A Biosensor Assay for Detecting Activation of Endogenous EGFR in Living Cells

Biowire Spring 2011 — Cell Lines — Models of Disease


By Dmitry Malkov, John Fetter, Nathan Zenser, and Keming Song


A fluorescent biosensor that binds to activated epidermal growth factor receptor (EGFR) tyrosine kinase was developed for detecting activity in live cells. Importantly, no modification of the EGFR or the EGF ligand is required for detection of EGFR activity. This achievement represents a significant advance over current biochemical assays or immunostaining methods that detect end point phenotypes and do not detect activity in live cells. It also improves on existing methods for detecting activity in live cells that require modification of the endogenous EGFR or EGF ligand. The biosensor is stably expressed in the A549 lung carcinoma cell line and is capable of detecting EGFR activation. Collaboration with Dr. Hakim Djaballah at Memorial Sloan- Kettering Cancer Center demonstrated that the biosensor is suitable for high content screening in a 384-well plate format.

back to top Back to Top


Receptor tyrosine kinases (RTKs) are a subclass of cell-surface growth-factor receptors that dimerize and autophosphorylate upon ligand binding.1,2 Dimerization stimulates the intrinsic intracellular protein-tyrosine kinase activity leading to the initiation of signaling transduction cascades. These signaling pathways regulate normal cellular processes and have been shown to play a critical role in the development and progression of many types of cancer.3 EGFR, along with other well-characterized RTKs, triggers several transduction pathways resulting in DNA synthesis and cell proliferation.4 Aberrant overexpression or activation of the EGFR is known to play a major role in cancer genesis.5,6

Current cell-based assays for detection of EGFR activation and internalization are either based on overexpression of EGFR-GFP fusion protein, or on immunostaining using antibodies specific for phosphorylated EGFR. A major limitation of these methods is their inability to detect activation of the endogenous EGFR in living cells. This deficiency signifies that the current methods are not suitable for analysis of dynamic changes of EGFR activation in real time. To fill this gap, we have developed an assay that utilizes a genetically encoded biosensor to detect activated EGFR in live cells.

back to top Back to Top

Materials and Methods

Reagents: RPMI-1640 (R0883), L-glutamine (G7513), trypsin (T3924), puromycin (P9620), fetal bovine serum (F2442), hexadimethrine bromide (H9268), EGF (E9644), and Tyrphostin AG 1478 (T4182)

A549 lung carcinoma cells were grown in RPMI-1640, 2 mM glutamine, 10% FBS in 5% CO2, 37 °C. The biosensor was constructed by linking two tandem SH2 domains with a GFP tag at the 5’ end (Figure 1). For transient transfection experiments, the biosensor was transfected into A549 cells using the recommended protocol on the Amaxa® Nucleofector®. To make a stable cell line, the biosensor was packaged in lentiviral particles and transduced into A549 cells at an MOI of 5 in media containing 8 μg/ml hexadimethrine bromide. The cells were imaged with an automated Nikon TE2000 microscope and MetaMorph® was used for image analysis. (GFP: excitation 450–490 nm / emission 500–550 nm; tagFP635: excitation 565–595 nm / emission 610–650 nm; DRAQ5: excitation 618–673 nm / emission 685–735 nm; 40x/1.4 oil.)

Structure of domain-based biosensors 

Figure 1. Structure of domain-based biosensors.
A. Generic structure of biosensors that consist of a target-specific binding domain and a fluorescent protein. B. The structure of the EGFR biosensors, consisting of one or two SH2 domains from adapter protein Grb2 and a green fluorescent protein tag.

back to top Back to Top

Results and Discussion

Characterization of EGFR Biosensor
The EGFR biosensor was transiently transfected into A549 cells expressing endogenous EGFR. Activation of the receptor was monitored in real time with an automated imaging system. As shown in Figure 2, within 3–5 minutes after EGF stimulation, the biosensor showed robust redistribution towards the cell membrane, then subsequent internalization (17–69 minutes) through endocytosis (Figure 2A). Imaging analysis was used to examine both steps of the movement of the biosensor. Analysis showed that, as the biosensor translocated to the plasma membrane, a decrease in fluorescence occurred rapidly in the perinuclear region and slowly in the nucleus. Granularity analysis showed an increase in the number of particles in the cell as the biosensor was internalized (Figure 2B). The assay specificity was assessed using a known EGFR inhibitor and a panel of RTK ligands. The results showed that the EGFR response is highly selective, as it could be induced only by specific EGFR ligands and was abolished by a selective inhibitor of EGFR, Tyrphostin AG 1478 (Figure 3).

Characterization of the EGFR biosensor 

Figure 2. Characterization of the EGFR biosensor.
A. EGF causes biosensor translocation to the plasma membrane followed by internalization. The nucleus was labeled with 1 μM DRAQ5.
B. The redistribution of the biosensor over time was quantified using translocation and granularity analysis in MetaMorph.

Selectivity and specificity of the biosensor 

Figure 3. Selectivity and specificity of the biosensor.
A. Tyrphostin AG 1478, a selective inhibitor of EGFR, blocked translocation of the biosensor to the plasma membrane and subsequent internalization. Heregulin-β1, a ligand for ErbB3 and ErbB4, did not show activity. HGF that binds HGFR also showed activity, but much less than EGF.
B. A panel of ligands to receptor tyrosine kinases were tested for activity with the biosensor. Strong activities were only seen with EGFR-specific ligands, EGF, and TGF-. Gas6, IGF-1, and Heregulin-β1 were tested at 1 g/ml. Insulin was tested at 2 g/ml. The other ligands were tested at 100 ng/ml.

Development of a Stable Cell Line for High-Content Screening
Lentiviral particles harboring the EGFR biosensor construct were used to transduce A549 cells, and several stable cell lines were selected based on expression levels of the EGFR biosensor. One of these cell lines was selected for further characterization. This stable cell line gave a similar but more uniform reporter response to ligand stimulation compared to that of transient expression (Figure 4). In collaboration with Dr. Hakim Djaballah’s group at Memorial Sloan- Kettering Cancer Center in New York, we assessed the feasibility of using this stable cell line for drug discovery. A high-content cell-based assay was successfully developed and miniaturized in 384-well microtiter plates. The assay was further validated in a pilot screen against a compound library containing pharmaceutically active compounds. The assay performance was deemed exceptional, in that a very low hit rate was observed with nearly all of the EGFR inhibitors present in the library identified as actives. The screen also identified novel modulators of the EGFR trafficking, the activity of which is currently being investigated.

Development of a stable cell line 

Figure 4. Development of a stable cell line.
Lentiviral particles containing the EGFR biosensor construct were used to transfect A549 cells. Single cell clones were selected and assayed for activity with EGF. This clone shows homogenous response to translocation and internalization upon EGF stimulation. 20x/0.75 objective, [EGF] = 100 ng/ml.

Ratiometric Imaging with the EGFR Biosensor
To improve the signal-to-noise ratio, a novel construct was created that expresses an RFP-labeled biosensor linked to GFP by a 2A peptide. Upon translation, the GFP domain is separated from the RFP-labeled biosensor and serves as a volume marker for ratiometric measurement (Figure 5A). The ratio of biosensor fluorescence over GFP enables improved detection of biosensor accumulation at the plasma membrane following stimulation with EGF (Figure 5B). This new biosensor enables researchers to obtain more accurate measurement of EGFR response with transiently transfected cells, and thus has a broad utility for studying EGFR activation in different types of cells, including primary cells.

Ratiometric imaging improves detection 

Figure 5. Development of a stable cell line.
A. The EGFR-specific binder domain was fused with RFP and linked to GFP through the 2A sequence. The translational skip at the 2A site resulted in expression of two separate proteins: GFP as a volume marker and RFP-B as the sensor.
B. The fluorescence signal from the RFP-labeled biosensor was divided by the GFP signal in MetaMorph®. This removes the effect of changes in cell thickness on the fluorescence intensity of the sensor and improves the sensitivity at the thinner periphery of the cell.

The data in this article was presented at the HCS 2011 conference. The data from Dr. Djaballah’s group was not presented in this Biowire article and will be published in a separate article.

back to top Back to Top


We describe the development of a biosensor derived from an SH2-domain protein specific for EGFR, and the characterization of its use for quantitation of EGFR activity. Upon activation with EGF in live A549 cells, the EGFR biosensor showed translocation to the receptor at the plasma membrane followed by internalization. The biosensor activity could be inhibited by a specific EGFR inhibitor and is highly specific for EGF ligand stimulation. A stable cell line constitutively expressing the biosensor was created and validated in an HCS laboratory using more than 1,000 pharmaceutically active compounds in a 384-well plate format. A dual reporter approach allowed an RFP-biosensor to be normalized to a background GFP, giving more robust data and a broader application. Our results demonstrate that this live cell biosensor-based assay provides a novel research tool to study the kinetics of endogenous EGFR activation and internalization. Compared to current assays on the market, this assay should be of great value in pharmaceutical screening for modulators of wild-type EGFR and its mutants.

back to top Back to Top


  1. Pawson T. Regulation and targets of receptor tyrosine kinases. Eur J Cancer. 2002;38(Suppl5):S3–10.
  2. Li E, Hristova K. Role of receptor tyrosine kinase transmembrane domains in cell signaling and human pathologies. Biochemistry. 2006;45:6241–51.
  3. Zwick E, et al. Receptor tyrosine kinase signalling as a target for cancer intervention strategies. Endocr Relat Cancer. 2001;8:161–73.
  4. Oda K, et al. A comprehensive pathway map of epidermal growth factor receptor signaling. Mol Syst Biol. 2005;1:2005.0010.
  5. Wang SC, Hung MC. Nuclear translocation of the epidermal growth factor receptor family membrane tyrosine kinase receptors. Clin Cancer Res. 2009;15:6484–89.
  6. Lemmon MA, Schlessinger J. Cell signaling by receptor tyrosine kinases. Cell. 2010;141:1117–34.



back to top Back to Top

Related Links