From In Vitro to In Cellulo: Evaluation of Anti-TNFα Activity of a New Series of Small Molecules

The Tumor Necrosis Factor alpha (TNF) is a relevant clinical target for the treatment of chronic inflammatory diseases as rheumatoid arthritis or Crohn’s disease. Anti-TNF biotherapies are used for the treatment of these diseases, improving considerably patient living conditions but they are not without drawbacks. Small molecules inhibitors of TNF could present fewer disadvantages than existing biotherapies, with less side effects, no resistance, oral administration and would probably lead to less expensive costs. Today only few small molecules are known as direct inhibitors of TNF, SPD304 was the first small molecule described by He et al in 2005. None of these molecules showed both an efficient activity and a low toxicity, necessary to yield them into clinical trial. We have set-up a program aiming at finding new small molecules inhibitors of TNF. A preview in silico docking study led to the identification of potential anti-TNF molecules. Based on the docking results, new small molecules have been designed, synthetized and biologically evaluated. Herein we describe the biological evaluation of a series of thirty new synthetized compounds for their capacity to inhibit the TNF. These molecules were evaluated in vitro using ELISA and cellular tests and appear promising compared to previously described small molecules.


Introduction
The Tumor Necrosis Factor alpha (TNF) is a major cytokine of immunity. It is know that a dysregulation of TNF expression is involved in many diseases as diabetes, cancer and especially in chronic inflammatory diseases such as Crohn's disease [1] or rheumatoid arthritis [2] in which TNF is a prime target for treatment. The commercialization of anti-TNFα biotherapies, mainly monoclonal antibodies: infliximab (Remicade ® ), adalimumab (Humira ® ), certolizumab pegol (Cimzia ® ) and golimumab (Simponi ® ) and one soluble receptor: etanercept (Enbrel ® ) have significantly improved the living conditions of patients for more than 15 years. However, those biotherapies are not without drawbacks. These biomolecules can promote resistance to treatment or side effects such as weakening of the immune system [3][4] [5] . In addition, they are expensive (approximately $ 15,000 per year per patient) and the administration route is restrictive (intravenous or subcutaneous injections). Taking to account all this, it would be useful to find an alternative to those biotherapies. The use of small synthetic molecules would have several advantages. Production and treatment costs are significantly lower than for biotherapies.

SciForum
Mol2Net, 2018, 1(Section A, B, C, etc.), 1-x, type of paper, doi: xxx-xxxx 2 Oral administration will be easier to implement. A decrease of the undesirable effects is also possible as no immune response directed to the treatment is expected. Moreover, in case of appearance of serious side effects, the treatment can be stopped immediately as the half-life of a small molecule is shorter in comparison to biotherapies [3] . SPD304 was the first small molecule describe for the direct inhibition of the TNF by He et al in 2005 [6] . The authors demonstrated that the SPD304 promotes the formation of the inactive dimeric form of the TNF by displacing a subunit of active trimeric form. Today, only few small molecules are known as direct inhibitors of TNF and SPD304 remains a reference as TNF inhibitor despite its toxicity [7] [8] .
Recently, we identified a new scaffold of anti-TNF small molecules though combined in silico/in vitro studies. Based on the proposed scaffold, 30 new compounds have been synthesized and then evaluated for their TNF inhibitory capacity.

Results and Discussion
First of all, the 30 compounds are evaluated via two ELISA assays, using SPD304 as a reference. A binding test, allowed us to determine the inhibitory activity of compounds for the binding of the TNF to its receptor TNFRI (IC50 values  table 1). The shifting test, compares the activity of the evaluated molecules to the inhibitory activity of SPD304 (shifting values table 1).
According to the in vitro results, we can classify our compounds into three groups. The first one comprises the non-active molecules with IC50 ≥ 50 µM (compounds 1, 16, 17 and 18). The second one includes 15 compounds with low activity, 50 µM > IC50 > 12 µM (compounds 2-8,  12, 13, 15, 22, and 27-30). The third group concerns the active compounds with IC50 ≤ 12 µM and Shifting ≥ 100 % (9-11, 14, 19-21 and 23-26). Taking into account these results, 11 compounds of our series are active, with a better or comparable activity to SPD304.  Then, these best 11 compounds were evaluated in cellulo using the HEK-Blue™ TNFα cells. At 100 µM, the small molecules are cytotoxic, with survival percentages comprising between 0 and 18%. At lower concentrations, the cytotoxicity decreases. The results for 6.25 µM and 1.56 µM are presented in Figure 1. At 6.25 µM, 5 compounds (10, 11, 14, 19 and 23) displayed a survival percentage higher than 60%. Those molecules have an inhibitory activity at this concentration, with neutralization percentages comprising between 57 and 80%, except for compound 23 with 17% of neutralization. However, 5 compounds remain highly toxic at 6.25 µM (20, 21, 24, 25 and 26), with survival percentages lower than 35%. At 1.56 µM, the 11 compounds are not cytotoxic with survival percentages higher than 80%. Nevertheless, the inhibitory activity decreases at this concentration. Only 3 compounds (20, 25 and 26) are still active with 60% of neutralization on cells for the inhibition of the interaction of the TNF with its receptor TNFRI, at this low concentration of 1.56 µM.
TNFα-TNFRI binding ELISA. Microtiter plates were coated with 10 ng of TNFRI per well and incubated one night at 37 °C. The wells were washed, blocked with PBS/BSA 2% for two hours and washed as before. Serial dilutions of compounds were mixed with a fixed quantity of TNFα in PBS/BSA 1% and incubated two hours at 37 °C. 100 μL of the mix were added to the wells and plates were incubated overnight at 4 °C. Wells were washed incubated with 30 ng of TNFα biotinylated antibody in 100 μL of PBS/BSA 1% for two hours at 37 °C. Wells were washed and incubated with avidin-HRP (1:500) in 100 μL of PBS/BSA 1% for 30 minutes at 37 °C. After washing, TMB solution was added to wells, quenched with 50 μL of 1 M H2SO4 solution. Absorbance was measured at 450 nm.
TNFα-TNFRI shifting ELISA. Microtiter plates were coated with 10 ng of TNFRI per well, incubated one night at 37 °C. The wells were washed, blocked with PBS/BSA 2% for two hours and washed as before. Serial dilutions of TNFα in PBS/BSA 1% were mixed with a fixed quantity of compounds and incubated two hours at 37 °C. 100 μL of the mix were added to the wells and plates were incubated overnight at 4 °C. Wells were washed incubated with 30 ng of TNFα biotinylated antibody in 100 μL of PBS/BSA 1% for two hours at 37 °C. Wells were washed and incubated with avidin-HRP (1:500) in 100 μL of PBS/BSA 1% for 30 minutes at 37 °C. After washing, TMB solution was added to wells, quenched with 50 μL of 1 M H2SO4 solution. Absorbance was measured at 450 nm.
Survival percentages were calculated using equation 1: Neutralization percentages were calculated using equation 2:

Conclusions
We evaluated the TNFα inhibitory activity of 30 new small molecules. Eleven compounds displayed a better or comparable activity to SPD304 used as reference. A test using HEK-Blue™ TNFα cells was used to confirm this inhibition ability. At low concentration (1.56 µM), 3 compounds are non cytotoxic and still active for the inhibition of the interaction of the TNF with its receptor TNFRI. Optimization of these 3 compounds is currently investigated in our laboratory and will be reported in due course.