The mechanisms by which cancer cells migrate to metastasise are not fully understood. Breast cancers are accompanied by electrical depolarisation of tumour epithelial cells. The electrical changes can be detected on the skin and are used to differentiate malignant from benign breast tumours. Could the electrical signals play a role in metastasis by promoting tumour cell migration? We report that electric fields stimulate and direct migration of human breast cancer cells. Importantly, these effects were more significant in highly metastatic tumour cells than in low metastatic tumour cells. Electric-field-enhanced directional migration correlates well with the expression level of EGF receptor (EGFR/ErbB1). To confirm this, we transfected low metastatic clone MTC cells with human ErbB1, which significantly increased the electrotactic response. Inhibition of ErbB1 completely abolished the directional response of MTLn3 cells to an electric field. Transfection of MTLn3 cells and MDA-MB-435 cells with expression vectors for ErbB family members ErbB1, ErbB2 and ErbB3 also significantly enhanced EF-induced migration. These results suggest that electric signals might play a role in metastasis of breast cancers by enhancing cell migration through the ErbB-signalling pathway.

Introduction

Breast cancer is the most prevalent malignancy in women, with 90% of mortality arising from metastasis (Christofori, 2003). The molecular events in breast cancer metastasis are not fully understood. To establish secondary tumours, cancer cells have to invade surrounding tissues and blood vessels and migrate away from the primary site (Banyard and Zetter, 1999; Lawrence and Steeg, 1996). The metastatic potency of tumours is associated with protrusive and motile activity of tumour cells (Chicoine and Silbergeld, 1995; Verschueren et al., 1994). Various environmental factors including extracellular matrix, cytokines, and growth factors from autocrine and paracrine sources have been demonstrated to create chemotactic signals to stimulate migration of mammary cancer cells (Segall et al., 1996; Kundra et al., 1994; Levine et al., 1995). Overexpression of epidermal growth factor (EGF) receptor in breast cancer cell results in significantly increased intravasation and metastasis from the primary tumour (Xue et al., 2006a; Hirsch et al., 2006).

Many motile cells sense naturally occurring physiological electrical gradients in extracellular spaces and show directional migration (galvanotaxis or electrotaxis) (McCaig and Zhao, 1997; McCaig et al., 2002; McCaig et al., 2005; Pu and Zhao, 2005; Robinson, 1985; Wang et al., 2003a; Wang et al., 2003b; Zhao et al., 1996; Zhao et al., 1999; Pullar and Isseroff, 2005; Pullar et al., 2006). A role for electric fields (EFs) in directing cell migration during development and in wound healing has been suggested (Nuccitelli, 2003; McCaig and Zhao, 1997; McCaig et al., 2005); however, accumulating evidence suggests that EFs may have a much more important role in wound healing than previously thought (Zhao et al., 2006). Normal breast epithelial cells maintain an asymmetric potential gradient of +30 mV between apical and basolateral surfaces (a transepithelial electrical potential difference or TEP; inside negative with respect to the lumen, outside) because of the generation of ionic gradients. As epithelial cells divide, the charge gradient across the epithelial layer is dissipated, resulting in electrical depolarisation (Faupel et al., 1997). When a cancer develops, epithelial cells in certain areas of the breast divide more rapidly than cells in normal areas, and this is accompanied by membrane depolarisation, which can produce a significantly greater electropotential difference on the skin surface over invasive cancer than benign lesions (Marino et al., 1994; Binggeli and Weinstein, 1986; Faupel et al., 1997). The electrical signal is measurable on the skin surface above the breast lesion and this non-invasive electropotential measurement has been used as a new clinical method for breast cancer diagnosis and evaluation of invasive potency (Cuzick et al., 1998; Faupel et al., 1997; Marino et al., 1994). In addition, a strongly metastatic phenotype is associated with the expression of voltage-gated Na+ channels (VGSCs) and inhibition of these channels strongly inhibits lateral motility and invasiveness of metastatic cancer cells. A voltage-gated Na+ channel opener enhances electrotaxis of highly metastatic cells, but modulation of Na+ channel activity does not affect electrotaxis in weakly metastatic cells (Grimes et al., 1995; Laniado et al., 1997; Fraser et al., 1999; Djamgoz et al., 2001). High-level VGSC expression is usually accompanied by fast inward Na+ currents in metastatic breast cancer cells. By contrast, normal breast epithelial cells and weakly metastatic breast cancer cells show no inward current (Fraser et al., 2005; Carter and Coffey, 1988).

Because there is a steady electrical gradient between tumour and normal tissues during metastasis (Cuzick et al., 1998; Marino et al., 1994), we asked whether tumour cells detect and respond to such signals. In chemotaxis, tumour cells move up a chemical gradient. Ligands bind to specific receptors, and signal transduction cascades are activated, leading to directional migration. The EGF receptor, ErbB1 is involved in the metastatic process. Overexpression of ErbB1 can increase chemotaxis to EGF (Bailly et al., 1998; Bailly et al., 2000; Gschwind et al., 2002; Dittmar et al., 2002). Activation of ErbB2/ErbB3 heterodimers can affect both proliferation and motility (Xue et al., 2006b).

To dissect the mechanisms underlying the direction of tumour cell migration in electrotaxis and explore whether ErbB1 signalling (receptor tyrosine kinase pathways) plays an important role during tumour invasion and metastasis, we compared the electrotactic responses of breast cancer cell lines with different ErbB expression levels. We show that the ErbB1 is a key element mediating tumour cell galvanotaxis and that its expression levels correlate with metastatic potential.

Results

Human breast cancer cells and rat mammary adenocarcinoma cells show strong electrotactic responses

The human breast cancer cell line MDA-MB-231 responded to applied electric fields of 1-4 V/cm with strong directional migration towards the anode (Fig. 1A; supplementary material Movie 1). Significant directional migration was observed at a field strength of 1 V/cm with migration directedness of –0.47±0.03 (n=195; P<0.001 compared with no EF control: 0.04±0.02 n=230, Fig. 1B). Cell migration rate was significantly increased when exposed to the EF (P<0.001 compared with no EF control) (Fig. 1C). The increased directedness of MDA-MB-231 cells was voltage dependent and peaked around 3 V/cm. Migration rate did not increase significantly between 1 and 3 V/cm (P>0.05) (Fig. 1C), but further increasing EF >3 V/cm increased migration speed (P<0.01), but did not increase directedness.

We also tested the electrotaxis of the rat mammary adenocarcinoma cell line MTLn3. In the absence of an applied EF, MTLn3 cells migrated randomly with an average net directedness of –0.06±0.08 and displacement speed of 10.2±1.0 μm/hour. Cells had clear responses toward the anode at a low EF strength of 0.5 V/cm and reached a peak at 1.5 V/cm (Fig. 1E,F). Cells extended anode-directed lamellipodia and began directed migration towards the anode within 5 minutes of switching the EF on. Following this, cells reoriented to lie perpendicular to the EF vector. Migrating cells extended membrane protrusions preferentially toward the anode, either from the leading edge, or at both ends of the long axis (Fig. 1D; supplementary material Movie 2). A series of coefficients of movement efficiency showed that Td/Tt and Dx/Tt increased by 68±3% and 63±3% respectively, but Dx/Td was 92±3%; this indicates that cells turned to move almost directly along the EF vector toward the anode at EF strength of 1.5 V/cm. When the EF polarity was reversed, cells rapidly changed direction to move towards the new anode (Fig. 1G; supplementary material Movie 3). In each case, there was significant directedness toward the anode, although migration speed remained virtually the same.

Electrotaxis of tumour cells correlates with metastatic potential and ErbB1 expression levels

To assay whether tumour cells of varying metastatic potential show altered galvanotaxis in a small electric field, we compared non-metastatic (MTC) and highly metastatic (MTLn3) cells lines from a common chemically induced tumour. MTLn3 cells showed strong electrotaxis towards the anode at a low EF strength of 0.5 V/cm. The directedness of MTLn3 was three times greater than MTC cells (P<0.001). Highly metastatic cells also moved 2.5 times faster than non-metastatic cells in response to 0.5 V/cm electric field (P<0.001) (Fig. 2A). This indicates that the responsiveness of mammalian breast cancer cells to direct current EFs is correlated with the metastatic potential of the clone.

ErbB1 expression in breast cancer cell lines increases chemotactic migration (Wyckoff et al., 1998). To test whether ErbB1 expression is also associated with the electrotactic response, we detected ErbB1 protein levels in human and rat breast cancer lines on western blot and compared them with cell directedness in EFs. We found that highly metastatic cell lines MTLn3 (rat) and MDA-MB-231 (human) have high ErbB1 levels and show strong anodal migration to an EF (0.5 V/cm for rat cell lines, 1.0 V/cm for human cell lines, t=2 hours). Rat MTC cells show a low level of ErbB1 expression and cell directedness was as low as 30% of that in MTLn3 cells. Expression of the human EGF receptor in MTC cells (MTC-B1) can enhance cell directedness to a level similar to MTLn3. ErbB1 levels in the non-metastatic human MCF7 cell line were particularly low and cells displayed a low cathodal response to the EF (Fig. 2B,C). In general, therefore, high levels of ErbB1 expression correlated with strong electrotaxis of breast cancer cells.

ErbB1 expression enhances electrotactic response

To determine whether expression of ErbB1 is involved in the signalling of tumour cell electrotaxis, two cell lines were compared: MTC cells express very low levels of ErbB1, and MTC-B1 cells are MTC cells transfected with an expression vector for the human ErbB1. MTC cells had a weak electrotactic response at 0.5 V/cm (directedness: –0.24±0.07 n=65), but MTC-B1 showed a threefold greater directedness toward the anode at the same EF strength (P<0.001). MTC-B1 cells also migrated faster than MTC cells (Tt/t 18.85±0.81; Td/t 14.10±0.56 P<0.001 n=105, compared with MTC cells) (Fig. 3A,B; supplementary material Movie 4). Transfecting cells with an expression vector for the human ErbB1 therefore markedly enhanced their responsiveness to a small EF.

Inhibition of ErbB1 activity completely abolished electrotaxis of MTLn3 cells

To precisely measure the effect of ErbB1 on tumour cell electrotaxis, we observed the effect of ErbB1 blockade on the ability of MTLn3 cells to sense and respond to an applied electric field of 1 V/cm. Exposing MTLn3 cells to 2 μM AG1478 (a specific inhibitor of ErbB1) resulted in a 90% reduction in directional migration, from an average directedness of –0.82±0.06 (n=141) in EF alone, to –0.07±0.11 (n=195) in EF + ErbB1 inhibitor, which was similar to the no EF control (P<0.001, compared with EF alone; P>0.05, compared with no EF control) (Fig. 4C,D). This indicated that cells had lost EF-induced directionality (Fig. 4A,B; supplementary material Movie 5). The ErbB1 inhibitor did not affect random motility; tracking of migrating cells shows no significant difference in trajectory speed between EF control and EF + AG1478 groups (P>0.05). The displacement speed, which shows how efficiently cells moved in a given direction, was suppressed by 34% to 15±0.9 μm/hour (n=195, P<0.001), but still moved faster than the no EF control (P<0.001) (Fig. 4C,D). The above results indicate that AG1478 virtually abolished electrotaxis of adenocarcinoma cells in a small EF (1.0 V/cm).

Fig. 1.

Breast cancer cells migrate anodally in a small physiological electric field. (A) Human breast cancer cells (MDA-MB-231) migrate to the anode (left) (see supplementary material Movie 1). The cells were starved in serum-free medium containing 0.35% HSA overnight before exposure to a direct current EF of 1.5 V/cm. White lines with blue arrowheads represent trajectories and direction of cell movement. (B) Directedness of cell migration shows voltage dependence of the directional migration with the threshold voltage inducing directional migration below 1 V/cm. (C) Small electric fields significantly increased the migration speed. *P<0.001 compared to no EF control, ^P<0.01 compared to 1-3V/cm EF. (D) Rat mammary cancer cells (MTLn3) migrate to the anode in an electric field of 1.5 V/cm (see supplementary material Movie 2). White lines and blue arrowheads represent trajectories and direction of cell movement. (E,F) Voltage dependence of the migration directedness and speed. Data are mean ± s.e.m. of three independent experiments. *P<0.01 compared with no EF control. Bar, 50 μm. (G) MTLn3 cells migrate toward the anode in EF. After 40 minutes, the polarity of electric field was reversed. Cells of the same field continued to be recorded for the indicated period. White arrows represent displacement distances and direction of cell movement (see supplementary material Movie 3). EF=1.5 V/cm.

Fig. 1.

Breast cancer cells migrate anodally in a small physiological electric field. (A) Human breast cancer cells (MDA-MB-231) migrate to the anode (left) (see supplementary material Movie 1). The cells were starved in serum-free medium containing 0.35% HSA overnight before exposure to a direct current EF of 1.5 V/cm. White lines with blue arrowheads represent trajectories and direction of cell movement. (B) Directedness of cell migration shows voltage dependence of the directional migration with the threshold voltage inducing directional migration below 1 V/cm. (C) Small electric fields significantly increased the migration speed. *P<0.001 compared to no EF control, ^P<0.01 compared to 1-3V/cm EF. (D) Rat mammary cancer cells (MTLn3) migrate to the anode in an electric field of 1.5 V/cm (see supplementary material Movie 2). White lines and blue arrowheads represent trajectories and direction of cell movement. (E,F) Voltage dependence of the migration directedness and speed. Data are mean ± s.e.m. of three independent experiments. *P<0.01 compared with no EF control. Bar, 50 μm. (G) MTLn3 cells migrate toward the anode in EF. After 40 minutes, the polarity of electric field was reversed. Cells of the same field continued to be recorded for the indicated period. White arrows represent displacement distances and direction of cell movement (see supplementary material Movie 3). EF=1.5 V/cm.

Overexpression of ErbB2 or ErbB3 increases tumour-cell-directed migration in electrotaxis

To explore further the effects of other relevant ErbB family members ErbB2 and ErbB3 on the electrotaxis of MDA-MB-435 and MTLn3 mammary tumour cells, we used a series of retroviral vectors based on the PLXSN retrovirus (Riese et al., 1995). MDA-MB-435 and MTLn3 cells were transduced with either empty vector (PLXSN alone) or PLXSN containing the cDNAs for ErbB1, ErbB2 or ErbB3 followed by selection for geneticin resistance. Cells generated by transduction of MDA-MB-435 or MTLn3 cells with PLXSN, ErbB1, ErbB2 or ErbB3 retrovirus were named 435-PL, 435-B1, 435-B2, 435-B3, or MTLn3-PL, MTLn3-B1, MTLn3-B2, MTLn3-B3, respectively (Xue et al., 2006b). We found that MTLn3-B1, -B2 and -B3 showed increased cell directedness to 0.5 V/cm EF compared with MTLn3-PL (P<0.05); MTLn3-B1 also significantly increased migration rate (P<0.01) (Fig. 5A). We also tested the electrotaxis of MDA-MB-435 transfected with different vectors. Upon exposure to an applied EF of 2 V/cm, both 435-B2 and 435-B3 increased cell directedness and displacement speed (P<0.05), but not trajectory rate (P>0.05 compared with 435-PL). 435-B1 not only markedly enhanced cell directedness, but also increased trajectory and displacement speed (P<0.01 compared with 435-PL) (Fig. 5B).

Fig. 2.

Strong correlation between electrotactic migration and EGFR (ErbB1) expression levels. (A) A small EF of 0.5 V/cm was applied to cells for 2 hours. Non-metastatic rat mammary adenocarcinoma MTC clone (13762NF) showed weak electrotaxis, whereas the high metastatic clone, MTLn3 showed robust electrotaxis. There were significant quantitative differences in migration speeds and in cell directedness. *P<0.001 compared with MTC cells. Data are mean ± s.e.m. of three independent experiments. (B) Significantly varied expression levels of ErbB1 in five breast cancer cell lines were detected by western blot. α-tubulin is loading control. (C) ErbB1 expression level (upper histogram) correlates strongly with the directionality of EF-directed migration (lower histogram). Cells were starved for 3 hours, and then a small EF was applied for 2 hours (0.5 V/cm for rat cell lines, 1.0 V/cm for human cell lines).

Fig. 2.

Strong correlation between electrotactic migration and EGFR (ErbB1) expression levels. (A) A small EF of 0.5 V/cm was applied to cells for 2 hours. Non-metastatic rat mammary adenocarcinoma MTC clone (13762NF) showed weak electrotaxis, whereas the high metastatic clone, MTLn3 showed robust electrotaxis. There were significant quantitative differences in migration speeds and in cell directedness. *P<0.001 compared with MTC cells. Data are mean ± s.e.m. of three independent experiments. (B) Significantly varied expression levels of ErbB1 in five breast cancer cell lines were detected by western blot. α-tubulin is loading control. (C) ErbB1 expression level (upper histogram) correlates strongly with the directionality of EF-directed migration (lower histogram). Cells were starved for 3 hours, and then a small EF was applied for 2 hours (0.5 V/cm for rat cell lines, 1.0 V/cm for human cell lines).

Signalling of ErbB: involvement of tyrosine kinases, PI3K, Rho GTPases and ERK

MTLn3 cells were exposed to the tyrosine kinase inhibitor Genistein (100 μM) and an EF of 1.5 V/cm. The cells showed significant decreases in anodal directedness. The trajectory and displacement speed also were reduced by 72% and 83% to 12.6±0.6 and 5.1±0.8, respectively (n=56 in EF; n=60 in EF + Genistein, P<0.001). Directedness decreased to –0.21±0.1 (Fig. 6A,B). These results demonstrate that MTLn3 cells lost directional migration and slowed down markedly in the presence of Genistein. Compared with ErbB1 inhibition (AG1478), Genistein was more effective in reducing migration rates during tumour cell electrotaxis, indicating that tyrosine phosphorylation not mediated by the EGF receptor is important for basal cell motility.

Fig. 3.

ErbB1 expression enhances electrotactic responses. MTC and MTC-B1 cell lines were starved for 3 hours, and then 0.5 V/cm EF was applied for 2 hours. (A) Non-metastatic MTC cell line did not show electrotaxis at 0.5 V/cm (upper panel), after ErbB1 transfection (MTC-B1), cells showed significant strong electrotaxis (lower panel). Red lines and yellow arrowheads represent trajectories and direction of cell movement. (B) Bar graphs showing that ErbB1-transfected cells (MTC-B1) move faster and with greater anodal directedness than MTC cells. *P<0.01 compared with MTC parental cells. Data are mean ± s.e.m. from three independent experiments.

Fig. 3.

ErbB1 expression enhances electrotactic responses. MTC and MTC-B1 cell lines were starved for 3 hours, and then 0.5 V/cm EF was applied for 2 hours. (A) Non-metastatic MTC cell line did not show electrotaxis at 0.5 V/cm (upper panel), after ErbB1 transfection (MTC-B1), cells showed significant strong electrotaxis (lower panel). Red lines and yellow arrowheads represent trajectories and direction of cell movement. (B) Bar graphs showing that ErbB1-transfected cells (MTC-B1) move faster and with greater anodal directedness than MTC cells. *P<0.01 compared with MTC parental cells. Data are mean ± s.e.m. from three independent experiments.

Treatment with 50 μM Ly294002, a PI3K inhibitor, significantly decreased EF-directed anodal migration in MTLn3 cells. However, the inhibition was not complete and the EF-directed migration largely remained (directedness value: –0.78±0.07, n=70, Fig. 6A). Displacement speed was reduced by 28% to 22.2±2.0 (n=70) (P<0.01) (Fig. 6B), indicating partial involvement of PI3K signalling.

Small GTPases contribute to cancer progression primarily through their effects on cell migration, thereby influencing invasion and metastasis (Fiordalisi et al., 2006; Jeong et al., 2005). Toxin B (10 ng/ml) as a general inhibitor of Rho, Rac and Cdc42, significantly decreased MTLn3 cells directedness and migration speed (P<0.001), but the inhibition was incomplete, and significant anodal electrotaxis still occurred (Fig. 6C). Three custom-designed, cell-permeable short peptides were used to inhibit RhoA, Rac and Cdc42 (LS201, LS202 and LS203, respectively) (Vastrik et al., 1999). Each of these peptides reduced tumour cell migration speed (P<0.01) to a similar extent as did Toxin B. LS201 and LS203, but not LS202, also decreased cell directedness (Fig. 6D), suggesting partial involvement of RhoA and Cdc42.

Fig. 4.

ErbB1 dependence of electrotactic migration. (A) Inhibition of EGF receptor signalling with AG1478 abolished electrotactic response in mammary cancer cells (see supplementary material Movie 5). Cells were plated on collagen-1 coated dishes overnight, starved in serum-free α-MEM for 3 hours before the experiment, with or without 2 μM AG1478. An EF of 1 V/cm was applied for 2 hours. Red lines and blue arrowheads represent trajectories and direction of cell movement. (B) Trajectories of cells over 2 hours with the starting points positioned at the origin. x- and y-axes give distance in μm. (C) AG1478 inhibited displacement speed in EF-induced MTLn3 electrotaxis (*P<0.001 compared with EF alone), but cells in AG1478 group still moved faster than no EF control, (^P<0.01 compared with no EF control). There are no differences in trajectory speed between the two groups. (D) Anodal directedness was largely suppressed by AG1478 to the same value as no EF control *P<0.001 compared with EF alone. Results were calculated from the means of three independent experiments.

Fig. 4.

ErbB1 dependence of electrotactic migration. (A) Inhibition of EGF receptor signalling with AG1478 abolished electrotactic response in mammary cancer cells (see supplementary material Movie 5). Cells were plated on collagen-1 coated dishes overnight, starved in serum-free α-MEM for 3 hours before the experiment, with or without 2 μM AG1478. An EF of 1 V/cm was applied for 2 hours. Red lines and blue arrowheads represent trajectories and direction of cell movement. (B) Trajectories of cells over 2 hours with the starting points positioned at the origin. x- and y-axes give distance in μm. (C) AG1478 inhibited displacement speed in EF-induced MTLn3 electrotaxis (*P<0.001 compared with EF alone), but cells in AG1478 group still moved faster than no EF control, (^P<0.01 compared with no EF control). There are no differences in trajectory speed between the two groups. (D) Anodal directedness was largely suppressed by AG1478 to the same value as no EF control *P<0.001 compared with EF alone. Results were calculated from the means of three independent experiments.

The MAP kinase ERK1/2 plays an important role in EF-directed migration of corneal and lens epithelial cells (Zhao et al., 2002; Wang et al., 2003b). Exposure to an EF activated ERK1/2 in MTLn3 cells (Fig. 7A,B). We observed the effect of ERK1/2 blockade on the electrotaxis of MTLn3 cells in an applied electric field of 0.5 V/cm. U0126 (50 μM) (a specific inhibitor of ERK1/2) resulted in a 32% reduction in directional migration to –0.49±0.05 (n=46) (P<0.001), but did not completely abolish MTLn3 electrotaxis (P<0.001 compared with no EF control) (Fig. 7C). These results demonstrated that ERK signalling partially contributes to tumour cell electrotaxis.

Fig. 5.

ErbB2/ErbB3 expression enhances directed migration in tumour cell electotaxis. MTLn3 and MDA-MB-435 cells were transduced with either empty vector (PLXSN alone) or PLXSN containing the cDNAs for ErbB1, ErbB2 or ErbB3. Cells were plated on collagen-1-coated dishes overnight and starved in serum-free α-MEM for 2-3 hours before EF exposure. (A) An EF of 0.5 V/cm was applied to MTLn3 cells for 2 hours. *P<0.05, **P<0.01 compared with MTLn3-PL cells. (B) An EF of 2 V/cm was applied to MDA-MB-435 cells for 2 hours. Directedness of cell migration in an EF was increased for all three lines transduced with ErbB1, ErbB2 or ErbB3. *P<0.05; **P<0.01 compared with 435-PL cells. Data are mean ± s.e.m from three independent experiments on more than 200 cells.

Fig. 5.

ErbB2/ErbB3 expression enhances directed migration in tumour cell electotaxis. MTLn3 and MDA-MB-435 cells were transduced with either empty vector (PLXSN alone) or PLXSN containing the cDNAs for ErbB1, ErbB2 or ErbB3. Cells were plated on collagen-1-coated dishes overnight and starved in serum-free α-MEM for 2-3 hours before EF exposure. (A) An EF of 0.5 V/cm was applied to MTLn3 cells for 2 hours. *P<0.05, **P<0.01 compared with MTLn3-PL cells. (B) An EF of 2 V/cm was applied to MDA-MB-435 cells for 2 hours. Directedness of cell migration in an EF was increased for all three lines transduced with ErbB1, ErbB2 or ErbB3. *P<0.05; **P<0.01 compared with 435-PL cells. Data are mean ± s.e.m from three independent experiments on more than 200 cells.

Fig. 6.

PI3K and Rho family small GTPases are partly involved in the directional migration of tumour cells in electrotaxis. (A,B) MTLn3 cells were pretreated with Ly294002 (50 μM) or Genistein (100 μM; Geni) for 2 hours before EF application. PI3K inhibitor Ly294002 (Ly) only slightly (but statistically significantly) inhibited displacement speed and anodal directedness (*P<0.01 compared with EF alone), but Genistein can significantly block migration speeds and directedness in electrotaxis (**P<0.001 compared with EF alone). (C,D) Cells were pretreated with Toxin B (10 ng/ml), or the peptide LS201 (RhoA inhibitor, 100 ng/ml), LS202 (Rac inhibitor, 100 ng/ml) or LS203 (Cdc42 inhibitor, 100 ng/ml) for 2 hours before EF stimulation. *P<0.01 compared with no inhibitor control; ^P< 0.05 compared with LS201, LS202; ^^P<0.01 compared with LS202. Data are presented as mean ± s.e.m. Data shown are from three independent experiments. EF=1.5 V/cm, 2 hours.

Fig. 6.

PI3K and Rho family small GTPases are partly involved in the directional migration of tumour cells in electrotaxis. (A,B) MTLn3 cells were pretreated with Ly294002 (50 μM) or Genistein (100 μM; Geni) for 2 hours before EF application. PI3K inhibitor Ly294002 (Ly) only slightly (but statistically significantly) inhibited displacement speed and anodal directedness (*P<0.01 compared with EF alone), but Genistein can significantly block migration speeds and directedness in electrotaxis (**P<0.001 compared with EF alone). (C,D) Cells were pretreated with Toxin B (10 ng/ml), or the peptide LS201 (RhoA inhibitor, 100 ng/ml), LS202 (Rac inhibitor, 100 ng/ml) or LS203 (Cdc42 inhibitor, 100 ng/ml) for 2 hours before EF stimulation. *P<0.01 compared with no inhibitor control; ^P< 0.05 compared with LS201, LS202; ^^P<0.01 compared with LS202. Data are presented as mean ± s.e.m. Data shown are from three independent experiments. EF=1.5 V/cm, 2 hours.

Fig. 7.

Phosphorylation of ERK1/2 is required in EF-induced tumour cell migration. (A,B) An EF of 1.5 V/cm enhanced activation of ERK 1/2 in mammary breast cancer cells (MTLn3). The expression of active ERK1/2 increased within 15 minutes (m) of EF application, reached a maximum at 30 minutes that lasted for at least 1 hour, whereas the total level of ERK remained unchanged. Each membrane was a representative of two to three repeated experiments. (C) Inhibition of ERK partly inhibited EF-induced MTLn3 cell migration. Cells were pretreated with 50 μM U0126 (ERK-specific inhibitor) for 2 hours before application of 0.5 V/cm EF. *P<0.01 compared with EF alone; ^P<0.01 compared with no EF control.

Fig. 7.

Phosphorylation of ERK1/2 is required in EF-induced tumour cell migration. (A,B) An EF of 1.5 V/cm enhanced activation of ERK 1/2 in mammary breast cancer cells (MTLn3). The expression of active ERK1/2 increased within 15 minutes (m) of EF application, reached a maximum at 30 minutes that lasted for at least 1 hour, whereas the total level of ERK remained unchanged. Each membrane was a representative of two to three repeated experiments. (C) Inhibition of ERK partly inhibited EF-induced MTLn3 cell migration. Cells were pretreated with 50 μM U0126 (ERK-specific inhibitor) for 2 hours before application of 0.5 V/cm EF. *P<0.01 compared with EF alone; ^P<0.01 compared with no EF control.

Discussion

Cell migration is a key element in tumour progression. Understanding whether naturally occurring or applied electrical signals contribute to the control of tumour cell migration might offer some novel strategies for preventing cancer metastasis. Cancer cell electrotaxis could be important in vivo. Endogenous direct current EFs have been measured in many biological systems, including tumours (Szatkowski et al., 2000), and EFs enhance growth of cancer spheroids (Wartenberg et al., 1997; Sauer et al., 1997). We show that breast tumour cells detect a small electrical signal and migrate anodally. The directional migration of breast epithelial cells could be influenced significantly by endogenous TEPs. Normal breast epithelium has a TEP of about +30 mV (duct lumen positive). Such a voltage gradient is larger than the applied direct current EF used to induce electrotaxis in this study (Fig. 8), because given a lumenal wall ∼50 μm wide, this would represent a substantial voltage gradient of about 6 V/cm (30 mV/50 μm), around ten times greater than we studied (Faupel et al., 1997). When a cancer develops, malignant epithelial cells choose to move and grow into the lumen rather than into surrounding tissue (Olivotto and Levine, 2001). Furthermore, as metastasis progresses, the duct deforms and the charge gradient across the epithelial layer may dissipate, resulting in collapse of the EF across the epithelial layer. Cancer cells then breach the basement membrane and invade the surrounding tissue and metastasise through the blood and lymph systems (Faupel et al., 1997; Djamgoz et al., 2001).

Tumour cell electrotaxis correlates with metastatic potential

When exposed to a small EF some cell types migrate cathodally, some anodally and others show no responsiveness (Nuccitelli, 2003). This highlights the cell-specific subtlety of response to electrical cues (McCaig et al., 2005). A further indicator of this complexity and refinement is that lens epithelial cells migrate cathodally below 1.0 V/cm and anodally at and above 1.5 V/cm; both values are within the physiological range (Wang et al., 2000; Wang et al., 2003b). In epithelial cells and human keratinocyte electrotaxis, directedness generally is around 0.6-0.7 at 1.5 V/cm. However, at the same strength, over 90% of MTLn3 tumour cells show perfect migration directly along the EF vector toward the anode, giving a remarkable directedness close to 1. Some tumour cells therefore show much stronger electrotaxis than other cell types. This phenomenon correlated closely with tumour metastatic potential. The 13762NF mammary-tumour-derived clones have been widely used to examine mechanisms of breast cancer metastasis (Segall et al., 1996). The highly metastatic rat breast cancer cell line MTLn3 showed a robust anodal galvanotactic response, whereas non-metastatic MTC cells responded three times less to an electric field of the same strength (directness 0.72±0.01 compared with 0.24±0.07 at 0.5 V/cm). The migration speed of MTLn3 cells was 2.5 times faster than MTC cells. The highly metastatic human breast cancer cell line MDA-MB-231 showed strong anodal electrotaxis and a fast migration speed, whereas weakly metastatic MCF-7 cells moved slowly and surprisingly in the opposite direction – cathodally. Djamgoz et al. (Djamgoz et al., 2001) compared the galvanotactic responses of two prostate cancer cell lines with different metastatic potential. The highly metastatic MAT-lyLU cells moved toward the cathode with strong directness of 0.82±0.01. By contrast, the weakly metastatic AT-2 cells responded weakly (directness is –0.34±0.01) and moved in the opposite direction. These data accentuate the variation in electrotactic responses in different types of tumour cells with different metastatic abilities.

Mechanisms of breast cancer cell electrotaxis – the role of the EGF receptor and other ErbB members

The EGF receptor has been suggested to play a critical role in the electrotactic response of corneal epithelial cells and keratinocytes (Zhao et al., 2002; Fang et al., 1999). Application of a small EF upregulates the expression of EGF receptors and induces asymmetrical distribution of EGF receptors, and their downstream signalling element, MAP kinase ERK1/2 and filamentous actin to the cathode-facing leading lamellipodia (Zhao et al., 1999; Zhao et al., 2002). Internalisation of EGF receptors remained concentrated behind the leading edge of cells, potentially continuing to contribute to polarisation of the tumour cell (Bailly et al., 2000). Kaufmann et al. (Kaufmann et al., 1996) showed that induction of apoptosis by EGF receptors in rat mammary adenocarcinoma cells coincides with enhanced spontaneous tumour metastasis. ErbB1 signalling plays a crucial role in metastatic migration of breast cancer cells (Goswami et al., 2005; Wyckoff et al., 2004). Implantation of MTC-B1 clone intravenously into mice led to enhanced metastasis (Lichtner et al., 1995). We demonstrated that ErbB1 expression levels correlate well with the electrotactic response. MTC cells with low ErbB1 expression levels showed a weak response. Transfection of the human EGF receptor into MTC cells enhanced EF-directed migration, with both migration speed and directedness markedly increased (Fig. 3A,B; supplementary material Movie 4). This restoration and amplification of EF responsiveness indicates that the EGF receptor signal transduction pathway is involved in electrotactic responses in these cancer cells. In further support of this, MTLn3 cells treated with AG1478, a specific inhibitor of ErbB1, showed a 90% reduction of EF-directed migration, leaving directedness close to zero, although trajectory speed remained unaffected (Fig. 4; supplementary material Movie 5). Genistein – a broad spectrum receptor tyrosine kinase inhibitor – completely inhibited EF-directed cell migration. These results indicate that EGF receptor activation is required for electrotaxis of these tumour cells.

Fig. 8.

Schematic diagram of the electrical gradient in the breast duct. A 30 mV transepithelial electrical potential difference (TEP) (the difference between the apical and the basolateral potentials) exists, with the lumen side positive (Faupel et al., 1997). This would generate a voltage of 6 V/cm across the epithelium layer (30 mV over 50 μm), which is up to ten times greater than the EF strengths used in our experiments. The vector of this EF (minus to plus) with the anode in the lumen coincides with the direction of the first metastasis of breast cancer cells, in which the cells migrate into the lumen (Wellings and Jensen, 1973).

Fig. 8.

Schematic diagram of the electrical gradient in the breast duct. A 30 mV transepithelial electrical potential difference (TEP) (the difference between the apical and the basolateral potentials) exists, with the lumen side positive (Faupel et al., 1997). This would generate a voltage of 6 V/cm across the epithelium layer (30 mV over 50 μm), which is up to ten times greater than the EF strengths used in our experiments. The vector of this EF (minus to plus) with the anode in the lumen coincides with the direction of the first metastasis of breast cancer cells, in which the cells migrate into the lumen (Wellings and Jensen, 1973).

In addition to the key role of ErbB1 in tumour metastasis, a human breast cancer tumour tissue microarray revealed a significant association between ErbB2 and ErbB3 expression and metastasis. Suppression of ErbB2 or ErbB3 expression significantly reduced intravasation and metastasis (Xue et al., 2006b). To identify the potential contributions of ErbB2- or ErbB3-dependent motility responses to tumour metastasis, we evaluated the effects of overexpressing of ErbB on the electrotaxis of MDA-MB-435 and MTLn3 tumour cells. We found that enhancing ErbB2 or ErbB3 expression markedly increases tumour cell directedness and displacement speed in EFs although the effects of ErbB2 and ErbB3 are not as strong as that of ErbB1 (Fig. 4). Our results propose that ErbB2- and ErbB3-dependent signalling can contribute to metastasis through tumour cell motility in electrotaxis.

Downstream signalling of EGF receptor and ErbB family

Activation of ErbB1 stimulates several downstream pathways. The phosphatidylinositol 3-kinase (PI3K) signalling pathway plays a crucial role in the electrotactic response of epithelial cells in wound healing (Zhao et al., 2006). It is a key regulator of the actin cytoskeleton and cell migration in various cancer cells (Verbeek et al., 1998; Ellerbroek et al., 2001; Hill et al., 2000). In squamous cell carcinoma of the head and neck (HNSCC), PI3K is required for ErbB1-mediated cell movement but not for EGF-induced mitogenesis (Chen et al., 1994; O-charoenrat et al., 2002). However, PI3K inhibition only partially decreased the electrotaxis of MTLn3 cells (Fig. 6A,B). Similarly, the Rho family small GTPases Rho and Cdc42, and the MAP kinase ERK1/2 are involved, but they only played a partial role in EF-induced migration in MTLn3 cells.

In summary, we demonstrated that breast cancer cells migrate directionally in an EF; this response correlates with metastatic potential and the expression level of the EGF receptor. Highly metastatic tumour cells show robust electrotaxis and move anodally. Transfection and expression of the EGF receptor and three ErbB family members in weak metastasis tumour cells significantly enhanced the electrotactic response. Inhibition of the EGF receptor activation abolished the electrotactic response. We thus conclude that EGF receptor signalling is essential for the electrotactic responses of breast cancer cells. Identification of potential roles for endogenous or applied EFs during metastasis and their underlying molecular mechanisms could lead to novel potential therapeutic strategies.

Materials and Methods

Cell cultures

MTLn3 is a metastatic cell line derived from the 13762NF rat mammary adenocarcinoma and MTC is a non-metastatic cell line from the same tumour (Neri et al., 1982; Welch et al., 1983). MTC cells were transfected with retroviral expression vectors containing the human ErbB1 gene (EGFR) with a neomycin-resistance gene, termed MTC-B1. The human breast cancer cells MDA-MB-435 and MTLn3 cells were subjected to retroviral transduction of constructs containing the empty retroviral expression vector pLXSN alone or pLXSN containing the human cDNAs for ErbB1, ErbB2 and ErbB3 (Vectors came from David Stern, Yale University) (Riese et al., 1995). All experiments were conducted within 10 passages. Expression of transduced ErbBs measured by fluorescence-activated cell sorting (FACS) showed no change with passage in vitro (Xue et al., 2006b). Cells were grown in α-MEM (Gibco), supplemented with 5% fetal bovine serum (FBS) and 50 U/ml penicillin and 50 μg/ml streptomycin at 37°C in a humidified atmosphere of 5% CO2. For all experiments, unless otherwise mentioned, cells were plated in α-MEM supplemented with 5% FBS for 16-20 hours at low density in a specially made culture chamber formed by two parallel strips of glass coverslip (4 cm long) adhering to the base of collagen 1 coated plastic dishes (Zhao et al., 1996). Cells were starved for 3 hours prior to the experiment in α-MEM supplemented with 0.35% BSA (starvation medium). Human breast cancer cell lines MDA-MB-231 and MCF 7 were grown in RPMI 1640 medium supplemented with 5% FCS and antibiotics. Cells were plated on collagen-1-coated dishes and starved for 3 hours prior to experimentation. Experiments were performed with subcultured cells in passage 5. All chemicals were added to the cells 2 hours before exposure to the electric field. AG1478, LY294002 and Toxin B were purchased from Calbiochem (La Jolla, CA) and Genistein was obtained from Sigma. LS201, LS202, LS203 (which inhibit RhoA, Rac and Cdc42, respectively) are custom designed peptides synthesised and prepared by the Proteomics Facility, University of Aberdeen (Vastrik et al., 1999).

Electrotaxis assay

Cell motility was assayed using a galvanotaxis apparatus described in detail by Zhao et al. (Zhao et al., 1996). Briefly, direct current was applied through agar-salt bridges connecting silver/silver chloride electrodes in beakers of Steinberg's solution, to pools of culture medium on either side of the chamber. A roof of No. 1 coverglass was applied and sealed with silicone grease (Corning DC4). The final dimensions of the chamber, through which current was passed, were 40 mm × 10 mm × 0.3 mm. Immediately before EF application, CO2-independent medium (including 5% FBS) was transferred into the culture chambers, using a push-pull technique. Cells were exposed to 0.5-4 V/cm EF for 1-3 hours at 37°C and viewed on an inverted microscope in a temperature-controlled chamber. The serial time-lapse phase-contrast images were recorded using Metamorph (Universal Imaging Corporation, PA).

Western blot analysis

Tumour cells were rinsed with cold PBS and lysed with lysis buffer [10 mM Tris-HCl, 50 mM NaCl, 5 mM EDTA, 50 mM sodium fluoride, 1% Triton X-100, 30 mM Na4P2O7, 1 mM sodium orthovanadate and protease inhibitor cocktail (Boehringer)]. Equal amounts of protein lysates were electrophoresed in 6% SDS-PAGE for ErbB1 or 4-12% SDS-PAGE for ERK detection, followed by electroblotting onto nitrocellulose membranes (Invitrogen). The gels were stained for remaining proteins with Coomassie Blue as a loading control. Membranes were stained with Ponceau S for detection of transfer efficiency, then were blocked with 5% milk TBS (pH 7.4) with 0.1% (w/v) Tween 20 for 1 hour. Membranes were incubated with relevant primary antibodies (1:1000) in 5% BSA TBS-T overnight at 4°C (anti-ErbB1 was from BD Biosciences, anti-active and anti-total ERK 1/2 antibody was from Cell Signaling). Relevant secondary antibodies with horseradish peroxidase (1:3000, Amersham Pharmacia Biotech, Amersham, UK) were incubated for 1 hour at room temperature, and the immunoblots were detected by an enhanced chemiluminescence (ECL) detection system (Amersham).

Quantitative analysis of cell behaviour

Cell migration was analysed to determine directedness and average migration speed by tracing the position of cell nuclei at frame intervals of 2 minutes or 5 minutes using Metamorph software. The directedness of migration was assessed as cosine θ (Zhao et al., 1996), where θ is the angle between the EF vector and a straight line connecting the start and end position of a cell. A cell moving directly along the field lines toward the cathode would have a directedness of 1; a cell moving directly toward the anode would have a directedness of –1; a value close to 0 represents random cell movement. The cosine θ will provide a number between –1 and +1 and the average of all of the separate cell events yields the directedness index. The average directedness of a population of cells gives an objective quantification of the direction cells have moved. The trajectory speed (Tt/T) is the total migration distance of cells per hour, and the displacement speed (Td/T) is the straight-line distance between the start and end positions of a cell each hour. Dx/T is an x-axis displacement speed, which represents the ability of cells to migrate along the EF vector. Td/Tt, Dx/Tt and Dx/Td are a series of coefficients of movement efficiency. A value of 1 indicates cells moving persistently along one straight line in one direction, and 0 for random movement. Statistical analyses were made using unpaired, two-tailed Student's t-test. Data are expressed as mean ± s.e.m.

Acknowledgements

We thank the Wellcome Trust for support (058551, 068012). We thank expert referees and editors for comments and suggestions that greatly improved the manuscript.

References

Bailly, M., Yan, L., Whitesides, G. M., Condeelis, J. S. and Segall, J. E. (
1998
). Regulation of protrusion shape and adhesion to the substratum during chemotactic responses of mammalian carcinoma cells.
Exp. Cell Res.
241
,
285
-299.
Bailly, M., Wyckoff, J., Bouzahzah, B., Hammerman, R., Sylvestre, V., Cammer, M., Pestell, R. and Segall, J. E. (
2000
). Epidermal growth factor receptor distribution during chemotactic responses.
Mol. Biol. Cell
11
,
3873
-3883.
Banyard, J. and Zetter, B. R. (
1999
). The role of cell motility in prostate cancer.
Cancer Metastasis Rev.
17
,
449
-458.
Binggeli, R. and Weinstein, R. C. (
1986
). Membrane potentials and sodium channels: hypotheses for growth regulation and cancer formation based on changes in sodium channels and gap junctions.
J. Theor. Biol.
123
,
377
-401.
Carter, H. B. and Coffey, D. S. (
1988
). Cell surface charge in predicting metastatic potential of aspirated cells from the Dunning rat prostatic adenocarcinoma model.
J. Urol.
140
,
173
-175.
Chen, P., Gupta, K. and Wells, A. (
1994
). Cell movement elicited by epidermal growth factor receptor requires kinase and autophosphorylation but is separable from mitogenesis.
J. Cell Biol.
124
,
547
-555.
Chicoine, M. R. and Silbergeld, D. L. (
1995
). The in vitro motility of human glioma increases with increasing grade of malignancy.
Cancer
75
,
2904
-2909.
Christofori, P. (
2003
). Changing neighbours, changing behaviour: cell adhesion molecule-mediated signaling during tumour progression.
EMBO J.
22
,
2318
-2323.
Cuzick, J., Holland, R., Barth, V., Davies, R., Faupel, M., Fentiman, I., Frischbier, H. J., LaMarque, J. L., Merson, M., Sacchini, V. et al. (
1998
). Electropotential measurements as a new diagnostic modality for breast cancer.
Lancet
352
,
359
-363.
Dittmar, T., Husemann, A., Schewe, Y., Nofer, J. R., Niggemann, B., Zanker, K. S. and Brandt, B. H. (
2002
). Induction of cancer cell migration by epidermal growth factor is initiated by specific phosphorylation of tyrosine 1248 of c-erbB-2 receptor via EGFR.
FASEB J.
16
,
1823
-1825.
Djamgoz, M. B. A., Mycielska, M., Madeja, Z., Fraser, S. P. and Korohoda, W. (
2001
). Directional movement of rat prostate cancer cells in direct-current electric field: involvement of voltage gated Na+ channel activity.
J. Cell Sci.
114
,
2697
-2705.
Ellerbroek, S. M., Halbleib, J. M., Benavidez, M., Warmka, J. K., Wattenberg, E. V., Stack, M. S. and Hudson, L. G. (
2001
). Phosphatidylinositol 3-kinase activity in epidermal growth factor-stimulated matrix metalloproteinase-9 production and cell surface association.
Cancer Res.
61
,
1855
-1861.
Fang, K. S., Ionides, E., Oster, G., Nuccitelli, R. and Isseroff, R. R. (
1999
). Epidermal growth factor receptor relocalization and kinase activity are necessary for directional migration of keratinocytes in DC electric fields.
J. Cell Sci.
112
,
1967
-1978.
Faupel, M., Vanel, D., Barth, V., Davies, R., Fentiman, I. S., Holland, R., Lamarque, J. L., Sacchini, V. and Schreer, I. (
1997
). Electropotential evaluation as a new technique for diagnosing breast lesions.
Eur. J. Radiol.
24
,
33
-38.
Fiordalisi, J. J., Keller, P. J. and Cox, A. D. (
2006
). PRL tyrosine phosphatases regulate rho family GTPases to promote invasion and motility.
Cancer Res.
66
,
3153
-3161.
Fraser, S. P., Ding, Y., Liu, A., Foster, C. S. and Djamgoz, M. B. A. (
1999
). Tetrodotoxin suppresses morphological enhancement of the metastatic MAT-LyLu rat prostate cancer cell line.
Cell Tissue Res.
295
,
505
-512.
Fraser, S. P., Diss, J. K., Chioni, A. M., Mycielska, M. E., Pan, H., Yamaci, R. F., Pani, F., Siwy, Z., Krasowska, M., Grzywna, Z. et al. (
2005
). Voltage-gated sodium channel expression and potentiation of human breast cancer metastasis.
Clin. Cancer Res.
11
,
5381
-5389.
Goswami, S., Sahai, E., Wyckoff, J. B., Cammer, M., Cox, D., Pixley, F. J., Stanley, E. R., Segall, J. E. and Condeelis, J. S. (
2005
). Macrophages promote the invasion of breast carcinoma cells via a colony-stimulating factor-1/epidermal growth factor paracrine loop.
Cancer Res.
65
,
5278
-5283.
Grimes, J. A., Fraser, S. P., Stephens, G. J., Downing, J. E., Laniado, M. E., Foster, C. S., Abel, P. D. and Djamgoz, M. B. (
1995
). Differential expression of voltage-activated Na+ currents in two prostatic tumour cell lines: contribution to invasiveness in vitro.
FEBS Lett.
369
,
290
-294.
Gschwind, A., Prenzel, N. and Ullrich, A. (
2002
). Lysophosphatidic acid-induced squamous cell carcinoma cell proliferation and motility involves epidermal growth factor receptor signal transactivation.
Cancer Res.
62
,
6329
-6336.
Hill, K., Welti, S., Yu, J. H., Murray, J. T., Yip, S. C., Condeelis, J. S., Segall, J. E. and Backer, J. M. (
2000
). Specific requirement for the p85-p110 alpha phosphatidylinositol 3-kinase during epidermal growth factor-stimulated actin nucleation in breast cancer cells.
J. Biol. Chem.
275
,
3741
-3744.
Hirsch, D. S., Shen, Y. and Wu, W. J. (
2006
). Growth and motility inhibition of breast cancer cells by epidermal growth factor receptor degradation is correlated with inactivation of Cdc42.
Cancer Res.
66
,
3523
-3530.
Jeong, H. W., Nam, J. O. and Kim, I. S. (
2005
). The COOH-terminal end of R-Ras alters the motility and morphology of breast epithelial cells through Rho/Rho-kinase.
Cancer Res.
65
,
507
-515.
Kaufmann, A. M., Lichtner, R. B., Schirrmacher, V. and Khazaie, K. (
1996
). Induction of apoptosis by EGF receptor in rat mammary adenocarcinoma cells coincides with enhanced spontaneous tumour metastasis.
Oncogene
13
,
2349
-2358.
Kundra, V., Soker, S. and Zetter, B. R. (
1994
). Excess early signaling activity inhibits cellular chemotaxis toward PDGF-BB.
Oncogene
9
,
1429
-1435.
Laniado, M. E., Lalani, E. N., Fraser, S. P., Grimes, J. A., Bhangal, G., Djamgoz, M. B. and Abel, P. D. (
1997
). Expression and functional analysis of voltage-activated Na+ channels in human prostate cancer cell lines and their contribution to invasion in vitro.
Am. J. Pathol.
150
,
1213
-1221.
Lawrence, J. A. and Steeg, P. S. (
1996
). Mechanisms of tumor invasion and metastasis.
World J. Urol.
14
,
124
-130.
Levine, M. D., Liotta, L. A. and Stracke, M. L. (
1995
). Stimulation and regulation of tumor cell motility in invasion and metastasis.
EXS
74
,
157
-179.
Lichtner, R. B., Kaufmann, A. M., Kittmann, A., Rohde-Schulz, B., Walter, J., Williams, L., Ullrich, A., Schirrmacher, V. and Khazaie, K. (
1995
). Ligand mediated activation of ectopic EGF receptor promotes matrix protein adhesion and lung colonization of rat mammary adenocarcinoma cells.
Oncogene
10
,
1823
-1832.
Marino, A. A., Morris, D. M., Schwalke, M. A., Iliev, I. G. and Rogers, S. (
1994
). Electrical potential measurements in human breast cancer and benigh lesions.
Tumour Biol.
15
,
147
-152.
McCaig, C. D. and Zhao, M. (
1997
). Physiological electric fields modify cell behaviour.
BioEssays
19
,
818
-826.
McCaig, C. D., Rajnicek, A. M., Song, B. and Zhao, M. (
2002
). Has electrical growth cone guidance found its potential?
Trends Neurosci.
25
,
354
-359.
McCaig, C. D., Rajnicek, A. M., Song, B. and Zhao, M. (
2005
). Controlling cell behaviour electrically: Current views and future potential.
Physiol. Rev.
85
,
943
-978.
Neri, A., Welch, D., Kawaguchi, T. and Nicolson, G. L. (
1982
). Development and biologic properties of malignant cell sublines and clones of a spontaneously metastasizing rat mammary adenocarcinoma.
J. Natl. Cancer Inst.
68
,
507
-517.
Nuccitelli, R. (
2003
). A role for endogenous electric fields in wound healing.
Curr. Top. Dev. Biol.
58
,
1
-26.
O-charoenrat, P., Rhys-Evans, P. H., Modjtahedi, H. and Eccles, S. A. (
2002
). The role of c-erbB receptors and ligands in head and neck squamous cell carcinoma.
Oral Oncol.
38
,
627
-640.
Olivotto, I. and Levine, M. (
2001
). Clinical practice guidelines for the care and treatment of breast cancer: the management of ductal carcinoma in situ.
Can. Med. Assoc. J.
165
,
912
-913.
Pu, J. and Zhao, M. (
2005
). Golgi polarization in a strong electric field.
J. Cell. Sci.
118
,
1117
-1128.
Pullar, C. E. and Isseroff, R. R. (
2005
). Cyclic AMP mediates keratinocyte directional migration in an electric field.
J. Cell Sci.
118
,
2023
-2034.
Pullar, C. E., Baier, B. S., Kariya, Y., Russell, A. J., Horst, B. A., Marinkovich, M. P. and Isseroff, R. R. (
2006
). β4 Integrin and EGF coordinately regulate electric field-mediated directional migration via Rac1.
Mol. Biol. Cell
17
,
4925
-4935.
Riese, D. J., van Raaij, T. M., Plowman, G. D., Andrews, G. C. and Stern, D. F. (
1995
). The cellular response to neuregulins is governed by complex interactions of the erbB receptor family.
Mol. Cell. Biol.
15
,
5770
-5776.
Robinson, K. R. (
1985
). The responses of cells to electrical fields: a review.
J. Cell Biol.
101
,
2023
-2027.
Sauer, H., Hescheler, J., Reis, D., Diedershagen, H., Niedermeier, W. and Wartenberg, M. (
1997
). DC electrical field-induced c-fos expression and growth stimulation in multicellular prostate cancer spheroids.
Br. J. Cancer
75
,
1481
-1488.
Segall, J. E., Tyerech, S., Boselli, L., Masseling, S., Helft, J., Chan, A., Jones, J. and Condeelis, J. (
1996
). EGF stimulates lamellipod extension in metastatic mammary adenocarcinoma cells by an actin-dependent mechanism.
Clin. Exp. Metastasis
14
,
61
-72.
Szatkowski, M., Mycielska, M., Knowles, R., Kho, A. L. and Djamgoz, M. B. (
2000
). Electrophysiological recordings from the rat prostate gland in vitro: identified single-cell and transepithelial (lumen) potentials.
BJU Int.
86
,
1068
-1075.
Vastrik, I., Eickholt, B. J., Walsh, F. S., Ridley, A. and Doherty, P. (
1999
). Sema3A-induced growth-cone collapse is mediated by Rac1 amino acids 17-32.
Curr. Biol.
9
,
991
-998.
Verbeek, B. S., Adriaansen-Slot, S. S., Vroom, T. M., Beckers, T. and Rijksen, G. (
1998
). Overexpression of EGFR and c-erbB2 causes enhanced cell migration in human breast cancer cells and NIH3T3 fibroblasts.
FEBS Lett.
425
,
145
-150.
Verschueren, H., Van der Taelen, I., Dewit, J., De Braekeleer, J. and De Beastlier, P. (
1994
). Metastatic competence of BW5147 T-lymphoma cell lines is correlated with in vitro invasiveness, motility and F-actin content.
J. Leukocyte Biol.
55
,
552
-556.
Wang, E., Zhao, M., Forrester, J. V. and McCaig, C. D. (
2000
). Re-orientation and faster, directed migration of lens epithelial cells in a physiological electric field.
Exp. Eye Res.
71
,
91
-98.
Wang, E., Yin, Y., Zhao, M., Forrester, J. V. and McCaig, C. D. (
2003a
). Physiological electric fields control the G1/S phase cell cycle checkpoint to inhibit endothelial cell proliferation.
FASEB J.
17
,
458
-460.
Wang, E., Zhao, M., Forrester, J. V. and McCaig, C. D. (
2003b
). Electric fields and MAP kinase signaling can regulate early wound healing in lens epithelium.
Invest. Ophthalmol. Vis. Sci.
44
,
244
-249.
Wartenberg, M., Hescheler, J. and Sauer, H. (
1997
). Electrical fields enhance growth of cancer spheroids by reactive oxygen species and intracellular Ca2+.
Am. J. Physiol.
272
,
R1677
-R1683.
Welch, D. R., Neri, A. and Nicolson, G. L. (
1983
). Comparison of “spontaneous” and “experimental” metastasis using rat 13762 mammary adenocarcinoma metastatic cell clones.
Invasion Metastatis
3
,
65
-80.
Wellings, S. R. and Jensen, H. M. (
1973
). On the origin and progression of ductal carcinoma in the human breast.
J. Natl. Cancer Inst.
50
,
1111
-1118.
Wyckoff, J. B., Insel, L., Khazaie, K., Lichtner, R. B., Condeelis, J. S. and Segall, J. E. (
1998
). Suppression of ruffling by the EGF receptor in chemotactic cells.
Exp. Cell Res.
242
,
100
-109.
Wyckoff, J., Wang, W., Lin, E. Y., Wang, Y., Pixley, F., Stanley, E. R., Graf, T., Pollard, J. W., Segall, J. and Condeelis, J. (
2004
). A paracrine loop between tumor cells and macrophages is required for tumor cell migration in mammary tumors.
Cancer Res.
64
,
7022
-7029.
Xue, C., Wyckoff, J., Liang, F., Sidani, M., Violini, S., Tsai, K. L., Zhang, Z. Y., Sahai. E., Condeelis, J. and Segall. J. E. (
2006a
). Epidermal growth factor receptor overexpression results in increased tumor cell motility in vivo coordinately with enhanced intravasation and metastasis.
Cancer Res.
66
,
192
-197.
Xue, C., Liang, F., Mahmood, R., Vuolo, M., Wyckoff, J., Qian, H., Tsai, K. L., Kim, M., Locker, J., Zhang, Z. Y. et al. (
2006b
). ErbB3-dependent motility and intravasation in breast cancer metastasis.
Cancer Res.
66
,
1418
-1426.
Zhao, M., Agius-Fernandez, A., Forrester, J. V. and McCaig, C. D. (
1996
). Orientation and directed migration of cultured corneal epithelial cells in small electric fields are serum dependent.
J. Cell Sci.
109
,
1405
-1414.
Zhao, M., Forrester, J. V. and McCaig, C. D. (
1999
). A small, physiological electric field orients cell division.
Proc. Natl. Acad. Sci. USA
96
,
4942
-4946.
Zhao, M., Pu, J., Forrester, J. V. and McCaig, C. D. (
2002
). Membrane lipids, EGF receptors, and intracellular signals colocalize and are polarized in epithelial cells moving directionally in a physiological electric field.
FASEB J.
16
,
857
-859.
Zhao, M., Song, B., Pu, J., Wada, T., Reid, B., Tai, G., Wang, F., Guo, A., Walczysko, P., Gu, Y. et al. (
2006
). Electrical signals control wound healing through phosphatidylinositol-3-OH kinase-gamma and PTEN.
Nature
442
,
457
-460.

Supplementary information