Assaying kinase activity of the TPL-2/NF-κB1 p105/ABIN-2 complex using an optimal peptide substrate

Abstract

The MKK1/2 kinase tumour progression locus 2 (TPL-2) is critical for the production of tumour necrosis factor alpha (TNFα) in innate immune responses and a potential anti-inflammatory drug target. Several earlier pharmaceutical company screens with the isolated TPL-2 kinase domain have identified small-molecule inhibitors that specifically block TPL-2 signalling in cells, but none of these have progressed to clinical development. We have previously shown that TPL-2 catalytic activity regulates TNF production by macrophages while associated with NF-κB1 p105 and ABIN-2, independently of MKK1/2 phosphorylation via an unknown downstream substrate. In the present study, we used a positional scanning peptide library to determine the optimal substrate specificity of a complex of TPL-2, NF-κB1 p105 and ABIN-2. Using an optimal peptide substrate based on this screen and a high-throughput mass spectrometry assay to monitor kinase activity, we found that the TPL-2 complex has significantly altered sensitivities versus existing ATP-competitive TPL-2 inhibitors than the isolated TPL-2 kinase domain. These results imply that screens with the more physiologically relevant TPL-2/NF-κB1 p105/ABIN-2 complex have the potential to deliver novel TPL-2 chemical series; both ATP-competitive and allosteric inhibitors could emerge with significantly improved prospects for development as anti-inflammatory drugs.

Keywords: TPL-2; high-throughput assay; inflammation; kinase.

Figure 1.. The primary amino acid sequence specificity of TPL-2.

The peptide library comprised 198 individual biotinylated peptide mixtures. Each peptide contained a central phosphoracceptor Ser or Thr, flanked by degenerate positions, comprising an equimolar mixture of the 17 amino acids, excluding Cys, Ser and Thr. In each peptide, one position was fixed (fixed residue) with one of the 20 naturally occurring unmodified amino acids, phosphor-Thr (pT) or phosphor-Tyr (pY). Peptides were incubated with 30 nM recombinant TPL-2/NF-κB1 p105/ABIN-2 complex at a final substrate concentration of 50 µM at 30°C for 1 h. Assays were performed in TPL-2 kinase buffer plus 10 µM ATP and 3 nCi/µL [γ-³²P]ATP in the absence (A) or presence (B) of C34 TPL-2 inhibitor (10 µM). Following incubation, reactions were transferred onto SAM2 membranes, which were then washed extensively. Incorporation of ³²P into peptides was quantified by phosphorimaging.

Figure 2.. Testing an optimized peptide substrate for TPL-2/NF-κB1 p105/ABIN-2.

(A) The primary and secondary amino acid preferences for phosphorylation by the recombinant TPL-2/NF-κB1 p105/ABIN-2 complex are shown. (B) The sequences and scansite scores for the MKK1 activation loop and optimized TPL-2tide peptide substrates. (C) Time-course experiment comparing TPL-2/NF-κB1 p105/ABIN-2 complex phosphorylation of MKK1 and TPL-2tide peptides (50 µM final concentration). Assays were performed with 30 nM the recombinant TPL-2/NF-κB1 p105/ABIN-2 complex in TPL-2 kinase buffer plus 1 mM ATP and 0.02 µCi/µL [γ-³²P]ATP. Peptides were transferred onto streptavidin-coated membranes, which were then extensively washed. Incorporation of ³²P into peptides was quantified by phosphorimaging. Linear regression was fitted with GraFit version 7.0.3. Values are means ± SD for three replicate reactions. (D) Kinase assays as in (C) comparing TPL-2tide peptide phosphorylation by wild-type (WT) or kinase-inactive (D270A) TPL-2/NF-κB1 p105/ABIN-2 complex (15 min). (E and F) TPL-2^30–404 and TPL-2/NF-κB1 p105/ABIN-2 were titrated (1 : 1.66) in 384-well plates. A substrate solution containing ATP and TPL-2tide (E) or S5 peptide (F) was then added. The plates were analyzed using a Sciex API6500 (E) or API4000 (F) Triple Quad with RapidFire™ Technology. The graph shows total peak area counts for the MRM transition 953.6/904.3 Da (E; phosphorylated TPL-2tide peptide) and 612.9/120 Da (F; phosphorylated S5 peptide) for the indicated enzyme at increasing concentrations. Linear regression was fitted using GraFit software. In E and F, representative graphs from one experiment are shown, including standard deviation of triplicate assays.

Figure 3.. High-throughput Rapid Fire™ mass spectrometry assay of TPL-2/NF-κB1 p105/ABIN-2 complex kinase activity.

(A) The efficacy of known TPL-2 inhibitors was tested for TPL-2^30–404 and the TPL-2 complex. 5 μL of the enzyme solutions were added to each well (two replicates for each using final concentrations of 80 nM TPL2 30–404 and 80 nM TPL-2/NF-κB1 p105/ABIN-2). The compounds were then dispensed using a HP digital dispenser. The enzyme/compound mixture was incubated for 10 min before the addition of 5 µL of substrate buffer containing final concentrations of 100 µM TPL-2tide and 400 µM ATP to initiate the reaction. The reactions were incubated for 30 min quenched with 50 µL of 1% formic acid. The plates were analyzed using a Sciex API6500 Triple Quad with RapidFire™ Technology. The data was analyzed using product counts for the MRM transition 953.6/904.3 Da (phosphorylated peptide) using a four-parameter logistical equation fitted using GraFit software to determine IC₅₀ values. For each compound, multiple replicates were run for each experiment and the experiment was repeated four to five times. The IC₅₀ values shown are generated from all datasets analyzed and include standard errors (SE). (B) Representative graphs of one experiment for each compound are shown, including standard deviation from replicates.