The utility of pathway selective estrogen receptor ligands that inhibit nuclear factor-κB transcriptional activity in models of rheumatoid arthritis

Rheumatoid arthritis (RA) is a chronic, debilitating condition affecting 0.5 to 1% of the world's population. The major goals of treatment of RA are to reduce pain and discomfort, to prevent deformities and loss of joint function, and to maintain a productive and active lifestyle. RA is characterized by chronic joint inflammation mediated by inflammatory cell infiltration into synovial tissues as well as joint destruction through the overexpression of matrix metalloproteinase (MMP) in articular synoviocytes and chondrocytes. The pathologic lesions of RA are driven, in part, by the production of inflammatory mediators in synoviocytes and macrophages, probably involving the transcription factor NF-κB. Because NF-κB is localized in the nuclei of synovial cells in patients with RA [1, 2] and the inducers and targets of NF-κB almost perfectly match the list of pivotal mediators increased in RA [3], an important role for activated NF-κB in human RA is likely.

NF-κB is a dimeric transcription factor composed of homodimeric and heterodimeric complexes of the Rel family of proteins, p65 (Rel A), p50/105, c-Rel, p52/100, and Rel B. Binding of cytoplasmic inhibitory protein-κB (IκB) to NF-κB masks the NF-κB nuclear localization signal and sequesters NF-κB in a non-activated form in the cytoplasm. Cell activation by a variety of extracellular signals such as oxidative stress, cytokines, and lipopolysaccharide induces a cascade of events that leads to the degradation of IκB; activated NF-κB then translocates to the nucleus, where it binds to DNA elements in the promoters of several proinflammatory gene families [4].

Activation of NF-κB has been observed in synovial cells from patients with RA [5] and results in the induction of proinflammatory genes such as tumor necrosis factor-α (TNF-α), IL-1β, IL-6, MMP-1, and MMP-3 in ex vivo synovial membrane cultures [6]. Moreover, NF-κB activation might also be a pivotal factor protecting cells from apoptosis, thus contributing to synovial hyperplasia (reviewed in [7]). Inactivation of NF-κB in transgenic mice expressing a 'super-repressor' IκBα or in rel- /- and nfkb1- /- knockout mice rendered the animals refractory to development of collagen-induced arthritis [8, 9]. In another study performed in the rat adjuvant-induced arthritis model, intra-articular injection of an adenoviral construct encoding a dominant-negative from of IκB kinase-2 significantly ameliorated the severity of the adjuvant arthritis and was correlated with a decrease in NF-κB DNA binding in the nucleus of synovial cells [10]. Because NF-κB is involved in normal immune and homeostatic processes, its prolonged inhibition might be harmful. Therefore, more indirect methods of targeting NF-κB might provide a safer pharmacological profile.

In tissues that express estrogen receptor (ER), 17β-estradiol inhibits NF-κB-driven transcription through multiple mechanisms that might include direct protein–protein interactions [11, 12], inhibition of NF-κB binding to DNA [13, 14], induction of IκB expression [15], or coactivator sharing [16, 17]. Two nuclear estrogen receptors have been identified (ERα and ERβ). Both receptors are widely distributed throughout numerous organs [18] and are present in T cells, monocytes, dendritic cells, synovial macrophages, articular chondrocytes, and proliferating fibroblasts present in the RA joint [19–22]. These two receptors have a nearly identical DNA-binding domain, both activate transcription through binding to identical ER response elements [23, 24], and both can antagonize NF-κB transcriptional activity [25, 26]. Taken together, these findings identify RA as a disease amenable to treatment with ER-selective NF-κB inhibitors.

The selective inflammatory modulator WAY-169916 is a non-steroidal ER-dependent inhibitor of NF-κB transcriptional activity. Although it inhibits the expression of a range of inflammatory proteins, including cytokines, chemokines, and cell adhesion molecules that are expressed after activation of NF-κB, WAY-169916 lacks estrogenic activity such as the stimulation of uterine proliferation [27]. Here we demonstrate that WAY-169916 is efficacious in two models of arthritis, the HLA-B27 transgenic rat and a Lewis rat model of adjuvant-induced arthritis. The activity of WAY-169916 is related to its ability to suppress inflammatory processes globally, as demonstrated by the decrease in serum acute-phase protein levels of haptoglobin, α1-acid glycoprotein (α1-AGP), and C-reactive protein (CRP) as well as the inhibition of adjuvant-induced gene expression in the spleen, liver, and popliteal lymph nodes in the rat adjuvant arthritis model. Moreover, WAY-169916 was also active in suppressing cytokine and adhesion molecule expression in fibroblast-like synoviocytes (FLS) isolated from patients with RA. Taken together, these data suggest the potential utility of the pathway-selective ER ligand WAY-169916 and other compounds in its class in the treatment of RA.

RA might occur as a result of an autoimmune response, and recent studies suggest that hypersensitivity to microbial antigens contributes to the development of the arthritis. Microbial or self-antigen presentation to T lymphocytes results in chronic activation of the immune system. Multiple proinflammatory mediators, including IL-1, TNF-α, interferon-γ, and MMPs mediate the inflammation of the joints. Biochemical and histological changes in synovial tissue, cartilage, and bone have been documented in various animal models of arthritis [34, 35]. In many respects the synovial and cartilage lesions that develop in these models closely resemble those seen in rheumatoid arthritis. We have investigated the role of WAY-169916 in two such models.

The HLA-B27 transgenic rats spontaneously develop arthritis similar to the human spondyloarthropathies associated with the HLA-B27 and β₂-microglobulin genes through a T cell-mediated process [34]. In this model, WAY-169916 restored the clinical joint scores to baseline after 10 days. Histological scoring of synovitis and cartilage damage in the tarsal joints after 29 days of treatment was also significantly improved with WAY-169916 treatment.

In the rat adjuvant-induced arthritis model [36], 3 to 6 days after the injection of adjuvant, induction of an αβ T cell response occurs and leads to clinical lesion development in the tarsal joints within 5 to 8 days. Because activated NF-κB was detected in the synovial lining layer and around blood vessels in the inflamed synovium as early as day 3 after adjuvant injection in the Lewis rats and is thought to be correlated with disease development [37], this model was used to test the therapeutic treatment with WAY-169916. We demonstrated that WAY-169916 was effective in improving both joint and histology scores at doses as low as 0.3 mg/kg given orally once daily. Improvement in the synovitis and Mankin scores did occur with higher doses of WAY-169916 even though the joint score reduction was already maximal at 0.3 mg/kg. When the rats were dosed at 5 mg/kg the total synovitis score decreased to 4.14 (data not shown). The beneficial effects of WAY-169916 on joint histology might therefore continue with increasing dose or longer exposure.

A benefit on arthritis progression with non-selective estrogens such a 17β-estradiol has also been demonstrated in both the rat adjuvant-induced arthritis model [38] and the collagen-induced mouse model [39, 40]. Indeed, 17β-estradiol has been shown to affect several processes involved in the pathogenesis of RA, including immunoregulation, regulation of adhesion molecules, and modulation of cytokine signaling. However, the role of estrogen has not been well defined in patients with RA. There is evidence that gender might affect the occurrence and progression of RA. Women have a higher risk of developing RA than men. During pregnancy, the disease activity is ameliorated in 75% of women, whereas after delivery, flares occur in up to 90% of patients [41]. The highest incidence of developing RA coincides with menopause, indicating that a decrease in estrogen production might increase the risk of joint inflammation. In a recent randomized clinical trial, post-menopausal women taking hormone therapy had suppressed signs of inflammation and significantly improved disease severity scores (DS28) after 12 months of treatment, which was consistent with previous trials [42].

With the identification of selective NF-κB transcriptional inhibitors such as WAY-169916, the expectation is to accentuate the anti-inflammatory, anti-rheumatic activity observed with the non-selective estrogens. Whereas non-selective estrogens have been documented to contain an anti-inflammatory activity through the suppression of NF-κB transcriptional activity [11, 12], hormone therapy can simultaneously elicit both proinflammatory and anti-inflammatory activities as exemplified by the decrease in haptoglobin and α1-AGP levels in women taking hormone therapy [43] while also inducing MMP9 and CRP levels [44, 45]. In the rat adjuvant model, WAY-169916 inhibited the adjuvant induction of CRP levels and those of haptoglobin and α1-AGP. Moreover, WAY-169916 had no effect on the basal levels of CRP whereas treatment with estradiol has been shown to increase rat CRP serum levels [46], suggesting a potential differential effect of WAY-169916 in comparison with estradiol. This differential activity has been demonstrated on several classic estrogenic effects. WAY-169916 neither stimulates creatine kinase gene expression driven via an estrogen receptor response element in vitro nor promotes uterine proliferation in vivo [27] while retaining the anti-inflammatory activity as demonstrated here.

The anti-inflammatory activity of WAY-169916 was further demonstrated in a series of gene-profiling experiments. In the spleens from the adjuvant-treated rats, 36 genes were identified that were induced greater than twofold by the adjuvant treatment. WAY-169916, when dosed at 5 mg/kg, repressed 29 of those genes by at least 50%, and 17 of them by more than 75%. An attractive hypothesis for WAY-169916-mediated activity in the spleen involves the downregulation of LBP and CD14 expression on monocytes and macrophages, resulting in a diminished immune response and ultimately resulting in the observed decreases in MMP9, IL1R2, chemokine-like factor 1, S100A8, and S100A9 through the repression of NF-κB activity [47, 48]. In spleen cell cultures, WAY-169916 treatment did not interfere with ConA-stimulated NF-κB DNA binding; however, the downregulation of S100A9 mRNA was confirmed (data not shown), which was consistent with our hypothesis that ER regulates NF-κB at the transcriptional level [16, 27]. A similar suppression of adjuvant-induced inflammatory gene expression was also observed in liver and lymph node studies. These data demonstrate that WAY-169916 can have an effect on a global level, both in terms of the tissues targeted and the different inflammatory signaling pathways, to suppress adjuvant-induced gene expression.

Infiltration of inflammatory cells into the synovial tissue and lining layer results in the formation of pannus, a highly vascularized tissue comprising FLS, macrophages, and lymphocytes. FLS are known for their role in joint destruction through the production of cytokines and MMP, which contribute to cartilage degradation (reviewed in [49]). Expression of ER has been detected in synovial tissues from patients with RA [50] and localized to synoviocytes in the synovial lining [22], providing another potential cell type by which WAY-169916 functions. Synoviocytes isolated from a male RA patient were confirmed to express ERα mRNA [33], but no ERβ mRNA was detected (data not shown). The ERα was functional in these cells, because WAY-169916 could effectively block the TNF-α-mediated inflammatory gene expression of IL-6, TNF-α, and ICAM-1. The potential involvement of NF-κB in mediating TNF-α gene induction was demonstrated with the use of a general NF-κB inhibitor, as shown previously [5]. Given the importance of TNF-α signaling in RA disease progression, the ability of WAY-169916 to interfere with this signaling pathway in human synoviocytes suggests a potential clinical benefit for WAY-169916 in patients with RA.

We detailed the activity of the first pathway-selective ER ligand, WAY-169916, in two models of RA. This compound selectively inhibits NF-κB activity via the ER and imparts significant efficacy in the HLA-B27 and Lewis rat adjuvant-induced models of arthritis. More importantly, no evidence for classic estrogenic activity has been observed with this compound [27]. These data provide evidence that the non-steroidal, pathway-selective ER ligand, WAY-169916, and other compounds in its class might be therapeutically useful in the treatment of RA.