Synthesis and properties of a selective inhibitor of homeodomain-interacting protein kinase 2 (HIPK2)

Abstract

Homeodomain-interacting protein kinase 2 (HIPK2) is a Ser/Thr kinase controlling cell proliferation and survival, whose investigation has been hampered by the lack of specific inhibitors able to dissect its cellular functions. SB203580, a p38 MAP kinase inhibitor, has been used as a tool to inhibit HIPK2 in cells, but here we show that its efficacy as HIPK2 inhibitor is negligible (IC₅₀>40 µM). In contrast by altering the scaffold of the promiscuous CK2 inhibitor TBI a new class of HIPK2 inhibitors has been generated. One of these, TBID, displays toward HIPK2 unprecedented efficacy (IC₅₀ = 0.33 µM) and selectivity (Gini coefficient 0.592 out of a panel of 76 kinases). The two other members of the HIPK family, HIPK1 and HIPK3, are also inhibited by TBID albeit less efficiently than HIPK2. The mode of action of TBID is competitive with respect to ATP, consistent with modelling. We also provide evidence that TBID is cell permeable by showing that HIPK2 activity is reduced in cells treated with TBID, although with an IC₅₀ two orders of magnitude higher (about 50 µM) than in vitro.

Figure 1. Structures of 2-aryl-4,5,6,7-tetrabromoisoindoline-1,3-dione derivatives.

The 2-aryl-4,5,6,7-tetrabromoisoindoline-1,3-dione scaffold is shown on the left, in the upper part of the figure, where R was variably replaced in the individual compounds, as indicated on the right. In the lower part of the figure the formulae of TBB, TBI and SB203580 are reported for comparison.

Figure 2. In silico analysis of HIPK2-TBID complex.

Molecular docking of TBID (yellow) was performed in the active site of the human HIPK2 model (green).

Figure 3. Kinetic analysis of HIPK2 inhibition by TBID.

Inhibition of HIPK2 by TBID is competitive with respect to the phosphodonor substrate ATP. Kinetics were performed as described in Materials and Method either in the absence or in the presence of the indicated TBID concentrations. The data represent means of triplicate experiments with SEM never exceeding 15%.

Figure 4. Selectivity profile of TBID (10 µM) against 70 kinases panel.

Inhibition assays were performed with 10 µM TBID under conditions described or referenced in the Experimental section. Residual activity expressed as per cent of activity in the absence of TBID is shown.

Figure 5. Lorenz curves, Gini coefficients and hit rates for TBID and TBI.

For details see experimental section and . Lorenz curves were drawn from the selectivity profile of Figure 4 and from analogous data published in ref. for TBID and TBI, respectively.

Figure 6. Cell treatment with TBID inhibits endogenous HIPK.

A. HIPK2 was immunoprecipitated from lysates of HepG2 cells treated with different concentrations of TBID or the inactive analog 5e, as indicated; HIPK2 activity was measure towards the specific peptide substrate, as detailed in Materials and Methods. The amount of immunoprecipitated HIPK2 is shown by WB. B. CEM cells were treated as indicated, then 10 µg of cell lysate proteins were analysed by WB with an antibody against pS46 of p53, anti-total p53, or anti-actin, as loading control. A representative experiment is shown on the left, while a histogram is presented on the right where Sp46 p53 of three separate experiments has been quantified, normalized to the p53 total level, and reported as means±SEM. C. Cell viability was assessed by the method of MTT, treating HepG2 or CEM cells as indicated (left panel) or by evaluating the amount of the caspase substrate PARP (HepG2 cells, right panel), whose reduction would denote apoptosis. The recognized band corresponds to the full length PARP protein of 116 kDa.