Background/hypothesis: The co-stimulatory CD40-CD40L receptor-ligand dyad is a well-known driver of chronic inflammatory diseases, including atherosclerosis. Upon activation, CD40 recruits tumor necrosis factor receptor-associated factors (TRAFs) to elicit intracellular signaling. Murine studies with mutated CD40-TRAF binding sites have shown that CD40-TRAF6 interactions in macrophages, but not CD40-TRAF2/3/5 interactions, are important for atherosclerosis. We thus synthesized a small molecule inhibitor called 6877002, that selectively blocks CD40-TRAF6 interactions. 6877002 reduced de novo and established atherosclerosis in ApoE^-/- mice without causing immunosuppressive side effects. We hypothesized that 6877002 is an excellent candidate to pass the translational pipeline towards a clinical application to treat atherosclerosis. Thus, the present study aimed at assessing the toxicity and pharmacological properties of the TRAF-STOP 6877002.

Methods/results: Toxicological analysis using in silico methods predicted that the structure of 6877002 is non-mutagenic and can be assigned to class 5 under ICH M7b. Together with Cyprotex, we performed elaborate in vitro experiments assessing the ADME profile of 6877002. Using a bidirectional Caco-2 assay, we found that 6877002 had low to moderate intestinal permeability (A2B direction) with no significant efflux, suggesting it is not a substrate for efflux transporters such as Pgp/BCRP or MRP2. Plasma protein binding assessment showed very low unbound fractions, indicating high plasma protein binding. The potential to inhibit cytochrome P450 isoforms was evaluated using human liver microsomes in the presence of specific probe substrates. Isoform CYP1A2 inhibition was observed with IC₅₀ of 4.01 μM. For all other CYP isoforms, no potential for inhibition was detected, suggesting low potential for drug-drug interactions. To screen for putative cardiovascular risks, we assessed inhibition of the hERG ion channel; no IC₅₀ could be determined for 6877002.

Hepatic clearance was very quick in mice, rats and dogs, while minipig and human hepatocytes showed low intrinsic clearance. In vivo pharmacokinetics studies in rats showed low bioavailability (8.43%) and rapid clearance of orally administered 6877002 compared to i.v. injected 6877002, likely – and in line with our in vitro data – due to a strong and rapid first-past effect in hepatocytes and uptake by macrophages. To evaluate the possibility that one or more of the 6877002 metabolites are potent binders in the CD40-TRAF6 pocket, we performed LC-MS/MS-based metabolite profiling. Structural elucidation was provided for the most abundant metabolites before subjecting them to virtual ligand modeling. Metabolite 4 (M4), the most abundant 6877002 metabolite, showed the highest binding mode (-19.6; defined as Binding Free Energy derived from 40-50ns of MD run) to the CD40-TRAF6 pocket, which was even higher than that of 6877002 (-16.5). Subsequently, M4 was administered i.p. to Western diet-fed ApoE^-/- mice for 6 weeks and we observed that, similar to 6877002, M4 reduced atherosclerosis by 54%, suggesting that M4 is an active metabolite.

Conclusion/future directions: These data validate that CD40-TRAF6 inhibitors are a promising therapeutic approach for atherosclerosis treatment with high translational potential. In follow-up studies, we will assess the potential therapeutic benefit of 6877002 and M4 on atherosclerosis in large animal models, such as minipigs.