Mechanisms that guide directional migration of neuroblasts from the subventricular zone (SVZ) are not well understood. We report here that endogenous electric currents serve as a guidance cue for neuroblast migration. We identify the existence of naturally occurring electric currents (1.5±0.6 μA/cm2, average field strength of ∼3 mV/mm) along the rostral migration path in adult mouse brain. Electric fields of similar strength direct migration of neuroblasts from the SVZ in culture and in brain slices. The purinergic receptor P2Y1 mediates this migration. The results indicate that naturally occurring electric currents serve as a new guidance mechanism for rostral neuronal migration.
There is a Hot off the Press (February 2013) associated with this Scientific Report.
During development, some neurons migrate long distances from their origin to their final destination. Migration of some types of new‐born neuron persists in adult mammalian brains [1, 2, 3, 4]. One well‐studied example in the brain is the highly directional migration of neuroblasts from the subventricular zone (SVZ) to the olfactory bulb (OB). This pathway spans more than several millimetres or even more and is called the rostral migratory pathway or rostral migratory stream [1, 3, 5, 6, 7, 8, 9, 10, 11]. Defects in neuronal migration might lead to important diseases, including lissencephaly, epilepsy and mental retardation [12, 13, 14].
Genetic and molecular analyses have suggested several molecules that might guide neuronal migration. These molecules form extracellular molecular gradients (for example gradients of stromal‐derived factor‐1, slits and reeli) and participate in guiding the rostral migration [2, 15, 16, 17]. In addition, biophysical cues might have a role in the rostral migration. A recent elegant work demonstrated an important role of flow of cerebro spinal fluid in guiding neuronal migration. The beating cilia of ependymal cells generates directional flow of cerebro spinal fluid, which helps to build slit gradients that contribute to the rostral migration of neuroblasts [18[.
Another potential candidate cue for guiding long distance migration is the electric field (EF). Directional fluxes of ions generate electrical potential gradients, endogenous EFs, which have been measured in the embryo and at wounds [19, 20, 21, 22, 23, 24, 25, 26]. Many types of cells respond to small EFs by directional polarization, growth and migration (galvanotropism, galvanotaxis or electrotaxis) [20, 27, 28, 29, 30, 31, 32]. Schwann cells from chick embryo, for example, migrate directionally in an EF as low as 3 mV/mm . Electric potential gradients have also been reported in the normal and injured brains. The cortical surface in adult rabbit was measured 0.5 to 5.5 mV positive to the ventricle . The potential difference increased with age, being only 1∼1.6 mV at age 5 days but 20 mV in the mature rat . Injury induces electrical potential gradients between normal cortical surface and a damaged area of cortex in the rat . Infarction of the brain generates large negative extracellular potential changes lasting on the order of minutes [36, 37].
We suggest that there are endogenous currents along the rostral migration pathway that contribute to the guidance of neuroblasts to migrate from the SVZ to the OB.
Results and Discussion
Endogenous currents in rostral migratory pathway
We first mapped the electric current flow at the lateral ventricle walls, and at the surface of the OB, using a vibrating probe system, which measures electric currents noninvasively with high sensetivity and spatial resolution [38, 39]. At lateral ventricle of adult mouse brain, very consistent inward currents were detected at all five measuring positions from frontal to occipital horn (see detailed Methods in Supporting Materials). The average magnitudes of the inward electric currents (Ii) were Ii=−1.6±0.4 μA/cm2, whereas at the OB, five measuring positions showed uniformly outward currents (Io=1.5±0.6 μA/cm2) (Fig 1A,B).
We next used the four‐point probe technique with Ag/AgCl electrodes to measure the direct current (DC) electrical resistance in brain tissues (Fig 2A) [40, 41]. A DC resistance of 5.8±1.4 kOhm existed between the SVZ and the OB. The extracellular voltage is the product of the resistance of interstitial fluid and the electric currents. The voltage can be calculated according to Ohm's law: V=R × I (I=J*Aout), where J is the ionic current density measured using vibrating probe; R is interstitial resistance measured with the four‐point probes; and Aout is the total surface area of the OB. The average distance from the SVZ to the OB in adult mice is 3 mm, therefore an EF of ∼2 mV/mm exists naturally along the rostral migration pathway.
Third, we directly measured the electric potential in the interstitial spaces along the rostral migration pathway, using a voltmeter. The Ag/AgCl electrodes were insulated except the tips (diameter 100 μm). The detection electrode was inserted in brain tissue ∼200 μm under the lateral ventricle (LV) wall at different positions, in the centre of the OB, or in the centre of the rostral migration pathway. The reference electrode was placed in artificial cerebral spinal fluid, which bathed the hemisphere with same distance to different measurement points (Fig 2B). The SVZ had a positive potential: 7.9±5.3 mV and the centre of OB had a negative potential: −2.0±5.6 mV in relation to the bathing artificial cerebral spinal fluid. The voltage drop between SVZ and OB thus is 9.9±1.8 mV (Fig 2C). This voltage drop divided by the distance between the measuring points in the SVZ and the OB gave an average voltage gradient of ∼3.3 mV/mm. We also measured the voltage drop when detection electrodes were simultaneously inserted into the SVZ and OB, respectively. These showed a voltage drop of 5.7±1.2 mV/mm. We measured the voltage at several points along the rostral migratory pathway; consistent positive potentials were measured along the rostral migration path relative to the potential at the OB (Fig 2C). Electrical potential gradients therefore exist naturally along the rostral migratory pathway (Fig 2D).
How is the voltage between the SVZ and the OB generated? The extracellular space might occupy as much as 20% of brain volume [42, 43]. The concentration of ions (K+, Na+, Ca2+, and so on) in the extracellular space contributes to the strength of the extracellular fields [44, 45, 46]. The field can provide a built‐in driving force to move ion currents through cells and organs, and these currents are integrally involved in all of our sensory systems . Electrogenic pumps such as Na+/K+‐ATPases are expressed at high levels at the superficial layers of the OB . We used antibodies to label the β1 subunits of Na+/K+‐ATPases, and confirmed very strong and exclusive apical expression (Fig 1C). The apical localization of these pumps would drive electric currents (flow of positive charges) outwards, generating a voltage sink in the OB. At the ependyma of the lateral ventricles, a layer of multi‐ciliated epithelial cells separates the SVZ from the ventricular lumen. The apical surfaces of these ependymal cells, including the microvilli, had little or no ATPase activity. Very strong ATPase reaction product was found in the slender processes terminating on the basal laminae of perivascular spaces . Our immunofluorescence staining verified the basal location of the Na+/K+‐ATPases, which would drive currents into the brain (Fig 1C). Ouabain, an inhibitor of Na+/K+‐ATPases, effectively reduced the electric current at both ventricular wall and the OB (Fig 1D). Thus, flow of electric currents into the brain tissue (flux of positive ions) at the SVZ and outward flow of currents at the OB orchestrate in establishing gradients of ions and EFs of 3–5 mV/mm along the rostral migratory pathway from the SVZ to the OB (Fig 2C,D).
Applied EFs guide neuroblast migration
We then tested the response of neuroblasts from the SVZ to EFs of similar strength that we detected in the brain.
We confirmed that the cells obtained from the SVZ expressed neuroblast markers (Fig 3A). These neuroblasts migrated actively in culture (Fig 3B–E). In an applied EF, SVZ neuroblasts migrated more actively with significantly increased migration rate (Fig 3E). Both trajectory and displacement rate (Tt/t and Td/t) increased significantly in an applied EF of 10, 50 or 250 mV/mm (Fig 3E). Most importantly, the SVZ neuroblasts migrated directionally towards the cathode (Fig 3B–D; supplementary Movies S1–S3 online). This is the direction of the endogenous electrical potential gradients from the SVZ to the OB (see Fig 2D). An EF as low as 3.5 mV/mm induced significant directional migration of neuroblasts towards the cathode (Fig 3C,D). This is the EF strength naturally occurring along the rostral migratory pathway in adult mouse brains. The physiological EFs that exist between the SVZ and OB therefore might direct and stimulate migration of the SVZ neuroblasts along the rostral migratory path.
Electrotaxis of neuroblasts in brain slices
We next investigated migration of neuroblasts transplanted into brain slices. SVZ neuroblasts from adult mice were labelled with cell membrane dye DiI. We transplanted the neuroblasts into the SVZ in the brain of neonatal mice (P3–P7). After recovery of 1–2 days, brain slices were obtained from the neonatal mice for tracking migration of transplanted cells (Fig 4A) (see supplementary information online for details). We monitored migration of transplanted cells with fluorescence time‐lapse video imaging (MetaMorph Imaging System) . In the brain slices, the primary migratory direction of neuroblasts was towards the OB and towards the cortex (Fig 4B). We used an established index—directedness—to quantify how directional the neuroblasts migrated towards the OB. The brain slices were placed in a defined orientation. If the cells migrated in random directions, the average directedness value would be 0. If all the cells migrated perfectly towards the right (cathode), the average directedness value would be 1. If cells migrated preferentially towards the right, an average directedness would be larger than 0 and approaching 1 . When the brain slices were orientated as in Fig 4B and no EF was applied, labelled neuroblasts migrated with a directedness value of 0.24, which indicated moderately directional migration towards the right (towards the OB) (Fig 4B). When the brain slices were subjected to EFs of 10 or 50 mV/mm, the transplanted neuroblasts migrated more pronouncedly towards the right (the cathode). In EF, the directedness value of migration was 0.48 and 0.6, significantly higher (2–2.5‐fold) than that of no‐EF control (Fig 4B). Trajectories of cell migration showed significantly more directional migration in an applied EF towards the cathode (Fig 4B). Reversal of the field polarity reversed the migration direction of neuroblasts in the brain slices (Fig 4C; supplementary Movie S4 online).
P2Y1 mediates EF‐guided neuroblast migration
We finally investigated possible molecular mediators of the electrically guided migration. We tested purinergic receptors P2Y1 as these receptors are transiently and specifically expressed in neuroblasts that migrate out of the SVZ. Extracellular ATP signalling is essential for the migration of intermediate neuronal progenitors and formation of the SVZ. P2Y1 receptors mediate Ca2+ fluctuations in neuronal progenitors at the time they leave the SVZ and are required for the formation of the SVZ [49, 50]. ATP signalling might also contribute to neuronal differentiation and process outgrowth; both contributed to the radial migration of postmitotic neurons .
We used two inhibitors of P2Y1 receptor: a general inhibitor for purinergic receptor suramin and a more specific P2Y1 receptor inhibitor MSR2179. Suramin significantly reduced EF‐directed migration of neuroblasts. Suramin at concentration of 5 μM significantly inhibited the directional migration without obvious effect on migration rate. Cells migrated in an EF of 50 mV/mm with a directedness value of 0.28, significantly lower than the control (Fig 5A,D; supplementary Figs S1A,S2 online and supplementary Movie S5 online).
As suramin is a general inhibitor of purinergic receptors, we used two approaches to narrow down specific type of receptors: (1) pharmacological inhibitor MSR2179, a specific P2Y1 inhibitor, and (2) knockdown with small interfering RNA (siRNA)P2Y1. MSR2179 inhibited the EF‐directed migration in a dose‐dependent manner without significantly affecting cell migration rate (Fig 5B,D; supplementary Fig S1B,C online, supplementary Fig S3 online and supplementary Movie S6 online). siRNAP2Y1 reduced P2Y1 expression level significantly as confirmed with western blot (Fig 5D). Knockdown of P2Y1 with siRNA resulted in significant reduction in the directedness of neuroblasts migration in an EF (Fig 5C,D; supplementary Fig S1C online).
In summary, we detected with different techniques the existence of DC EF gradients between the SVZ and the OB. Applied EFs of the same strength guided migration of the SVZ neuroblasts towards the cathode. P2Y1 receptors that are expressed transiently in the SVZ neuroblasts seem to mediate the EF‐guided migration. Our data suggest that the naturally occurring electric potential gradients are a long‐distance directional signal for neuroblast rostral migration. This mechanism might offer new approaches in guiding neuroblast migration in stem cell therapies. These results indicate an alternative view of cell migration that complements, but is not mutually exclusive from, the present well known roles of chemokines, matrix metalloproteinases and adhesion molecules in the rostral migration stream.
Endogenous EFs were measured with a vibrating probe, a four‐point probe system, and microelectrodes. Migration of neuroblasts was recorded with an imaging system. For detailed Methods, please see supplementary Information online.
Conflict of Interest
The authors declare that they have no conflict of interest.
Supplementary Movie 1
Supplementary Movie 2
Supplementary Movie 3
Supplementary Movie 4
Supplementary Movie 5
Supplementary Movie 6
This work was supported by a grant (RB1‐01417) from California Institute of Regenerative Medicine (to M.Z.). M.Z. is also supported by grants from NSF MCB‐0951199, and NIH 1R01EY019101 and UC Davis Dermatology Developmental Fund. Part of the work was supported by Grant S‐CIRMTG1‐GSTDW, RS1‐00453, NIH grant RO1‐DC010917 and DC007592 (to EY).
Author contributions: M.Z. conceived and designed the project. L.C., D.W., B.R. and S.Z. performed the experiments and prepared figures. M.Z. and L.C. wrote the main manuscript text. T.P. and E.Y. provided facilities and commented on the manuscript. All authors reviewed the manuscript.
- Copyright © 2013 European Molecular Biology Organization