Systemic lupus erythematosus (SLE) is an autoimmune disease involving multi-organ systems with a widely heterogeneous clinical presentation (Durcan et al., 2019). Renal involvement, observed mainly in lupus nephritis (LN), reportedly develops in approximately 50% of patients with SLE and increases the risks of renal failure and death (Aragon et al., 2020; Bertsias et al., 2012; Karasawa et al., 2018). However, despite medical advances, treatment of LN remains controversial and challenging (Kagawa et al., 2018; Mok, 2016). LN is typically treated with immunosuppressive drugs, such as glucocorticoids, mycophenolate mofetil (MMF) or cyclophosphamide. However, these treatments are not uniformly effective, do not lead to sufficient complete remission, are associated with safety concerns, and have not reduced the incidence of end-stage renal disease due to LN (Anders et al., 2020; Appel et al., 2009; Croca et al., 2011). Therefore, alternative therapeutic approaches for LN are required.

To facilitate the development of drugs for LN, one useful tool is a clinically relevant LN animal model, especially a therapeutic model which closely replicates clinical cases of LN. NZB/W F1 mice are a widely used mouse model of SLE/LN which spontaneously develops various severe lupus-like manifestations including lymphadenopathy, splenomegaly, elevated serum anti-dsDNA IgG and immune complex-mediated glomerulonephritis. The spontaneous elevation of proteinuria and nephritis become apparent between 5 and 12 months of age (Perry et al., 2011). However, because the timing of proteinuria onset is inconsistent and heterogeneous, selecting a sufficient number of animals with nephritis with similar pathology on which to test promising drug candidates can be a difficult and lengthy process. Thus, in the present study, we examined whether an accelerated model (rather than a spontaneous model) that can develop nephritis in a shorter period of time might be more useful for evaluating the therapeutic efficacy of candidate drugs.

Administration or induction of pathogenic cytokines is often used to control the progression of spontaneously occurring disease manifestations in animals. Polyinosinic-polycytidylic acid (Poly (I:C)), an analog of double-stranded RNA (dsRNA), has been reported to accelerate the emergence of lupus manifestations in several lupus-prone mice by inducing type-I interferons (IFNs) in macrophages and kidney cells via the stimulation of toll-like receptor 3 (TLR3) (Jorgensen et al., 2006; Lahita et al., 2011; Patole et al., 2005; Triantafyllopoulou et al., 2010). Given the critical role of IFN in the pathogenesis of LN (Baccala et al., 2007; Baechler et al., 2003; Iwamoto et al., 2021), Poly (I:C) may be an appropriate compound with which to accelerate the onset and reduce the pathological variability of LN in animals. However, no LN model has yet been established using Poly (I:C). Further, the optimal experimental conditions for evaluating candidate drugs in such a model are unknown.

The present study aimed to establish a new murine LN model with faster disease onset and lower pathological variability at the start of treatment for screening clinical candidates using Poly (I:C). We also evaluated on this model the therapeutic effects of MMF, a recommended first-choice agent for LN (Fanouriakis et al., 2019), and the cathepsin S (CatS) inhibitor ASP1617, which prevents the development of lupus-like glomerulonephritis in the spontaneous NZB/W F1 mouse model (Kawato et al., 2022).