Background and Clinical Significance

Autoimmune-inflammatory diseases afflict more than 10 million Americans, often with devastating financial and life-altering consequences. Women are particularly vulnerable since (i) about 75% of autoimmune-inflammatory diseases patients are women and (ii) autoimmune-inflammatory diseases are among the top ten leading causes of death in women. The spectrum of autoimmune-inflammatory diseases is broad, and includes more than 80 diseases. Taking just one example to illustrate their socioeconomic importance, if we were to group diabetes, atherosclerosis, and vascular complications of Types 1 and 2 diabetes as a single disease spectrum, as current research in TLRs suggest, this group of autoimmune-inflammatory diseases alone would exceed a national prevalence of 6.0%. This does not even account for other autoimmune-inflammatory diseases such as those of the thyroid, inflammatory bowel diseases, systemic lupus erythematosus, rheumatoid arthritis, toxic shock to name a few.

Like autoimmune-inflammatory diseases, cancer is a major health concern affecting millions of people worldwide. It was estimated that 1,368,030 new cases of cancer were diagnosed in 2008 in the United States alone. Malignant melanoma and pancreatic cancer are particularly important health concerns among cancers, because few therapies exist. Malignant melanoma has a marked propensity for metastasis and a prognosis for survival of <20% with distant metastasis. Malignant melanoma exceeds many other types of cancers in lost "years of life", because it is most prevalent in younger individuals. Pancreatic cancer also has an alarmingly high mortality rate, as well as an increasing incidence over the past several decades. It is the fourth leading cause of cancer deaths in the United States and has an overall survival rate of less than 4%; most die within 6 months to 1 year from time of diagnosis. The poor prognosis is attributable to a highly invasive nature, metastases before discovery, and a generally poor response to chemotherapeutic and/or surgical intervention. Uncovering a new or additional potentially effective treatment for malignant melanoma and pancreatic cancer is, therefore, of importance, particularly if the therapy has a novel molecular basis and is applicable to both autoimmune-inflammatory disease and cancer.

The underlying genetic susceptibilities of both autoimmune-inflammatory diseases and cancer are poorly defined for the most part. Nevertheless, in both cases it is recognized that there is a genetic as well as an environmental component to both diseases and that the environmental agents inducing or worsening disease expression are similar (Fig. 1). The organ specificity of autoimmune-inflammatory diseases and cancers implicates non-immune cells in both cases. A bystander immune response that results in autoantibodies to non-immune cell antigens as well as the production of cytokines that are important in disease progression is common to both (Fig. 1). Overwhelming evidence suggests that chronic inflammation is crucial to the onset/progression of a multiplicity of human cancers and autoimmune-inflammatory diseases and that abnormal activation of Toll-like receptors and/or Toll-like receptor signaling in non-immune cells is involved in the initiation and progression of disease expression. Our preliminary data supported this model and suggested inhibition of Toll-like receptor signaling in non-immune cells might be a new paradigm of therapy.

FIGURE 1

In both autoimmune-inflammatory diseases and cancer, the evidence reported in the literature indicates genes that are involved in disease susceptibility are multiple, are often not disease specific, and, are not readily amenable to therapy at this time. Immunosuppressive agents that largely target the immune cell response to date are only partially effective (humanized monoclonal antibodies), too toxic for broad use (cyclosporine), or simply palliative anti-inflammatory agents (corticosteroids). In this context, since we now know that many autoimmune diseases and cancers are induced or worsened by an environmental agent, e. g. smoking or viral infections, and that pathologic expression of TLR/TLR signaling in non-immune cells is associated with expression of autoimmune-inflammatory diseases and cancer, a therapeutic opportunity and demand exists for developing drugs to block this environmental induction process without harming normal immune responsiveness. Thus, the ability to develop a broad platform of drugs to attack these diseases has enormous socioeconomic impact.

Phenlymethimazole (Compound 10; C10)

Phenylmethimazole (C10) is a tautomeric cyclic thion (TCT) derivative of methimazole (MMI). MMI is a drug used to treat autoimmune Graves' disease and is well known to inhibit thyroid hormone formation. However, evidence accumulated that it also had immunosuppressive activity. Phenylmethimazole was developed in our laboratory as a compound that was more effective than MMI in suppressing abnormal MHC class I and aberrant MHC class II expression in non-immune cells (thyrocytes) stimulated with IFN-b, but was less able or unable to inhibit thyroid hormone formation. Search for a better derivative of MMI followed the demonstration that (i) class I knockouts did not develop experimental systemic lupus erythematosus or type 1 diabetes, (ii) MMI suppressed abnormal MHC gene expression and mimicked the effect of the class I knockouts in preventing experimental SLE or type 1 diabetes, (iii) abnormal MHC gene expression was important to the development of autoimmune Graves' disease, as exemplified in the hallmark development of the Shimojo model of that disease wherein dsDNA or dsRNA transfection of non-immune cells could induce abnormal MHC gene expression, cause thyrocytes with native TSHR or fibroblasts transfected with the TSHR to take on the characteristics of antigen presenting cells

Importantly, C10 did not develop in vivo toxicity (i) in these models, (i) in a partial pancreatectomy rat model of type 1 diabetes (preliminary data), (iii) a monkey model of islet cell transplant rejection (unpublished data). When given daily for as long as 3 months, there has been no evidence of thyroid function changes, changes in TSH levels, liver or kidney toxicity. Thus, C10 could modulate and suppress mediators of cellular and humeral autoimmunity by inhibiting abnormal TLR expression/signaling in non-immune cells without apparent toxicity in vivo, including inhibition of thyroid hormone formation, which is a therapeutic feature of its parental analog methimazole. Additionally, neither C10 nor MMI inhibited the ability of mice to develop immune responses to tetanus. In short, C10 and its TCT analogs appeared to agents that met the new therapeutic paradigm requested by Davendra and Eisenbarth as well as by others.

Toll-like receptors (TLR)
Toll-like receptors (TLR) are a family of known cell surface receptors that are related to IL-1 receptors and have been well characterized on immune cells (Fig. 2). They are policemen that protect mammals from pathogenic organisms, such as viruses, by generating an "innate immune" response to products or signature molecules of the pathogenic organism. This response results in increases in genes for multiple inflammatory cytokines and chemokines, major histocompatibility (MHC) genes I and II as well as costimulatory molecules and is critical for the development of antigen-specific adaptive immunity. Perhaps two of the most well studied TLRs are TLR3 and TLR4. Both TLR3 and TLR4 have been found in human leukocytes, with TLR3 selectively expressed in dendritic cells and TLR4 expressed in monocytes and neutrophils. Recent work has shown that TLR4 is also expressed on non-immune cells, such as endothelial and epithelial cells. Additionally, there is increasing evidence that TLR3 is expressed on somatic cells.

TLR signal transduction
TLR3 recognizes double stranded RNA (dsRNA) released into the extracellular space by viral killing of cells and may also recognize signal stranded mRNA. Arguably, the most important ligand for TLR4 is lipopolysaccharide (LPS). The dsRNA binding to TLR3 or LPS interaction with TLR4 activates at least two distinct pathways. One pathway signals through the adapter molecule, MyD88 in the case of TLR4 to activate NF-kB, MAPK, and various cytokines and chemokines. The second pathway couples through an adapter molecule, termed TIR domain-containing molecule adapter inducing Type I IFN/TIR-containing adapter molecule (TRIF/TICAM)-1, to activate IFN regulatory factor (IRF)-3 that leads to the synthesis and release of type I IFNs (INF-a or INF-b) and other chemokines. The type I IFNs can initiate an autocrine/paracrine loop, further upregulating TLR3 on the secreting cell and on neighboring cells. TLR4 signaling has a MyD88 independent and dependent mechanism, the former involves TICAM-2 as well as TICAM-1.

C10 and TLR Signaling
C10 decreased abnormal expression of the IRF-3, IFN-β, STAT-1, and IRF-1 pathways, in particular, in multiple in vitro and in vivo models, as well as downstream markers of the NF-kB signal pathway. This broad action could be construed as non-specificity. However, since pathologic TLR signaling involving both the IRF-3/Type 1 IFN/STAT-1/IRF-3 pathway and the NF-kB pathway is critical for expression of both autoimmune-inflammatory diseases and cancer, broad applicability can used as a "proof of principle". Thus, an inhibitor of these pathologic pathways should be mechanistically effective in a multiplicity of diseases involving pathologic TLR expression in non-immune cells, including Graves', Type 1 diabetes, type 2 diabetes, atherosclerosis, and multiple cancer models associated with abnormal expression of TLR3. C10 appears to act initially to inhibit IRF3 activation (Fig. 2) and the type I IFN (IFN-a or INF-b) pathway and the secondary effects of cytokine activation through that path on NF-kB signaling.

Figure 2

Proof of Principal
TLR3 and TLR4 have been implicated in autoimmunity/pathological inflammation. One disease implicating TLR4 on nonimmune cells is endotoxic shock. C3H/HeJ mice with a point mutation in the TLR4 gene and a defect in TLR4 signaling become hyporesponsiveness to LPS challenge. The Kubes group found strong evidence that endothelial TLR4, as opposed to leukocyte TLR4, is a critical player in endotoxic shock. Thus, mice deficient in endothelial TLR4, but not leukocyte TLR4, had significantly attenuated leukocyte sequestration in the lungs subsequent to challenge with LPS, whereas mice deficient in leukocyte TLR4 still exhibited dramatic sequestration.

A second disease implicating TLR4 is inflammatory bowel disease (IBD). Cario et al. reported that TLR4 was upregulated in intestinal epithelial cell lines isolated from patients with IBD. Using the dextran sodium sulfate (DSS)-induced murine model of colitis, Ortega-Cava et al. found that TLR4 is upregulated in the colon of colitic mice. Also, in the myeloid cell-specific Stat3-deficient mouse model of enterocolitis used to study the mechanisms of IBD, enterocolitis significantly improved in TLR4/Stat3-deficient mice, whereas TNF-a/Stat3 deficient mice still had severe enterocolitis, indicating the importance of TLR4 in enterocolitis. A third disease implicating TLR4 is atherosclerosis. Michelsen et al. found that mice deficient in TLR4 had a significant reduction in aortic plaque development in atherosclerosis-prone apolipoprotein E-deficeint (ApoE-/-) mice, suggesting an important role for TLR4. They also demonstrated that lack of TLR4 signaling results in reduced monocyte adhesion to TLR4-/-endothelium, suggesting that endothelial TLR4 may be a key player in atherogenesis and associated thrombocytic events.

Another autoimmune-inflammatory disease where TLR3 has been implicated is Type 1 diabetes. Thus, TLR3 is overexpressed in pancreatic β cells in association with destructive changes in Type 1 diabetes and dsRNA could induce insulinitis and type 1 diabetes in animals, consistent with the animal models wherein coxsacki virus induces Type 1 diabetes in NOD mice. Devendra and Eisenbarth point out that a wide variety of studies implicate enteroviruses as a potential agent in the pathogenesis of type 1 diabetes and the mechanism of viral infection leading to B cell destruction involves the cytokine interferon alpha (IFN-a) [a Type I IFN]. They hypothesize that activation of TLR by dsRNA and induction of IFN-a may activate or accelerate immune-mediated beta cell destruction. They conclude that, "therapeutic agents targeting IFN-a may potentially be beneficial in preventing type 1 diabetes and autoimmunity." Preventing or correcting pathologic expression TLR3, and its pathologic signaling in islet cells, becomes a "desired" new therapy.

Non-immune cell TLR3 may also be involved in thyroid autoimmune-inflammatory diseases. We have found that functional TLR3 is present on FRTL-5 cells, a rat thyroid cell line, and that TLR3 is overexpressed on thyrocytes in patients with Hashimoto's thyroiditis, one of two major thyroid autoimmune-inflammatory diseases with different pathologies. One, Graves' disease, is a hyperthyroid condition caused by over stimulation of the thyroid due to autoantibodies to the thyroid stimulating hormone receptor. Hashimoto's is a hypothyroid condition characterized by a dense lymphocytic infiltrate and destruction of the thyroid cells. Thyrocytes from 100% of samples from Hashimoto’s patients stained similarly positive for TLR3 whereas none of the thyrocytes from the Graves’ patients or normal controls stained positive. These observations are strikingly similar to those presented by Wen et al. who postulated a role for nonimmune cell TLR3 in the destruction of B cells in Type 1 diabetes.

C10 has been shown to inhibit pathologic TLR expression and signaling. C10 can reverse TLR3 over-expression in dsRNA transfected thyrocytes, inhibit cytokine (TNFα)-induced IRF-1, VCAM-1, and leukocyte adhesion in vascular endothelial cells in vitro. It is also effective in vivo in animal models of toxic shock and in chemically-induced (DSS) colitis where it inhibits TLR-4 expression and signaling, particularly the IRF-3/Type I IFN/ STAT1/IRF-1 pathways as well as downstream markers of the NF-κB signal pathway, where TLR-3 and TLR-4 are over-expressed.

The recognition that MMI or C10 could prevent these phenomena was followed by the demonstration they acted (C10 50-100-fold better than MMI) to inhibit TLR expression and signaling. Thus, C10 could reverse overexpression of TLR3 and TLR3 signaling in dsRNA transfected thyrocytes, inhibit LPS activation of TLR4 signaling in RAW (mouse macrophages) cells, and inhibit cytokine (TNF-a)-induced IRF-1, VCAM-1, and leukocyte adhesion in vascular endothelial cells in vitro. C10 was also effective in vivo in animal models of toxic shock and in chemically-induced (DSS) colitis where it could induce 100% relief of symptoms and 100% survival. Both of those diseases are recognized today as diseases of abnormal TLR4 signaling in non-immune cells. Efficacy in tumor models originated from studies in papillary thyroid cancer cells, now in fact known to be melanoma cells.

Figure 3