TLRs are among the most widely expressed recognition receptors of the innate immune system and recognize a very diverse set of molecules derived from pathogenic microorganisms known as pathogen-associated molecule patterns. 10 different TLRs have been identified in humans that recognize conserved microbial components, initiate specific biological responses, and are thus essential components of the innate response to infection. Interestingly, 4 of these 10 TLRs have been implicated in the binding of nucleic acids. TLR3 recognizes dsRNA from viruses and can also be stimulated by Poly I:C; TLR7/8 recognize ssRNA; and TLR9 recognizes bacterial and viral DNA and synthetic oligonucleotides containing unmethylated CG dinucleotides (1
). Activation of TLRs without appropriate control, however, can lead to substantial inflammation resulting in tissue damage and autoimmunity (9
). Recently, Wang et al. reported that the production of proinflammatory cytokines in response to West Nile virus, a dsRNA virus, was reduced in TLR3-deficient mice. This reduction did not affect viral load in the mice but rather protected them from the TLR3-dependent inflammation that promoted destruction of the blood–brain barrier (8
). Examples of TLR4 involvement in the induction of colitis or microbial sepsis are well defined (7
). Other examples link TLR stimulation and atherosclerosis (10
). Various mechanisms of control have been established to avoid unwanted TLR activation. Their expression is restricted to certain subsets of cells, and their level of expression is usually down-regulated after activation through a negative feedback loop. Furthermore, TLR and regulatory T cells cross-regulate each other in order to balance TLR-induced inflammation and suppression by regulatory T cells. As shown by Pasare et al., APCs, when stimulated through their TLRs, induce signals that will block the suppressive action of regulatory T cells (40
). Conversely, regulatory T cells can secrete cytokines such as IL-10 that are known to inhibit TLR stimulation (41
In addition, there is increasing evidence that TLRs can also be activated by endogenous ligands (9
), and it is essential that these receptors can discriminate self from nonself to prevent undesirable inflammation. For TLRs recognizing nucleic acids, the distinction between nucleic acids from mammalian versus microbial origin might be very difficult. The major difference between microbial and mammalian DNA is the level of methylation, whereas there are no clear sequential or structural differences for RNA. All four receptors that recognize RNA and DNA products are localized in intracellular compartments that could be one of the mechanisms to prevent undesirable TLR activation by endogenous RNA and DNA. Collectively, this suggests that TLR stimulation can be an important factor in the development of autoimmunity, and the ability to control TLR activation may have great therapeutic potential.
We and others have reported oligonucleotide sequences that can inhibit TLR9 activation mediated by CpG-containing oligonucleotides (30
). These sequences are active on both mouse and human cells in vitro and in mice in vivo. We recently extended this to show that IRS could prevent IFN-α production from PDCs after TLR9 activation (36
). Furthermore, coinjection of IRS with ISS in mice treated with D-galactosamine prevented death, demonstrating that IRS are potent enough in vivo to prevent massive systemic inflammation caused by TLR9 activation (36
). Some IRS have been shown to have efficacy in vivo in severe autoimmune models (34
). However, the use of complete Freund's adjuvant in some of these models, which includes mycobacterial DNA, renders these experiments difficult to interpret, because the observed effect of IRS may be by inhibiting the adjuvant and not the resulting pathology. Further experiments using complete Freund's adjuvant–free models will be necessary to address the potential of IRS to prevent autoimmunity. We have now developed three series of IRS, including inhibitors of TLR9, inhibitors of TLR7 but not TLR9, and, most interestingly, sequences that inhibit signals through these two receptors. The mechanism of action of IRS is not known; however, evidence to date and the pattern of specificity of these inhibitors suggest that the effect is targeted to the receptor either through competitive antagonism or another mechanism. We have yet to find any activity of IRS when used alone, and broader assays such as microarray analysis may be necessary to conclude whether IRS have direct activities.
SLE affects >1,000,000 people in the United States alone, primarily young and middle-aged women. The etiology of the disease is unknown, but a strong genetic component appears to be involved. SLE is a relapsing, remitting disease with devastating consequences and is poorly treated or prevented with existing therapies. Patients suffer from kidney dysfunction leading to renal failure and a wide and variable range of symptoms, including arthritis, fever, skin rashes, and brain inflammation. ICs of autoantibodies to chromatin and RNA protein particles (snRNP) are diagnostic for SLE and are thought to play an important role in the pathogenesis of the disease. Increased serum levels of IFN-α have been observed in many SLE patients and correlate with both disease activity and key disease markers, such as anti-DNA antibodies (15
). Furthermore, a set of characteristic IFN-α–inducible genes are constitutively up-regulated in blood cells of SLE patients (20
). These elevated IFN-α levels may have a direct role in the pathology of lupus because patients with nonautoimmune disorders who are treated with IFN-α can develop antinuclear antibodies, anti-dsDNA antibodies, and occasionally SLE as well. Viral infections, UV skin injury, or other events leading to IFN-α induction are known to be activators of flares of SLE. In addition, NZB mice, which spontaneously develop a lupus-like disease, have less severe disease with delayed onset when made deficient for the IFN-α receptor (47
There is considerable evidence that chronically activated PDCs and the IFN-α that they produce in response to TLR stimulation are involved in the pathogenesis of SLE. Patients with SLE have a 50–100-fold decrease in the number of PDCs circulating in the blood (20
). This decrease appears caused by in vivo activation followed by cell migration into peripheral lymphoid tissues and sites of inflammation. Indeed, activated PDCs have been observed to accumulate in cutaneous lupus erythematosus lesions (22
). These cells, when activated with viruses, can produce large amounts of IFN-α. In addition, it was recently shown that ICs present in serum samples from SLE patients can cause the production of IFN-α by PBMCs in vitro. A growing body of evidence supports the idea that TLR activation plays a central role in the maintenance and progression of the disease by promoting elevated IFN-α levels. TLR7 and TLR9 are particularly relevant to SLE, as they are expressed by human PDCs, and stimulation through these receptors leads to very high levels of IFN-α production by PDCs. Exogenous viruses acting through these TLRs would be expected to induce IFN-α and thus exacerbate the disease, and this is consistent with the observed association of lupus flares with viral infections. We show that ICs associated with self-DNA and RNA can directly activate PDCs to make IFN-α. The recognition by TLRs is likely facilitated by the expression of FcγRII on PDCs, allowing efficient uptake of the self–nucleic acid into the TLR-containing endosomal compartments (29
). These ICs are thus endogenous IFN-α inducers, and the resulting IFN-α production could perpetuate the autoimmune process. In addition, the role of ICs in the pathogenesis of lupus is complex, and a similar mode of activation could occur for B cells, as previously shown in mice (14
We show that immunoregulatory sequences such as IRS 954 can inhibit signaling through TLR7 and TLR9 and decrease IFN-α produced by PDCs either in response to DNA and RNA viruses or to both types of circulating ICs isolated from lupus patients. IRS 954 also inhibits the activation of B cells through TLR7/8 and TLR9, a process that, in concert with a signal through the B cell receptor, can lead to production of autoantibodies specific for both the nucleic acid and immunoglobulin components of ICs (14
). Our work demonstrates that mammalian RNA and DNA, when complexed with autoantibodies, can represent potent self-antigens for TLR7 or TLR9, respectively, and that this inappropriate self-recognition by the innate immune system may play a substantial role in autoimmune diseases such as lupus. Treatment with an IRS has the potential to modulate the major source of excessive IFN-α in SLE, without completely preventing the acute IFN-α responses to viral infection mediated by other recognition mechanisms, such as TLR3 and protein kinase R. This approach could thus be less immunosuppressive than therapies aimed at blocking IFN-α interaction with its receptor; however, this will have to be evaluated in further studies. Using inhibitors of TLR represents a new approach to treatment of SLE with the potential to reduce symptoms and prevent relapses through inhibition of a key step in disease pathogenesis.