A-KO mice manifested less severe CIA. (A) Reduced numbers of immune cells in KO dLN during CIA induction. Mice were immunized with BTIIC/CFA at the tail base on day 0. On day 14 after the immunization, dLN were collected and photographed (first panel), and cells from dLN of WT (n = 10) and TL1A KO (n = 11) mice were isolated, enumerated, and analyzed by flow cytometry for subpopulations. The numbers of total dLN cells per mouse (second panel), CD4 T cells (CD4+) of the dLNs per mouse (third panel), and B cells (B220+) of the dLNs per mouse (fourth panel) are shown. The horizontal bars indicate the means. Student t test was used to assess the statistical significance of the difference between WT and TL1A KO mice, and the p values are indicated in the second, third, and fourth panels. (B) CIA clinical scores of WT and KO mice after BTIIC/CFA immunization. WT (n = 8) and TL1A KO (n = 9) mice were immunized with BTIIC/CFA on day 0 and scored for their RA symptoms daily starting from day 21 until day 50. Mean ± SEM scores are plotted. Scores for the WT and TL1A KO mice were evaluated statistically by two-way ANOVA.The p value is indicated. (C) CIA incidence in WT and TL1A KO mice. The CIA incidence of the WT and TL1A KO groups in (A) is plotted. (D and E) Histology of mouse joints. WT and TL1A KO mice from (B) were sacrificed on day 50, and all the paws were sectioned and stained with H&E or safranin O (for cartilage staining) as indicated. Representative histology images from each group are shown (D). Boxed areas in the images in the upper panels are shown with a higher magnification in the bottom panels in (D). Asterisks indicate synovial hyperplasia, and arrowheads indicate cartilage damage. The histology of each paw (one section per paw) was evaluated by blinded raters and scored semiquantitatively based on lining hyperplasia, bone erosion, and cell infiltration. Each parameter was scored on a 0–3 scale, and all four paws of each animal were scored. The overall score for each animal, which is the sum of the three parameters of the four paws, is shown in the plot and the horizontal bars represent the mean scores (E). Student t test was performed to assess the statistical significant of the difference between WT and TL1A KO mice, and the p value is indicated. (F) Reduced numbers of infiltrating immune cells in KO paws during CIA induction. On day 19 after BTIIC/CFA immunization, before the onset of CIA symptoms, cells infiltrating the paws of WT and TL1A KO mice were isolated, enumerated, and analyzed by flow cytometry for subpopulations. The total numbers of infiltrating cells of hematopoietic origin (CD45.2+) per paw are shown in the first panel; percentage of DC (CD11c+) in the second panel; percentage of neutrophils/monocytes/macrophages (CD11b+) in the third panel; percentage of T cells (Thy1.2+) in the fourth panel; and percentage of B cells (B220+) in the fifth panel. Mouse numbers per group are indicated. Original magnification ×1 (A), ×40 (D). Student t test was used to assess the statistical significance of the difference between WT and TL1A KO mice, and the p values are indicated. KO-CIA, Joints from KO mice immunized with BTIIC/CFA; Naïve, joints from normal unimmunized DBA/1 mice; WT-CIA, joints from WT mice immunized with BTIIC/CFA.

The TL1A KO mice displayed slower onset of CIA and significantly lower clinical scores (Fig. 3B). Incidence of CIA in the TL1A KO mice was also lower compared with that among the WT controls before day 31 after the immunization, although all TL1A KO mice eventually developed CIA (Fig. 3C). At the end of the experiment, on day 50, the arthritic paws of the TL1A KO and WT groups were collected and then assessed histologically. As illustrated in Fig. 3D, the TL1A KO paws had less synovial membrane hyperplasia (asterisks), less immune cell infiltration and milder cartilage erosion (arrowheads). The pathological scores of the TL1A KO groups were significantly lower compared with the WT controls (Fig. 3E). We isolated cells from paws on day 19 after BTIIC/CFA immunization, which was chosen because it was shortly before the onset of the disease and paws were still normal in appearance so that the selection of the paws was in no way biased. The number of cells of the hematopoietic origin isolated from the paws was significantly smaller in TL1A KO mice than in WT mice, but the percentages of DC, neutrophils/macrophages/monocytes, T cells, and B cells among the total isolated cells from TL1A KO and WT paws showed no statistically significant differences (Fig. 3F).

These results clearly show that a lack of TL1A diminishes the severity of CIA.

TL1A KO mice had lower collagen-specific Ab titers after BTIIC/CFA immunization

The nonimmunized TL1A KO and WT mice had comparable levels of total serum IgG, IgM, IgA, and IgG subtypes IgG1, IgG2a IgG2b, and IgG3 (Supplemental Fig. 3). Autoantibodies are pathogenic in RA (42, 43), and anti-collagen Ab is especially so in CIA. We measured the BTIIC-specific Ab during the course of CIA from day 14 to day 50 after BTIIC/CFA immunization. The BTIIC-specific total IgG titers in the KO mice were significantly lower than their WT counterparts at all time points starting from day 21 (Fig. 4A). Among the IgG isotypes, BTIIC-specific IgG2a and IgG2b but not IgG1 titers in the TL1A KO mice were significantly lower than in the WT mice (Fig. 4B–D). The reduced titers of collagen-specific Abs titers, especially the IgG2a and IgG2b isotypes, which are major pathogenic Ab isotypes in CIA (44), is likely a relevant factor for the observed reduced CIA severity in the TL1A KO mice.

FIGURE 4.
FIGURE 4.

Reduced collagen-specific Ab production in KO mice with CIA. Sera were obtained from WT (n = 7) and TL1A KO (n = 7) mice on days 14, 21, 28, 35, 42, and 50 after BTIIC/CFA immunization. Arbitrary titers of collagen-specific total IgG (A), IgG1 (B), IgG2a (C), and IgG2b (D) were determined by ELISA. Serum from a WT mouse exhibiting typical arthritis was arbitrarily given a titer of 1 and used as reference in each assay. For (A)–(D), mean ± SEM is presented and evaluated statistically by two-way ANOVA. The p values are indicated.

CIA-related T cell functions in TL1A KO mice

T cells are a major type of target cells of TL1A and play vital role in RA pathogenesis including producing inflammatory lymphokines such as IFN-γ and IL-17 and providing help to B cells for pathogenic Ab production. We assessed the Ag-specific dLN T cells for their IFN-γ and IL-17 production and proliferation upon in vitro collagen restimulation shortly before the onset of CIA on day 14 after the immunization, a time point these cells reportedly become detectable in this location (45, 46). However, no significant difference between TL1A KO and WT dLN was found (Supplementary Fig. 4A, 4B). This finding is in agreement with a previous study concerning the role of TL1A in Ag-specific responses. In the OT-II system, deficiency of TL1A/DR3 signal does not affect naive T cells differentiation under Ag-specific stimulation, and DR3 KO dLN cells from EAE mice proliferate normally to the stimulation of MOG peptides in vitro (15). It is also compatible with the data of polyclonal T cell activation as shown in Fig. 2.

CRCXR5+/PD-1+ Tfh cells express CD40L and ICOS on the cell surface and produce IL-21, all of which are essential in B cell survival, differentiation, and eventually Ab production. We examined Tfh in the dLN on day 18 after the immunization, shortly before the lower titers of BTIIC-specific Abs in the KO CIA mice became apparent. The TL1A KO and WT dLN harbored similar percentages of CRCXR5+/PD-1+ Tfh (Fig. 5A, left column); TL1A KO and WT Tfh cells expressed comparable levels of CD40L and ICOS (Fig. 5A, right column). Probably as a consequence, the Fas+GL7+ GC B cells percentages in the dLN of TL1A KO and WT were not different (Fig. 5B). With that said, because the absolute number of T cells in the KO dLN was reduced (Fig. 3A), the absolute number of Tfh in the KO dLN was still smaller than that of WT dLN. We confirmed that undifferentiated CD4 T cells (Th0) did not expression DR3, but Th1, Th17, and Tfh cells all expressed DR3, providing a basis for them to respond to TL1A (Fig. 5C). In vitro–differentiated Tfh cells from naive TL1A KO and WT CD4 cells were all capable of produce high levels of IL-21, compared with KO Th0, Th1, and Th17 cells (Fig. 5D). Exogenous TL1A did not enhance IL-21 production by KO Tfh cells (Fig. 5D). The results of this section suggest that the deletion or exogenous TL1A does not have significant effect on Tfh cells with regard to their B cell helper functions, such as their CD40L/ICOS expression and IL-21 production, although the diminished absolute number of Tfh in the KO dLN might be a factor contributing to the reduced level of collagen-specific Abs in KO CIA mice.

FIGURE 5.
FIGURE 5.

TL1A KO T cells presented normal help to B cells. The experiments in this figure were repeated three times, and representative data are shown. (A) Flow cytometry analysis of Tfh in dLN of mice after BTIIC/CFA immunization. WT or TL1A KO mice were immunized with BCTII/CFA and then sacrificed on day 18. CD4+ T cells from dLN were purified and stimulated with solid anti-CD3ε (2 μg/ml) and anti-CD28 (2 μg/ml) Ab for 30 min. The percentage of CD4+PD-1+CXCR5+ Tfh cells among CD4+ cells was determined by flow cytometry and shown (left column). The gated Tfh were further analyzed for their CD40L and ICOS expressions (right column), and the percentages of CD40L+ and ICOS+ cells among Tfh (PD-1+/CXCR5+) cells are indicated. (B) Flow cytometry analysis of GC B cell populations. B220+ B cells from the dLN of WT and TL1A KO mice on day 18 after BTIIC/CFA immunization were analyzed for GL7 and Fas expression. The percentages of B220+GL7+Fas+ GC B cells among B220+ B cells are indicated. (C) DR3 expression on Th and Tfh cells. Naive CD4+ T cells were cultured on anti-CD3ε (2 μg/ml for coating) and anti-CD28 (2 μg/ml for coating) Ab-coated wells under Th1, Th17, and Tfh-like cell differentiation conditions. After 3 d, DR3 expression on CD4+ cells was assessed by flow cytometry. Shaded area, isotypic controls. Thick lines, DR3 signals. (D) IL-21 production by in vitro–differentiated WT and KO Tfh cells. WT or TL1A KO naive CD4+ T cells were cultured in anti-CD3ε– and anti-CD28 Ab–coated wells under the Tfh-like cell differentiation condition. They were also cultured under Th0 (without reagents for Th differentiation except anti-CD3ε, anti–IFN-γ, and anti–IL-4 Abs), Th1, or Th17 conditions as controls. TL1A at 10 and 50 ng/ml was added to some cultures as indicated. On day 3, the cells were restimulated with PMA/ionomycin for 9 h, and IL-21 in the supernatants was measured by ELISA. Samples in ELISA were in duplicate and means ± SD are shown.

Plasma cells are direct target cells of TL1A

It is puzzling that, in TL1A KO CIA mice, there was an obvious decrease in Ag-specific Ab production, yet T cell’s B cell–facilitating function showed no signs of compromise, with the exception of a decrease of absolute number of Tfh. To date, B cells and their derivative plasma cells are not known target cells of TL1A because an earlier study demonstrates that B cells do not express or express very little DR3 (3, 13), although Pelletier et al. (47) did reveal that in vitro generated human plasma cells express DR3 at the protein level. We confirmed that indeed, B220+CD138− B cells did not express DR3, nor did B220+CD138+ plasmablasts (Fig. 6A, left and middle panels). However, we discovered that plasma cells from dLN of BTIIC/CFA-immunized WT mice expressed very high levels of DR3 (Fig. 6A, right panel). DR3 has a full-length isoform and also truncated isoforms (10). According to RT-qPCR, T cells and plasma cells expressed similar levels of the full-length DR3 isoform Fig. 6B, left panel), whereas the former expressed more isoforms with deletions (Fig. 6B, middle and right panels). The expression of the full-length DR3 isoform on plasma cells provided a basis for them to respond to TL1A. We then interrogated the effect of TL1A on plasma cells. B cells and plasma cells from dLN of KO mice 21 d after BTIIC/CFA immunization were negatively selected using B cell negative selection kit (StemCell Technologies) by eliminating T, NK, monocyte/macrophages, and DC. The remaining B cells and plasma cells were cultured for 6 d in the presence of exogenous TL1A, because there was no endogenous TL1A in this system. In the presence of TL1A, significantly more plasma cells survived after the 6-d culture (Fig. 6C, left panel), and there were higher levels of collagen-specific IgG secreted into the supernatants (Fig. 6C, right panel), compared with the culture without TL1A. In support of such in vitro results, on day 28 after the BTIIC/CFA immunization, when the anti-BTIIC Ab production was actively produced and WT mice had significantly higher titers of serum anti-BTIIC Abs, dLN cells and bone marrow cells from WT mice presented significantly higher numbers of anti-BTIIC Ab–secreting plasma cells compared with those from TL1A KO dLN, according to ELISPOT (Fig. 6D–F). Collectively, the data from this section suggest that TL1A can directly promote plasma cell survival and function in terms of Ab production, and this effect in turn contributes to the elevated pathogenic Ab production in CIA. Conversely, in the absence of TL1A, the well-being of the plasma cells is compromised, and this is one of the factors leading to decreased levels of pathogenic anti-collagen Ab in TL1A KO mice.

FIGURE 6.
FIGURE 6.

The effect of TL1A on plasma cells. The experiments in (A)–(C) were repeated three times and representative data are shown. (A) DR3 expression in plasma cells. dLN cells from WT mice on day 15 after BTIIC/CFA immunization were gated on B220+CD138− regular B cells (left panel), B220+CD138+ plasmablast, and Dump−B220−CD138+ plasma cells and analyzed for their DR3 expression by three-color or four-color flow cytometry. Shaded areas, isotypic controls. Thick lines, DR3 signals. Dump staining used anti-CD8, anti-CD11b, anti-CD11c, anti-F4/80, anti-IgM, anti-IgD, and anti-CD4 mAbs. (B) DR3 isoform expression in CD4 cell and plasma cells. RT-qPCR was used to detect the mRNA of full-length DR3 and all DR3 isoforms (including the full-length and all truncated isoforms) in flow cytometer–sorted CD4+ T cells (as controls) and Dump−B220−CD138+ plasma cells. The results are expressed as ratios of DR3 (full-length or all isoforms) versus β-actin signals with means ± SD indicated. Left panel, full-length DR3 isoform expression in CD4 T and plasma cells; middle panel, the expression all DR3 isoforms (full-length plus all truncated isoforms) in CD4 T and plasma cells; and right panel, ratios of full-length DR3 versus all isoforms of DR3. (C) Higher plasma cell numbers in and collagen-specific Ab production by WT dLN cells cultured in the presence of TL1A. dLN cells from WT mice 21 d after BTIIC/CFA immunization were cultured in the absence or presence of TL1A (50 ng/ml) for 6 d. The number of Dump−B220−CD138+ plasma cells was counted by flow cytometry four times on day 6, and the plasma cell numbers per 1 × 106 total dLN cells (means ± SD) are shown (left panel). The culture supernatants were harvested on day 6, and collagen-specific IgG levels in the supernatants were determined by ELISA, which was conducted in triplicate samples. Arbitrary titers of anti-collagen IgG Abs (means ± SD) are shown (right panel). Experiments were repeated at least twice and representative data are shown. Student t test was used to assess the statistical significance of the difference, and the p values are indicated. (D–F) Reduced collagen-specific IgG-producing cells (CSIGGPC) in the dLN and bone marrow (BM). Twenty-eight days after BTIIC/CFA immunization, CSIGGPC in the dLN and BM was enumerated by ELISPOT. Representative images of spots in wells are shown in (D). The spots in eight replicate wells of each experiment were counted, and the data of three similar experiments were pooled and summarized by bar graphs [(E) for dLN and (F) for BM]. Mean ± SEM are presented. *p < 0.05; **p < 0.01 (Student t test).

Discussion

In this study, we revealed that TL1A KO mice had ameliorated CIA compared with WT mice. No significant T cell dysfunction including Th17 and Treg differentiation was apparent in the absence of TL1A, and the TL1A KO mice had normal anti-LCMV immune responses. The KO mice presented reduced production of pathogenic anti-collagen Ab. A novel finding in this study is that plasma cells were a direct target of TL1A, which these cells expressed high levels of the full-length isoform of TL1A receptor DR3. TL1A promoted plasma cell’s survival and Ab production.

The results of our current study and several previous studies support the general concept that TL1A can intensify inflammatory response. For example, DR3 KO mice present reduced EAE and Ag-induced arthritis (15). The administration of exogenous TL1A aggravates CIA, whereas TL1A-neutralizing Ab reduces the severity of CIA (5, 23). Transgenic overexpression of TL1A worsens dextran sodium sulfate–induced colitis and causes intestinal mucosal inflammation (9, 20, 21, 48). Several mechanistic aspects of these disease models corroborate each other. For example, our data using TL1A KO mice showed that TL1A was not required for constitutive or induced Th1, Th2, Th17, and Treg cell development; studies using deletion of DR3 reached the same conclusion (15). However, there are also discrepancies between the conclusions obtained from DR3 KO mice versus TL1A KO mice and from TL1A administration/overexpression versus TL1A/DR3 deletion studies. The discrepancies and possible explanations are discussed as follows.

It seems that excessive TL1A, either from exogenous recombinant protein or endogenous transgenic overexpression, often results in enhanced IFN-γ and IL-17 expression, but TL1A KO or DR3 KO has minimal effect on constitutive or inducible Th1 and Th17 cell development (5, 9, 15, 48). A possible explanation for this discrepancy is that such Th cell development is vital to the biological system, and there is sufficient redundancy to compensate for the missing DR3 or TL1A in the KO mice. As a consequence, in the absence of DR3 or TL1A, Th cells could still be developed to near-normal levels. Alternatively, DR3 and TL1A are not utterly essential in Th cell development; the enhanced Th cell development in the presence of nonphysiological high doses of exogenous or transgenic TL1A is due to its cross-reaction with other TNFR superfamily members (12). The results from CIA mice treated with TL1A-neutralizing Ab favor the first explanation (23). In this model, there is no excessive exogenous TL1A, and the reduction of the bioactive TL1A occurs at the adult stage and is not complete, less likely triggering a drastic compensation. Such a neutralizing Ab treatment can hamper the development of Th1 and Th17 cells, which are reported essential in CIA pathogenesis, and results in reduced CIA severity (49).

We demonstrated that TL1A KO mice showed no abnormality in antiviral immune responses against LCMV. However, DR3-KO mice manifest compromised anti-murine CMV immunity (24). Although TL1A is a ligand of DR3, there are ∼20 other TNF superfamily members that share some degrees of homology with TL1A. In the absence of TL1A, other members of the TNF superfamily might individually or collectively bind to DR3, albeit at a lower affinity, and such binding might trigger some low level DR3 signaling, which could be sufficient to compensate for the missing TL1A. In contrast, a missing DR3 will totally eliminate the DR3 signaling, resulting more drastic phenotype including compromised antivirus immune responses. Although we favor this hypothesis, we cannot exclude the possibility that the observed difference in DR3 KO and TL1A KO regarding the antivirus immune responses is caused by different viruses (i.e., murine CMV versus LCMV).

Another interesting point is the role of TL1A in Treg development. In our TL1A KO and WT mice, no difference was found in constitutive or induced Treg populations. Not surprisingly, there was no aggravation but amelioration of CIA in the TL1A KO mice. In mice with transgenic TL1A expression or mice administered with exogenous recombinant TL1A, the immune/inflammation responses were aggravated but not ameliorated (5, 9, 21, 48). In all the studies using either TL1A or DR3 KO mice models or TL1A neutralizing Abs (3, 15, 23–25), the immune responses were abated. Although the Treg status of the above mentioned studies are not always assessed, at least we could conclude that excessive TL1A does not lead to reduced immune response, a possible functional consequence of Treg upregulation. Conversely, TL1A or DR3 deletion does not lead to exuberant immune responses, which are a possible functional consequence of Treg downregulation.

Our ex vivo data indicated that the percentages of Th1 and Th17 cells in dLN were comparable between CIA WT and CIA KO mice. However, because there was a drastic reduction of cellularity and absolute T cell number in the dLN of KO mice, the absolute numbers of these effector cells were also reduced. Such reduction could likely cause a compromised CIA development in the TL1A KO mice.

Several members of TNF family are involved in regulating the humoral immune responses. They could either affect the B cells directly or indirectly through T cells. A well-studied member is CD40L, which is essential to generation of plasma cells (50). CD40/CD40L signaling initiates GC responses, GC B cell proliferation, isotype switching, and differentiation of B cells into plasma cells (51). Because DR3 was reportedly expressed mainly on T cells in the immune system, we assessed the effect of TL1A on Tfh cells. Our ex vivo data indicated that the percentages of Tfh cells in dLN were comparable between WT and KO mice with CIA. The functions of KO Tfh cells, in terms of CD40L expression and IL-21 production, also were not compromised. However, because there was a significant reduction of absolute T cell number in the dLN of KO mice, comparing to that of WT mice, it is conceivable that absolute number of Tfh and GC B cells in the dLN of KO mice was reduced. Such reduction might contribute to compromised collagen-specific Ab production in CIA TL1A KO mice.

We found that the plasma cells also expressed functional DR3. This finding pinpointed such plasma cells as a target cell population for TL1A’s effect. We demonstrated that TL1A could promote dLN plasma cell survival in vitro. The direct effect of TL1A on plasma cell survival might be one of the contributing mechanisms for the following observation: 1) TL1A KO results in reduced number of collagen-specific Ab-producing plasma cells ex vivo and reduced collagen-specific Ab titers in vivo; and 2) recombinant TL1A enhanced collagen-specific Ab production in vitro. However, we noticed that in unimmunized KO mice, their total serum IgG levels and plasma cell population were comparable to those of WT mice. It is possible that the benefit of TL1A to plasma cells is only for a limited time during a humoral response and perhaps even specific to a restricted stage of the plasma cells.

We demonstrated that reduced pathogenic Ab production is a mechanism contributing to milder CIA in the TL1A-KO mice. However, it is probably not the only mechanism responsible for this phenotype. We have assessed several T cells function in the absence of TL1A and did not find any anomaly, but such an assessment is by no means exhaustive. We did find that there was a significant reduction of inflammatory immune cell infiltration in the CIA paws, and this phenomenon is consistent with the findings with several disease models in that there is always a reduction of inflammatory cells in the diseased organ or tissues when DR3 or TL1A is missing (15, 22). This is certainly an additional mechanism by which TL1A contributes to the pathogenesis of CIA. We tested TL1A KO T cell chemotaxis toward several CIA-related chemokines in vitro using Transwell, but no significant defect was detected. More comprehensive investigations in this aspect including assessing local chemokine secretion by T cells, monocyte/macrophages, and DC, and the endothelium permeability in the excessive or absence of TL1A are warranted.

Disclosures

The authors have no financial conflicts of interest.

Footnotes

  • This work was supported by Canadian Institutes of Health Research Grants MOP97829 (to H.L.), MOP69089 and MOP123389 (to J.W.), and MOP89797 (to A.L.). This work was also supported by grants from the Heart and Stroke Foundation of Quebec and the Jean-Louis Levesque Foundation (to J.W. and A.L.) and Natural Sciences and Engineering Research Council of Canada Grant 203906-2012 (to J.W.).

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    BTIIC
    bovine type II collagen
    CD40L
    L-selectin
    CIA
    collagen-induced arthritis
    DC
    dendritic cell
    dLN
    draining lymph node
    DR3
    death receptor 3
    EAE
    experimental autoimmune encephalomyelitis
    ES
    embryonic stem
    GC
    germinal center
    KO
    knockout
    LCMV
    lymphocytic choriomeningitis virus
    LN
    lymph node
    PD-1
    programmed death-1
    RA
    rheumatoid arthritis
    RT-qPCR
    reverse transcription-quantitative PCR
    Tfh
    follicle T help
    TL1A
    TNF-like ligand 1A
    Treg
    regulatory T
    WT
    wild-type.

  • Received June 5, 2013.
  • Accepted September 17, 2013.

References

  1. ↵
    1. Migone T. S.,
    2. J. Zhang,
    3. X. Luo,
    4. L. Zhuang,
    5. C. Chen,
    6. B. Hu,
    7. J. S. Hong,
    8. J. W. Perry,
    9. S. F. Chen,
    10. J. X. H. Zhou,
    11. et al
    . 2002. TL1A is a TNF-like ligand for DR3 and TR6/DcR3 and functions as a T cell costimulator. Immunity 16: 479–492.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Bamias G.,
    2. M. Mishina,
    3. M. Nyce,
    4. W. G. Ross,
    5. G. Kollias,
    6. J. Rivera-Nieves,
    7. T. T. Pizarro,
    8. F. Cominelli
    . 2006. Role of TL1A and its receptor DR3 in two models of chronic murine ileitis. Proc. Natl. Acad. Sci. USA 103: 8441–8446.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Fang L.,
    2. B. Adkins,
    3. V. Deyev,
    4. E. R. Podack
    . 2008. Essential role of TNF receptor superfamily 25 (TNFRSF25) in the development of allergic lung inflammation. J. Exp. Med. 205: 1037–1048.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Takedatsu H.,
    2. K. S. Michelsen,
    3. B. Wei,
    4. C. J. Landers,
    5. L. S. Thomas,
    6. D. Dhall,
    7. J. Braun,
    8. S. R. Targan
    . 2008. TL1A (TNFSF15) regulates the development of chronic colitis by modulating both T-helper 1 and T-helper 17 activation. Gastroenterology 135: 552–567.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Zhang J.,
    2. X. Wang,
    3. H. Fahmi,
    4. S. Wojcik,
    5. J. Fikes,
    6. Y. Yu,
    7. J. Wu,
    8. H. Luo
    . 2009. Role of TL1A in the pathogenesis of rheumatoid arthritis. J. Immunol. 183: 5350–5357.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Prehn J. L.,
    2. S. Mehdizadeh,
    3. C. J. Landers,
    4. X. Luo,
    5. S. C. Cha,
    6. P. Wei,
    7. S. R. Targan
    . 2004. Potential role for TL1A, the new TNF-family member and potent costimulator of IFN-γ, in mucosal inflammation. Clin. Immunol. 112: 66–77.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Kim S.,
    2. L. Zhang
    . 2005. Identification of naturally secreted soluble form of TL1A, a TNF-like cytokine. J. Immunol. Methods 298: 1–8.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Biener-Ramanujan E.,
    2. R. Gonsky,
    3. B. Ko,
    4. S. R. Targan
    . 2010. Functional signaling of membrane-bound TL1A induces IFN-γ expression. FEBS Lett. 584: 2376–2380.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Meylan F.,
    2. Y.-J. Song,
    3. I. Fuss,
    4. S. Villarreal,
    5. E. Kahle,
    6. I.-J. Malm,
    7. K. Acharya,
    8. H. L. Ramos,
    9. L. Lo,
    10. M. M. Mentink-Kane,
    11. et al
    . 2011. The TNF-family cytokine TL1A drives IL-13‑dependent small intestinal inflammation. Mucosal Immunol. 4: 172–185.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Wang E. C.,
    2. J. Kitson,
    3. A. Thern,
    4. J. Williamson,
    5. S. N. Farrow,
    6. M. J. Owen
    . 2001. Genomic structure, expression, and chromosome mapping of the mouse homologue for the WSL-1 (DR3, Apo3, TRAMP, LARD, TR3, TNFRSF12) gene. Immunogenetics 53: 59–63.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Bu R.,
    2. C. W. Borysenko,
    3. Y. Li,
    4. L. Cao,
    5. A. Sabokbar,
    6. H. C. Blair
    . 2003. Expression and function of TNF-family proteins and receptors in human osteoblasts. Bone 33: 760–770.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Al-Lamki R. S.,
    2. J. Wang,
    3. A. M. Tolkovsky,
    4. J. A. Bradley,
    5. J. L. Griffin,
    6. S. Thiru,
    7. E. C. Y. Wang,
    8. E. Bolton,
    9. W. Min,
    10. P. Moore,
    11. et al
    . 2008. TL1A both promotes and protects from renal inflammation and injury. J. Am. Soc. Nephrol. 19: 953–960.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Craxton A.,
    2. G. Shu,
    3. J. D. Graves,
    4. J. Saklatvala,
    5. E. G. Krebs,
    6. E. A. Clark
    . 1998. p38 MAPK is required for CD40-induced gene expression and proliferation in B lymphocytes. J. Immunol. 161: 3225–3236.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Screaton G. R.,
    2. X. N. Xu,
    3. A. L. Olsen,
    4. A. E. Cowper,
    5. R. Tan,
    6. A. J. McMichael,
    7. J. I. Bell
    . 1997. LARD: a new lymphoid-specific death domain containing receptor regulated by alternative pre-mRNA splicing. Proc. Natl. Acad. Sci. USA 94: 4615–4619.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. Meylan F.,
    2. T. S. Davidson,
    3. E. Kahle,
    4. M. Kinder,
    5. K. Acharya,
    6. D. Jankovic,
    7. V. Bundoc,
    8. M. Hodges,
    9. E. M. Shevach,
    10. A. Keane-Myers,
    11. et al
    . 2008. The TNF-family receptor DR3 is essential for diverse T cell‑mediated inflammatory diseases. Immunity 29: 79–89.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Papadakis K. A.,
    2. J. L. Prehn,
    3. C. Landers,
    4. Q. Han,
    5. X. Luo,
    6. S. C. Cha,
    7. P. Wei,
    8. S. R. Targan
    . 2004. TL1A synergizes with IL-12 and IL-18 to enhance IFN-γ production in human T cells and NK cells. J. Immunol. 172: 7002–7007.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Bamias G.,
    2. C. Martin III.,
    3. M. Marini,
    4. S. Hoang,
    5. M. Mishina,
    6. W. G. Ross,
    7. M. A. Sachedina,
    8. C. M. Friel,
    9. J. Mize,
    10. S. J. Bickston,
    11. et al
    . 2003. Expression, localization, and functional activity of TL1A, a novel Th1-polarizing cytokine in inflammatory bowel disease. J. Immunol. 171: 4868–4874.
    OpenUrlAbstract/FREE Full Text
    1. Bamias G.,
    2. S. I. Siakavellas,
    3. K. S. Stamatelopoulos,
    4. E. Chryssochoou,
    5. C. Papamichael,
    6. P. P. Sfikakis
    . 2008. Circulating levels of TNF-like cytokine 1A (TL1A) and its decoy receptor 3 (DcR3) in rheumatoid arthritis. Clin. Immunol. 129: 249–255.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Bamias G.,
    2. G. Kaltsa,
    3. S. I. Siakavellas,
    4. K. Papaxoinis,
    5. E. Zampeli,
    6. S. Michopoulos,
    7. I. Zouboulis-Vafiadis,
    8. S. D. Ladas
    . 2010. High intestinal and systemic levels of decoy receptor 3 (DcR3) and its ligand TL1A in active ulcerative colitis. Clin. Immunol. 137: 242–249.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Taraban V. Y.,
    2. T. J. Slebioda,
    3. J. E. Willoughby,
    4. S. L. Buchan,
    5. S. James,
    6. B. Sheth,
    7. N. R. Smyth,
    8. G. J. Thomas,
    9. E. C. Y. Wang,
    10. A. Al-Shamkhani
    . 2011. Sustained TL1A expression modulates effector and regulatory T-cell responses and drives intestinal goblet cell hyperplasia. Mucosal Immunol. 4: 186–196.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Shih D. Q.,
    2. R. Barrett,
    3. X. Zhang,
    4. N. Yeager,
    5. H. W. Koon,
    6. P. Phaosawasdi,
    7. Y. Song,
    8. B. Ko,
    9. M. H. Wong,
    10. K. S. Michelsen,
    11. et al
    . 2011. Constitutive TL1A (TNFSF15) expression on lymphoid or myeloid cells leads to mild intestinal inflammation and fibrosis. PLoS One 6: e16090.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Pappu B. P.,
    2. A. Borodovsky,
    3. T. S. Zheng,
    4. X. Yang,
    5. P. Wu,
    6. X. Dong,
    7. S. Weng,
    8. B. Browning,
    9. M. L. Scott,
    10. L. Ma,
    11. et al
    . 2008. TL1A-DR3 interaction regulates Th17 cell function and Th17-mediated autoimmune disease. J. Exp. Med. 205: 1049–1062.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Bull M. J.,
    2. A. S. Williams,
    3. Z. Mecklenburgh,
    4. C. J. Calder,
    5. J. P. Twohig,
    6. C. Elford,
    7. B. A. J. Evans,
    8. T. F. Rowley,
    9. T. J. Slebioda,
    10. V. Y. Taraban,
    11. et al
    . 2008. The death receptor 3-TNF-like protein 1A pathway drives adverse bone pathology in inflammatory arthritis. J. Exp. Med. 205: 2457–2464.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Twohig J. P.,
    2. M. Marsden,
    3. S. M. Cuff,
    4. J. R. Ferdinand,
    5. A. M. Gallimore,
    6. W. V. Perks,
    7. A. Al-Shamkhani,
    8. I. R. Humphreys,
    9. E. C. Y. Wang
    . 2012. The death receptor 3/TL1A pathway is essential for efficient development of antiviral CD4+ and CD8+ T-cell immunity. FASEB J. 26: 3575–3586.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Buchan S. L.,
    2. V. Y. Taraban,
    3. T. J. Slebioda,
    4. S. James,
    5. A. F. Cunningham,
    6. A. Al-Shamkhani
    . 2012. Death receptor 3 is essential for generating optimal protective CD4+ T-cell immunity against Salmonella. Eur. J. Immunol. 42: 580–588.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Gay S.,
    2. R. E. Gay,
    3. W. J. Koopman
    . 1993. Molecular and cellular mechanisms of joint destruction in rheumatoid arthritis: two cellular mechanisms explain joint destruction? Ann. Rheum. Dis. 52(Suppl. 1): S39–S47.
    OpenUrlFREE Full Text
  27. ↵
    1. van der Heijden I. M.,
    2. B. Wilbrink,
    3. I. Tchetverikov,
    4. I. A. Schrijver,
    5. L. M. Schouls,
    6. M. P. Hazenberg,
    7. F. C. Breedveld,
    8. P. P. Tak
    . 2000. Presence of bacterial DNA and bacterial peptidoglycans in joints of patients with rheumatoid arthritis and other arthritides. Arthritis Rheum. 43: 593–598.
    OpenUrlCrossRefPubMed
    1. Mauri C.,
    2. M. R. Ehrenstein
    . 2007. Cells of the synovium in rheumatoid arthritis. B cells. Arthritis Res. Ther. 9: 205.
    OpenUrlCrossRefPubMed
    1. Cooke T. D.,
    2. E. R. Hurd,
    3. H. E. Jasin,
    4. J. Bienenstock,
    5. M. Ziff
    . 1975. Identification of immunoglobulins and complement in rheumatoid articular collagenous tissues. Arthritis Rheum. 18: 541–551.
    OpenUrlCrossRefPubMed
    1. Vetto A. A.,
    2. M. Mannik,
    3. E. Zatarain-Rios,
    4. M. H. Wener
    . 1990. Immune deposits in articular cartilage of patients with rheumatoid arthritis have a granular pattern not seen in osteoarthritis. Rheumatol. Int. 10: 13–19.
    OpenUrlCrossRefPubMed
    1. Mollnes T. E.,
    2. T. Lea,
    3. O. J. Mellbye,
    4. J. Pahle,
    5. O. Grand,
    6. M. Harboe
    . 1986. Complement activation in rheumatoid arthritis evaluated by C3dg and the terminal complement complex. Arthritis Rheum. 29: 715–721.
    OpenUrlCrossRefPubMed
    1. Scrivo R.,
    2. M. Di Franco,
    3. A. Spadaro,
    4. G. Valesini
    . 2007. The immunology of rheumatoid arthritis. Ann. N. Y. Acad. Sci. 1108: 312–322.
    OpenUrlCrossRefPubMed
    1. Manzo A.,
    2. C. Pitzalis
    . 2007. Lymphoid tissue reactions in rheumatoid arthritis. Autoimmun. Rev. 7: 30–34.
    OpenUrlCrossRefPubMed
    1. Lundy S. K.,
    2. S. Sarkar,
    3. L. A. Tesmer,
    4. D. A. Fox
    . 2007. Cells of the synovium in rheumatoid arthritis. T lymphocytes. Arthritis Res. Ther. 9: 202.
    OpenUrlCrossRefPubMed
    1. Stamp L. K.,
    2. M. J. James,
    3. L. G. Cleland
    . 2004. Interleukin-17: the missing link between T-cell accumulation and effector cell actions in rheumatoid arthritis? Immunol. Cell Biol. 82: 1–9.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Imamura F.,
    2. H. Aono,
    3. T. Hasunuma,
    4. T. Sumida,
    5. H. Tateishi,
    6. S. Maruo,
    7. K. Nishioka
    . 1998. Monoclonal expansion of synoviocytes in rheumatoid arthritis. Arthritis Rheum. 41: 1979–1986.
    OpenUrlCrossRefPubMed
  37. ↵
    1. Liu P.,
    2. N. A. Jenkins,
    3. N. G. Copeland
    . 2003. A highly efficient recombineering-based method for generating conditional knockout mutations. Genome Res. 13: 476–484.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Nagy A.,
    2. J. Rossant,
    3. R. Nagy,
    4. W. Abramow-Newerly,
    5. J. C. Roder
    . 1993. Derivation of completely cell culture-derived mice from early-passage embryonic stem cells. Proc. Natl. Acad. Sci. USA 90: 8424–8428.
    OpenUrlAbstract/FREE Full Text
  39. ↵
    1. Luo H.,
    2. T. Charpentier,
    3. X. Wang,
    4. S. Qi,
    5. B. Han,
    6. T. Wu,
    7. R. Terra,
    8. A. Lamarre,
    9. J. Wu
    . 2011. Efnb1 and Efnb2 proteins regulate thymocyte development, peripheral T cell differentiation, and antiviral immune responses and are essential for interleukin-6 (IL-6) signaling. J. Biol. Chem. 286: 41135–41152.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    1. Brand D. D.,
    2. K. A. Latham,
    3. E. F. Rosloniec
    . 2007. Collagen-induced arthritis. Nat. Protoc. 2: 1269–1275.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Kojima F.,
    2. M. Kapoor,
    3. L. Yang,
    4. E. L. Fleishaker,
    5. M. R. Ward,
    6. S. U. Monrad,
    7. P. C. Kottangada,
    8. C. Q. Pace,
    9. J. A. Clark,
    10. J. G. Woodward,
    11. L. J. Crofford
    . 2008. Defective generation of a humoral immune response is associated with a reduced incidence and severity of collagen-induced arthritis in microsomal prostaglandin E synthase-1 null mice. J. Immunol. 180: 8361‑8368.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    1. Smolen J. S.,
    2. D. Aletaha,
    3. K. Redlich
    . 2012. The pathogenesis of rheumatoid arthritis: new insights from old clinical data? Nat Rev Rheumatol 8: 235–243.
    OpenUrlCrossRefPubMed
  43. ↵
    1. McInnes I. B.,
    2. G. Schett
    . 2011. The pathogenesis of rheumatoid arthritis. N. Engl. J. Med. 365: 2205–2219.
    OpenUrlCrossRefPubMed
  44. ↵
    1. Brand D. D.,
    2. A. H. Kang,
    3. E. F. Rosloniec
    . 2003. Immunopathogenesis of collagen arthritis. Springer Semin. Immunopathol. 25: 3–18.
    OpenUrlCrossRefPubMed
  45. ↵
    1. Sheibanie A. F.,
    2. T. Khayrullina,
    3. F. F. Safadi,
    4. D. Ganea
    . 2007. Prostaglandin E2 exacerbates collagen-induced arthritis in mice through the inflammatory interleukin-23/interleukin-17 axis. Arthritis Rheum. 56: 2608–2619.
    OpenUrlCrossRefPubMed
  46. ↵
    1. Bouaziz J.-D.,
    2. K. Yanaba,
    3. G. M. Venturi,
    4. Y. Wang,
    5. R. M. Tisch,
    6. J. C. Poe,
    7. T. F. Tedder
    . 2007. Therapeutic B cell depletion impairs adaptive and autoreactive CD4+ T cell activation in mice. Proc. Natl. Acad. Sci. USA 104: 20878–20883.
    OpenUrlAbstract/FREE Full Text
  47. ↵
    1. Pelletier N.,
    2. M. Casamayor-Pallejà,
    3. K. De Luca,
    4. P. Mondière,
    5. F. Saltel,
    6. P. Jurdic,
    7. C. Bella,
    8. L. Genestier,
    9. T. Defrance
    . 2006. The endoplasmic reticulum is a key component of the plasma cell death pathway. J. Immunol. 176: 1340–1347.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    1. Barrett R.,
    2. X. Zhang,
    3. H. W. Koon,
    4. M. Vu,
    5. J.-Y. Chang,
    6. N. Yeager,
    7. M. A. Nguyen,
    8. K. S. Michelsen,
    9. D. Berel,
    10. C. Pothoulakis,
    11. et al
    . 2012. Constitutive TL1A expression under colitogenic conditions modulates the severity and location of gut mucosal inflammation and induces fibrostenosis. Am. J. Pathol. 180: 636–649.
    OpenUrlCrossRefPubMed
  49. ↵
    1. McInnes I. B.,
    2. G. Schett
    . 2007. Cytokines in the pathogenesis of rheumatoid arthritis. Nat. Rev. Immunol. 7: 429–442.
    OpenUrlCrossRefPubMed
  50. ↵
    1. Elgueta R.,
    2. M. J. Benson,
    3. V. C. de Vries,
    4. A. Wasiuk,
    5. Y. Guo,
    6. R. J. Noelle
    . 2009. Molecular mechanism and function of CD40/CD40L engagement in the immune system. Immunol. Rev. 229: 152–172.
    OpenUrlCrossRefPubMed
  51. ↵
    1. King C.,
    2. S. G. Tangye,
    3. C. R. Mackay
    . 2008. T follicular helper (TFH) cells in normal and dysregulated immune responses. Annu. Rev. Immunol. 26: 741–766.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section