Full length human TRKA, TRKB and TRKC cDNAs were synthesized by Blue Heron Biotechnology (Bothell, MA) according to the sequences of NM_002529 for TRKA, NM_006180 for TRKB and NM_002530 for TRKC, and cloned into pcDNA6.2/V5-DEST (Invitrogen, Carlsbad, CA) via Gateway™ LR reaction.

NFAT beta-lactamase reporter line NFAT-bla CHO-K1 cells (Invitrogen, Carlsbad, CA, catalog number K1078) were transfected using Lipofectamine™ LTX transfection reagent (Invitrogen, Carlsbad, CA) with pcDNA6-TRKA, pcDNA6-TRKB and pcDNA6-TRKC and selected for blasticidin-resistance. Blasticidin-resistant TrkA transfected cells were stimulated with NGF (Invitrogen, catalog number 13257-019), TrkB transfected cells with BDNF (Invitrogen, catalog number PHC7074) and TrkC transfected cells with NT-3 (Invitrogen, catalog number PHC7034) for 5 hours before neurotrophin stimulated beta-lactamase expressing cells were collected by FACS as single cells onto 96-well plates. The survived clones that responded to the respective ligand were selected and expanded. The clone with the best response to each ligand was further tested in a 384-well format for the dose response of neurotrophins and inhibitors and in a 96-well format for RNAi knock-down.

Cells in a sub-confluency state (or cryopreserved cells) were resuspended in Assay Medium (DMEM with GlutaMAX™ (Invitrogen, catalog number 10569) supplemented with 0.1% dialyzed FBS, 0.1 mM NEAA, 25 mM HEPES, 100 U/mL Penicillin and 100 µg/mL Streptomycin) and plated in a 384-well assay plate (Corning, Lowell, MA, catalog number 3712) at 10,000 cells (32 μL) per well. 32 μL of Assay Medium without cells were plated in Cell-free Control wells on the same plate. 4 µL 1% DMSO was added to each well to reach final DMSO concentration of 0.1% to mimic the screening situation where compounds dissolved in DMSO are added to test wells. Cells were then stimulated with 4 µL/well 10 x ligands (NGF, BDNF, NT-4 (Invitrogen, catalog number PHC7024), NT-3 or Thapsigargin (Sigma, St. Louis, MO)) over the indicated concentration range (1 pM ~ 50000 pM) for 5 hours before adding 8 µL/well 6X LiveBLAzer™-FRET B/G Substrate mixture (Invitrogen, catalog number K1096) for 2 hours. Fluorescence intensity at excitation 406 nm and emission 460 nm and 530 nm were obtained using Tecan Safire² fluorescence plate reader (Tecan, Durham, NC) (Table 1). After subtracting the average fluorescence intensity from the Cell-free Control wells, the 460nm/530nm emission ratio was calculated. Response Ratio is a measurement of the assay window and is calculated as the 460nm/530nm Emission Ratio of the stimulated wells divided by the 460nm/530nm Emission Ratio of the unstimulated wells. Response Ratios were plotted against test ligand concentrations in log scale and then analyzed using Prism software (GraphPad Software, Inc. San Diego, CA). Sigmoidal dose-response equation with varying slope was used to fit the data and generate EC₅₀ values. Z’-factor values were calculated as: Z’-factor = 1 - [(3 x stdev_unstim + 3 x stdev_maxstim) / (avg_maxstim – avg_unstim)].

Table 1

Beta-Lactamase Assay Protocol for Trk-NFAT-bla CHO-K1 Cells in 384-Well Format

Cells were seeded into a 384-well assay plate at 10,000 cells/well as described above and treated for 30 minutes with 4 µL of AG879 (EMD, San Diego, CA, catalog number 658460), or GW441756 (Tocris Bioscience, Ellisville, MO, catalog number 2238) or K252a (Invitrogen, catalog number PHZ1131), each at 10 x final concentrations. Cells were then stimulated with 4 µL of neurotrophins at 10 x EC₈₀ concentrations (EC₈₀ concentrations are: NGF, 0.77 nM; BDNF, 1 nM; NT-3, 0.72 nM; Thapsigargin, 8 nM) for 5 hours before beta-lactamase assay was performed as described above.

Cells were plated in 96-well assay plates (Corning, catalog number 3603) at 6,000 cells/well in 100 µL growth medium (DMEM containing 10% dialyzed FBS, 0.1 mM NEAA and 25 mM HEPES), and incubated in 37°C, 5% CO₂ incubator overnight. Next morning, Stealth™ RNAi (Invitrogen, catalog number 10620318) and Lipofectamine^TM RNAiMAX (Invitrogen, catalog number 13778) complexes were prepared according to the manufacturer’s protocol and added to the wells (50 nM RNAi oligos and 1µL RNAiMAX/well). RNAi oligo targeting for beta-lactamase was used as the positive control and a random-sequenced oligo with 48% GC content (Med GC) as the negative control. A set of 2 oligos for each Trk was used:

TRKA1 (ACAUCAUCGAGAACCCACAAUACUU)

TRKA2 (ACGCUGCUCCUUGUGCUCAACAAAU)

TRKB1 (UGGUAAUGCUGUUUCUGCUUAAGUU)

TRKB2 (ACACCACGAACAGAAGUAAUGAAAU)

TRKC1 (GCCAAGUGUAGUUUCUGGCGGAUUU)

TRKC2 (CCAGACCAAUCUGAACUGGACCAAU)

Cells were incubated with RNAi oligos at 37°C for 32 hours, followed by a medium change. Cells were incubated in Assay Medium at 37ºC for 16 hours and then stimulated with respective ligands (NGF, 0.77 nM; BDNF, 1 nM; NT-3, 0.72 nM and Thapsigargin, 30 nM) for 5 hours before beta-lactamase assay was performed as described above.

For the purpose of developing cell-based assays for each Trk kinase, we generated stable clones of NFAT-bla CHO-K1 expressing full length human TrkA, TrkB or TrkC as described in Methods. Each clone was validated for the dose response to NGF, BDNF, NT-3 and NT-4 (Fig. 2, Table 2). TrkA-NFAT-bla CHO-K1 cells responded well to NGF with a maximum response ratio of 5.5 fold (Z’ factor of 0.75) and an EC₅₀ of 0.044 nM, consistent with previously published results [12Ahamed J, Venkatesha RT, Thangam EB, Ali H. C3a enhances nerve growth factor-induced NFAT activation and chemokine production in a human mast cell, line HMC-1 J Immunology 2004; 172: 6961-8.]. TrkA-NFAT-bla CHO-K1 cells showed a 3-fold response over baseline to NT-3 with an EC₅₀ around 16 nM, suggesting that NT-3 binds TrkA with a lower affinity than NGF [13Ryden M, Ibanez CF. Binding of neurotrophin-3 to p75NTR, TrkA, and TrkB mediated by a single functional epitope distinct from that recognized by TrkC J Biol Chem 1996; 271: 5623-7.]. BDNF and NT-4 had a minimal effect on TrkA-NFAT-bla CHO-K1 cells, but significantly stimulated beta-lactamase reporter activity in TrkB-NFAT-bla CHO-K1 cells with a maximum response ratio of 14 (Z’ value of 0.82) and an EC₅₀ of 0.39 nM for BDNF and a maximum response ratio of 13 (Z’ value of 0.8) and an EC₅₀ of 3.5 nM for NT-4 (Fig. 2B). TrkB-NFAT-bla CHO-K1 cells also responded to NT-3 with a maximum response ratio of 13 and an EC₅₀ of 2.5 nM, but did not respond at all to NGF, which is in agreement with previous findings [13Ryden M, Ibanez CF. Binding of neurotrophin-3 to p75NTR, TrkA, and TrkB mediated by a single functional epitope distinct from that recognized by TrkC J Biol Chem 1996; 271: 5623-7.]. TrkC-NFAT-bla CHO-K1 cells responded to TrkC preferred ligand NT-3 with a maximum response ratio of 8.5 (Z’ value of 0.9) and an EC₅₀ of 0.093 nM (Fig. 2C). TrkC-NFAT-bla CHO-K1 cells responded to BDNF only at the highest concentration (50 nM) tested and did not show any response to NGF and NT-4. All three cell lines responded to thapsigargin treatment with a similar EC₅₀ (data not shown) to that of the parental NFAT-bla CHO-K1 cells (Fig. 2D). Taken together, TrkA-NFAT-bla CHO-K1 cells specifically responded to TrkA preferred neurotrophin, NGF, TrkB-NFAT-bla CHO-K1 cells to BDNF, NT-3 and NT-4, and TrkC-NFAT-bla CHO-K1 cells to TrkC ligand NT-3. The fact that NT-3 generated dose responses in both TrkC and TrkB cells with a 10 fold less EC₅₀ in TrkC cells confirms that NT-3 binds to both TrkC and B with TrkC being the preferred receptor. The fact that CHO cells lack p75^NTR expression [14Zapf-coby Z, Olefsky J. Nerve growth factor processing and trafficking events following trkA-mediated endocytosis Endocrinology 1998; 139: 3232-40.] and the parental NFAT-bla CHO-K1 cells do not respond to neurotrophins (Fig. 2D) suggests that the effect we observed for each neurotrophin was mediated by the engineered Trk receptor. Our results confirmed differential affinity and rank order potency of each neurotrophin to Trk receptors as reported in the literature, which suggests that these cell-based reporter assays for Trk kinases can be used for studying the potency and selectivity of Trk modulators.

Fig. (2) Neurotrophin dose responses of TrkA, TrkB, TrkC and parental NFAT-bla CHO-K1 cell lines. TrkA-NFAT-bla CHO-K1 (A), TrkB-NFAT-bla CHO-K1 (B), TrkC-NFAT-bla CHO-K1 (C), and parental NFAT-bla CHO-K1 (D) cells were stimulated with NGF (■), BDNF (▲), NT-4 (♦), NT-3 (▼) and Thapsigargin (●, parental cells only) over the indicated concentration range in the presence of 0.1% DMSO for 5 hours before beta-lactamase assay was performed as described in Methods. Response Ratios were plotted for the indicated concentrations of each ligand (n=5 for each data point).

Table 2

EC₅₀ Values of Neurotrophins in the Three Trk Assays

We further validated these assays with a commonly used Trk kinase inhibitor, K252a. Dose response experiment showed K252a inhibited NGF-induced beta-lactamase activity in TrkA-NFAT-bla CHO-K1 cells (Fig. 3A), BDNF-induced beta-lactamase activity in TrkB-NFAT-bla CHO-K1 cells (Fig. 3B) and NT-3 induced TrkC-NFAT-bla CHO-K1 cells (Fig. 3C) with low nano-molar IC₅₀ values (Table 3), which are in accordance with literature values using various types of assays [7Berg MM, Sternber DW, Parada LF, Chao MV. K-252a inhibits nerve growth factor-induced trk-proto-oncogene tyrosine phosphorylation and kinase activity J Biol Chem 1992; 267: 13-6., 15Groth RD, Mermelstein PG. Brain-derived neurotrophic factor activation of NFAT (nuclear factor of activated T-cells)-dependent transcription: A role for the transcription factor NFATc4 in neurotrophin-mediated gene expression J Neurosci 2003; 23: 8125-34., 16Rende M, Brizi E, Conner J, et al. Nerve growth factor (NGF) influences differentiation and proliferation of myogenic cells in vitro via TrkA Int J Dev Neurosci 2000; 18: 869-5.]. Interestingly, K252a had a partial inhibitory effect on Thapsgargin induced reporter activity in all three cell lines at the highest concentration tested (Fig. 3, Table 3). AG879, also known as an EGFR family kinase inhibitor [17Zhou Y, Li S, Hu YP, et al. Blockade of EGFR and ErbB2 by the novel dual EGFR and ErbB2 kinase inhibitor GW572016 sensitizes human colon carcinoma GEO cells to apoptosis Cancer Res 2006; 66: 404-11.], has been reported in the literature to inhibit NGF induced TrkA autophosphorylation and downstream PLC-γ phosphorylation in micro-molar concentrations [18Ohmichi M, Pang L, Ribon V, Gazit A, Levitzki A, Saltiel AR. The tyrosine kinase inhibitor tyrophostin blocks the cellular actions of nerve growth factor Biochemistry 1993; 32: 4650-8.]. When tested in our reporter assays, AG879 inhibited all three Trk-mediated NFAT activation with an IC₅₀ around 1 μM (Table 3). Interestingly, this compounds inhibited Thapsigargin-induced NFAT reporter activity with much lower IC₅₀s, suggesting that AG879 affects other targets than Trks. GW441756, a potent TrkA inhibitor in a biochemical assay [19Wood ER, Kuyper L, Petrov KG, Hunter RN, Harris PA, Lackey K. Discovery and in vitro evaluation of potent TrkA kinase inhibitors: oxindole and aza-oxindoles Bioorg Med Chem Lett 2004; 14: 953-7.], also inhibited all three Trk kinases induced NFAT-bla reporter activity with an IC₅₀ value of 0.23 μM for TrkA, 0.14 μM for TrkB and 0.46 μM for TrkC. It also affected Thapsigargin-induced NFAT reporter activity, but only at much higher concentrations (Fig. 3, Table 3). These data suggest that GW441756 is a more potent and selective Trk inhibitor than AG879. Our data provides first evidence that GW441756 is a potent inhibitor for all three Trk receptor kinases. This inhibitor experiment demonstrated how one can use these three cell assays to identify compounds based on both potency and selectivity.

Fig. (3) Inhibition of neurotrophin-induced beta-lactamase reporter activity in the cognate Trk receptor expressing cells. Cells were pretreated with indicated concentrations of K252a (▼), or AG879 (■) and GW441756 (▲) and then stimulated with NGF (TrkA cells, A), BDNF (TrkB cells, B), or NT-3 (TrkC cells, C), or Thapsgargin for all three cell lines for 5 hours before beta-lactamase assay was performed as described in Methods. Response Ratios were plotted for the indicated concentrations of each compounds (n=3 for each data point).

Table 3

IC₅₀ Values (in nM) of Inhibitors in the Three Trk Assays

To further demonstrate the specificity of each assay, we examined the effect of Trk specific RNAi oligos on neurotrophin-induced NFAT-bla reporter activity. We also determined the specificity of each oligo by examining the effect of the oligos on thapsigargin-induced reporter activity. Cells were treated with Stealth™ Select RNAi oligos and stimulated with neurotrophins as described in Methods. Beta-lactamase gene-specific RNAi (b-lac) was used as a positive control and showed complete inhibition of neurotrophin-induced reporter activity in all three cell lines (Fig. 4). RNAi oligo with 48% GC content (Med GC) was used as a negative control and had no inhibitory effect in any of the three assays. Treatment with two TrkA RNAi oligos significantly knocked down NGF induced beta-lactamase activity in TrkA expressing cells (Fig. 4A) and had little effect on BDNF induced beta-lactamase activity in TrkB cells (Fig. 4B) and NT-3 induced in TrkC cells (Fig. 4C). Treatment with the two TrkB oligos completely inhibited BDNF induced reporter activation in TrkB cells (Fig. 4B) and had minimal effect on the TrkA and TrkC cells induced by NGF and NT-3 respectively (Fig. 4A and C). Finally, treatment with the two TrkC oligos completely knocked down NT-3 induced beta-lactamase reporter activity in TrkC cells and had minimal inhibitory effect on TrkA and TrkB cells (Fig. 4C). Longer incubation (72 hour) with these RNAi oligos gave a similar inhibitory pattern (data not shown). No cytotoxicity effect was observed with any of the RNAi oligos used even after 72 hour incubation.

Fig. (4) Effects of various RNAi oligos on neurotrophin-induced beta-lactamase reporter activity. Cells were incubated with indicated RNAi oligos for 32 hrs, followed by a medium change and additional 16 hour-incubation. Cells were then stimulated with respective ligands (NGF for TrkA cells, A; BDNF for TrkB cells, B; NT-3 for TrkC cells, C) for 5 hours before beta-lactamase assay was performed as described in Methods. Response Ratios were plotted for each RNAi oligo (n=3 for each data point). Unstim: untransfected cells were left unstimulated with ligands; Mock: cells were treated with transfection reagent (no RNAi oligo) and stimulated with ligands.

To further address the specificity of these RNAi oligos, we examined the effect of these oligos on thapsigargin-induced NFAT-bla reporter activity. Thapsigargin works downstream of the Trk receptors by directly increasing intracellular Ca²⁺ concentration leading to the activation of NFAT independent of Trk activity. Accordingly, specific Trk RNAi oligos should not knockdown thapsigargin induced reporter activity. Indeed, the two TrkA, TRKC2 and TRKB1 RNAi oligos did not inhibit thapsigargin effect in all three cell lines (Fig. 5). TRKB2 oligo and TRKC1 to a less extent, however, had about 50% inhibitory effect in all three cell lines stimulated with thapsigargin, suggesting that TRKB2 and TRKC1 may have effects on other targets that contribute to thapsigargin induced-signaling.

Fig. (5) Effects of various RNAi oligos on thapsigargin-induced beta-lactamase reporter activity. TrkA-NFAT-bla CHO-K1 (A), TrkB-NFAT-bla CHO-K1 (B) and TrkC-NFAT-bla CHO-K1 (C) cells were incubated with indicated RNAi oligos for 32 hrs, followed by a medium change and additional 16 hour-incubation. Cells were then stimulated with thapsigargin for 5 hours before beta-lactamase assay was performed as described in Methods. Response Ratios were plotted for each RNAi oligo (n=3 for each data point). Unstim: untransfected cells were left unstimulated with ligands; Mock: cells were treated with transfection reagent (no RNAi oligo) and stimulated with thapsigargin.