CTLA-4 is a negative regulator of the immune response expressed by regulatory T (Treg) cells and activated T cells. Polymorphisms in the CTLA4 gene have been associated with autoimmune diseases, including systemic lupus erythematosus. Disease-associated polymorphisms have been shown to affect the production of the different CTLA-4 variants through an effect on alternative splicing. This study was undertaken to evaluate the role of the 1/4 CTLA-4 isoform in lupus-prone mice.
We generated an MRL/lpr mouse strain that transgenically overexpresses a short isoform of CTLA-4 (1/4 CTLA-4) by backcrossing C57BL/6.1/4CTLA-4–transgenic mice to the MRL/lpr strain for 9 generations. A new antibody was generated to detect the expression of the 1/4 CTLA-4 isoform. Routine methods were used to evaluate kidney damage, humoral immunity, and cellular immunity.
Expression of the 1/4 CTLA-4 isoform accelerated autoimmune disease. Transgenic mice died earlier, had more severe renal disease, and had higher titers of anti–double-stranded DNA antibodies than wild-type MRL/lpr mice. The acceleration of autoimmunity and disease pathology associated with the presence of the short (1/4) isoform of CTLA-4 was linked to increased numbers of activated T cells and B cells and heightened interferon-γ production, but not to altered expression of the full-length CTLA-4 molecule or Treg cell numbers.
Our results indicate that the presence of the alternatively spliced 1/4 CTLA-4 isoform can further promote autoimmunity and autoimmune pathology in lupus-prone mice and suggest that altered splicing of CTLA4 contributes to the expression of autoimmune disease.
CTLA-4, a costimulatory receptor in the immunoglobulin superfamily, is closely related to CD28 and inducible costimulator and binds to CD80 and CD86 (1, 2). Expression of CTLA-4 is constitutive on regulatory T (Treg) cells (3) and is induced following activation on effector T (Teff) cells (1). CTLA-4 exerts an essential inhibitory role, and its absence causes an early and fatal autoimmune disease in mice (4).
The CTLA4 gene is highly conserved (76% homology between humans and mice) (5). It is composed of 4 exons. Exon 1 codes for the signal peptide, exon 2 codes for the ligand-binding domain, exon 3 codes for the transmembrane region, and exon 4 codes for the intracytoplasmic tail (6) (Figure 1A). Human peripheral blood lymphocytes express 3 isoforms of CTLA-4 produced by alternative splicing: full-length CTLA-4 (all 4 exons), soluble CTLA-4 (exons 1, 2, and 4), and a short variant that lacks both the ligand-binding domain and the transmembrane domain (1/4 CTLA-4) (6). Allelic variations and single-nucleotide polymorphisms in the CTLA4 gene have been associated with several human autoimmune diseases, including autoimmune thyroid disease (6), rheumatoid arthritis (7), and systemic lupus erythematosus (SLE) (8, 9). Interestingly, the polymorphisms associated with autoimmune disease affect CTLA4 splicing and thus the relative expression of each variant isoform (6). How the differential expression of CTLA-4 isoforms impacts susceptibility to autoimmune disease is not yet clear.
The 1/4 CTLA-4 isoform lacks the CD80/CD86-binding domain and the transmembrane portion, and thus its function remains unclear. Forced expression of the 1/4 CTLA-4 isoform in T cells was shown to induce spontaneous autoimmune disease and facilitate the development of experimental autoimmune encephalomyelitis in C57BL/6 mice through an unknown mechanism (10). In this study, we demonstrate that increased expression of 1/4 CTLA-4 in lupus-prone mice accelerates autoimmunity, exacerbates disease pathology, and causes early death.
MATERIALS AND METHODS
Female MRL/MpJ-Tnfrsf6lpr (MRL/lpr) mice were purchased from The Jackson Laboratory. The generation of C57BL/6.1/4CTLA-4–transgenic mice has been described previously (10). MRL/lpr.1/4CTLA-4–transgenic mice were generated by backcrossing C57BL/6.1/4CTLA-4–transgenic mice to MRL/lpr mice (for 9 generations). Mice were killed at the end of the 12th or 15th week of life. All mice were maintained in a specific pathogen–free animal facility, and all experiments were approved by the Institutional Animal Care and Use Committee of Beth Israel Deaconess Medical Center.
Proteinuria and pyuria were measured in a semiquantitative manner. Briefly, mice in each group (n = 4) were placed together overnight in a Nalgene metabolic cage for the collection of urine. This procedure was repeated in 3 independent experiments, so that the data presented are the average from a total of 12 mice per group.
Equal aliquots of mouse serum samples diluted 1:100 were fractionated on NuPAGE 4–12% Bis-Tris Gel (Invitrogen) and transferred to a 0.2-μm PVDF membrane (Millipore). The membrane was blocked for 1 hour with 3% skim milk in Tris buffered saline–Tween (TBST), probed with anti–1/4 CTLA-4 antibody (custom antibody from Yenzym Antibodies), washed with TBST, and incubated with a 1:5,000 dilution of goat anti-rabbit IgG coupled with horseradish peroxidase (Jackson ImmunoResearch). The Amersham ECL system was used for detection.
Measurement of cytokine levels in cell supernatants.
Two million splenocytes were incubated in 1 ml of RPMI 1640 supplemented with 10% fetal calf serum and stimulated with anti-CD3 (0.25 μg/ml) and anti-CD28 (0.5 μg/ml) for 24 hours. At the end of the culture period, supernatants were collected. The interferon-γ (IFNγ) concentration was determined by enzyme-linked immunosorbent assay kits according to the recommendations of the manufacturer (R&D Systems).
Flow cytometric analysis.
The thymus, spleen, axillary and inguinal lymph nodes, mesenteric lymph nodes, and Peyer's patches were excised from MRL/lpr mice, and single-cell suspensions were obtained by teasing the organs through a nylon mesh. Cells were stained for flow cytometry with antibodies against CD25, CD80, CD86, and CTLA-4 (all from eBioscience), CD3e, CD4, CD44, CD62L, CD19, F4/80, and CD11c (all from BioLegend), and FoxP3 (BD Biosciences) for 30 minutes at 4°C. For intracellular staining of FoxP3, a FoxP3-staining buffer set (fixation/permeabilization and permeabilization buffers) was used according to the manufacturer's protocol (eBioscience). Total cell numbers were determined by counting live cells. Absolute cell numbers were calculated on the basis of the percentage of each population and presented as the median. Samples were acquired in an LSRII flow cytometer (BD Biosciences). Analysis was performed with FlowJo version 7.6.1 (Tree Star).
Real-time polymerase chain reaction (PCR).
Total messenger RNA (mRNA) was isolated from mouse spleen cells using an RNeasy Mini kit (Qiagen). Complementary DNA was produced using random primers from an equal amount of RNA. TaqMan quantitative PCR was performed on an Applied Biosystems StepOnePlus system, and gene expression was assessed by the comparative Ct method. Primer sequences are available from the corresponding author upon request.
Values are expressed as the mean ± SD or the mean ± SEM. Kruskal-Wallis test with post hoc comparisons using Scheffe's test was used for intergroup comparisons of multiple variables. Survival was analyzed by the Kaplan-Meier method and log rank test. Statistical analyses were performed using StatView software (Abacus Concepts). P values less than 0.05 were considered significant.
The 1/4 CTLA-4 variant isoform codes for a secreted protein.
To determine whether increased expression of 1/4 CTLA-4 affects the onset and/or severity of systemic autoimmunity and related pathology, we generated MRL/lpr mice that transgenically express the 1/4 CTLA-4 variant isoform in T cells by backcrossing C57BL/6.1/4CTLA-4–transgenic mice to MRL/lpr mice. To ensure that the transgene produced a protein product, we generated a rabbit polyclonal antibody directed against the protein domain encoded by exon 4. Consistent with the fact that it lacks the transmembrane domain (exon 3), we detected a protein corresponding to the size of 1/4 CTLA-4 in the sera of transgenic mice (Figure 1B). Serum levels of this protein were significantly higher in MRL/lpr.1/4CTLA-4–transgenic mice than in nontransgenic MRL/lpr and C57BL/6 mice (Figure 1C). Consistent with these results, we detected 1/4 CTLA-4 transcripts in spleen cells from transgenic mice by using a TaqMan assay that amplifies only this species (and not other Ctla4 splice variants) (10) (Figure 1D).
Exacerbation of autoimmunity in mice expressing the 1/4 CTLA-4 variant.
The presence of the 1/4 CTLA-4–coding transgene was associated with accelerated disease onset and early death in MRL/lpr mice (Figure 2A). Lymphoproliferation secondary to Fas deficiency is a prominent feature of the disease in MRL/lpr mice. To determine if early death was associated with increased lymphoproliferation and/or accelerated autoimmunity, we analyzed the secondary lymphoid organs of 12- and 15-week-old mice. Spleen weight and cellularity were modestly increased in transgenic mice as compared to their nontransgenic littermates. However, no differences in lymph node weight or size were detected (data not shown).
In contrast, MRL/lpr.1/4CTLA-4–transgenic mice expressed a more severe glomerulonephritis than did MRL/lpr mice. This was manifested by increased levels of proteinuria (P = 0.01) and pyuria (P = 0.03), as well as significantly more damage to the glomeruli, as indicated by the presence of more hyaline thrombi, which is a sign of extensive immune complex deposition in the capillary loops. Moreover, inflammatory cell infiltration into tubulointerstitial zones was more severe in MRL/lpr.1/4CTLA-4 mice than in their MRL/lpr littermates (Figure 2B). Humoral autoimmunity was likewise more intense in the 1/4CTLA-4–transgenic mice, as manifested by significantly higher levels of serum IgG and anti–double-stranded DNA antibodies (Figure 2C).
Increased T and B cell activation in 1/4 CTLA-4–expressing mice.
The increased antibody and autoantibody production, as well as increased signs of renal inflammation, observed in the MRL/lpr.1/4CTLA-4 mice suggested that the presence of 1/4 CTLA-4 further facilitated the autoimmune response and kidney damage in these mice. We analyzed the activation state of T and B lymphocytes in 12-week-old MRL/lpr and MRL/lpr.1/4CTLA-4 mice. Consistent with the pathology findings, transgenic mice had significantly higher numbers of activated (CD44+) T cells (P < 0.01) along with a reciprocal decrease in naive (CD62L+CD44−) T cells (P = 0.01). After in vitro stimulation, cells from MRL/lpr.1/4CTLA-4 mice secreted significantly higher levels of the proinflammatory cytokine IFNγ (Figure 2D). The activation state of B cells was also increased, as demonstrated by augmented expression of the costimulatory molecules CD80 and CD86 (Figure 2D). Expression of CD86 was also increased in spleen macrophages from MRL/lpr.1/4CTLA-4 mice (Figure 2D).
Lack of an effect of 1/4 CTLA-4 overexpression on Treg cells.
CTLA-4 is constitutively expressed by Treg cells (3) and has been shown to be necessary for their suppressive function (11, 12). Moreover, 1/4 CTLA-4 is highly expressed by Treg cells since its transcription closely resembles that of full-length CTLA-4 in terms of kinetics and cellular distribution (10). For this reason, we considered that Treg cell numbers and/or homeostasis could be affected by the presence of the 1/4 CTLA-4–coding transgene. Using flow cytometry, we quantified the abundance of CD4+FoxP3+ cells in the thymus, spleen, axillary, inguinal, and mesenteric lymph nodes, and Peyer's patches from transgenic and wild-type MRL/lpr mice. There were no differences in the number or distribution of Treg cells between MRL/lpr mice that expressed the 1/4 CTLA-4 transgene and those that did not (Figure 3A).
Lack of an effect of 1/4 CTLA-4 overexpression on the expression of full-length CTLA-4.
Certain autoimmune disease–associated polymorphisms in CTLA4 are located in noncoding regions and have been shown to affect splicing of CTLA4 (6). To determine whether forced expression of 1/4 CTLA-4 altered the transcription of the full-length native molecule, we quantified CTLA-4 levels at the mRNA and protein levels. As shown in Figure 3B, levels of mRNA for full-length Ctla4 were not modified in cells freshly isolated from mouse lymph nodes. Likewise, the expression of full-length CTLA-4 on the surface membrane of lymph node cells, determined as the percentage of CTLA4+CD3+ cells (Figure 3C) or as mean fluorescence intensity (Figure 3D), was comparable between MRL/lpr and MRL/lpr.1/4CTLA-4 mice.
To assess if the transgene affected CTLA-4 expression in Treg cells, we performed flow cytometry to compare CTLA-4 levels in CD4+CD25+ (Treg) cells and CD4+CD25− (Teff) cells from transgenic and nontransgenic mice. As expected, CTLA-4 expression was significantly higher in Treg cells than in Teff cells (Figure 3D). However, no difference in CTLA-4 Treg cell expression was detected in the presence of 1/4 CTLA-4. To rule out differential regulation upon T cell stimulation, T cells isolated from the peripheral lymph nodes of MRL/lpr and MRL/lpr.1/4CTLA-4 mice were activated with anti-CD3 and anti-CD28 antibodies. No differences were observed in CTLA-4 mRNA expression after 72 hours of cell stimulation (Figure 3B, right panel).
In this study, we investigated whether forced expression of an alternatively spliced CTLA-4 variant isoform (1/4 CTLA-4) in T cells alters systemic autoimmune disease in a murine model of SLE. Our results showed that increased levels of 1/4 CTLA-4 exacerbated autoimmunity and caused premature death of MRL/lpr mice. These data have clinical relevance, because single-nucleotide polymorphisms of CTLA4 that have been associated with human autoimmune disease have been shown to regulate alternative splicing of this gene (6).
CTLA-4 is an essential negative regulator of the immune system. Its expression (13) and function (14, 15) have been suggested to be abnormal in patients with SLE. However, the relationship between these defects and the polymorphisms that have been associated with SLE in genetic studies is unknown. Moreover, the function(s) of the splice variants of CTLA-4 in health and autoimmune disease are unknown. Our results show that 1/4 CTLA-4 may exist in the serum as a secreted protein. Its lack of the CD80/CD86-binding domain suggests that it does not block the binding of CTLA-4 but rather exerts its effects through an independent mechanism. Our data indicate that the presence of 1/4 CTLA-4 does not affect the expression of full-length CTLA-4. Moreover, the numbers and distribution of FoxP3+ Treg cells were not affected by the transgene. This confirms previous findings showing that the numbers and in vitro function of Treg cells were not altered in C57BL/6 mice expressing 1/4 CTLA-4 (10).
Our work indicates that the presence of the 1/4 CTLA-4 variant facilitates the activation of T and B cells, and this probably explains the acceleration in autoimmune disease observed in the transgenic mice. Further work will investigate more thoroughly the mechanism(s) by which CTLA-4 promotes lymphoid activation. This study highlights the importance of the regulation of the expression of distinct splice variants of complex genes.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Drs. Crispín and Tsokos had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Ichinose, Zhang, Juang, Sharpe, Kuchroo, Crispín, Tsokos.
Acquisition of data. Ichinose, Zhang, Koga, Juang, Kis-Tóth, Kuchroo, Crispín.
Analysis and interpretation of data. Ichinose, Zhang, Koga, Juang, Kuchroo, Crispín, Tsokos.