**5.1 Active targeting strategies**

Current drug delivery systems for rheumatoid arthritis have almost exclusively been developed based upon the principles of passive targeting. Specificity and, consequently, efficacy can be further improved by employing an active targeting moiety. As indicated previously, RASFs and RASMs possess receptors that can potentially be used to increase the ability of the drug carrier systems to discern pannus tissue from healthy tissue. Those receptors that have the highest potential for success are discussed in this section.

Folate receptor β (FRβ) has been identified as a viable candidate for active targeting of RASMs. Several transport mechanisms exist whereby folate and folate antagonists, including methotrexate, can enter a cell. Numerous cell types throughout the body constitutively express the reduced folate carrier (RFC), a transmembrane protein. In contrast, membrane associated folate receptors (MFRs) are restricted in expression and facilitate uptake by endocytosis. The beta isoform of MFR (FRβ) is expressed selectively by activated macrophages within the pannus tissue (Nagayoshi et al., 2005; van der Heijden et al., 2009). Thus, folate may serve as an effective means of actively targeting the pannus tissue. In support of this, folate-tagged cationic and anionic poly(amidoamine) (PAMAM) dendrimers loaded with indomethacin were more effective at treating arthritic rats than indomethacin-loaded dendrimers that were folate-free (Chandrasekar et al., 2007a; Chandrasekar et al., 2007b; Chauhan et al., 2004). Furthermore, a recent study demonstrated that PAMAM dendrimers covalently linked to both folate and methotrexate could be used to selectively deliver methotrexate to synovial macrophages (Thomas et al., 2011). Although the antifolate MTX does not have a high affinity for any MFRs, several other folate analogs have been identified that possess high affinity for FRβ with low affinities for other MFRs and the RFC (van der Heijden et al., 2009). These analogs, therefore, have strong potential to be used as active targeting moieties in the future and may possess inherent anti-rheumatic properties.

RASFs also possess an active targeting receptor in the cell surface adhesion molecule CD44, the hyaluronic acid receptor. Several studies have indicated that CD44 is upregulated in the pannus tissue relative to healthy, normal tissue (Haynes et al., 1991; Johnson et al., 1993). Furthermore, RASFs express numerous CD44 alternatively spliced variants, including long isoforms CD44v3 and CD44v6 that are associated with an enhanced invasive capacity (Croft et al., 1997). CD44 is critical to rheumatoid arthritis pathogenesis, facilitating inflammatory cell migration and signaling activation of lymphocytes (Naor & Nedvetzki, 2003). Despite evidence of upregulation and/or selective expression, the CD44 receptor has not yet been used as an active target for drug delivery in the treatment of rheumatoid arthritis. Some cancer cells also upregulate CD44, and the potential for success using this active targeting strategy has previously been illustrated by anti-cancer drug carrier systems modified with hyaluronic acid oligomers (Ossipov, 2010).

As indicated by the feasibility of passive targeting, neovascularization is a critical component of rheumatoid arthritis pathogenesis (Szekanecz & Koch, 2008). Adhesion molecules, including intercellular cell-adhesion molecule-1 (ICAM-1) and E-selectin, are upregulated within the newly formed vasculature and; therefore, the endothelium serves as another possibility for active targeting (Banquy et al., 2008; Zhang et al., 2008). Preliminary studies demonstrated that a large number of PLGA-PEG nanoparticles surface modified with a peptide specific for ICAM-1 were endocytosed by vascular endothelial cells as compared to unmodified particles (Zhang et al., 2008). A similar phenomenon was observed for polyester-based polymeric nanoparticles labeled with ligand specific for E-selectin (Banquy et al., 2008). Likewise, liposomes surface modified with the polysaccharide Sialyl Lewis X, which is known to bind selectively to E-selectin, were shown to accumulate within inflamed regions when administered intravenously to arthritic mice (Hirai et al., 2007). The vitronectin receptor, overexpressed by both angiogenic endothelial cells and RASMs (Tandon et al., 2005; Wilder, 2002), is an attractive target based upon the extensive use of Arg-Glyc-Asp (RGD) tripeptide labeled carrier systems for drug delivery to invasive tumor tissue (Hsu et al., 2007). An antibody to V3, Vitaxin (ME-522), is currently being clinically developed for rheumatoid arthritis (Wilder, 2002); however, the RGD peptide sequence has not yeen been applied towards increasing the targeting ability and efficacy of existing DMARDs.
