High-Throughput High-Content Imaging Assays for Identification and Characterization of Selective AXL Pathway Inhibitors

Abstract

Receptor tyrosine kinases (RTKs) regulate a wide range of important biological activities, including cell proliferation, differentiation, migration, and apoptosis. Abnormalities in RTKs are involved in numerous diseases, including cancer and other proliferative disorders. AXL belongs to the TAM (Tyso3, AXL, and Mer) family of RTKs. The AXL signaling pathway represents an attractive target for the treatment of diseases, such as cancer. Using phospho-AKT as readout, a high-throughput 384-well cell-based assay was established in the NCI-H1299 human non–small cell lung carcinoma cell line to evaluate compound potency in inhibiting AXL pathway activation. In addition, a counter screen assay was established in the same cellular background to differentiate AXL kinase inhibitors from AXL receptor antagonists, which block the interaction of AXL and its natural ligand GAS6. These cell-based functional assays are useful tools in the identification and optimization of small molecules and biological reagents for potential therapeutics for the treatment of GAS6/AXL-related diseases.

Introduction

Receptor tyrosine kinases (RTKs) are a large family of transmembrane proteins that contain a highly conserved intracellular tyrosine kinase domain.¹ Ligand binding of the extracellular domain triggers intracellular signaling by the activation of the tyrosine kinase domain and its subsequent phosphorylation of a series of substrates.¹ RTKs are major effectors of a diverse range of biological activities, such as cell proliferation, differentiation, migration, and apoptosis.¹

The TAM family of RTKs is composed of three closely related members: Tyso-3, AXL, and Mer.² TAM activation and signaling have been implicated in a variety of cellular functions, including growth arrest, survival, proliferation, adhesion, and inflammation.² TAM receptors have been demonstrated to be involved in the regulation of multiple aspects of tumorigenesis, control of oligodendrocyte cell survival, and regulation of osteoblast function.² In addition, TAM signaling has been shown to play essential roles in innate immunity, where they regulate the inhibition of inflammation in dendritic cells and macrophages, stimulate differentiation of natural killer cells, and promote clearance of apoptotic cells.³ TAMs regulate most of these biological activities through the PI3K/AKT pathway.⁴

AXL was originally identified from patients with chronic myelogeneous leukemia and a chronic myeloproliferative disorder.^5,6 Recently, AXL has been shown to be a potential therapeutic target of cancer.⁷ The extracellular region of the AXL receptor consists of two immunoglobin-related domains and two fibronectin type III repeats, followed by a single transmembrane domain and an intracellular catalytic kinase domain.⁸ The natural ligands for AXL include two very similar vitamin K-dependent proteins, GAS6 and protein S, which exhibit 42%–43% sequence identity. Each protein has an N-terminal Gla domain consisting of 11 γ-carboxyglutamic acid residues and four epidermal growth factor-like modules, followed by a C-terminal sex hormone-binding globulin-like structure that contains two tandem laminin G domains.⁹

The GAS6/AXL system represents an attractive target for the treatment of cancer. Both small molecule kinase inhibitors and biological reagents that block the binding of GAS6 to the AXL receptor have been reported at various stages of drug development.¹⁰ Given its involvement in other diseases, such as autoimmune disease and thrombosis, this pathway may emerge as a target for treating these diseases as well.^3,11 However, it has been a challenge to develop a selective and high-throughput cell-based assay to monitor the activation of the GAS6/AXL signaling pathway. All reported assays are in 96-well format and utilize technologies, such as ELISA or In-Cell Western, with limited throughput.^12
–14 Using the NCI-H1299 human non–small cell lung carcinoma cell line and a phospho-AKT readout, a 384-well high-content imaging cell-based assay was established to assess compound potency in inhibiting GAS6-induced AXL pathway activation. In addition, the novelty of our approach was the establishment of a high-throughput counter screen assay to differentiate AXL kinase inhibitors from AXL receptor antagonists, which block the interaction of GAS6 and AXL. These tools will be useful in the screening of small molecules and biological reagents for potential therapeutics for the treatment of GAS6/AXL-driven diseases.

Materials and Methods

Materials

The human non–small cell lung carcinoma cell line H1299 and the human glioblastoma cell line A172 were obtained from ATCC (Manassas, VA). RPMI 1640 medium, DMEM (high glucose), penicillin–streptomycin (Pen/Strep), and 0.05% trypsin were from Gibco (Carlsbad, CA). Heat-inactivated fetal bovine serum (FBS) was purchased from Hyclone (South Logan, UT). pAKT (S437) antibody and goat anti-rabbit Alexa488 antibody were purchased from Cell Signaling Technology (Danvers, MA). MAB154 anti-AXL monoclonal antibody was from R&D Systems (Minneapolis, MN).

hAXL Ig1 Domain Preparation

His-Tb-hAxl (E26-G130) was expressed in Escherichia coli Shuffle T7 Express lysY competent cells (M9 minimal media+CAA/20 h induction at 16°C) and purified by Ni affinity chromatography followed by size exclusion chromatography (S75). Purified fusion protein was cleaved with Human Alpha Thrombin (Enzyme Research Lab, South Bend, IN), quenched with 1 mM PMSF, and further purified on size exclusion chromatography (S75). The final peptide has amino acids of E26-G130 of the human AXL protein, with four additional amino acids (GSHM) at the N-terminus.

Cell Culture

H1299 cells were grown in RPMI with 10% heat-inactivated FBS plus 1% Pen/Strep at 37°C with 5% CO₂. A172 cells were grown in DMEM (high glucose) with 10% heat-inactivated FBS plus 1% Pen/Strep at 37°C with 5% CO₂. Both cell lines were split every 3–4 days using 0.05% Trypsin-EDTA.

High-Content Imaging Assays

H1299 or A172 cells were plated in a black-walled, clear bottom, 384-well tissue culture plate (BD Biosciences, San Jose, California; Cat# 353962) at the density of 1,500 cells/well in 50 μL of growth medium and incubated at 37°C with 5% CO₂ overnight (Table 1). Cells were serum-starved in 45 μL of serum-free medium with 100 nL of compound prediluted in DMSO, where applicable, for 2 h. The final assay concentration of DMSO was 0.2%. Five microliters of either GAS6 protein (4 μg/mL) or MAB154 anti-AXL antibody (2 μg/mL) in serum-free medium were added to activate the AXL signaling pathway for 45 min at 37°C with 5% CO₂. Cells were then fixed with 4% formaldehyde at room temperature for 30 min. Fixed cells were permeabilized by incubating with 20 μL of permeabilization buffer (0.2% Triton X-100 in Dulbecco's phosphate-buffered saline [DPBS]) for 10 min at room temperature, blocked by incubation in blocking buffer (0.1% BSA in DPBS) for 30 min at room temperature, and stained by primary anti-pAKT antibody (1:300 in blocking buffer) at 4°C overnight. Cells were then washed three times with DPBS and then stained by the secondary antibody (1:1,000 in blocking buffer) together with 5 μg/mL of Hoechst 33342 for 2 h at room temperature. The Cellomics ArrayScan™ VTI system (Cellomics, Pittsburgh, PA) was used to acquire and analyze images using the Cell Health Profiling BioApplication. The XF100-Hoechst [365(50)/535(45)]nm and XF100-FITC [475(40)/535(45)]nm filters were used with an X-Cite^® 120Q excitation light source (Lumen Dynamics, Mississauga, Ontario, Canada) to collect images of the Hoechst and fluorescein isothiocyanate (FITC) channels, respectively, by an ORCA-ER camera and a 10× objective. Images of at least 500 cells were collected in each well. Images in the Hoechst channel were subjects to identify objects with fixed threshold. A circle was drawn 40 pixels outside the object boarder, and the total intensities of the FITC channel inside the circle were used to plot curves.

Table 1.

H1299 Cell AXL Pathway pAKT High-Content Assay Protocol

Step	Parameter	Value	Description
1	Plate cells	50 μL	1,500 cells in the growth medium
2	Incubation	Overnight	37°C with 5% CO₂
3	Serum starvation	2 h	In 45 μL serum-free medium
4	Dose compounds	100 nL	Preserially diluted in DMSO, 2 h preincubation during serum starvation
5	Add stimuli	5 μL	GAS6 protein or anti-AXL antibody in serum-free medium
6	Incubation	45 min	37°C with 5% CO₂
7	Fix	30 min	Fixing buffer
8	Permeabilization	10 min	Permeabilization buffer
9	Block	30 min	Blocking buffer
10	Primary antibody	Overnight	Anti-pAKT antibody
11	Wash	Three times	DPBS
12	Secondary antibody	2 h	Goat anti-rabbit FTIC with Hoechst 33342
13	Wash	Three times	DPBS
14	Image acquisition and data analysis	[365(50)/535(45)] nm; [475(40)/535(45)] nm	Hoechst and FITC channels

Step Notes

1. Black-walled, clear bottom, 384-well tissue culture plate.

2. Humidified incubator.

3. In humidified 37°C incubator with 5% CO₂.

4. Top stock concentration is 10 mM; 3× serial dilution; final DMSO concentration is 0.2%.

5. At room temperature, the final GAS6 concentration is 400 ng/mL and the final MAB154 concentration is 200 ng/mL.

6. Humidified incubator.

7. 4% formaldehyde; 20 μL at room temperature.

8. 0.2% Triton X-100 in DPBS; 20 μL at room temperature.

9. 0.1% BSA in DPBS; 20 μL at room temperature.

10. 1:300 in blocking buffer; 20 μL at 4°C.

11. 50 μL each time.

12. 20 μL at room temperature; goat anti-rabbit FTIC (1:1,000) and 5 μg/mL of Hoechst 33342 in blocking buffer.

13. 50 μL each time.

14. The Cellomics ArrayScan VTI system using the Cell Health Profiling BioApplication; 10× objective; images of at least 500 cells were collected in each well. A circle was drawn 40 pixels outside the object boarder, and the total intensities of the FITC channel inside the circle were used to plot curves.

DMSO, dimethyl sulfoxide, DPBS, Dulbecco's phosphate-buffered saline; FITC, fluorescein isothiocyanate.

Data Analysis

The positive control (0% inhibition) consists of the average values in assay wells containing cells stimulated with either 400 ng/mL GAS6 or 200 ng/mL MAB154 in the presence of 0.2% DMSO. Unstimulated cells containing 0.2% DMSO were used as negative control or 100% inhibition. Results were plotted as percent inhibition of the average positive control signal. The IC₅₀ was defined as the concentration of test compound corresponding to 50% inhibition derived from the 11-point fitted curve as determined using a four-parameter logistic regression model.

Results and Discussion

H1299 Cells Showed Robust AXL Pathway Activation Upon GAS6 Stimulation

The GAS6/AXL signaling pathway has emerged as a target for treating multiple diseases. For drug discovery efforts, it would be essential to have a robust cell-based assay to monitor the activation of the AXL signaling pathway in a relevant cellular background. AKT is a downstream component in the GAS6/AXL pathway, and its phosphorylation can be used as an indication of AXL pathway activation. To examine AKT phosphorylation, a high-content imaging analysis approach was used as a cost-effective detection method easily adaptable to established robotic platform infrastructure. High-content imaging also offers the capability to examine compound toxicity and general cell health in the same experiment. In an effort to design a high-throughput AXL activation assay, several cell lines were evaluated that are known to have high levels of AXL receptor expression, including MDA-MB-231, RKO-PM, HeLa, and HUVEC. However, these cells displayed high background pAKT levels, potentially due to constitutive phosphorylation of the AXL receptor (data not shown). To avoid this problem, cell lines needed to be identified that have high levels of AXL protein but low constitutive AXL phosphorylation. Li et al. measured the expression level of AXL as well as its phosphorylation status with or without GAS6 stimulation in 11 cancer cell lines. All 11 cell lines had high levels of total AXL and significant constitutive pAXL. However, several cell lines showed significantly elevated pAXL level after GAS6 stimulation, including H1299 and A172.¹⁵ Using high-content imaging analysis, the pAKT levels in H1299 cells were found to increase by fourfold after a 15 min treatment of 500 ng/mL GAS6 (Fig. 1A–C). In contrast, the pAKT levels in A172 cells were only slightly increased (∼1.5-fold) after GAS6 stimulation (Fig. 1C). Therefore, H1299 cells were chosen for GAS6/AXL cellular assay development.

Fig. 1.

H1299 cells show robust AXL pathway activation upon GAS6 stimulation. (A) Representative images of pAKT in H1299 cells with and without GAS6 stimulation. (B) Representative image analysis. Blue circles are nuclei. Yellow circle are 40 pixels away from the blue circle or the middle line between cells. The intensity of the pixels inside the yellow circles is used for data analysis. Scale bars are 100 μm. (C) Comparison of selective pAKT activation in H1299 and A172 cells. Cells were seeded at 2.5×10³ per well in a 384-well assay plate for overnight incubation, treated with (□) or without (■) 500 ng/mL GAS6 for 15 min stimulation and fixed by formaldehyde. pAKT was stained with an anti-pAKT primary antibody and a fluorescein isothiocyanate (FITC)-conjugated secondary antibody. Images were taken and analyzed by the Cellomics ArrayScan VTI system. (D) Effect of serum starvation on pAKT activation in H1299 cells. The experiment was performed as in (B) with or without 2 h of serum starvation before GAS6 induction. Results are shown as means of three wells, and error bars are standard deviations. Three independent experiments were conducted, of which the representative results are shown (*P<0.05).

Although a large assay window was obtained, the background pAKT levels in H1299 cells still remained high. Serum starvation can typically reduce constitutive PI3K/AKT pathway activation. Indeed, serum starvation of H1299 cells for 2 h before GAS6 stimulation significantly reduced background pAKT levels, whereas GAS6-induced pAKT induction was not affected, which helped to increase the signal to background window to >20-fold (Fig. 1D).

Characterization of GAS6-Induced AXL Pathway Activation in H1299 Cells

To characterize the kinetics of the GAS6-stimulated AXL pathway activation, a GAS6 induction time course analysis as well as a GAS6 concentration titration was conducted. In the GAS6-induced pAKT activation time course study, pAKT levels in H1299 cells start to increase at 4 min after the addition of 500 ng/mL GAS6, plateau at around 16 min, and are stable to at least 60 min (Fig. 2A). Based on these data, a 45-min stimulation was used for the ensuing experiments. In the GAS6 concentration titration experiment, GAS6 induces AKT phosphorylation in a dose-dependent manner (Fig. 2B).

Fig. 2.

Characterization of GAS6-induced AXL pathway activation in H1299 cells. (A) Time course analysis of AXL pathway activation by GAS6 in H1299 cells. H1299 cells were seeded overnight at 1×10³ per well in a 384-well plate and fixed after 500 ng/mL GAS6 stimulation. pAKT was stained with an anti-pAKT specific primary antibody and a FITC-conjugated secondary antibody. Images were taken and analyzed by the Cellomics ArrayScan VTI system. (B) GAS6 ligand dose response for pAKT activation in H1299 cells. Serum-starved cells were fixed after 45 min incubation with designated concentration of GAS6 stimulation. Results are representative of at least three independent experiments, and the bar graphs are mean values of triplicate wells with error bars representing standard deviations.

AXL Antibody MAB154 Induces AXL Pathway Activation in H1299 Cells

The GAS/AXL activation assay described above allows the evaluation of compound inhibition along the AXL pathway from receptor binding to AKT activation. This assay can identify compounds that bind to GAS6 and disrupt its interaction with AXL (i.e., AXL receptor antagonists), as well as compounds that bind to the AXL receptor and block ligand binding or inhibit kinase activity of AXL (i.e., AXL kinase inhibitors). However, this assay is not able to differentiate AXL receptor antagonists from AXL kinase inhibitors.

To differentiate the two, a counter screen assay was established using a cross-linking monoclonal antibody against AXL, MAB154 from R&D Systems. A time course study revealed fast kinetics of MAB154 AKT phosphorylation starting at 2 min with a plateau at 4 min (Fig. 3A). The pAKT level was sustained until at least 45 min after the start of stimulation (Fig. 3A). In a concentration titration experiment, MAB154-induced pAKT in a clear dose-dependent manner (Fig. 3B).

Fig. 3.

Counter screen assay of AXL antibody MAB154-induced AXL pathway activation in H1299 cells. (A) Time course analysis of pAKT activation by MAB154 in H1299 cells. H1299 cells were seeded overnight at 1×10³ per well in a 384-well plate and fixed after 200 ng/mL MAB154 stimulation. pAKT was stained with an anti-pAKT-specific primary antibody and a FITC-conjugated secondary antibody. Images were taken and analyzed in the Cellomics ArrayScan VTI system. (B) MAB154 agonist dose response for AXL pathway activation in H1299 cells. Serum-starved cells were fixed after 45 min of incubation with designated concentration of MAB154. Results are representative of at least three independent experiments. Data presented are shown as means of triplicate wells and error bars are standard deviations.

Performance Assessment of Optimized Assays

To measure the DMSO tolerance of the assays, H1299 cells were stimulated for 45 min with 400 ng/mL of GAS6, 200 ng/mL of MAB154, or assay medium alone with the presence of different concentrations of DMSO. pAKT levels were measured by high-content imaging after the cells were fixed. No significant differences in pAKT levels were observed in either unstimulated or stimulated cells with DMSO concentrations up to 1% (Fig. 4A). 0.2% DMSO was therefore used in our assays, which does not affect assay results.

Fig. 4.

Performance assessment of optimized assays. (A) H1299 cells were stimulated with 400 ng/mL of GAS6, 200 ng/mL of MAB154, or assay medium alone for 45 min with the presence of different concentrations of dimethyl sulfoxide (DMSO). No significant differences in pAKT levels were observed with up to 1% of DMSO (P>0.05). (B, C) Cells in every other well in a 384-well plate were stimulated by either 400 ng/mL of GAS6 or 200 ng/mL of MAB154. pAKT levels were measured in all wells. Signal to background (S/B) window was calculated as the ratio of the average values of the signals of the stimulated wells and the unstimulated wells. Z′ factor was calculated as described before.¹⁶ Results are representative of at least three independent experiments, and the bar graphs are mean values of triplicate wells with error bars representing standard deviations.

To assess the quality of the pAKT high-content assays, the Z′ factor was determined for the two assays.¹⁶ In this experiment, half of the wells in a 384-well plate were stimulated by either 400 ng/mL of GAS6 or 200 ng/mL of MAB154, and the other half of the wells were mock stimulated with assay buffer alone. The stimulation was performed in a way that every other well is stimulated, so that bias in signals across the whole plate was minimized. The GAS6 assay yielded a 28-fold assay window and a Z′ factor of 0.7, and the MAB154 assay had a 36-fold window and a Z′ factor of 0.6 (Fig. 4B, C). These results indicate that the well-to-well intra-plate variability is small and confirmed the robustness of the two assays.

Screening for Selective AXL Receptor Antagonists Via Counter Screen Against MAB154 AXL Antibody-Induced pAKT Activation in H1299 Cells

To confirm that GAS6 and MAB154 induce AKT activation through the AXL receptor in H1299 cells, a small molecule AXL kinase inhibitor was used. The IC₅₀ of this compound in an in-house AXL biochemical kinase assay is 1.1±0.3 nM. At a concentration of 100 nM, this compound was able to completely inhibit AKT phosphorylation stimulated by 400 ng/mL of GAS6 or 200 ng/mL of MAB154 (Fig. 5).

Fig. 5.

Representative images of pAKT levels under different conditions. H1299 cells were stimulated with 400 ng/mL of GAS6, 200 ng/mL of MAB154, or assay medium alone for 45 min with or without the presence of 100 nM AXL kinase inhibitor. Images of the Hoechst and FITC channels were collected by an ORCA-ER camera with a 10× objective. Scale bar is 100 μm.

The establishment of the GAS6-induced pAKT assay together with the MAB154-induced pAKT assay permits functional screening of selective AXL receptor antagonists versus AXL kinase inhibitors (Fig. 6A). To identify the best GAS6 concentration for an inhibition assay, the small molecule AXL kinase inhibitor was tested at different GAS6 concentrations (Fig. 6B). The IC₅₀ of the AXL kinase inhibitor remained similar across the GAS6 titration range and a final concentration of 400 ng/mL was used in ensuing experiments to produce a robust assay window (>20-fold). The AXL kinase inhibitor was also tested with different MAB154 concentrations. At 200 ng/mL of MAB154, we obtained a robust assay window and an IC₅₀ that is consistent with that of 400 ng/mL of GAS6 (Fig. 6C). We therefore used 200 ng/mL of MAB154 antibody in subsequent experiments for the counter screen assay. Under these conditions, GAS6 and MAB154 robustly induced AKT phosphorylation, which is inhibited by the AXL kinase inhibitor (Fig. 5).

Fig. 6.

Screening for selective AXL antagonists via counter screen against MAB154 AXL antibody-induced AXL pathway activation in H1299 cells (A) Schematic depiction of the GAS6-induced AXL pathway activation assay (left) and the MAB154-induced AXL pathway activation counter screen assay (right). Although small molecule AXL kinase inhibitors (⎔) can block the activation of pAKT induced by both GAS6 ligand and the cross-linking antibody MAB154, AXL antagonists, such as soluble AXL-Ig1 (⊂), are only able to inhibit GAS6-induced AKT activation. (B) The IC₅₀ of the AXL kinase inhibitor was measured at different concentrations of GAS6. Similar IC₅₀ values were obtained at 400, 800, or 1,600 ng/mL of GAS6. (C) Concentration–response curves showing dose-dependent inhibition of GAS6- or MAB154-induced pAKT activation by an AXL kinase inhibitor versus Axl-Ig1 receptor antagonist. Y-axis is percentage inhibition compared to controls (0% inhibition: 0.2% DMSO; 100% inhibition: 100 nM of the AXL inhibitor with 0.2% DMSO). The AXL inhibitor is potent in both assays with nanomolar IC₅₀s, whereas soluble AXL-Ig1 is active only in the GAS6 assay, due to its lack of binding to the MAB154 antibody. IC₅₀ values are mean±one standard deviation with n=3. Results in this figure are representative of at least three independent experiments with duplicate wells.

As a proof-of-concept, we tested the AXL kinase inhibitor and the Ig1 domain of AXL in both assays. The AXL kinase inhibitor should inhibit both GAS6-induced as well as MAB154-induced AKT activation, as illustrated in Figure 6A. Indeed, the AXL kinase inhibitor displayed single-digit nM potency in both assays (Fig. 6C). The AXL Ig1 domain binds to the GAS6 ligand and disrupts the GAS6/AXL interaction to inhibit pathway activation as an AXL antagonist.⁸ As expected, AXL-Ig1 is a potent inhibitor of GAS6-induced AKT activation with an IC₅₀ of 6.4±2.3 nM (Fig. 6C). In contrast, the AXL Ig1 domain cannot disrupt MAB154 binding and subsequent AXL pathway activation because the antibody was raised against an epitope outside the Ig1 domain. Consistent with our hypothesis, AXL-Ig1 is not active in the MAB154-induced pAKT assay (Fig. 6C). These results suggest that the GAS6-induced pAKT assay can be used to evaluate compound potency in inhibiting the AXL signaling pathway in a H1299 cellular background. The MAB154-induced pAKT assay can further distinguish AXL receptor antagonists from AXL kinase inhibitors.

Conclusion

The GAS6-AXL signaling pathway has emerged as a target to treat diseases, including cancer, immune disorders, and cardiovascular diseases. It is essential to establish a robust system to evaluate compound potency in a cellular background for drug discovery efforts. The H1299 cell line was identified as a good model system for a cell-based AXL signaling pathway activation assay. We focused not only on the AXL expression level in various cell lines but also on their AKT phosphorylation status and its change before and after GAS6 stimulation, which was the key factor leading to the successful identification of the H1299 cell line for the assay.

The GAS6-induced pAKT assay in H1299 cells can identify inhibitors of GAS6-induced AKT activation. However, the assay cannot distinguish between compounds that block GAS6/AXL binding (i.e., AXL receptor antagonists) and AXL kinase inhibitors. To differentiate compounds, a counter screen assay was established using the monoclonal antibody MAB154 against AXL for induction. This counter screen assay can only identify AXL kinase inhibitors but not AXL receptor antagonists (i.e., compounds that block the binding of GAS6 and AXL). As a proof-of-concept, we tested an AXL kinase inhibitor and soluble AXL-Ig1 in both assays. The AXL kinase inhibitor inhibited both assays, whereas the soluble AXL-Ig1 only inhibited the GAS6-induced pAKT but not the counter screen assay for selectivity. These data demonstrate the utility of the assays.

One limitation of these assays is that pAKT is not a direct phosphorylation target of AXL kinase. As a result, one cannot differentiate compounds inhibiting AXL and compounds inhibiting any signaling components between AXL activation and AKT phosphorylation. A pAXL readout would help to distinguish the two. Although there are a few pAXL antibodies commercially available, these antibodies had high background staining in our high-content assays. This high background was probably due to antibody cross-reactivity against other proteins instead of high levels of constitutive pAXL levels because the pAKT levels were low when measured in parallel (data not shown).

In summary, a robust, high-throughput high-content imaging assay was established to assess the inhibition of GAS6-induced AXL pathway activation in 384-well format. Furthermore, a counter screen assay using an MAB154 AXL cross-linking antibody enabled the differentiation of AXL receptor antagonists from AXL kinase inhibitors. These tools enable the screening and optimization of small molecules as well as biological reagents as therapeutics for the treatment of GAS6/AXL-driven diseases.

Footnotes

Acknowledgments

The authors thank Liang Schweizer, Ming Lei, Lisa Elkins, Margaret Prack, and Jennifer Qiao for stimulating discussions. We are grateful to John Hunt for critical review of this article.

Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Lemmon

, Schlessinger

: Cell signaling by receptor tyrosine kinases. Cell, 2010; 141:1117–1134.

Linger

, Keating

, Earp

, Graham

: TAM receptor tyrosine kinases: biologic functions, signaling, and potential therapeutic targeting in human cancer. Adv Cancer Res, 2008; 100:35–83.

Lemke

, Rothlin

: Immunobiology of the TAM receptors. Nat Rev Immunol, 2008; 8:327–336.

Fernandez-Fernandez

, Bellido-Martin

, Garcia de Frutos

: Growth arrest-specific gene 6 (GAS6). An outline of its role in haemostasis and inflammation. Thromb Haemost, 2008; 100:604–610.

Liu

, Hjelle

, Bishop

: Transforming genes in chronic myelogenous leukemia. Proc Natl Acad Sci USA, 1988; 85:1952–1956.

Janssen

, Schulz

, Steenvoorden

, et al.: A novel putative tyrosine kinase receptor with oncogenic potential. Oncogene, 1991; 6:2113–2120.

Verma

, Warner

, Vankayalapati

, Bearss

, Sharma

: Targeting Axl and Mer kinases in cancer. Mol Cancer Ther, 2011; 10:1763–1773.

Sasaki

, Knyazev

, Clout

, et al.: Structural basis for Gas6-Axl signalling. Embo J, 2006; 25:80–87.

Stitt

, Conn

, Gore

, et al.: The anticoagulation factor protein S and its relative, Gas6, are ligands for the Tyro 3/Axl family of receptor tyrosine kinases. Cell, 1995; 80:661–670.

10.

Linger

, Keating

, Earp

, Graham

: Taking aim at Mer and Axl receptor tyrosine kinases as novel therapeutic targets in solid tumors. Expert Opin Ther Targets, 2010; 14:1073–1090.

11.

Maree

, Jneid

, Palacios

, Rosenfield

, MacRae

, Fitzgerald

: Growth arrest specific gene (GAS) 6 modulates platelet thrombus formation and vascular wall homeostasis and represents an attractive drug target. Curr Pharm Des, 2007; 13:2656–2661.

12.

Rushton

, Winter

JJG

: 2-Pyraz inylbenz imidazole derivatives as receptor tyrosine kinase inhibitors. U.S. Patent WO 2009/007390 A2. January 15, 2009.

13.

Goff

, Zhang

, Singh

, et al.: N3-heteroaryl substituted triazoles and n5-heteroaryl substituted triazoles useful as axl inhibitors. U.S. Patent WO 2008/083354 A1. July 10, 2008.

14.

Pave

, Firth

, Stewart

, Rigoreau

LJM

, Wynne

: [1,2,4]Triazolo[1,5-a]pyridine and [1,2,4]triazolo[1,5-c]pyrimidine compounds and their use. U.S. Patent WO 2009/047514 A1. April 16, 2009.

15.

, Ye

, Tan

, et al.: Axl as a potential therapeutic target in cancer: role of Axl in tumor growth, metastasis and angiogenesis. Oncogene, 2009; 28:3442–3455.

16.

Zhang

, Chung

, Oldenburg

: A simple statistical parameter for use in evaluation and validation of high throughput screening assays. J Biomol Screen, 1999; 4:67–73.