Lidocaine Inhibits Tyrosine Kinase Activity of the Epidermal Growth Factor Receptor and Suppresses Proliferation of Corneal Epithelial Cells

Antiphosphotyrosine antibody (4G10) and anti-EGFR rabbit polyclonal antibody were obtained from Upstate Biotechnology, Inc. (Lake Placid, NY). EGFR from human carcinoma A431 cells purified by affinity chromatography, and chemicals including lidocaine hydrochloride were from Sigma Chemical Co. (St. Louis, MO).

Human corneal epithelial cells were generously provided from Dr. Kaoru Sasaki (Department of Ophthalmology, Osaka University Medical School, Osaka, Japan) 13and were maintained with Dulbecco’s modified Eagle’s medium (GIBCO, Grand Island, NY) containing 10% (vol/vol) fetal bovine serum, penicillin, streptomycin, choleratoxin, insulin, and EGF. The cells were grown in 25-cm²culture flasks, and the medium was changed every other day.

Purified EGFR (1 μg protein) was phosphorylated with 0.2 mm adenosine triphosphate for 5 min at 37°C in a 50-μl incubation buffer (50 mm HEPES, pH 7.4, 125 mm NaCl, 1 mm EDTA, 10 mm MgCl², 5 mm MnCl², 5 mm dithiothreitol, and 1 mm phenylmethylsulfonyl fluoride). EGF (100 ng/ml), lidocaine (4 μM, 40 μM, 400 μM, 4 mM, 40 mM), or both were also added before incubation. In an in vitro  study using purified EGFR, 100 ng/ml EGF was an appropriate concentration to stimulate EGFR. 14After the incubation, the samples were prepared for sodium dodecyl sulfate-polyacrylamide gel electrophoresis by adding Laemmli sample buffer (Bio-Rad Laboratories, Hercules, CA) and boiling for 5 min, followed by Western blot analysis with antiphosphotyrosine antibody.

The cells were detached with a 0.25% trypsin/1 mm EDTA-4Na solution for 20 min. Subsequently, the cells were seeded onto 96-well plates (1 × 10³cells/well), and test compounds were added to the wells (100 μl in total for each well). Control medium contained Dulbecco’s modified Eagle’s medium, 1% (vol/vol) fetal bovine serum, penicillin, streptomycin, choleratoxin, and insulin. Test compounds consisted of the control medium and drugs of various concentrations (10 ng/ml EGF and 4 μM, 40 μM, 400 μM, 4 mM, and 40 mm lidocaine). Eight wells each were used for control and each drug concentration. Previous studies showed that 4 –40 ng/ml EGF was appropriate to induce cell proliferation and tyrosine phosphorylation in a cell culture study using corneal epithelial cells. 15,16 

After incubation in a 37° C, 5% CO²environment for 5 days, the number of cells in each well was measured colorimetrically by CellTiter 96^®AQueous One Solution Cell Proliferation Assay (Promega Corporation, Madison, WI) according to the manufacturer’s protocol. Briefly, 20 μl CellTiter 96^®AQueous One Solution Reagent was added into each well. After 1 h of incubation, the absorbance of each well at 492 nm was recorded by a 96-well plate reader (Multiskan BICHROMATIC; Labsystems, Helsinki, Finland). After subtraction of the background absorbance (i.e. , the average absorbance of wells containing medium without any cells), the ratios of the absorbance can be interpreted as those of cell number of each well.

The cells were cultured on dishes to be confluent. Media were removed, and test compounds (10 ng/ml EGF; 4 μM, 40 μM, 400 μM, 4 mM, or 40 mm lidocaine; or both) were added (5 ml each). After incubation in a 37°C, 5% CO²environment for 5 min, the compounds were replaced with 5 ml phosphate-buffered saline. Each sample was scraped and harvested into the tubes. The supernatant was removed by centrifugation at 1,500 rpm for 5 min. Each sample was suspended in 0.6 ml lysis buffer (50 mm HEPES, pH 7.5, 150 mm NaCl, 2 mm EDTA, 1%[vol/vol] Nonidet P-40 [Nacalai tesque, Kyoto, Japan], 10%[vol/vol] glycerol, 10 mm sodium fluoride, 2 mm sodium vanadate, 1 mm phenylmethylsulfonyl fluoride, 10 mm sodium pyrophosphate, 5 μg/ml aprotinin, 0.5 μg/ml pepstatin) and laid on ice for 30 min.

Insoluble material was removed by centrifugation at 15,000 rpm for 15 min. Aliquots of the supernatants containing equal amounts of protein, as determined using the Bradford protein assay with Bradford reagent (Sigma Chemical), were subjected to immunoprecipitation for 1 h at 4°C with anti-EGFR antibody. After the addition of protein A-Sepharose CL-4B (Pharmacia Biotech, Piscataway, NJ), the immunoprecipitates were washed three times in a wash buffer (50 mm HEPES, pH 7.5, 150 mm NaCl, 2 mm EDTA, 0.1%[vol/vol] Nonidet P-40, 10%[vol/vol] glycerol, 10 mm sodium fluoride, 2 mm sodium vanadate, 1 mm phenylmethylsulfonyl fluoride, 10 mm sodium pyrophosphate, 5 μg/ml aprotinin, 0.5 μg/ml pepstatin). The samples were prepared for sodium dodecyl sulfate-polyacrylamide gel electrophoresis by adding Laemmli sample buffer and boiling for 5 min.

The immunoprecipitates were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis in 7.5% (vol/vol) acrylamide solving gels and transferred electrophoretically to nitrocellulose membrane (Bio-Rad Laboratories). The membranes were then blocked in 5% (wt/vol) dried milk in phosphate-buffered saline containing 0.1% (vol/vol) polyoxymethylenesorbitan monolaurate (Tween 20; Sigma Chemical) for 1 h at room temperature and were then immunoblotted with appropriate antibody. The antigen antibody complexes were visualized by chemiluminescence luminol reagent (Santa Cruz Biotechnology, Santa Cruz, CA). Bands of interest were scanned and quantified by using LightCapture AE-6960 (ATTO Corporation, Tokyo, Japan).

Data were analyzed by one-way analysis of variance with Bonferroni-corrected post hoc  analysis. The statistical significance was established at the level of P < 0.05. All values are reported as mean ± SD.

Epidermal growth factor increased autophosphorylation of purified EGFR (fig. 1). EGF-stimulated responses of EGFR were taken as 100%. Unstimulated basal (absence of EGF) phosphorylation of EGFR was 20.8 ± 2.4% of the EGF-stimulated response. During EGF stimulation, lidocaine suppressed tyrosine phosphorylation of EGFR in a dose-dependent manner (73.8 ± 25.0, 69.7 ± 18.2, and 20.8 ± 10.1%, respectively, with 400 μM, 4 mM, and 40 mm lidocaine). These results suggest that lidocaine inhibited the EGF-stimulated tyrosine kinase activity of EGFR.

Figures 2 and 3show the results of the HCEC study. EGF alone greatly stimulated HCEC proliferation, and EGF-stimulated responses of the proliferation were taken as 100% (fig. 2). Lidocaine (400 μM) significantly suppressed EGF-stimulated HCEC proliferation (84.9 ± 8.2%). In both 4 and 40 mm lidocaine, HCECs showed greatly reduced survival.

To assess tyrosine phosphorylation of EGFR in HCECs, equal amounts of protein from HCECs were subjected to immunoprecipitation with anti-EGFR antibody followed by immunoblotting with antiphosphotyrosine antibody or anti-EGFR antibody. EGF stimulation resulted in a marked increase in tyrosine phosphorylation of EGFR in HCECs. EGF-stimulated responses of EGFR were taken as 100% (fig. 3). Lidocaine (400 μM, 4 mM, and 40 mM), however, significantly attenuated EGF-stimulated tyrosine phosphorylation of EGFR (44.6 ± 26.1, 21.4 ± 12.9, and 21.7 ± 10.9%, respectively) relative to the control level (100%). EGFR protein levels did not differ among every lane.

Epidermal growth factor exists as a constant component of human tear fluid, 17and both messenger RNA (mRNA) for EGF and mRNA for EGFR are present in HCECs. 18EGF is a potent mitogen for corneal epithelial proliferation. 7Therefore, the current study suggests that the attenuation of biologic functions of EGF by lidocaine is attributable to the inhibitory effect on the proliferation of HCECs.

Local anesthetic lidocaine is expected to interact with acidic, basic, and aromatic amino acids around several autophosphorylation sites in the EGFR by a variety combinations of noncovalent interactions, such as electrostatic, π-π stacking,19cation-π,20C-H-π,21and aromatic C-H···O hydrogen bonding22interactions. The π-π stacking provides interactions between the aromatic ring of lidocaine and aromatic amino acids (F and Y).19The aromatic ring of lidocaine can interact with the side chains of basic amino acids (K and R) through cation-π interaction.20Tertiary amine nitrogen of lidocaine interacts electrostatically with any of the negatively charged acidic amino acids (D and E). We observed that several autophosphorylation sites of tyrosine residues in EGFR are surrounded by acidic, basic, or aromatic amino acids. Moreover, lidocaine is an amphiphilic molecule and forms micelles. Therefore, it can be considered that lidocaine binds to tyrosine residues as a micelle by interacting with tyrosine itself and/or with acidic, basic, or aromatic amino acids by noncovalent interactions (table 1). Taking these facts together, we suggest that lidocaine inhibits EGF-stimulated tyrosine kinase activity of EGFR through the interaction with tyrosine residues themselves or with the residues residing around them in the autophosphorylation sites of the EGFR.

Kuroda et al.  8,9studied interactions between local anesthetics and sodium channel inactivation gate–related peptides, and reported that local anesthetics, dibucaine and lidocaine, interact with the aromatic ring of phenylalanine (F1489) and with the negatively charged amino acids (D1487, E1492) around the sodium channel inactivation gate (DIFMTE 1487–1492). On the other hand, etidocaine is known to bind with aromatic amino acids, phenylalanine (F1764), and tyrosine (Y1771) in the IVS6 segment facing the pore of the sodium channel. 23Therefore, in the previous study, we suggested that local anesthetics interact both with the inactivation gates and with the S6 segments in the sodium channel. 11We found the close resemblance between the amino acid sequence around the sodium channel inactivation gate and that around the autophosphorylation site of insulin receptor (table 1) 10,11and reported that lidocaine interacted with this site of insulin receptor. 10Although there are no homologous alignments in hypothetical binding sites for lidocaine among EGFR, sodium channel, and insulin receptor in table 1, lidocaine would bind to autophosphorylation sites of EGFR, which are surrounded by acidic, basic, and aromatic amino acids, with noncovalent interactions. Other local anesthetics, which are amines having aromatic rings, can also interact with these hypothetical binding sites.

Various kinds of growth factors, such as EGF, keratinocyte growth factor, insulin-like growth factor, fibroblast growth factor, transforming growth factor, hepatocyte growth factor, and platelet-derived growth factor, play a key role in corneal wound healing. 6,7,24Autophosphorylation sites of both keratinocyte growth factor and insulin-like growth factor receptors are surrounded by acidic, basic, or aromatic amino acids (EEY  ⁷⁷⁰L D  L in keratinocyte growth factor receptor, RD  I Y  ¹¹³¹E  T DY  ¹¹³⁵Y  ¹¹³⁶R  in insulin-like growth factor receptor). 25,26Insulin, which was contained in our medium for the HCEC study, may also play a mitogenic role on corneal epithelial cells through insulin receptor, 27and its autophosphorylation site is surrounded by these amino acids (RD  I Y  ¹¹⁵⁸E  T DY  ¹¹⁶²Y  ¹¹⁶³R ). We suggest that lidocaine interacts not only with EGFR but also with keratinocyte growth factor receptor and insulin-like growth factor receptor in addition to insulin receptor on the surface of HCECs (table 1). Moreover, the limitation of the current study is that we could not exclude the possible mechanisms of the inhibitory effect of lidocaine either upstream of the tyrosine phosphorylation sites of EGFR, such as the EGF/EGFR ligand binding site, or at downstream sites. Although autophosphorylation site of EGFR is an important target of lidocaine, other possible sites than EGFR would also play a role in the suppression of HCEC proliferation.

Clinical application of lidocaine, which ranges in concentration from 0.5 to 2% (approximately 20 – 80 mM), is associated with delay of corneal reepithelialization after corneal wounding. 1,2Instillation of these concentrations of lidocaine, however, does not provide a measurable steady state concentration because it is diluted and washed away. To evaluate the effect of steady state lidocaine concentrations on corneal epithelial wound healing, Bisla and Tanelian 3performed a tissue culture study using rabbit cornea with subepithelial wounds, and reported that continuous perfusion of 250 μg/ml (approximately 1 mM) lidocaine for 75 h delayed wound healing, but 100 μg/ml (approximately 400 μM) lidocaine did not. In the current study, 400 μm lidocaine suppressed HCEC proliferation slightly, but 40 μm lidocaine did not. We suggest that prolonged application of a low concentration of lidocaine (< 400 μM) can be used safely on the cornea.

High concentrations of lidocaine (4 and 40 mM) showed greatly reduced survival of HCECs in the current study. Tissue culture study using rabbit cornea also showed that 500 and 1,000 μg/ml (approximately 2 and 4 mM) lidocaine completely halted reepithelialization during corneal wound healing. 3In cell culture studies, lidocaine (> 3 mM) induced apoptosis and necrosis in both dose-dependent and time-dependent manners. 28–30Taken together, these results show that prolonged application of high concentrations of lidocaine induces cell death. The inhibitory effect of lidocaine on EGFR might not be a single cause of the halt in cell proliferation. Other mechanisms, which induce apoptosis or necrosis in cultured HCECs, might also be causes of the effect of high-concentration (4 and 40 mM) lidocaine on HCEC proliferation.

In conclusion, lidocaine directly inhibits tyrosine kinase activity of EGFR. This mechanism may be one of the causes of corneal toxicity of topical lidocaine after corneal wounding.