**4.2 S1P expression and SphK activity in RA synovium and synovial fluids**

Several studies reveal high levels of S1P in RA synovial tissues and fluids (Kitano et al., 2006; Lai et al., 2008). Kitano et al. (2006) reported elevated levels of S1P in synovial fluids of RA patients (1078 pM/ml) in comparison to those of osteoarthritis (OA) patients (765 pM/ml). The S1P content in RA synovial fluids was significantly higher than those in serum (400 pM/ml) or plasma (100 pM/ml) from normal people. Lai et al. (2008) also measured the amount of S1P by a competitive ELISA and detected up to 17 M S1P in RA synovial fluids, which was more than five fold higher than that found in OA synovial fluids (3 M). The increased level of S1P could be responsible for the recruitment and the infiltration of immune cells into the synovium (see section 4.2.2). Pi et al. (2006) also showed that peripheral blood B lymphoblastoid cell lines derived from patients with RA exhibit a high level of S1P synthesis.

SphK expression and activity were also demonstrated in RA. Increased SphK1 expression and activity were found in RA B lymphoblastoid cell lines, which were implicated as the underlying mechanism of impaired Fas-mediated death signalling in RA (Pi et al., 2006). More recently, SphK2 was shown to be expressed in rheumatoid synovial fibroblasts *in vivo* and *in vitro* and associated with the up-regulation of S1P production (Kamada et al., 2009). Surprisingly, FTY-720 (2-amino-2-[2-(4-octylphenyl) ethyl] propane-1, 3-diol hydrochloride), a sphingosine analog that is phosphorylated to the active metabolite FTY-720-phosphate (FTY-720-P) by SphK2 (Brinkmann et al., 2002; Mandala et al., 2002), induces apoptosis of synovial fibroblasts. This effect could be due, at least in part, to FTY-720-P-mediated degradation of specific S1P receptors (Matloubian et al., 2004). Corroborating this observation, we reported that S1P protects synovial fibroblasts from ongoing apoptosis (Zhao et al., 2008). Using the mouse model of collagen-induced arthritis (CIA), Lai et al. (2009) highlighted distinct roles for SphK1 and SphK2 in regulating cell growth and survival. The *in vivo* administration of SphK1 siRNA reduced inflammation whereas mice treated with SphK2 siRNA resulted in a more aggressive disease and greater secretion of proinflammatory cytokines (IL-6, TNF- and IFN-) by immune cells. How and why SphK1 and SphK2 exert distinct opposing roles in the regulation of inflammatory arthritis remains unclear and may be related to the location of S1P production (Maceyka et al., 2005). The role of SphK2 in chronic inflammatory diseases remains ambiguous since the selective inhibitor of SphK2 ABC249640 has been reported to reduce bone and cartilage degradation in the mouse models of adjuvant-induced arthritis (AIA) and of CIA (Fitzpatrick et al., 2011). The SphK1 signalling pathway is activated in RA (Limaye et al., 2009). In the mouse model of CIA, administration of a non-specific pharmacological inhibitor of SphKs, *N*,*N*dimethylsphingosine (DMS), and a siRNA approach to knockdown the SphK1 isoform, markedly suppressed joint pathologies such as adjacent cartilage and bone erosion, synovial hyperplasia, and infiltration of inflammatory cells into the joint compartment (Lai et al., 2008). Moreover, SphK1 deficiency in hTNF- transgenic mice that develop arthritis at an early age was related to less synovial inflammation and bone erosion (Baker et al., 2010). To

Targeting the Metabolism and Receptors of

mediators, which is relevant to RA pathology.

**lymphoblastoid cells and chondrocytes** 

Sphingosine-1-Phosphate for the Treatment of Rheumatoid Arthritis 201

inflammatory mediators, including cytokines, chemokines, and prostaglandin E2 (PGE2) (Kitano et al., 2006; Zhao et al., 2008). S1P administration strongly stimulated the secretion of IL-8 (interleukin-8), IL-6, MCP-1 (monocyte chemotactic protein-1), and RANTES (regulated on activation normal T cells expressed and secreted) via S1P2 and S1P3 receptors. The signalling pathways involved in S1P-mediated cytokine/chemokine secretion were dependent on ERK1/2, p38 MAPK, and Rho kinase activation. These cytokines/chemokines may subsequently increase the recruitment of inflammatory cells such as neutrophils into the synovium, as chemokines such as IL-8 exhibit selective chemotactic activity for neutrophils, whereas MCP-1, MIP-1 (macrophage inflammatory protein-1), MIP-1 (macrophage inflammatory protein-1), and RANTES primarily attract monocytes (Koch, 2005). S1P may therefore contribute to the regulation and maintenance of the inflammatory response in RA, in part through stimulation of multiple cytokine/chemokine secretion by synovial fibroblasts. SphK/S1P signalling was also implicated in cell-contact-mediated proinflammatory cytokine/chemokine secretion in RA synovium without additional exogenous stimulation. Lai et al. (2008) used peripheral blood mononuclear cells from RA patients in cell-cell contact experiments. They discovered that activated peripheral T lymphocytes from RA patients induced substantial production of TNF-, IL-1, IL-6, MCP-1 and MMP-9 by autologous peripheral monocytes. Importantly, treatment with a potent SphK inhibitor, DMS, significantly suppressed production of these cytokines. The results suggest the importance of SphK/S1P signalling in cell-contact-mediated inflammatory

Regarding PGE2, S1P was reported to stimulate COX-2 expression and super-production of PGE2 by RA synovial fibroblasts (Kitano et al., 2006; Nochi et al., 2008). PGE2 is an autocrine lipid mediator derived from arachidonic acid metabolism by cyclooxygenases (COX-1 or COX-2) (Ghosh et al., 2004). High levels of PGE2 were detected in synovial fluids of RA patients (Hidaka et al., 2001; Lettesjo et al., 1998). PGE2 has pro-inflammatory properties and contributes to the pathogenesis of arthritis, since PGE2 stimulates angiogenesis in the rheumatoid synovium (Ben-Av et al., 1995) and triggers bone resorption by osteoclasts (Mino et al., 1998). Thus, S1P may aggravate synovial hyperplasia, inflammation and angiogenesis through the induction of COX-2 and PGE2 in RA synovial tissues. Indeed, the COX-2-PGE2 axis has been suggested to play an important role in arthritic diseases by stimulating inflammation, angiogenesis, and osteoclastic bone destruction (Martel-Pelletier et al., 2003).

**4.4.3 S1P signalling in the proliferation and survival of RA synovial fibroblasts, B** 

and via the activation of ERK (Kim et al., 2006; Masuko et al., 2007).

Abnormal proliferation and resistance to apoptosis of synovial fibroblasts have been suggested to contribute directly to the hyperplasia of the rheumatoid synovium (Ospelt et al., 2004). Indeed, S1P appears capable of increasing cell survival and inhibiting apoptosis of various cell types, including B lymphoblastoid cells derived from RA patients (Pi et al., 2006). We previously demonstrated that S1P, through S1P1, protected synovial fibroblasts from apoptosis (Zhao et al., 2008). In our study we were not able to demonstrate a significant effect of S1P on RA synoviocyte proliferation, although Kitano et al. (2006) reported a small stimulatory effect of S1P on cell proliferation. Nonetheless these studies suggest that S1P can induce the proliferation of synovial fibroblasts, which may contribute to synovial hyperplasia (Knedla et al., 2007). S1P was also reported to aggravate arthritis by inducing chondrocyte proliferation through the stimulation of COX-2 and PGE2 production

our knowledge the effect of specific inhibitors of SphK1, such as BML-258 (Paugh et al., 2008), has not been tested in animal models of arthritis.

At present, evidence for roles of SPPs and SPL in RA is limited. However, up-regulation of SPP2 was detected in samples of skin lesions from patients with psoriasis, a chronic inflammatory skin disease (Mechtcheriakova et al., 2007). Interestingly, elevated mRNA expression of SPP1 and SPL was observed in RA synoviocytes compared to non-arthritic synoviocytes (Zhao et al., unpublished data).
