|Home | About | Journals | Submit | Contact Us | Français|
Epithelia of the cornea, lens and retina contain a vast array of ion channels and pumps. Together they produce a polarized flow of ions in and out of cells, as well as across the epithelia. These naturally occurring ion fluxes are essential to the hydration and metabolism of the ocular tissues, especially for the avascular cornea and lens. The directional transport of ions generates electric fields and currents in those tissues. Applied electric fields affect migration, division and proliferation of ocular cells which are important in homeostasis and healing of the ocular tissues. Abnormalities in any of those aspects may underlie many ocular diseases, for example chronic corneal ulcers, posterior capsule opacity after cataract surgery, and retinopathies. Electric field-inducing cellular responses, termed electrical signaling here, therefore may be an unexpected yet powerful mechanism in regulating ocular cell behavior. Both endogenous electric fields and applied electric fields could be exploited to regulate ocular cells. We aim to briefly describe the physiology of the naturally occurring electrical activities in the corneal, lens, and retinal epithelia, to provide experimental evidence of the effects of electric fields on ocular cell behaviors, and to suggest possible clinical implications.
The eye is an electrically active organ, displaying many different types of electrical properties. The best-studied electrical activity in the eye is found in the retina, where incoming light excites the rod and cone photoreceptors, resulting in an electrical signal which propagates through the optic nerve and travels to the visual cortex where it is converted by the brain into an image. The corneal nerves may be excited by many types of stimuli. These are fast-acting electrical pulses like the action potentials of the nervous system.
The eye also has an electrical feature that is much less-well studied - slow-acting, long-lasting, steady, direct current (dc) electric fields (EFs). These include the transcorneal potential difference (TCPD), ionic currents at the lens, the transepithelial potential difference (TEP) of the retinal pigment epithelium (RPE) and the EFs/currents associated with injury (wound EFs/injury currents). In the cornea, ionic transport is required to transport water from the stroma to the tear film. The lens drives a unique current flow through itself that creates an internal microcirculation (Mathias et al., 2007). In the RPE ionic transport is required to move water from the subretinal space into the blood, and transport nutrients in the opposite direction, while also maintaining an optimal ionic composition in the subretinal space, required for photoreceptor excitation (Wimmers et al., 2007).
Accumulating experimental evidence suggests that this latter type of EF may provide an important signal that contributes to the control of cell behavior. The cells of the ocular tissues are exposed to EFs in their natural setting, and together with the finding that many cellular behaviors are affected by applied EFs of physiological magnitude, the role of EFs as a novel cell signaling mechanism is strongly suggested. We will use “electrical signaling” here specifically to describe this type of endogenous signaling mechanism and for signaling that may be incited by externally applied electric fields. Electrical signaling may represent a novel therapeutic strategy in ocular tissue repair and regeneration. Indeed, electrical stimulation is being tested clinically for the treatment of bone fracture, wound healing, and spinal cord injury.
This review aims to focus on 1) how the endogenous electrical activities are generated - the ionic transport mechanisms in the cornea, lens, and RPE; 2) the responses of ocular cells to applied EFs; 3) potential intracellular signaling mechanisms; and 4) clinical implications. For general and relevant background on electrical control of development, wound healing, and regeneration, please refer to the following elegant reviews (Funk et al., 2009; Levin, 2007; McCaig et al., 2005; McCaig et al., 2009; Mycielska and Djamgoz, 2004; Nuccitelli, 2003a, b; Ojingwa and Isseroff, 2003; Robinson and Messerli, 2003; Stewart et al., 2007).
Electric current is the movement or flow of electrically charged particles, while an EF is an influence produced by an electric charge on other charges in its vicinity. In a biological context, ions (e.g. Na+, K+, Cl−, Ca2+, etc.) are the charged particles whose flux constitutes electric currents. In biological tissue, which is a resistive as well as conductive medium, EFs form due to the presence of charged particles and the flow of electric currents. The EFs produced basically follow Ohm’s law which states that voltage equals current multiplied by resistance: V = I × R, where V is the voltage (measured in volts: V), I is the current (measured in amps: A), and R is the resistance (measured in ohms: Ω), although in biological tissue the situation is far more complex.
The resistivity is a measure of how strongly a material opposes the flow of electric current. Biological fluids typically have a resistivity value of ~100 Ω-cm. The resistivity of tear and aqueous humor is ~70 Ω-cm and that for corneal stroma is ~100 Ω-cm. Endothelium and epithelium have much higher resistivity values of ~1,500 and 250,000 Ω-cm respectively (Gabriel et al., 1996; Guimera et al., 2010; Klyce, 1972; Lim and Fischbarg, 1981). When there is a voltage difference between any two points in conductive media current flows. In biological tissues or solutions, the EF and electric currents are therefore inseparable. The relationship between the current density and the EF is E = J × ρ, where ρ is the static resistivity (measured in ohm-metres, Ω-m), E is the magnitude of the electric field (measured in volts per metre, V/m), and J is the magnitude of the current density (measured in amperes per square metre, A/m2). Most often in literature, the resistance value is used to quantify electrical resistance across a membrane structure, such as the cornea, epithelial and endothelial layers. The resistance across a membrane structure with defined area is expressed as the resistance times the areas, and presented as Ω-square centimeter. For example, the resistance of rabbit corneal endothelium was measured to be ~70 Ω-cm2 (Lim and Fischbarg, 1981).
Live cells maintain electrical potentials of varying magnitude across their plasma membranes (Fig. 1A). This is achieved by the active, polarized transport of ions by pumps and transporters, together with the selective permeability of channels in the cell membrane. Excitable cells such as neurons and muscle fibers have large membrane potentials (~70 mV), a voltage difference negative inside relative to outside the cells. Epithelial cells in cornea, lens, and retina also have significant membrane potentials. However, the membrane potential of these non-excitable cells is smaller (~30 mV) (Lichey et al., 1974; Mergler and Pleyer, 2007; Rae, 1979; Wiederholt and Koch, 1979).
Analogous to the cell membrane potential, epithelial cells form a “membrane” that collectively transports ions across the whole epithelial layer. At the same time, the epithelium maintains tight junctions between the cells preventing free diffusion of ions. The energy consuming transport of ions coupled with the epithelial barrier establish a steady, long-lasting TEP, similar to the membrane potential in cells (Fig. 1B). In the cornea, the TEP contributes to the generation of the TCPD. In the lens, large and consistent currents flow out from the equator and enter the anterior and posterior poles. In the retina, the RPE also generates and maintains a TEP.
EFs and current are vectors and it is this directional quality that makes EFs a candidate spatial organizer that is capable of imposing directional movement on molecules, cells, and tissues. The facts that ion channels and pumps are often localized to specific cells or parts of cells, and that the local resistance of cells or tissues may vary due to spatial variation in cell packing or tight junction density, means that EFs could be regulated spatially and temporally (McCaig et al., 2005).
Electrical potential differences (PDs) exist between the three layers of the cornea: the epithelium, stroma, and endothelium. Transport of ions and segregation of the ions in different tissue compartments generates the PD or EFs. We will firstly describe some ion transport mechanisms, then the electric potentials in the normal cornea and at corneal wounds, followed by experimental results on how corneal cells may be affected by such EFs.
The human corneal epithelium is about 50 µm thick consisting of 5–7 layers of continuously renewing cells. The epithelial cells are joined by tight junctions and function as a semi-permeable, highly-electrically resistant membrane (Schermer et al., 1986; Tuli, 2005). The stroma makes up ~90% of the total corneal thickness and is primarily composed of collagen fibrils which are oriented in a parallel manner and organized in lamellae (Maurice, 1957). The endothelium consists of a monolayer of hexagonal cells (Konomi et al., 2005).
To remain transparent, corneal water balance/hydration level (deturgesence) must be regulated within a specific physiological range (Donn et al., 1959; Klyce, 1975, 1977; Klyce and Crosson, 1985; Klyce and Wong, 1977; Maurice, 1951, 1967, 1972; Reinach et al., 2008b). Transport of ions coupled with water transport regulate the osmolarity balance between the corneal stroma, the anterior chamber, and the tear fluid, thus regulating deturgesence (Edelhauser, 2006). The corneal endothelium is responsible for ~ 90% of the fluid transport of the rabbit cornea. However, ion transport activity in the corneal epithelium also plays a significant role in maintaining proper hydration. Most of the following description will be on the corneal epithelium, because it maintains a far greater PD and generates significant wound EFs. We emphasize those aspects more relevant to generation of electric currents/EFs in corneal tissues. For more comprehensive reviews on ion transport and hydration of the cornea, please refer to these in-depth reviews and papers (Bonanno, 2011; Candia and Alvarez, 2008; Dikstein and Maurice, 1972a, b; Fischbarg and Lim, 1974; Fischbarg and Maurice, 2004; Hodson, 1974; Hodson and Miller, 1976; Klyce, 1977; Klyce and Wong, 1977; Mergler and Pleyer, 2007; Riley et al., 1995).
Corneal epithelial cells (CECs) and endothelial cells express a wide spectrum of ion channels, transporters and pumps (Fig. 2). These include K+, Na+, and Cl− channels; transient receptor potential (TRP) channels; sodium proton exchangers (NHE); K+/Cl− co-transporters (KCC); Cl−/HCO3− exchangers (CBE); Na+/Ca2+ exchangers (SCE); Na+/K+/Cl− co-transporters (NKCC); Na+/K+-ATPase; and Ca2+/Mg2+-ATPase. The functional expression of these diverse ion transporters and channels has been demonstrated by genetic, molecular, biochemical, immunostaining, and electrophysiological studies. These channels and transporters play important roles in regulating hydration levels of the cells, as well as the cornea as a whole (Reinach et al., 2008b).
Na+/K+-ATPase is responsible for maintaining ionic gradients between the extracellular and intracellular spaces, required for cellular homeostasis and for the transepithelial movement of water and organic solutes in response to the differential osmotic gradients (Tsuji et al., 1992). Na+/K+-ATPase is a multigene family of proteins with three α isoforms (α1, α2, α3), three β isoforms (β1, β2, β3) (Gick et al., 1993; Sweadner, 1989), and sometimes a γ subunit (Geering, 2006). The α subunit hydrolyzes ATP to ADP and in doing so shifts three Na+ ions out of the cell and two K+ ions into the cell (Delamere and Tamiya, 2004), while the β and γ subunits appear to have modulatory roles (Mathias et al., 2007). The α1 subunit was found to be expressed predominantly in corneal endothelial cells and along the basolateral surfaces of basal epithelial cells. The α2 and α3 subunits distributed homogenously in the epithelial layer. Corneas affected by Fuch’s’ dystrophy, a corneal adematous disease, show reduced epithelial immunostaining for Na+/K+-ATPase subunits. Other alterations in epithelial Na+/K+-ATPase point to the importance of epithelial changes in the development of corneal adematous diseases (Ljubimov et al., 2002).
Two isoforms of the Na+/K+/Cl− co-transporter (NKCC1 and NKCC2) are expressed in corneal epithelial and endothelial cells (Bildin et al., 2001a). Na+/K+-ATPase and NKCC are suggested to be the major driving force for apical to basal transport of Na+ and basal to tear transport of Cl− (Reinach et al., 2008b).
Cl− channels are expressed in both corneal epithelium and endothelium. Messenger RNA coding ClC Cl− channels (ClC-2, -3, -5, -6, -7), cystic fibrosis transmembrane conductance regulator (CFTR), and Ca2+-activated Cl− channel-1 (CLCA-1) are positive in CECs (Fig. 3A). Some protein expression has also been confirmed (Fig. 3B), and Cl− currents are present in human CECs (Fig. 4) (Cao et al., 2010). ClC-2 and ClC-3 immunolabeling confirmed the presence of these proteins in corneal epithelium (Davies et al., 2004). CFTR and CLCA-1 are also expressed in human CECs (Al-Nakkash et al., 2004; Al-Nakkash and Reinach, 2001; Connon et al., 2006; Itoh et al., 2000). Significantly for directional ion transport, channels (e.g. ClC-2 and a CLCA) are distributed asymmetrically at the apical and basal layer of the corneal epithelium, respectively (Cao et al., 2010; Connon et al., 2006; Davies et al., 2004). CFTR plays a significant role in regulating the TCPD, as CFTR null mice fail to respond to perfusion of low Cl− solution (Levin et al., 2006).
K+ channel activities may be mediated by cholinergic receptor activation, Ca2+ level, arachidonic acid, and some fatty acids (Farrugia and Rae, 1992; Bockman et al., 1998; Takahira et al., 2001). K+ channel activity modulates adrenoceptor stimulation of net Cl− transport in CECs (Wolosin and Candia, 1987). A specific type of K+ channel (Kv3.4) is expressed widely in CECs and is activated by ultraviolet irradiation (Lu, 2006). The hyperactivation of Kv3.4 results in rapid intracellular K+ loss and cell volume shrinkage followed by apoptosis.
CECs express L-type Ca2+ channels and other Ca2+ influx pathways (Reinach et al., 2008a; Rich and Rae, 1995). Plasma membrane Ca2+-ATPase is expressed in human and rabbit corneal epithelium. During wound healing, Ca2+-ATPase changed expression distribution. More Ca2+-ATPase was expressed at the basal side of the epithelial cells migrating into the wound bed (Talarico, 2010).
Some channels and transporters have unique spatial localization. In contrast to the basolateral distribution of the α1 subunit of Na+/K+-ATPase, some Cl− channels showed mostly apical distribution, and importantly CLC-6 and CFTR had a gradient of high expression toward the apical side (Fig. 5). The spatial difference in Na+/K+-ATPase and Cl− channels suggests directional transport of cations and anions. This is consistent with the measurements of ion transport, demonstrating directional transport of different ions described above (Candia and Alvarez, 2008; Fischbarg and Maurice, 2004; Klyce and Wong, 1977).
Baso-lateral distribution of Na+/K+-ATPase suggests there are functionally inward (tear to basal) fluxes of positive charges making the basal side electrically positive. This is because, when activated, each transport cycle of Na+/K+-ATPase pumps two K+ ions into the cell and three Na+ ions out of the cell (Jorgensen et al., 2003).
Active Cl− transport is a secondary active process, because Cl− uptake from the stroma into the epithelial layer is dependent upon basolateral membrane-coordinated Na+/K+-ATPase and NKCC activity. Pump coupling to NKCC function is a result of Na+/K+-ATPase establishing a bath-to-cell inwardly directed Na+ gradient, which provides the chemical driving force for NKCC-mediated uphill intracellular Cl− accumulation. Na+/K+-ATPase activity has a variable load-dependent Na+/K+ stoichiometry (Candia et al., 1984; Reinach et al., 2008a).
The specific ionic fluxes that account for ion transport in corneal epithelia vary among species. In amphibians, it is essentially Cl− flux (Zadunaisky and Lande, 1971). An inward flow of Na+ ions has been observed in rabbit and frog cornea (Candia and Askew, 1968; Klyce, 1975; Reid et al., 2005; Van der Heyden et al., 1975), while Cl− is actively transported outward from the aqueous humor, across the stroma and epithelium, into the tear film (Klyce and Wong, 1977; Reid et al., 2005; Zadunaisky, 1966). In mammals, this mechanism is accounted for by approximately equal contributions by inwardly directed net Na+ transport toward the stroma and outwardly directed active Cl− transport into the tears (Klyce, 1975).
Transport of electrolytes is coupled with the transport of water between the stroma and tear fluid. Corneal epithelium transports Na+ from the tear side to the basal side, and Cl− in the opposite direction (Candia, 2004; Fischbarg et al., 2006; Klyce and Crosson, 1985; Levin and Verkman, 2005; Yang et al., 2001). Cl− is actively transported cellularly. The active transport of Cl− to the tear side creates a negative potential on this side. To maintain electroneutrality of solutions, the negative potential generated at the tear side draws Na+ to move passively through the paracellular space to the tears to produce a net transport of NaCl. This would create a net osmotic gradient. Together with fluxes of other ions, e.g. HCO3−, the net transport generates osmotic gradients that drive water transport, with Cl− secretion providing a primary driving force for water transport across the corneal epithelium (Fischbarg, 1997).
Aquaporins (AQPs) facilitate water transport in response to small osmotic gradients produced by salt pumping. The corneal epithelia of mouse, rat, and human express water-selective AQPs, AQP5 and AQP3. (Hamann et al., 1998; Levin and Verkman, 2004; Levin, 2005; Patil et al., 1997). The functional significance of various AQPs in ocular tissues has been confirmed in AQP-deficient mice. AQP deletion reduces transepithelial water permeability and hence the amount of fluid absorbed or secreted (Verkman et al., 2008). Altered AQP expression may play a role in various human corneal diseases associated with fluid imbalance (Levin and Verkman, 2006; Levin, 2005; Rabinowitz et al., 2005).
Neuronal and non-neuronal mechanisms may regulate the functioning of ion transporters. The regulatory mechanisms appear to have interspecies differences. Adrenergic, serotonergic, and cholinergic receptors contribute to the control of ion transport activity (Klyce et al., 1982; Pesin and Candia, 1982a, b; Zadunaisky et al., 1973). The intracellular Cl− accumulated through Cl− channels egresses into the tear film (Fig. 2). These receptors activate several signaling pathways that control the activity of Cl− channels in the corneal epithelium. Adrenergic signaling pathways elicit increases in fluid secretion into the tears through transient cytosolic increases in cAMP and intracellular Ca2+. In addition to neuronal control, the active transport is regulated through many mechanisms, including those that are very important for wound healing, e.g. epidermal growth factor (EGF) increases NKCC activity in rat CECs and the increase in its activity lasts for more than an hour (Yang et al., 2000). Within the ClC family, Cl− channels may be regulated by the cAMP-activated transmembrane conductance regulator (CFTR), Ca2+-activated Cl− channels (CaCC) - ClCA2, and volume-regulated anion channels (Reinach et al., 2008a).
Growth factors, cytokines, ATP, mechanical forces, and UV irradiation are all known to induce ion fluxes in CECs (Lu, 2006; Reinach et al., 2008a). For example, growth factor receptor stimulation induced the activation of other types of K+ channels through induction of CEC proliferation processes (Roderick et al., 2003). Therefore, selective K+ channel activation via diverse receptor-activated events is essential for determining whether this activation induces cell death or promotes cell proliferation.
The corneal endothelium transports water and electrolytes between the stroma and anterior aqueous chamber. The corneal endothelium provides a pathway for nutrient uptake and waste removal via simple and facilitated diffusion, and secondary active transport mechanisms such as facilitated glucose transporter, glut-1 (Bildin et al., 2001b), lactate/H+, and lactate/Na+ co-transporters (Giasson and Bonanno, 1994). The endothelium also secretes fluid via ionic transport in order to counterbalance the continuous leak of fluid into the stroma (Bonanno, 2003). The “pump-leak” mechanism maintains corneal deturgesence and transparency by ensuring no net fluid transport occurs, as active secretion out of the cornea counterbalances the leak of fluid into the cornea (Maurice, 1972).
The polarized transport of ions, selective permeability to different ions, and electrical resistance between different compartments of tissues establish electric potential gradients in the cornea. Electrical PDs across the whole cornea (TCPDs) have been demonstrated with in vitro and in vivo measurements in many different animals. This was first described over half a century ago (Friedman and Kupfer, 1960; Ito et al., 2000; Klyce, 1972; Modrell and Potts, 1959; Potts and Modrell, 1957; van der Bijl et al., 2004). In rabbit, mouse, and frog corneas, TCPD measurements range from 15–33 mV, with the tear side negative (Klyce, 1972; Levin et al., 2006).
The corneal epithelium is electrogenic, which is to say that it is capable of producing its own electrical activity. As a collective layer, the epithelium produces a uniform and consistent TEP (Candia et al., 1968; Chiang et al., 1992; Fischer et al., 1972; Foulds and Barker, 1983; Lindemann, 1968). Detailed measurements demonstrated that the TEP contributes to over 90% of the TCPD. In various species, measurements of the corneal TEP have been found to range from ~20 to 40 mV (basal side positive compared to the tear side) (Fig. 6) (Akaike and Kiyohara, 1977; Candia et al., 1968; Chiang et al., 1992; Fischer et al., 1972; Klyce, 1972; Klyce and Crosson, 1985; Lindemann, 1968; Maurice, 1967). In isolated rabbit eyes, the TEP was identical in polarity with almost the same magnitude, to the simultaneously measured TCPD (Klyce, 1972).
In rabbit corneal epithelium, Klyce showed three iso-potential regions: beneath the superficial layer, the region between squamous and wing cells, and the transition region between wing and basal cells (Fig. 6). These three iso-potential regions correspond to the three resistance regions, with the first superficial layer having the highest electrical resistance (Fig. 6C).
Corneal endothelium is extremely leaky with the electrical resistance could be as low as only 25 Ω-cm2, due to its discontinuous tight junctions (Bonanno, 2003). The corneal transendothelial potential is only 0.5–1 mV, ~30–60 times smaller than the corneal TEP (Fischbarg and Lim, 1974; Hodson, 1977; Hodson and Miller, 1976). The transendothelial potential is dependent on the activity of Na+/K+-ATPase, which like most epithelial cells is located on the basolateral membrane of corneal endothelial cells, (Guggenheim and Hodson, 1994), and HCO3− (Hodson, 1974; Hodson and Miller, 1976), but to a lesser extent on the activity of Cl− transporters (Barfort and Maurice, 1974; Fischbarg and Lim, 1974; Hodson, 1971).
Corneal transparency depends on regulation of the hydration of the corneal stroma. The stromal swelling pressure drives water into the cornea, while endothelial and epithelial layers counterbalance the pressure by their capacity to transport ions and water. Ion fluxes from the stromal to anterior chamber side, and to the tear film through channels and pumps, are responsible for the counterbalancing force. The identity and location of all of the components of this transport system are being gradually elucidated (Bonanno, 2003; Candia and Alvarez, 2008; Fischbarg and Maurice, 2004). In individual cells, the pumps and channels respond to osmotic challenge and regulate cell volume (Fig. 2) (Reinach et al., 2008a). The epithelium, with multiple layers and many cells as a functional unit contributes to regulation of stromal hydration through activation of asymmetrically distributed Na+/K+-ATPase, and other ion channels (Fig. 5, ,7)7) (Cao et al., 2010; Ljubimov et al., 2002).
While the role of ionic transport in maintaining corneal hydration has been the focus of most previous investigations, some less-recognized research suggests another potentially important role for ionic transport - in generating wound EFs. Recent experiments have produced strong evidence that the EFs thus generated may have an important role in corneal epithelial wound healing (Chiang et al., 1992; Reid et al., 2007; Reid et al., 2005; Zhao, 2009; Zhao et al., 2006). We will present a detailed account of wound-related EFs and cellular effects.
Injury disrupts the continuity of the epithelium, breaking down the tight junctions between cells that help maintain the PD, thus creating a short-circuit at the wound, allowing ions - Na+, Cl−, K+ etc. to flow into the wound freely from surrounding tissue (Fig. 7). Because ions can flow freely at the wound site, the PD here drops to 0 mV. The PD is maintained however, in the surrounding intact epithelium. The more positive potential underneath the intact epithelium relative to the wound drives electric currents (by convention, the flow of positive charge) into the wound, resulting in endogenous EFs pointing to the wound center (red arrow in Fig. 7B). The electrical differential between wounded and intact areas is known as the “wound EF”. Many different techniques have confirmed very consistently the existence of wound EFs (Chiang et al., 1992; Reid et al., 2007; Reid et al., 2005; Zhao, 2009; Zhao et al., 2006).
Chiang and co-workers, using micro-glass electrodes, measured a lateral potential drop along the outer surface of the cornea at the edge of wounds made in the corneal epithelium of the isolated bovine eye. Microglass electrodes measured lateral corneal wound EFs of ~42 mV/mm in the first 0.25 mm from the wound edge on the surface of the eye (Chiang et al., 1992). Because of the limitation of the spatial resolution of the measuring technique, the authors stated that “the 40 mV/mm average value reported here for the lateral EFs measured on the surface of the cornea at the edge of wounds made in the epithelium is almost certainly a minimum one”. They believe measurements with higher spatial resolution made closer to the edge of the wound would likely yield even higher values. Indeed, this value is smaller than those measured at skin wounds using other independent techniques such as the bioelectric field Imager. In mouse and human skin, 177 mV/mm has been measured with the bioelectric field imager (Nuccitelli et al., 2008).
Another different technique, the vibrating probe, is an ultra-sensitive micro-probe used to measure electric current non-invasively. This technique has permitted measurement of growth-, development-, and wound-related electric currents in a wide variety of organisms (Jaffe and Nuccitelli, 1974). Using the vibrating probe, we measured outward currents of up to 10 µA/cm2 in corneal wounds from rat, bovine, and human (Fig. 8) (Reid et al., 2010; Reid et al., 2007; Reid et al., 2005). In human corneal wounds, currents of 0.41 µA/cm2 were measured. The significantly weaker wound currents in human corneas compared to those in fresh rat and bovine corneas is likely due to the extended storage of the corneas.
It may first appear that the corneal wound site being positive (anode) is in conflict with the idea of the field guiding cells into the wounds, because epithelial cells in vitro migrate toward the negative cathode (please see section 3.4. Corneal cells respond to electric fields). Direct measurements of the lateral fields at corneal wounds showed that the wound was relatively positive (Chiang et al., 1992). It should be noted that the microglass electrodes Chiang et al. used measured the surface EFs in the tear film. This is the return path of the currents (flow of positive charge) (blue arrow in Fig. 7B) driven by the positive TEP below the epithelial layers, out of the wound margin. The bioelectric field imager and vibrating probe also measure EFs and currents on the surface of the wound and the outside vicinity of the wound.
However, what is more relevant is the lateral fields immediately below the upper epithelial layer, where the tight junctions exist and across which the TCPD is established (Klyce, 1972; Wolosin, 1988). Immediately below the migrating surface epithelia, cells are tightly packed, with little extracellular space. The electric currents, by the laws of physics, must form a complete circuit. The current flowing away from the wound on the outer surface of the epithelium (blue dashed arrow in Fig. 7B) must flow through the epithelium back to the basal side to complete the circuit. The ion channels and pumps in the cells at the wound edge thus keep transporting ions (black arrows in Fig. 7B). This transport is likely to have an active component in an effort to maintain the TEP, like a battery, to keep the current flowing in the direction of the wound in the healing epithelium (red arrow in Fig. 7B). This is the direction in which active cell migration occurs. Tissue resistance may be substantial and may establish larger lateral fields with the wound being negative (red arrow in Fig. 7B). Thus, the most superficial layer of epithelium may “see” the wound as an anode, whereas the lower surface would see a cathode at the wound.
A new technique, the microneedle array, has demonstrated such voltage gradients in the epidermis of human skin with the field vector pointing toward the wound. The microneedle arrays were placed on skin with the electrode tips piercing and recording in-depth below the skin surface so that they lie below the stratum corneum, but above the basement membrane. The results confirmed the existence of a variation in skin potential that is maximum near the wound, and a resulting transverse EF directed toward the wound (more negative potential approaching the wound). The field strength at the wound edge is ~40 mV/mm (Mukerjee et al., 2006). This recording confirmed the generation of a lateral EF in human skin with the field direction pointing toward the wound in epidermis. This measurement is consistent with Nuccitelli’s estimation based on measurements with the bioelectric field imager - about 40 mV/mm within the epidermis with the negative pole of the field at the wound site (Nuccitelli et al., 2008).
Thus, the EF is orientated in the same direction that CECs migrate and proliferate to heal the wound. Wounding establishes a steady, laterally-oriented EF with the cathode at the wound centre. The current runs laterally under the basal surfaces of the epithelial cells and returns laterally within the tear film across the apical surface. The TEP in the surrounding intact epithelium ‘pushes’ ions out of the wound, producing a large and rapidly increasing electric current (Reid et al., 2005). This ion flow generates EFs in the surrounding tissues, orientated toward the wound site (Fig. 7).
Mammalian skin also maintains a TEP, and the first person to measure injury currents was Emil Du-Bois Reymond who, using a galvanometer, noted a current exiting a wound on his finger (Du Bois-Reymond, 1843; Du Bois-Reymond, 1860). When the skin is cut, a wound-induced EF arises for the same reasons as outlined for corneal epithelium, with the current returning under the cornified layer. Importantly, wound-induced EFs persist until the migrating epithelial cells at the leading edges re-seal the wound and re-establish a high electrical resistance across the tissue (Danjo and Gipson, 1998; Hudspeth, 1982; Keese et al., 2004).
EFs have been described as a leakage at the wound site, i.e. a passive process. However, corneal wound EFs contain an actively-regulated component. This is supported by the following experimental evidence. Firstly, the time course of wound electric currents showed a gradual increase after wounding. It takes 60–90 min for the currents to reach a peak after wounding rat cornea (Reid et al., 2005). If it is a pure leakage, the currents should reach the peak immediately or very soon after the epithelial breakdown. Secondly, drug treatment with aminophylline and other drugs which enhance Cl− transport significantly enhanced the electric currents, while inhibition of ion transportation significantly reduced the currents (Reid et al., 2010; Reid et al., 2007; Reid et al., 2005). Our recent experiments showed that wounding induced changes in the expression level and distribution of Na+/K+ ATPase and some Cl− channels (Cao et al., unpublished results).
Electric currents at human skin wounds also have a dynamic time course. At accidentally amputated finger tips, the currents reached a peak 6–8 days after amputation in children who regenerated the tip (Illingworth and Barker, 1980). The regulation of the current size appeared to coincide with the expression level of Na+/K+ ATPase at skin wounds, which gradually increased until day 7 and diminished at day 9 (Dube et al., 2010). The wound electrical signaling thus, may be actively regulated, through expression and activity of the pumps and channels. It could also be regulated by modification of tight junctions.
At corneal wounds, ions such as Na+, K+, Ca2+, Cl− and Mg2+ among others may contribute to the overall net electric current. Using ion-substituted artificial tear solution and vibrating probe techniques, we found that Cl− and Na+ are the major components of electric currents in rat corneal wounds. Na+ appeared to be the major component of ionic transport in the resting (non-wounded) rat cornea and of the wound center leakage current, whereas Cl− was a more important component of the endogenous electrical current at the wound edges (Fig. 9).
Ion-selective probes allow direct measurement of the contribution of each specific ion to the wound electric current. At rat cornea wounds, Ca2+ efflux increased steadily, whereas K+ showed an initial large efflux which rapidly decreased. A significant observation was a persistent large influx of Cl−, which had a time course similar to the net wound electric currents (Fig. 10) (Vieira et al., 2011).
Therefore, the outward electric currents occurring naturally at rat corneal wounds are carried mainly by a large influx of Cl− ions, and in part by effluxes of Na+, Ca2+ and K+ ions. Ca2+ and Cl− fluxes appear to be mainly actively regulated, while K+ flux appears to be largely due to leakage. The dynamic changes of electric currents and specific ion fluxes after wounding suggest that electrical signaling is an active response to injury.
Corneal cells migrate and proliferate in the presence of endogenous EFs, especially when the cornea is wounded. Applied EFs have been used to mimic the endogenous EFs to test their effects on cultured CECs in vitro. We will not discuss the methods and techniques used to apply EFs to cultured cells and tissues, because they have been described in detail previously (Song et al., 2007; Tai et al., 2009). Experimental evidence suggests that endogenous wound EFs provide a powerful signal for corneal epithelial wound healing.
Cultured CECs respond to applied EFs of physiological magnitude by migrating toward the cathode (Fig. 11A). Soong et al. demonstrated that EFs induced directional migration (galvanotaxis or electrotaxis) of CECs in the cathodal direction, with a threshold of 4 V/cm and a latency of 20 min (Soong et al., 1990a). Subsequent investigation by other groups demonstrated a much smaller threshold (<100 mV/mm) and response time within 10 min after onset of the EF (Farboud et al., 2000; Zhao et al., 1996b).
Monolayers and sheets of epithelial cells also migrate directionally in an applied EF (Zhao et al., 1996a) (Fig. 11B). An EF as low as 12.5 mV/mm is able to direct the migration of bovine CECs in culture (Zhao, 2009; Zhao et al., 2006). This collective migration has not been shown in chemotaxis responses. Compared to dissociated cells, the monolayer or sheets of cells migrate more directionally in an EF of much lower voltage (Zhao, 2009; Zhao et al., 1996a, b). Cells coupled by gap junctions react as a single unit to an applied EF and it has been suggested that gap junction coupling may increase cellular sensitivity to weak EFs as the perturbations induced by an applied EF are greater for groups of cells that are electrically coupled than for single cells (Cooper and Keller, 1984). This is pertinent as the in vivo healing of corneal epithelial wounds is accomplished by en mass directional growth of epithelium into the wound. In vivo wound healing is led by closely associated groups of cells rather than isolated single cells (Imanishi et al., 2000; Lu et al., 2001; Suzuki et al., 2003).
Increasing field strength induces a stronger electrotaxis/galvanotaxis response in bovine as well as human CECs (Zhao et al., 1997). CECs from rat, bovine, and human respond to an EF as low as 25 mV/mm (0.5 mV across a cell of 20 µm diameter), well within the physiological range (Farboud et al., 2000; Zhao et al., 1996a, b; Zhao et al., 1997).
Multiple guidance cues exist at the wound to guide the cells to proliferate and migrate directionally into the wound. Generally accepted guidance cues include injury stimulation, contact inhibition release, free edge, chemotaxis, and mechanical forces. Injury disrupts the plasma membranes of the cells at the wound margin which induces Ca2+ entry and K+ release. Wounded cells and tissue may release chemical substances (growth factors, cytokines, ATP) that attract cells into the wound by chemotaxis (directional cell migration in response to chemical gradients). Contact inhibition release or free edge, and simply the availability of space may also provide cells with directional signals (Block et al., 2010; Zhao et al., 2006) (Fig. 12A).
We compared the above directional cues with applied EFs. Surprisingly we found that EFs of physiological strength guide CECs, overriding other co-existing directional signals. When applied EFs of physiological strength (as small as 25 mV/mm) are applied against those well accepted directional guidance cues, CECs migrate in the direction of the EFs and ignore the other cues (Zhao, 2009; Zhao et al., 2006) (Fig. 12B).
Stratified multi-layered epithelial cells in a healing wound also respond to an applied EF. The stratified CECs also move away from the wound or into the wound following an applied EF and ignore other directional cues (Fig. 13). No other known directional signals, either biochemical or biophysical, have such an overriding guidance effect.
CECs also elongate and reorientate so that their long axis comes to lie perpendicular to the EF vector (Soong et al., 1990a; Zhao et al., 1996b; Zhao et al., 1997) (see Fig. 11A, 1 and 6 hours). Non-dividing CECs normally established a long axis perpendicular to the EF vector and when they came to divide they did so with a cleavage plane perpendicular to the EF vector. In a group of eighteen dividing cells almost all aligned themselves with the division axis parallel to the EF and the cleavage plane perpendicular, suggesting that the mitotic spindle had aligned parallel to the EF vector (Zhao et al., 1999b).
Cell proliferation rate and division orientation correlate with the endogenous EFs at corneal wounds. The endogenous EFs at corneal wounds were manipulated using drugs with different pharmacological actions but sharing the common feature of affecting ion transport. The wound-induced EFs controlled the orientation of cell division; most epithelial cells divided with a cleavage plane parallel to the wound edge and perpendicular to the EF vector. Increasing or decreasing the EF pharmacologically, respectively increased or decreased the extent of oriented cell division. An endogenous EF also enhanced the frequency of cell division (Song et al., 2002).
Nerve sprouting in the healing epithelium also appears to be regulated by endogenous EFs. In a rat in vivo corneal model, similar pharmacological manipulation was used to enhance or inhibit the endogenous EF. The nerve sprouting, the direction of nerve growth, and the rate of epithelial wound healing correlated significantly with the wound-induced EFs (Song et al., 2004).
To achieve better wound healing in vivo is one ultimate goal of understanding the electric signaling at corneal wounds. Several attempts have been made with promising results. While experiments from different groups described above provide compelling and consistent evidence for a strong effect of EFs in guiding migration of CECs, in vivo experimental results are inconsistent and confusing.
Because the wound EFs are produced by ion transport, it is possible to pharmacologically manipulate the EFs. We used different pharmacological agents to stimulate or inhibit the TCPD and wound EFs. Increasing or decreasing wound EFs sped up or slowed down corneal epithelial wound healing, respectively (Reid et al., 2005; Song et al., 2004; Song et al., 2002; Zhao et al., 2006). For example, using AgNO3 which enhances epithelial ion transport and increases the TCPD (Klyce and Marshall, 1982), significantly increased the wound electric currents (Fig. 14A). Furosemide inhibits Cl− secretion and reduces the TCPD (Candia et al., 1986; Nagel and Carrasquer, 1989). Furosemide significantly reduced electric currents in rat corneal wounds. Application of these two, and several other similar drugs, consistently increased or decreased wound electric currents. Application of eye drops containing those drugs to rat corneal wounds in vivo sped up or slowed down wound healing, consistent with the effects on the wound electric currents (Fig. 14) (Reid et al., 2005; Song et al., 2004; Song et al., 2002; Zhao et al., 2006).
Sta Iglesia and Vanable used microglass electrodes to apply EFs to corneal wounds on bovine eyeball cultures (Sta Iglesia and Vanable, 1998). They found that the strength of the EFs in the vicinity of circular lesions in the bovine cornea influenced the rate of re-epithelialization. They treated the wounds with Na+-depleted Hanks' solution, or with the application of dc EFs. Both treatments increased the field strength and sped up wound healing significantly. When they reversed the field’s polarity, a similar pattern was also observed. Those results appear to suggest that the presence of the fields, regardless of polarity, is important for corneal wound healing. The authors suggested two possibilities. The first is that the fields promote epithelialization by some means other than guiding cell migration. The second is that populations of cells take part in the wound healing – one migrates toward the cathode, and the other migrates toward the anode.
A more recent study using Pax6+/− mice yielded some contradictory results. Pax6+/− mice have an expression dose-dependent phenotype (Dora et al., 2008). One study reported that Pax6+/− mice tended to heal corneal epithelial wound faster (Ramaesh et al., 2006). Other reports found that Pax6+/− mice healed corneal epithelial wounds significantly slower (Dora et al., 2008; Leiper et al., 2006). Surprisingly, in some Pax6+/− mice, the endogenous electric currents at corneal wounds flowed in the opposite direction, inward, albeit only at about one-10th of the magnitude. Wound healing in those corneas was not different from that in wild type controls (Kucerova et al., 2011).
The very consistent directional effects of EFs on the migration of CECs in vitro, significant effects of direct application of EFs at bovine corneal wounds, and the complexity of the Pax6+/− mouse model call for further and more definitive experiments. Sta Iglesia and Vanable argue for the necessity of the presence of EFs, while in Pax6+/− mice, significantly reduced EFs or even reversed polarity did not affect healing. Both the directional and non-directional processes, such as activation of multiple signaling pathways will need to be investigated.
An externally applied dc EF guides the movement of cultured rabbit corneal stromal fibroblasts. The cells became spindle shaped, underwent galvanotropism by aligning their long axes perpendicular to the field, and migrated in the anodal (upfield) direction. The field strength for stromal cell migration appeared to be significantly higher (~6 V/cm) (Soong et al., 1990a; Soong et al., 1990b).
Cultured rabbit corneal endothelial cells responded to steady EFs (2–6 V/cm) by elongating their somata perpendicular to the field vector and by migrating toward the anode. During these directional movements, pseudopodia and ruffled membranes formed preferentially on the anodal side of the cells, while they retracted on the cathodal side (Chang et al., 1996).
Can the above information on corneal wound EFs contribute to the promotion of corneal wound healing? One possible approach we have explored is to modulate the ion transport, and therefore also the wound EFs at corneal wounds. Using donor human corneas, we have shown that electric current at human corneal wounds can be significantly increased with aminophylline or Cl−-free solution (Reid et al., 2010). Because corneal wound currents may regulate cell migration, cell division and nerve sprouting at wounds, pharmacological modulation of the endogenous wound EFs may offer a practical approach for chronic corneal wounds and related conditions. Substituting certain ions in the bathing solution also has significant effects on the electrical properties of the cornea. Eye drops could be developed with defined ionic components to facilitate corneal wound healing. This approach offers bioelectric stimulation without electrodes and can be readily tested in patients.
A significant piece of information lacking in this field is whether chronic or non-healing corneal wounds have defective electric signaling. The Pax+/− mouse model did show abnormal electric currents at corneal wounds, but due to the complicated healing response, the relationship between the endogenous EFs and wound healing remain inconclusive. Elucidation of wound EFs at chronic corneal wounds in diabetes and other conditions will significantly strengthen our understanding of this biophysical factor.
Direct application of EFs to the cornea is possible using a contact lens with built-in circuitry. Recent advances in micro-fabrication have produced an integrated amperometric contact lens (Cong and Pan, 2008; Yao et al., 2011). Similar contact lenses with integral circuitry may harvest and deliver electric currents to the cornea.
The ocular lens has a unique electric current flow which has been suggested to provide a circulatory mechanism to facilitate metabolisms in this avascular organ. In addition, the electric currents may have other effects on lens cells, such as regulating migration, proliferation, and differentiation. However, the experimental evidence is scarce and we would like to summarize the limited information to suggest this mostly under-explored role.
In order to focus light onto the retina, the lens must be crystal clear (Mathias et al., 1997). The absence of light-scattering elements requires an absence of blood vessels which would normally bring nutrients to cells and remove waste products. The lens consists of two types of cells - long fiber cells make up the bulk of the lens, while a monolayer of epithelial cells covers the anterior surface of the fibers (McAvoy and Chamberlain, 1989). The lens epithelial cells (LECs) proliferate, migrate, orientate, elongate, and differentiate into fiber cells at the equator. New fiber cells are continuously added throughout life, with the interior cells being the oldest, and they must remain viable for the entire lifetime of the organism (Mathias et al., 2007; Mathias et al., 1997; McAvoy and Chamberlain, 1989).
It is now known that the lens is not inert like nail or hair. Instead, it is a living, dynamic organ that carries out a host of biochemical and physiological processes necessary for homeostasis and possibly for regeneration. The lens has a unique pattern of electric currents, which have been proposed to serve as an alternative circulatory system.
The vibrating probe technique revealed these electric currents. Large electric currents (~26 µA/cm2 in frog and rat lens) flow inward at the anterior and posterior poles, and outward at the equator (Robinson and Patterson, 1982; Wind et al., 1988b) (Fig. 15A). Experimental results on expression patterns of membrane ion pumps and channels across the lens, cell electrophysiological parameters, and measurements of ion currents at the lens, all support such an electrical circulation (Mathias and Rae, 2004; Mathias et al., 1997). The large outward electric currents at the equator, and inward currents at the anterior and posterior poles, have been confirmed by many labs, ranging from ~20–40 µA/cm2 (Candia and Zamudio, 2002; Mathias et al., 2007; Mathias and Rae, 1985a, b; Mathias et al., 1997; Parmelee, 1986; Parmelee et al., 1985; Patterson, 1988; Robinson and Patterson, 1982; Wind et al., 1988a) (Fig. 15).
The physiological role of the ionic circulation has been hypothesized to be a micro-internal circulatory system. The lens “circulation” is likely accompanied by fluid flow which will bring nutrients and remove waste (Mathias et al., 2007; Mathias et al., 1997). The ionic circulation in this way will compensate for the lack of vasculature (Mathias et al., 2007).
Studies using the vibrating probe and Ussing-like chambers showed that Na+ and K+ fluxes are the major contributors to the circulating electric currents (Candia and Zamudio, 2002; Wind et al., 1988b). K+ efflux plus the Na+/K+-ATPase -generated current comprise the currents at the equatorial surface and an area anterior to it. Passive influx of Na+ occurs at both the anterior polar region and posterior lens surface. Inhibition of Na+/K+-ATPase with ouabain significantly reduced this ion flow (Wind et al., 1988a). The lens currents can be regulated. Beta-adrenergic agonists increased the equatorial current of the frog lens by a mechanism that is mediated by cAMP (Walsh and Patterson, 1993), while Ca2+ level and oxidants in the bathing solution also regulated the lens currents (Walsh and Patterson, 1989, 1991).
Detailed mechanisms of lens current generation and the physiological role of the current as an internal circulatory system have been reviewed in detail (Beebe and Truscott, 2010; Donaldson et al., 2010; Mathias et al., 2007; Mathias et al., 1997). Here we will mainly focus on the electrical properties of the ionic circulation, and its possible effects on migration, proliferation, and homeostasis of lens cells.
LECs and fiber cells have drastically different Na+/K+-ATPase activities. Na+/K+-ATPase activity is found predominantly in the LECs and surface fiber cells, although Western blot analysis determined that LECs and fiber cells express a similar amount of Na+/K+-ATPase protein (Delamere and Dean, 1993). However, the older, more internal fiber cells have very little Na+/K+-ATPase activity (Baghieri and Garner, 1992; Candia and Alvarez, 2008; Candia et al., 1971; Neville et al., 1978; Palva and Palkama, 1976; Delamere and Tamiya, 2004).
The Na+/K+-ATPase activities in LECs are spatially distributed. The subcellular distribution of Na+/K+-ATPase in LECs is on the basolateral side. Therefore, Na+/K+-ATPase activity drives net current from the apical side to the basal side. The equatorial regions of the lens have the highest activities, while cells in the anterior epithelium have a low activity. Using an Ussing-like chamber, and by local application of the Na+/K+-ATPase inhibitor ouabain, Candia and Zamudio mapped the surface distribution of the Na+/K+-ATPase current (Ip, µA/cm2). They concluded that Ip was too small to be detected at either pole but reached a maximum of 10 µA/cm2 at the equator (Candia and Zamudio, 2002). Whole-cell patch-clamp studies provided similar results (Gao et al., 2000; Tamiya et al., 2003). The asymmetrical distribution of Na+/K+-ATPase activity in the epithelium and fiber cells (and the equatorial and anterior epithelium) may contribute to the ionic currents and therefore the electrical currents that flow out at the equator and in at the poles.
A high K+ conductance has been found in LECs (Mathias et al., 2007; Mathias and Rae, 1985a; Mathias et al., 1985). K+ that is transported into the lens via the Na+/K+-ATPase, therefore mostly leaks out of the same cells (predominantly anterior epithelial cells). The lens expresses at least three major K+ channel types: inward rectifiers, large-conductance Ca2+-activated channels, and delayed rectifiers (Mathias et al., 1997). Lens cells have a small Na+ permeability but this leads to a significant total influx of Na+ due to the large number of fiber cells (Mathias et al., 2007). Low resistance gap junctions connect all fiber cells so when Na+ enters them it flows from cell to cell through these gap junctions until it reaches the surface epithelial cells and is then transported out of the lens by Na+/K+-ATPase (Fig. 15B). In this model of circulating Na+ current, extracellular current flows inward from the surface of the lens to the center, while intracellular current flows outward from the center to the surface. As well as Na+/K+-ATPase and fiber cell gap junctions, this model also requires the presence of fiber cell Na+ channels. Cl− plays a significant role in lens cell volume regulation but not in lens circulation under normal conditions (Mathias et al., 2007).
A link has been found between abnormal elevation of lens Na+ and the opacification of the lens cortex that occurs in age-related human cataract (Delamere and Tamiya, 2004). The transparent human lens contains some 17 mM Na+ and 130 mM K+ (data as mmol/kg lens water) (Patmore and Duncan, 1980). The levels of Na+ and K+ in the lens are established by the action of Na+/K+-ATPase which exports Na+ from the lens and imports K+ inward to balance ion leakage in the opposite direction. However, in human age-related cortical cataracts and human diabetic cataracts, lens Na+ is abnormally elevated (often higher than 100 mmol/kg lens water) and the K+ is abnormally low (often below 80 mmol/kg lens water) (Davies et al., 1984; Duncan and Bushell, 1975). As a result of the osmotic disturbance caused by the ionic changes, water accumulates in lens cells. Eventually this leads to cell lysis and the appearance of fluid droplets that scatter light and impair transparency (Delamere and Tamiya, 2004).
Wounding the lens by removal of part of the anterior capsule induces significant changes in the electric current pattern of the lens. Normal lenses have a small inward current across the anterior surface (0.22–0.38 µA/cm2) but removal of part of the anterior capsule and the lens nucleus resulted in larger outward currents (1.8–2.6 µA/cm2). A small cut ~1 mm in length significantly reduced outward equatorial currents to 56% of their original size and removing a larger part of the anterior capsule ~1 mm in diameter reduced equatorial currents to almost 0 µA/cm2 (0.4 ± 0.56 µA/cm2) (Lois et al., 2010; Wang et al., 2003c) (Fig. 16). The ionic components of these currents have not been determined. It is likely that multiple ions originally transported by the equatorial LECs make up the currents.
The large leakage or short circuitry due to disruption of the lens capsule and epithelium is likely to significantly reduce the PD across the lens epithelium. As discussed below (section 4.5.2. Applied Electric Fields Regulate Proliferation of Lens Epithelial Cells) the presence of an EF inhibited cell proliferation. One may argue that release of this inhibition mechanism may contribute to the aberrant proliferation after cataract surgery, which results in posterior capsule opacification (PCO).
LECs proliferate and migrate in the presence of constant ionic currents and accompanying EFs. The PD across the lens capsule measured ~−79 mV and −64 mV in frog and rat lens, respectively (Wind et al., 1988a). The thickness of the epithelium is ~5 µm. Theoretically, the voltage drop may reach 50 mV/5 µm, ~10,000 mV/mm across the epithelium layer in resting status. This may have some effects on lens cell migration and proliferation as discussed in the following sections.
LECs proliferate at the equator to give rise to new lens fiber cells and migration is initiated at the equator where cells encounter increasing gradients of growth factors from the ciliary body and fibroblast growth factor (FGF) from the vitreous body. The cells first divide and then migrate and initiate differentiation. Migration continues until the fiber cell contacts the anterior tip of a symmetrically located migrating fiber cell and the two cells join, detach from the basement membrane (lens capsule), and form a lens suture (Sue Menko, 2002).
Cultures of bovine and human LECs respond to applied EFs with directional migration and increased migratory rate, elongation, reorientation of the long axis of the cell, and an increase in the flattening size of individual cells (Fig. 17). Primary human LECs and primary bovine LECs, and a transformed human cell line all showed directed cell migration in a voltage-dependent manner, with a response threshold between 100–150 mV/mm and 25–50 mV/mm, respectively. Small sheets of LECs also migrated directionally. Reversing the polarity reversed the direction of migration of all three cell types (Wang et al., 2000) (Fig. 17). Primary human LECs migrated cathodally with a response threshold less than 100 mV/mm. Interestingly, LECs from different regions of the lens (where they will experience different EFs) moved in different directions (Wang et al., 2003b).
The migration of LECs from peripheral and central regions showed an intriguing phenomenon; in an EF of 150–250 mV/mm, both central and peripheral LECs migrated anodally, however, in a low EF of 50 mV/mm peripheral LECs migrated cathodally, while central LECs showed no response (Wang et al., 2000; Wang et al., 2003b). This is intriguing as it shows that cells derived from the same line can show different responses to an EF depending on 1) their location relative to the EF vector, 2) slight differences in their anatomical location and morphology, and 3) EF strength, suggesting that a mechanism more complex than a simple directional cue is in play. This happens to have parallels with the in vivo situation. LECs which are generated in the peripheral (equatorial) regions may migrate in one of two directions in vivo. Either they migrate toward the equator, elongate, and trans-differentiate, to become lens fiber cells, or very few cells may migrate in the opposite direction, toward the front of the lens, to compensate for ongoing apoptosis of LECs, as the epithelium turns over (Wang et al., 2003b).
LECs migrated at more than double the migration rate (compared to no EF control) in an EF of 50 mV/mm or higher (Wang et al., 2000). LECs also re-orientated to align perpendicular to the applied EFs. In control cultures without an applied EF, the orientation of the long axis of each LEC was distributed randomly. In an applied EF however, LECs re-oriented to lie perpendicular to the field vector (Wang et al., 2000). We sometimes observed that in an EF, LECs expanded and flattened (Fig. 17). These responses mimic the alignment, expansion of cell surface area, and flattening of LECs in vivo.
LECs have a much slower proliferation rate than CECs. Applied EFs significantly inhibited proliferation of LECs over 24–48 hours. The simple presence of an EF of 200 mV/mm significantly reduced the growth rate by 50, 75, and 85% at 24, 48, and 72 hours, respectively, after onset of the field (Wang et al., 2003a). The reduction in cell numbers was due to inhibition of cell cycle progression, not apoptosis. Flow cytometry and Western blotting suggested that this occurred due to the EF affecting levels of important cell cycle control enzymes. The G1 specific cyclin, cyclin E, was suppressed whilst the cyclin-cdk complex inhibitor p27kip1 was enhanced, thus restricting G1 to S phase transition (Wang et al., 2003a).
There are endogenous EFs and electric currents in vivo, and applied EFs regulate migration, orientation, and proliferation of LECs. It may not be too far-fetched to propose that the ionic currents in the lens might contribute to controlling the cellular behaviors of LECs in vivo. The EF-regulated cell behavior may have significant implications in the normal lens, in the etiology of PCO, and in lens regeneration. The EFs may regulate cell proliferation, migration, elongation, and flattening. These are important cellular behaviors required for LEC maturation into proper fiber cells. Therefore, dis-regulated endogenous EFs may contribute to PCO.
Electrical signaling may regulate behavior of LECs in PCO. The mainstay treatment for cataract is phacoemulsification, which involves removal of most of the anterior lens capsule, anterior lens epithelium, disaggregated fibers, and surrounding cortex, leaving only the posterior capsule, lens bow (germinal area), and anterior capsule-lens epithelium periphery in place. An artificial lens is implanted as common belief dictates that human lens regeneration does not occur (Lois et al., 2010). However, this treatment is associated with a high incidence of PCO. PCO occurs due to inappropriate proliferation and migration of LECs, which normally line the residual anterior and equatorial areas of the lens, onto the intact posterior capsule, where they compromise vision (Apple et al., 1992). Phacoemulsification is expected to significantly alter the endogenous current (Fig. 16). The normal electric control of migration and proliferation may become aberrant, and contribute to the PCO. It is not known whether the EFs regulate differentiation of LECs. Future research in this direction may provide some insights.
The electrical signal may be a critical component in lens regeneration. Following cataract extraction a mini lens or “Soemmerring’s ring” develops in the majority of cases (Apple et al., 1992). The Soemmerring’s ring is formed by the gradual build up of lens-like material at the equator where the anterior lens capsule and its underlying epithelium, lens bow, and posterior lens capsule are present (Lois et al., 2003). We hypothesized an important role for the anterior lens capsule-anterior lens epithelium (the parts removed during cataract surgery), and proposed that preservation of the whole capsule with its attached anterior lens epithelium may allow for complete regeneration of the lens (Lois et al., 2010).
Cocteau and d’Etoille were the first to observe regeneration of mammalian lens material when the lens capsule and lens epithelium were left intact (Cocteau, 1824). Similar observations were made in rabbits (Gwon et al., 1989) and in mice (Call et al., 2004). Removal of the anterior lens capsule-anterior lens epithelium disrupts the inward currents at the anterior lens surface (Wang et al., 2005) and these electrical signals control the cellular events that lead to lens formation, i.e. elongation, migration, division, and cell cycle progression (Wang et al., 2003b). We studied the electrical events occurring following a modified method of lens extraction which left the anterior lens capsule-anterior lens epithelium in place to determine whether anterior capsule-anterior lens epithelium removal disrupts the endogenous electrical signals and whether the absence of these leads to de-regulated LEC migration and proliferation, and PCO (Lois et al., 2010).
When the anterior capsule was preserved, lens regeneration occurred. Despite the initial changes in EFs, by eight weeks (coinciding with the full development of the regenerated lens) inward currents had returned and resembled those seen in the controls, while outward currents remained at the site of capsulotomy (Lois et al., 2010). The interaction between biochemical signals is critical and should not be ignored in this model. Maintenance of the lens capsule structure to some extent by not removing the anterior flap may generate a biochemical/biophysical interaction which is optimal for lens regeneration. Research has demonstrated the importance of FGF in regulation of lens cell proliferation, differentiation, and migration. The interaction of those biochemical cues and electrical signaling could be a fertile research field.
The retina has many types of electric activities. Our focus here will be on the steady TEP generated by the RPE and their possible patho-physiological significance. The actual effects in vivo are likely to be very sophisticated. The RPE is a monolayer of cuboidal epithelial cells which lies in close association with the rod and cone photoreceptors. The RPE forms the outer blood-retinal barrier and regulates the environment of the outer retina. The RPE functions as three types of cell: epithelium, macrophage, and glia (Steinberg, 1985). The basolateral membrane faces Bruch’s membrane which separates the RPE from the fenestrated endothelium of the choriocapillaris (Strauss, 2005).
The RPE transports ions, water, and metabolites between the subretinal space and the blood and this is important in the prevention of edema (Hamann, 2002; Marmor, 1999). The RPE also transports nutrients such as glucose, retinol, and fatty acids from the blood to the photoreceptors (Baehr et al., 2003; Besch et al., 2003; Thompson and Gal, 2003). Another important function of the RPE is the maintenance of photoreceptor excitability, through the voltage-dependent ion conductance of the apical membrane which enables stabilization of the ionic composition of the subretinal space (Steinberg et al., 1983). Please refer to other comprehensive reviews on ion channels in RPE (Levin and Verkman, 2006; Mergler and Pleyer, 2007; Sparrow et al., 2010; Wimmers et al., 2007).
The RPE has a very different Na+/K+-ATPase distribution from other epithelial cells, including CECs and LECs (Gundersen et al., 1991; Rizzolo, 1999). The RPE contains high levels of Na+/K+-ATPase on the apical side rather than on the basolateral side as in other epithelia (Miller et al., 1978). Expression levels of cadherin and rearrangement of the cytoskeleton in RPE cells appear to be responsible for the assembly of Na+/K+-ATPase with opposite polarity to other epithelial cells (Burke et al., 2000).
The RPE expresses voltage- and ligand-gated K+-, Cl−-, and Ca2+-conducting channels. K+ and Cl− channels are involved in transepithelial ion transport and volume regulation. Voltage-gated Ca2+ channels act as regulators of secretory activity, and ligand-gated cation channels contribute to RPE function by providing driving forces for ion transport, or by influencing intracellular Ca2+ homeostasis. The apical membrane contains a relatively large K+ conductance (Quinn and Miller, 1992) and changes in intracellular K+ in the RPE alters the current produced by Na+/K+-ATPase on the apical membrane, consistent with the finding that intracellular K+ is a competitive inhibitor of Na+/K+-ATPase (Oakley et al., 1978).
Similar to the TEP in the corneal and lens epithelia, a TEP is present in the RPE (Griff et al., 1985; Hughes et al., 1988; Maminishkis et al., 2006; Miller et al., 1978; Oakley et al., 1978; Quinn et al., 2001; Scharschmidt et al., 1988). In contrast to the corneal epithelium and lens epithelium in which the TEPs are positive at the basement membrane side relative to the apical side, in RPE the basal potential is negative relative to the apical potential. In adult human RPE the apical membrane has been shown to have a mean resting cell membrane potential of ~49 ± 7 mV, a TEP of ~1.9 ± 0.6 mV, and a resistance of ~79 ± 48 Ω-cm2 (Quinn and Miller, 1992). The magnitude of the TEP compared to that in the corneal and lens epithelia is much smaller. The field strength across the RPE therefore, is likely to be 1.9 mV over 5 µm, ~400 mV/mm. This is one-50th of that in lens epithelium and smaller than that in corneal epithelium.
The TEP of the RPE is regulated by Na+/K+-ATPase inhibition, K+ concentration, HCO3− and taurine, which is necessary for normal vision. Acute exposure of the apical, but not basal membrane of the RPE to taurine reduced the normally apically positive TEP in bullfrog RPE. With continued taurine exposure, the initial reduction in TEP was sometimes followed by a recovery of the TEP toward baseline. This recovery was abolished by strophanthidin or ouabain, indicating an important role of the apical Na+/K+-ATPase. Increasing extracellular K+ hyperpolarized the apical membrane and increased the TEP. Taurine reversibly doubled these responses, but did not change total epithelial resistance or the ratio of apical-to-basal membrane resistance, and ouabain abolished these responses. Collectively, these findings indicate the presence of an electrogenic Na+/taurine co-transport mechanism in the apical membrane of the bullfrog RPE (Gallemore et al., 1997; Hillenkamp et al., 2006; McBee et al., 2001). They also provide direct evidence that taurine produces a Na+-dependent increase in electrogenic pumping by the apical Na+/K+-ATPase (Scharschmidt et al., 1988). Collectively, these ion channels determine the physiology of the RPE (Wimmers et al., 2007; Wollmann et al., 2006).
We have measured the resting electric currents at bovine retinal epithelium in vitro. Consistent with the distribution of Na+/K+-ATPase, the intact RPE presents small outward currents (flow of positive charge from basal to apical). Upon wounding which disrupts the RPE barrier, a large inward (apical to basal) current was detected (Reid et al., unpublished results).
Much higher EFs (600–1000 mV/mm) are required to guide migration of human RPE (hRPE) cells, as compared to most cell types studied (Han et al., 2009b; Sulik et al., 1992). Cell edges facing the anode retracted their extensions, while cathodal edges and longitudinal ends developed lamellipodia and ruffled membranes (Sulik et al., 1992). EF exposure also induced a cathodal accumulation of F-actin and β1 integrin in hRPE cells (Han et al., 2009b). The directional migration was directed toward the cathode.
We investigated the responses of hRPE cells to applied EFs of various strengths, under different conditions. Primary hRPE cells migrated toward the anode at all EF strengths of 50 mV/mm or stronger. ARPE19 cells showed significant anodal migration at 300 mV/mm. Sheets of hRPE cells also migrate in an EF. Cell clusters appeared to migrate significantly faster than single cells (Gamboa et al., 2010) (Fig. 18).
At all EF strengths of 100 mV/mm or stronger, primary hRPE cells migrated faster than controls. Translocation rate was voltage-dependent, with faster rates at higher field strengths. Translocation rate was significantly reduced in the absence of serum or gelatin and was significantly greater in elongated cells as compared to round or amorphous cells (Gamboa et al., 2010). In “early” confluence primary hRPE cell monolayers, control wounds showed centripetal migration into the centre of the wound with no significant difference in rate between cells at the anodal or cathodal sides. Exposure to an EF of 300 mV/mm induced anodal migration at both sides of the wound and when exposed to an EF of 600 mV/mm, the hRPE cells showed elongated spindle shapes and aligned perpendicularly to the EF (Gamboa et al., 2010).
Normal TEP appears to be essential for healthy RPE, but in addition to investigations of TEP in ion transportation, no research has been directed to test the role of TEP in various RPE functions. It is worthwhile to consider the TEP in RPE function and integrity, as malfunction or death of the RPE underlies many important diseases including age-related macular degeneration and diabetic retinopathy.
Disorders that result from malfunction of ion channels, either through gain of function, loss of function, or by dominant negative effects leading to a reduced number of channels in the cell membrane, are termed channelopathies (Wimmers et al., 2007). Degenerative diseases of the retina have been found to involve mainly Cl− channel functions. Inactivation of the ClC-2 gene in the mouse results in a phenotype very similar to that of retinitis pigmentosa in man and the RPE in these mice shows no TEP (Bosl et al., 2001). This suggests the importance of transepithelial transport in the maintenance of photoreceptor health and function (Wimmers et al., 2007).
Mutations in VMD2 (the gene coding for bestrophin-1) cause an inherited form of macular degeneration with juvenile onset, known as Best’s vitelliform macular dystrophy (Cross and Bard, 1974; Godel et al., 1986; Marquardt et al., 1998; Petrukhin et al., 1998; Weingeist et al., 1982). Best’s disease is characterized by a reduction of the light-rise or light-peak in the patient’s electro-osulogram (Cross and Bard, 1974). The light-rise results from an increase in the basolateral membrane Cl− conductance. Mutations in the VMD2 gene lead to a reduction in the basolateral membrane Cl− conductance (Wimmers et al., 2007). The VMD2 gene product, bestrophin-1, appears to function as a Ca2+-dependent Cl− channel (Sun et al., 2002). Of 15 different mutant bestrophins investigated to date, all lead to a loss in Cl− channel function with a dominant negative effect on the wild-type allele.
Exploiting bioelectric phenomena as a diagnostic and therapeutic tool is possible. Corneal epithelial wound healing can be stimulated with EFs. Healing of denuded areas in the RPE and restoring its function is desirable. Aberrant endogenous EFs could be a potential reason for defective healing. Applied EFs could be used for this purpose, stimulating and accelerating proliferation, directional migration, and subsequently repopulation of RPE cells in RPE-denuded areas of Bruch’s membrane.
The most obvious effects of EFs on cells are the directional (vectorial) responses – cells migrate, orientate, elongate, divide, and spread in a directional manner. This has been the focus of many studies. There is likely another aspect – non-vectorial effects. These include activation of important intracellular signaling pathways that may affect diverse cellular behaviors, such as motility, proliferation, and apoptosis. The vectorial and non-vectorial effects are similar to the chemotactic and chemokinetic effects of growth factors and chemokines. Because of limited understanding of how the weak extracellular EFs induce intracellular responses, these effects are much less recognized and studied. Elucidating the mechanisms of EF transmission will shed more light on both the vectorial effects (electrotactic) and non-vectorial effects (electrokinetic). The following will focus on mechanisms that may underlie EF-induced directional cell migration.
An EF induces epidermal growth factor receptors (EGFRs) and membrane lipids to colocalize and redistribute cathodally in bovine CECs. Electrotaxis of CECs requires serum growth factors, and migration also depends on the substrate. Living and fixed CECs which had been exposed to an EF showed asymmetric distribution of EGFRs as early as 10 min after EF onset. Membrane lipid stained with DiD showed that redistribution of EGFR and DiD was virtually identical, with greater DiD and EGFR staining cathodally than anodally (Zhao et al., 1999a; Zhao et al., 2002). EGFR signaling is also required for galvanotaxis of keratinocyte and breast cancer cells (Fang et al., 1999). EF-enhanced directional migration correlates well with the expression level of EGFR/ErbB1 (Pu et al., 2007).
In hRPE cells migrating toward the cathode, β1 integrin distributed together with F-actin on the cathodal side (Han et al., 2009a). Integrins α5, and α5β1 redistribute and aggregate differentially in fibroblasts cells that migrate directionally in an EF (Brown and Loew, 1994). Knockout of β4 abrogated the electrotactic response of keratinocytes, which could be recovered by transfection and expression of β4 integrin. EGF addition also recovered the electrotactic response of β4 integrin null cells (Pullar et al., 2006). It is therefore likely that multiple signaling mechanisms underlie EF-induced cell migration.
Cell membranes are fluid in structure and permit movement of proteins and lipids in the plane of the membrane. The majority of proteins and lipids externally protruding from the plasma membrane are negatively charged, whereas lecithin and sphingomyelin have a positive charge, while cholesterol and certain glycolipids are neutral (Jaffe, 1977). Indeed, the surfaces of most cells bear a net negative charge (McLaughlin and Poo, 1981). Jaffe suggested that application of an EF parallel to the surface of a cell should redistribute charged macromolecules that are free to move laterally in the plasma membrane through the process of electrophoresis (Jaffe, 1977).
Con A receptors (Poo and Robinson, 1977), Ach receptors (Orida and Poo, 1978), and Fcε receptors (McCloskey et al., 1984) have been shown to be redistributed to the cathodal side by EFs, as have most other receptors studied, which is surprising considering that most glycoproteins have a net negative charge at physiological pH (Robinson, 1985). These observations led to the suggestion of electro-osmosis - the imposition of an EF parallel to the surface of a cell should produce an electro-osmotic flow of fluid near the surface of the cell. The overall negative charge on the cell surface attracts mobile counterions (e.g. Na+) to accumulate in an “aqueous diffuse double layer” adjacent to the surface of the cell. The movement of counterions induces a movement of fluid parallel to the surface of the cell. This electro-osmotic flow exerts a hydrodynamic force on the mobile macromolecules, ultimately causing even negatively charged macromolecules to accumulate at the negative side of the cell (McLaughlin and Poo, 1981). In support of this theory, pretreatment of embryonic muscle cells with neuraminidase, a treatment that removes negatively charged sialic acid residues, or with diI lipid which adds positive charges to the surface of the cell, both resulted in anodal accumulation of Con A receptors (McLaughlin and Poo, 1981).
The role of electrical charge on redistribution of cell membrane molecules was supported by a recent study using mammalian cell lines (3T3, HeLa, and CHO cells). Neuraminidase treatment removes sialic acids on the cell membrane, which reduces total cell surface charge, while cationic avidin treatment actually reverses the directional polarization of sialic acids. However, neither treatment altered directionality or speed of motility in a consistent manner. Polarization of the bulk of charged cell surface proteins may not be necessary nor sufficient to cause motility in an EF (Finkelstein et al., 2007).
Application of an EF to cells cultured in serum-free medium activates multiple intracellular signaling pathways, including phosphatidylinositol 3-kinase (PI3K), mitogen-activated protein kinase (MAPK), and Src. Significantly, the activation is polarized toward the direction of cell migration (Zhao et al., 2006).
PI3Ks are a family of important intracellular signaling molecules involved in many cellular functions such as cell growth, proliferation, differentiation, motility, survival and intracellular trafficking. PI3Ks phosphorylate the 3 position hydroxyl group of the inositol ring of phosphatidylinositol (PtdIns). In many types of cells, PI3K signaling acts as a compass mechanism in chemotaxis (Firtel and Chung, 2000; Haugh et al., 2000; Iijima and Devreotes, 2002; Servant et al., 2000; Xu et al., 2003). Application of an EF to keratinocytes or neutrophils activated PI3K signaling pathways as indicated by increased phosphorylation of Akt (Fig. 19).
In neutrophil-like cells (DMSO-treated HL60 cells), we used a green fluorescence protein (PHAkt-GFP) to show polarized activation of PI3K pathways. When PI3K is activated, PtdIns(3,4,5)P3 (PIP3) is produced. PHAkt-GFP binds to PIP3 revealing the location of PI3K activation. We found that PHAkt-GFP moved to the plasma membrane and was polarized to the leading edge of cells migrating in an EF. When the field polarity was reversed the PI3K activation was polarized to the new cathodal side, and directional migration of the cells toward the new cathode ensued (Fig. 20) (Zhao et al., 2006). Genetic experiments support an important role for PI3K in electrotaxis. When the gene P110γ that codes for one catalytic subunit of a PI3K isoform (PI3Kγ) was disrupted, activation of Akt, Src, P38, and ERK1/2 were all impaired in response to an EF. Electrotactic migration of keratinocytes and neutrophils was inhibited by the loss of PI3Kγ (Zhao et al., 2006). Dermal fibroblasts from p110γ (the PI3K gene) null mice showed significantly reduced directional migration (Guo et al., 2010).
One important downstream signaling element of the EGFR, activated by EFs, is the MAPK - ERK1/2. Applied EFs activate ERK1/2 in CECs and LECs. Importantly, the activation appeared to be asymmetric. Marked cathodal asymmetry of dp-ERK was evident after 1.5 hours in an EF of 150 mV/mm, especially in the cathode-facing lamellipodia. In control cells ERK1/2 activation was largely perinuclear (Zhao et al., 2002).
In LECs it was shown that EF-induced activation of ERK1/2 required serum. ERK1/2 activation was enhanced in peripheral LECs at both 50 and 200 mV/mm, which stimulated migration in different directions. However, in central LECs which showed directional migration only at higher EF strengths, ERK1/2 activation was enhanced only at 200 mV/mm. U0126 (a MAPK inhibitor) prevented activation of ERK1/2 in LECs and inhibited EF-directed migration (Wang et al., 2003c).
U0126 inhibited wound healing in corneal and lens epithelial monolayer wounds with or without exposure to an EF. Wounding and application of an EF enhanced activation of ERK1/2 independently and U016 completely inhibited these activations. Immunocytochemical staining showed an asymmetric activation of ERK1/2 in EF-exposed wounds, with much weaker fluorescence in cathode-facing wounds (the LECs migrated anodally) (Wang et al., 2003c; Zhao et al., 1999a; Zhao et al., 2002).
Src tyrosine kinase activity is required for polarization in fibroblasts, and acts via the small GTPases Rac1 and Cdc42 (Timpson et al., 2001). The Src tyrosine kinase inhibitor, PP2, completely abolished Golgi polarization and changes in morphology. In an EF of 300 mV/mm, Src and Akt (the downstream effector of PI3K) were activated preferentially at the cells cathode-facing side (Pu and Zhao, 2005). In EF-exposed CECs and keratinocytes, Src activation was polarized in the direction of migration (Kucerova et al., 2010; Zhao et al., 2006).
While the majority of research found a significant role for Ca2+ in electrotaxis, some reported a non-essential role for Ca2+. Diverse stimuli increase intracellular Ca2+ which is associated with cytoskeleton remodeling. Myosin light chain kinase functions depend on Ca2+/calmodulin. Transient, graded increases in intracellular free Ca2+ are often seen in migrating cells and they correlate with the direction of movement (Cooper and Schliwa, 1985; Hong et al., 2000; Lee et al., 1999; Witke et al., 1993; Yamamoto et al., 1982; Zheng, 2000). Increasing Ca2+ levels have also been shown to be involved in the detachment of the uropod (Lee et al., 1999). Ca2+ release induces biased motility of growth cones (Hong et al., 2000; Zheng, 2000). Actin-binding proteins such as α-actinin and severin are sensitive to intracellular free Ca2+ (Witke et al., 1993; Yamamoto et al., 1982).
Ca2+ signaling therefore has long been proposed to play a role in the electrotactic response. Ca2+ has been shown to be required for electrotaxis in neural crest cells, mouse embryo fibroblasts, and fish and human keratocytes (Cooper and Schliwa, 1985; Onuma and Hui, 1985, 1988; Trollinger et al., 2002). Blockers of voltage-gated Ca2+ channels have been shown to inhibit galvanotaxis (Trollinger et al., 2002). However, except at very high voltages, physiological EFs which induce evident cell migration have not been shown to increase intracellular [Ca2+] (Brust-Mascher and Webb, 1998). In contrast, two strains of mouse fibroblasts exhibit electrotaxis independent of Ca2+ signaling (Brown and Loew, 1994). Several laboratories have tried to image the intracellular Ca2+ but no convincing data has been obtained. With faster imaging techniques, it may be possible to image in a more definitive way whether Ca2+ fluctuation is induced by physiological EFs.
Polarized distribution of organelles (Golgi apparatus, microtubule organization center) contribute to directional cell migration. Asymmetry of the Golgi apparatus plays a role in the anterograde supply of membrane components to the cell leading edge for membrane protrusion (Bershadsky and Futerman, 1994; Nabi, 1999; Prigozhina and Waterman-Storer, 2004; Ridley et al., 2003). In an EF, Golgi apparatus and actin polarized cathodally in CHO cells (Pu and Zhao, 2005).
Golgi polarization correlates linearly with electrotaxis. Brefeldin A which induces dispersal of Golgi completely abolished EF-induced Golgi polarization and significantly inhibited EF-directed migration (Cao et al., 2011; Pu and Zhao, 2005). Detailed time-lapse imaging however, demonstrated a novel insight. While Golgi polarization and directional cell migration depend on each other, initiation of directional migration does not require Golgi polarization. Golgi then polarizes and facilitates and augments the directional migration (Pu and Zhao, 2005).
The last few sections suggest that chemotaxis and electrotaxis share many common molecules and features. There are however some differences between electrotaxis and chemotaxis.
In chemotaxis of Dictyostelium cells, G protein-coupled receptors (GPCRs) bind chemoattractants and regulate or bias pseudopod formation at the leading edge (Chung et al., 2001; Parent and Devreotes, 1999). Electrotaxis of Dictyostelium cells however, appears to be independent of GPCRs, while GPCR signaling is essential for chemotaxis of Dictyostelium cells. The G protein Gα2 subunit and Gβγ complex couple together with receptors to transduce signals from chemoattractants. The Gβ subunit is essential for chemotaxis to all chemoattractants (Jin et al., 1998; Wu et al., 1995). However, similar to cAR1−/cAR3− and Gα2− cells, Gβ− cells maintained significant directional migration in an EF although it was markedly suppressed (Zhao et al., 2002).
Membrane potential may play a different role in electrotaxis than chemotaxis. One of the immediate changes induced by extracellular EFs is a change in membrane potential. The anodal side of the plasma membrane becomes hyperpolarized (more polarized), while the cathodal side of the plasma membrane becomes depolarized (less polarized). It is difficult to precisely control the membrane potential in mammalian cells. We recently exploited the tolerance of Dictyostelium cells to changes in membrane potentials. We drastically changed the membrane potential by changing [K+], or pH in the bathing buffer. Electrotaxis was significantly inhibited while chemotaxis remained largely unchanged (Van Duijn et al., 1990; Gao et al., unpublished results).
It has been suggested that cells align perpendicular to an EF in order to minimize the voltage drop across themselves (Robinson, 1985). However, not all cell types align perpendicularly in an EF. CHO cells and differentiated Dictyostelium elongate parallel to an EF (Pu and Zhao, 2005; Zhao et al., 2004). The role of EF-induced changes in membrane potential in cellular responses is yet to be elucidated.
Although we have discovered some of the intracellular signaling pathways that are activated following EF exposure, and that alterations are made to membrane electric potential, membrane lipid distribution, and membrane receptor distribution, as well as intracellular distribution of many organelles such as the actin cyctoskeleton and the Golgi apparatus, and that certain genes (i.e. PTEN, Pax6) play a role, exactly how cells sense the EF and transduce it into an intracellular signal remains one of the most important and elusive questions. We expect significant efforts in the coming years on elucidation of the “sensors”/sensing mechanisms for small EFs.
EFs resulting from ionic fluxes in ocular tissues appear to be an important regulator of many important aspects of ocular cell behavior. An important role for these EFs is likely to be in wound healing where naturally occurring wound EFs regulate cell migration, proliferation, and alignment of the cells. This biophysical signal may work together with other better-understood cues such as growth factors, other signals elicited by wounding, and the extracellular matrix, to modulate wound healing and regeneration. Further understanding of this aspect of electrically-controlled ocular cell behavior is likely to offer new insights into ocular tissue homeostasis and wound healing. It may also lead to development of tissue engineering techniques to repair damaged or diseased ocular tissues.
This work is supported by NIH 1R01EY019101. We thank the Wellcome Trust for continuous support (068012). This work was also supported in part by UC Davis Dermatology Developmental fund, and grants from California Institute of Regenerative Medicine RB1-01417, NSF MCB-0951199, and the Research to Prevent Blindness, Inc.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.