The fact that interleukin-2 (IL-2) production is decreased in patients with systemic lupus erythematosus (SLE) and in mouse models of lupus has been known for more than 30 years. Significant efforts have been made to understand the molecular events that lead to its decreased production, its impact on the dysregulated immune system, and, more importantly, the clinical consequences that may appear to be directly or indirectly related to low levels of IL-2 (1). Much has been learned about the link between autoimmunity and decreased production of IL-2 from studies of mice lacking IL-2 or the IL-2 receptor, an experimental model in which mice develop a severe, spontaneous autoimmune disease characterized by lymphoproliferation, splenomegaly, and anemia. The lymphoproliferation observed in these mice as well as the elevated serum autoantibody titers are thought to result from decreased numbers of peripheral Treg cells, similar to what has been reported in patients with SLE (2).

In studies of patients with SLE, it has been shown extensively that the levels of IL-2 are controlled at the transcriptional level, and that various transcription factors are involved. In particular, the imbalance between CREB and its modulator, CREM, is highly relevant. CREB, an enhancer, and CREM, a repressor, compete for a binding site in the IL2 promoter. T cells from patients with SLE express increased amounts of calcium/calmodulin-dependent protein kinase IV (CaMK4) in the nucleus, which phosphorylates CREM to bind to the IL2 promoter, whereas simultaneously, T cells express increased amounts of serine/threonine-specific protein phosphatase 2A (PP-2A), which dephosphorylates pCREB (the phosphorylated form of CREB, which acts as the actual transcriptional enhancer) and tilts the balance of pCREB/CREM toward CREM (3), resulting in suppression of IL2 promoter activity (Figure 1).

thumbnail image

Figure 1. Decreased production of interleukin-2 (IL-2) results in compromised immune functions and clinical manifestations in patients with systemic lupus erythematosus (SLE). Altered transcription factor binding to the promoter of the IL2 gene (indicated by the straight horizontal line) results in decreased IL-2 transcription. The enhancer pCREB is decreased (because of increased rates of dephosphorylation by protein phosphatase 2A) and enables the repressor CREM (which is phosphorylated at increased rates by calcium/calmodulin-dependent protein kinase IV [CaMK4]) to bind (to the same site) and limit the transcriptional activity of the promoter. CREM recruits enzymes that modify histones (His) in ways that close the IL2 locus (indicated by wavy lines). NF-AT binds to the promoter, but because the adjacent activator protein 1 (AP-1) site is empty (AP-1 is decreased in SLE T cells), it fails to increase transcription. In addition, microRNA-31 (miR-31), the levels of which are decreased in SLE T cells, fails to limit the amounts and activity of the inhibitor of NF-AT, RhoA, and thus provides another mechanism that limits NF-AT activity. Decreased IL-2 production leads to decreased Treg cell function, decreased cytotoxic cell activity, decreased activation-induced cell death (AICD), and increased production of IL-17. Altered immune function leads to increased rates of infection and tissue inflammation in patients with SLE.

Download figure to PowerPoint

These molecular events have independent value in whole animals, for the following reasons. 1) Inhibition of CaMK4 with a small drug in lupus-prone mice results in disease prevention and treatment, and genetic deletion of CaMK4 in MRL-lpr mice suppresses autoimmunity and lupus nephritis. 2) Overexpression of PP-2A in T cells of normal B6 mice leads to increased IL-17 production and makes the mice susceptible to an anti–glomerular basement membrane antibody–induced glomerulonephritis (4). 3) Overexpression of CREM in mice results in a predicted decrease in IL-2 production and an unexpected increase in IL-17 production (5).

Another important transcription factor in the regulation of IL-2 production is NF-AT, which is increased in the nucleus of SLE T cells but fails to transactivate the IL2 promoter (as it does the CD40 ligand promoter) because it binds to a cis site next to activator protein 1 (AP-1). The AP-1 site needs to be occupied for the NF-AT enhancer to exert its activity, and AP-1 levels are decreased in SLE T cells (1, 3). The role of NF-AT may be even more complicated, as is discussed below.

Epigenetic factors have already been implicated in the control of the production of IL-2 in SLE patients. First, CREM, after it binds to the IL2 promoter, recruits histone deacetylase 1 (HDAC-1), which deacetylates histones, leading to local closure of DNA and deterrence of binding of transcription factors to the promoter. HDAC inhibitors have been reported to mitigate lupus nephritis in lupus-prone mice. More recent data have demonstrated the involvement of CREM in silencing IL2 in T lymphocytes from SLE patients through a gene-wide, HDAC-1– directed deacetylation of histone H3K18 and DNA methyltransferase 3a–directed CpG-DNA hypermethylation. Therefore, gene-wide histone deacetylation and CpG-DNA methylation precede the control of IL-2 transcriptional regulation through the direct binding of cis factors (6).

MicroRNAs (miRNAs) are noncoding regulatory RNAs that are 21–23 nucleotides in length and function as posttranscriptional, or perhaps posttranslational, regulators of gene expression. Most mammalian miRNAs are transcribed by RNA polymerase II, the same polymerase that directs the transcription of coding genes. MicroRNAs are encoded in intergenic regions, in a sense or antisense orientation within introns of specific genes, or by noncoding transcripts. In the nucleus, miRNAs are derived from larger precursors, known as primary miRNAs, that are processed into 75 nucleotide–long precursor miRNA (pre-miRNA) hairpins by the RNase III enzyme Drosha. The hairpins are exported into the cytoplasm by exportin 5.

In the cytoplasm, the pre-miRNA hairpins are further processed by Dicer, an RNase III–like enzyme, to generate small RNA duplexes. One strand, and sometimes both of the strands, will be incorporated into the RNA-induced silencing complex (RISC) or micro- RNP (miRNP) complex containing argonaute 2, among other proteins. The RISC/miRNP complex executes miRNA functions, and miRNAs perform their functions by forming a duplex with the target gene(s) in the 3′-untranslated region of its messenger RNA (mRNA). Depending on the degree of complementarity between the miRNA:mRNA duplex, this interaction usually leads to the down-regulation of protein expression by translational repression, mRNA cleavage, or promotion of mRNA decay (6, 7).

Data from in vivo gain-of-function and loss-of-function studies in mouse models demonstrate, without a doubt, that miRNAs, along with coding genes, control diverse biologic processes, including cell differentiation, cell cycle programming, apoptosis, and immune regulation in mammals. The regulation of miRNA expression is tightly controlled, and often the same rules and regulations that govern coding gene expression also apply to miRNAs. Similar to coding genes, altering the levels and the temporal expression of a specific miRNA clearly affects the proper development and function of the tissue in which it is expressed. Therefore, it is reasonable to argue that the dysregulated control of miRNA expression would give rise to disease. Indeed, numerous studies have underscored the participation of miRNAs in disorders of immune regulation (7), including SLE (8).

In this issue of Arthritis & Rheumatism, Fan and colleagues (9) expand on their previous observations, showing that the expression of several miRNAs is dysregulated in SLE T cells. They found that miR-31 is down-regulated in SLE T cells, while the expression level of RhoA is up-regulated. The change in RhoA and miR-31 expression correlated with impaired IL-2 production. The authors suggest that miR-31 targets RhoA, a negative regulator of NF-AT, leading to the induction of IL-2 production. Based on these observations, the authors propose that the dysregulation of miR-31 expression in SLE T cells results in higher RhoA levels, leading to impaired IL-2 production. This proposal is tantalizing and may offer additional insights into the dysregulation of IL-2 expression in SLE T cells.

However, several issues remain unresolved. First, in contrast to the findings of Fan et al, RhoA has been shown to regulate the transcriptional activity, but not the expression, of NF-AT (10). Second, the expression levels of both miR-31 and RhoA are constitutively high in normal human T cells; therefore, it is not easy to conceive how miR-31 regulates RhoA. Third, NF-AT activation is an early event, taking place within minutes to hours of T cell activation, and yet the effects of miR-31 on IL-2 production only occur after 24–48 hours following stimulation of the cells with phorbol myristate acetate and ionomycin. These considerations are confounded by reports that the nuclear levels of NF-AT in stimulated SLE T cells are increased. In an attempt to reconcile the available data, we can consider it likely that the early increase in the levels of NF-AT is not sufficient to exert an enhancing effect (because of the absence of the adjacent AP-1) on the transcription of IL2, and IL-2 production thus remains low. At a later time point, the effect of low expression of miR-31 (through increased levels of the NF-AT inhibitor RhoA) contributes to decreased NF-AT transcriptional activity and prolongation of the suppression of IL-2 production.

Further work is needed to understand how miR- 31 is down-regulated in SLE T cells, whether RhoA is a direct target of miR-31, and if miR-31 has a direct effect on IL-2 expression. In addition, additional studies in animal models are necessary to definitively demonstrate the role of miR-31 in the regulation of IL-2 production in vivo. Decreased IL-2 levels in patients with SLE may account for a number of aberrant immune cell functions, including decreased cytotoxic responses, diminished activation-induced cell function, reduced Treg cell numbers and function, and decreased IL-17 production (1, 2, 11, 12). An array of cell cytotoxic responses, most of which depend on the presence of IL-2, are decreased in SLE patients. Natural killer cell activity, cell cytotoxic responses to alloantigens, virally modified cells, and non–antigen-specific cytolytic activity are all decreased in SLE.

IL-2 plays a necessary role in the activation-induced cell death (AICD) of T cells, the process of programmed cell death that T cells undergo after repeated antigenic stimulation. Therefore, IL-2 is needed not only for the ignition of the immune response but also for the elimination of activated T cells. In autoimmune diseases, autoantigen-stimulated T cells will live longer in the absence of IL-2 and will contribute to the propagation of the autoimmune response. AICD is defective in SLE patients, and the deficiency in IL-2 production may contribute to this process. IL-2 is also necessary for the generation of Treg cells. It has been established recently that IL-2 is indispensable for the generation of Treg cells, which are important in the control of the autoimmune response. The numbers of Treg cells are reportedly decreased in patients with SLE (2, 11, 12).

Finally, more recent information suggests that IL-2 is needed to control the expression of IL-17, which clearly contributes to the expression of lupus nephritis in patients with SLE and lupus-prone mice. In the absence of IL-2 signals, the numbers of Treg cells decline, whereas the numbers of IL-17–producing cells increase. It appears that IL-2 signals are crucial for the reciprocal balance between IL-17–producing cells and Treg cells. More specifically, while IL-2–induced STAT-5 promotes the Treg cell–determining transcription factor FoxP3, it suppresses the IL-17–determining transcription factor retinoic acid receptor–related orphan nuclear receptor γt induced by IL-6 (11). This balanced regulation between IL-2– and IL-17–dependent processes may be further complicated because, as mentioned above, in CREM-overexpressing mice, levels of IL-2 are low while levels of IL-17 are high (5).

What are the direct clinical repercussions of decreased IL-2 production? As was indicated above, IL-2 supports CD8+ T cell and natural killer cell activity, thus providing a proper defense against infectious agents. Infection is a leading cause, or perhaps the leading cause, of morbidity and mortality in SLE and may also change the natural course of the disease, since components of infectious agents may propagate the autoimmune response. Obviously, many other factors may contribute to the increased rates of infections in SLE, including the immunomodulatory effects of corticosteroids and other immunosuppressive drugs. Common, as well as opportunistic, pathogens are frequently encountered among patients with SLE, and an infection imposes a serious burden on lupus patients and on the caring physician (1).

Can IL-2 production be restored as an approach to control SLE? This question gained prime time recently in 2 clinical reports in which the authors claimed that infusion of small amounts of IL-2 into patients with vasculitis and patients with graft-versus-host disease resulted in a significant clinical benefit, which was attributed to a common mechanism involving the improvement of Treg cell function (12). In an earlier study, continuous administration of IL-2 by means of a vaccinia virus vector in lupus-prone MRL-lpr mice resulted in suppression of autoimmunity and lupus nephritis (13). These findings have prompted the consideration of additional modifiers that could enhance IL-2 production. For example, inhibition of CaMK4 ameliorates SLE activity in mice, as do HDAC-1 inhibitors (1). As it is unknown what controls the expression of miR-31 and what accounts for its decreased expression in SLE patients, it is still premature to consider the proper development of targeted intervention strategies.


  1. Top of page

Both authors were involved in drafting the article or revising it critically for important intellectual content, and both authors approved the final version to be published.


  1. Top of page