Identification of HMGA2 inhibitors by AlphaScreen-based ultra-high-throughput screening assays

Abstract

The mammalian high mobility group protein AT-hook 2 (HMGA2) is a multi-functional DNA-binding protein that plays important roles in tumorigenesis and adipogenesis. Previous results showed that HMGA2 is a potential therapeutic target of anticancer and anti-obesity drugs by inhibiting its DNA-binding activities. Here we report the development of a miniaturized, automated AlphaScreen ultra-high-throughput screening assay to identify inhibitors targeting HMGA2-DNA interactions. After screening the LOPAC1280 compound library, we identified several compounds that strongly inhibit HMGA2-DNA interactions including suramin, a century-old, negatively charged antiparasitic drug. Our results show that the inhibition is likely through suramin binding to the "AT-hook" DNA-binding motifs and therefore preventing HMGA2 from binding to the minor groove of AT-rich DNA sequences. Since HMGA1 proteins also carry multiple "AT-hook" DNA-binding motifs, suramin is expected to inhibit HMGA1-DNA interactions as well. Biochemical and biophysical studies show that charge-charge interactions and hydrogen bonding between the suramin sulfonated groups and Arg/Lys residues play critical roles in the binding of suramin to the "AT-hook" DNA-binding motifs. Furthermore, our results suggest that HMGA2 may be one of suramin's cellular targets.

Figure 1

The AlphaScreen primary assay for HMGA2-DNA interaction. The biotin-labeled FL814 (double-stranded DNA) and the His-tagged HMGA2 (green oval) were immobilized to streptavidin-coated donor beads and nickel chelate (Ni-NTA) acceptor beads, respectively.

Figure 2

HMGA2-DNA pilot screens using the Sigma LOPAC1280 compound library. Netropsin was used as positive controls.

Figure 3

The discovery of suramin as a potent inhibitor of HMGA-DNA interactions. (a) The AlphaScreen Primay assay with IC₅₀ of 2.6 μM. (b) The TR-FRET LANCE secondary screen. (c) The couterscreen assay using the BRD4 AlphaScreen Assay. (d) The cytotoxicity assay. All assays were described in “Methods”. The standard deviation was calculated according to three independent experiments. The curve represents the best fit of a four parameter logistic that was determined by nonlinear regression. Data points represent mean ± SEM .

Figure 4

Sample raw data from isothermal titration calorimetry (ITC) experiments for the titration of suramin to HMGA2 (a) and ATHP3 (b). ITC experiments were performed according to conditions as described in “Methods”. The ITC data were fit using the software supplied by the manufacturer to yield thermodynamic parameters.

Figure 5

Suramin induces the differentiation of brain tumor stem cells (BTSCs) #83 and #30p. (a) Light Microscopy of BTSC#83 and BTSC#30p treated with various concentrations of suramin for 6 days, compared to non-treated (NT) cells. (b) Western blot analysis (left panel) and densitometry (right panels) of HMGA2 expression after the 6-day treatment of suramin. This image was generated using two different blots of the same Western blot by antibodies against HMGA2 (the top panel) and actin (the bottom panel), respectively. The original images are provided in the supplemental information Fig. S11. (c) Expression of HMGA2, SNAIL, MSI1, ID2, and OLIG2 in suramin-treated BTSCs. qRT-PCR analyses of HMGA2 and SNAIL in BTSC#83 and BTSC#30p treated with suramin 100 µM for 48 h, compared to non-treated cells. qRT-PCR analyses for stemness (MSI1 and ID2) and differentiation markers (OLIG2) in BTSC#30p, treated with suramin 100 µM for 9 days. Fold changes are normalized for actin expression. Data represent the mean value +/− SD of two or three independent experiments performed in duplicate. (*p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; Student’s t test).

Figure 6

The simulated structure of the suramin-ATHP3 complex obtained by molecular docking and simulations.