NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Frostig RD, editor. In Vivo Optical Imaging of Brain Function. 2nd edition. Boca Raton (FL): CRC Press; 2009.

Cover of In Vivo Optical Imaging of Brain Function

In Vivo Optical Imaging of Brain Function. 2nd edition.

Show details

Chapter 6Imaging the Brain in Action: Real-Time Voltage- Sensitive Dye Imaging of Sensorimotor Cortex of Awake Behaving Mice

, , and .


Electrophysiological measurements have demonstrated that information can be processed on the time scale of milliseconds in the mammalian brain. Neuronal electrical changes are probably the fastest events occurring in the nervous system, and they are likely to orchestrate many of the subsequent slower events, such as changes in second-messenger concentration, structural alterations, and regulation of gene expression. Whereas electrophysiological recordings from individual electrodes have revealed many important aspects of brain function, it is also clear that complex neuronal processing of information does not derive from the activity of individual neurons, but rather results from the concerted actions of many neurons distributed across different brain areas. Indeed, considerable progress has been made toward increasing the number of electrodes in electrophysiological recordings in order to begin to understand the coordinated function of neuronal networks [1–2]. Despite these very important technical developments, it remains clear that the spatial organization of neuronal electrical activity will be difficult to study with electrophysiological approaches, even using arrays of more than 100 electrodes.

Optical methods allow high-resolution imaging and therefore provide a useful tool to explore the spatial distribution of neuronal activity, especially in superficial brain structures such as the cerebral cortex. Indeed, quite remarkably, there are intrinsic changes in reflected light from the cortical surface, which can be easily measured and provide high spatial resolution maps of cortical organization [3–6]. However, these intrinsic changes in light absorption and scattering of brain tissue do not relate directly to neuronal electrical activity, but instead result primarily from hemodynamic changes, in a similar way to the BOLD fMRI signal. In order to make optical measurements that relate directly to electrical activity, it has so far proven necessary to apply dyes that change their absorption or emission spectra in a manner depending upon membrane potential. Such compounds are termed voltage-sensitive dyes.


6.2.1. Optical Measurement of Membrane Potential

The first VSDs applied to invertebrate preparations showed that it was possible to optically record fast membrane potential changes like action potentials in single neurons [7–8]. These pioneering recordings inspired the next decades of research, much of which has been targeted toward making measurements from the intact brain during the processing of sensory information. The first in vivo VSD recordings of sensory processing in the vertebrate brain were obtained in the optic tectum of the frog [9] and subsequently in the rat somatosensory and visual cortex [10]. These early investigations were made difficult by small signal amplitudes, large hemodynamic signals, phototoxicity, and limitations in detector and computer technology. Many of these problems have begun to be solved and, in this chapter, we try to provide information on the current state-of-the-art techniques for in vivo VSD imaging with the specific goal of measuring neuronal activity in awake behaving animals.

6.2.2. Voltage-Sensitive Probes for in Vivo Applications

One of the largest problems for in vivo VSD imaging is the prominence of heartbeat-related artifacts. With each heartbeat, the recorded signals are contaminated by movement artifacts from the physical pulsation of blood vessels and, in addition, by the shift of the hemoglobin absorption spectrum, which depends strongly on its oxygenation level. The most frequently used wavelengths for fluorescence measurements are indeed strongly absorbed by hemoglobin with large differences between deoxyhemoglobin and oxyhemoglobin. Heartbeat-related signals can therefore contribute significantly to optical fluorescence measurements.

To minimize these artifacts, Grinvald and coworkers suggested the application of fluorescent dyes with absorption and emission spectra in the far red, longer than 600 nm. In these wavelengths, hemoglobin absorbs little light, and the difference between oxygenated and deoxygenated forms is not so dramatic. In order to test this idea, they generated a new series of VSDs (RH1691, RH1692, and RH1838), which have revolutionized in vivo VSD imaging [11–12]. With these new dyes, together with advances in camera technology, imaging cortical dynamics at high temporal and spatial resolution have become routine measurements in many laboratories [13–38]. In our own studies, we have only used RH1691, and this review will therefore focus on this VSD. Although RH1691 works well in the rodent neocortex, it does not work in the rodent olfactory bulb [16] so care is needed in choosing and validating the VSD in individual applications.

RH1691 (Figure 6.1A) is a water-soluble aromatic anionic oxonol compound based on a vinylogous carboxylate conjugate system; its conformation changes are executed via transferring electrons. Upon application to cells, it is thought to insert mostly into the lipid bilayer of plasma membranes (Figure 6.1B) where it presents the properties of a fluorescent dye: it can be excited by light at ~630 nm, and emits fluorescent photons with wavelengths >665 nm. The electrical potential difference (on the order of a hundred millivolts) across the plasma membrane (with a thickness of ~5 nm) generates a strong electric field, which affects its fluorescence properties. Changes in the membrane potential are linearly related to the measured fluorescence of RH1691 (Figure 6.1C,D), when it is applied to Xenopus oocytes under two-electrode voltage clamp [27]. Voltage-clamp jumps are accompanied by rapid fluorescence changes (Figure 6.1E), which follow the membrane potential with submillisecond resolution [27]. The fluorescence of RH1691 is therefore rapidly and linearly related to membrane potential, similar to other voltage-sensitive dyes [39], making interpretation of measurements relatively straightforward.

FIGURE 6.1. RH1691 is a fast and linear voltage-sensitive dye.


RH1691 is a fast and linear voltage-sensitive dye. (A) Chemical structure of the voltage sensitive dye RH1691. (B) The VSD RH1691 is a fluorescent dye excited at ~630 nm and emitting fluorescence >665 nm. The dye is thought to insert primarily (more...)


6.3.1. Surgery and Cortical Staining With RH1691

In order to make in vivo measurements, the dye must be introduced into the brain. The simplest way to obtain an even staining of the cortical surface is to apply the dye dissolved in Ringer’s solution directly to the cortical surface. A craniotomy must therefore be prepared over the cortical region of interest, and the dura subsequently removed (although a recent technical report from Lippert and collaborators [30] shows that RH1691 can stain through the rat dura). This must be done with extreme care not to damage the cortex, especially during removal of the dura. The VSD RH1691 is applied at a concentration of 1 mg/ml directly to the cortical surface and left to diffuse into the brain for a period of ~1 hour. Subsequently, the unbound dye must be washed away; the craniotomy can eventually be covered with agarose (~1–2%), which stabilizes the exposed cortex and reduces pulsation-related movement artifacts. A coverslip is then placed on the top of the preparation, forming a sealed chamber that, in addition, prevents the agarose from drying during the recording session. To combine VSD imaging with other experimental approaches (i.e., electrophysiological recordings, microstimulation, or drug injections), it is also possible to preserve an open access to the cortex on the side of the coverslip, but in this case, the experimenter might need to add extra Ringer’s solution during the experiment to keep the preparation hydrated.

6.3.2. Acquisition of the Fluorescence Signals

In order to image VSD signals in vivo it is necessary to collect large amounts of light at low magnification. To obtain a high numerical aperture with a large working distance, it is useful to use tandemlens optics [40]. The magnification is then given by the ratio of the focal length of the two lenses, which are mounted front-to-front, sandwiching the dichroic mirror. The VSD RH1691 is excited with band pass filtered (630 ± 15 nm) light, reflected using a 650 nm dichroic, and focused on the cortical surface with the lower lens. The excitation light is typically delivered by a halogen light source, which has very high stability compared to an arc lamp. The emitted fluorescent light is collected via the same optical pathway, but without the reflection of the dichroic, long-pass filter (>665 nm), and is focused onto a detector via the upper lens. Because VSD signals are typically small (on the order of 0.1%), it is critical to collect many photons to get beyond the physical limitations of shot noise, which follows the square root of the number of collected photons. In order to resolve a 0.1% change in light, it is necessary to collect ~1,000,000 photons. Specialized CMOS (complementary metal oxide semiconductor) cameras allow the acquisition of images at high time resolution with each pixel having a very large well depth. Fortunately, there are now a number of commercially available imaging systems that are well-suited for in vivo VSD imaging.

6.3.3. Data Analysis

The recorded VSD signals are usually affected by the bleaching of fluorescence which can be corrected offline. Subtraction of a best-fit double-exponential is well-suited when analyzing single trials. For experiments carried out with anesthetized animals, subtraction of heartbeat-synchronized and averaged blank sweeps is an alternative that has the additional advantage of removing the artifacts arising from blood pulsation–related movements. However, experiments performed with anesthetized mice revealed that the heartbeat-related signals often have a minimal impact on VSD recordings with RH1691 (Figure 6.2A,B). No modulation of the optical signals was observed with respect to the respiratory rhythm in these recordings [24] (Figure 6.2A,C).

FIGURE 6.2. Heartbeat- and respiration-related artifacts have a minimal impact on the VSD recordings using RH1691 in anesthetized mice.


Heartbeat- and respiration-related artifacts have a minimal impact on the VSD recordings using RH1691 in anesthetized mice. (A) Simultaneous recording of the voltage-sensitive dye signal, respiration (from recording movement of the rib cage with an optical (more...)


6.4.1. Origins of Recorded Optical Signals

Brain slices can be prepared in order to investigate what the dye stains under these conditions. In both rat [19,30] and mouse [24] barrel cortex, the RH1691 primarily labels the supragranular layers (Figure 6.3A). The dye is found in the neuropil, consistent with being bound to plasma membranes rather than being internalized (Figure 6.3B). The peak of the fluorescence profile with respect to cortical depth is found in layer 2/3 (Figure 6.3C). Presumably because the dye is water soluble, it diffuses out of layer 1 during the rinsing period after the staining, although other unknown structural features in layer 1 could also decrease RH1691 binding. Based on the linear relationship of fluorescence and membrane potential (Figure 6.1), in addition to the dye distribution following in vivo labeling (Figure 6.3), the fluorescence signals of RH1691 are therefore most likely to relate to changes of membrane potential in layer 2/3.

FIGURE 6.3. Applying RH1691 to the cortex stains the supragranular layers.


Applying RH1691 to the cortex stains the supragranular layers. (A) Topical application of RH1691 to the mouse barrel cortex labels mainly the supragranular layers. Parasagittal slices (100 μm thick) from stained barrel cortex, counterstained with (more...)

6.4.2. RH1691 Reports Supragranular Membrane Potential changes in Anesthetized Mice

In order to directly correlate the fluorescence signals from RH1691 with membrane potential, we have combined whole-cell recordings with VSD imaging in the barrel cortex of urethane anesthetized rodents [19,20,24,27]. Remarkably close relationships were observed between the membrane potential of individual neurons and the VSD signal for both evoked and spontaneous cortical activity (Figure 6.4).

FIGURE 6.4. VSD signals under anesthesia correlate with subthreshold membrane potential changes in the supragranular layers.


VSD signals under anesthesia correlate with subthreshold membrane potential changes in the supragranular layers. (A) Simultaneous whole-cell (WC) recording and VSD imaging in vivo under urethane anesthesia from the mouse somatosensory barrel cortex. The (more...)

The rodent barrel cortex is a region of the primary somatosensory cortex, which is closely involved in the processing of tactile information from the mystacial vibrissae [41]. This cortical area is highly organized with an obvious anatomical mapping of each individual whisker. Because the whiskers and the corresponding barrel cortex are laid out in a stereotypical pattern, they have been given labels to allow identification. In our studies we have focused primarily on the so-called C2 whisker. The simplest experiment that one can perform with respect to comparing VSD signals with membrane potential is to briefly deflect a single whisker and measure the evoked cortical response. A brief deflection of the C2 whisker evokes activity in the somatosensory cortex, which is signaled via the trigeminal nerve to the brain stem, which in turn sends information to the ventrobasal thalamus from where thalamocortical neurons project to the primary somatosensory cortex. There are only two synapses between the whisker and the cortex, the first in the brain stem and the second in the thalamus. Sensory information can therefore reach the neocortex rapidly following a whisker deflection, with latencies typically around 10 ms. In vivo whole-cell recordings of layer 2/3 neurons show that a single brief 2 ms whisker deflection evokes a depolarizing sensory response lasting some tens of milliseconds (Figure 6.4A). Simultaneous optical measurement indicates that the kinetics of the VSD signals are remarkably close to the subthreshold membrane potential changes (Figure 6.4A): plotting VSD fluorescence as a function of membrane potential shows a linear relationship (Figure 6.4B). This correlation is particularly remarkable because the VSD signals result from changes in a large population of neurons, whereas the whole-cell recording relates to only a single neuron. From this observation, it would therefore seem likely that most neurons, within a few hundred microns of each other, have similar dynamics in their subthreshold membrane potentials. Dual whole-cell recordings have indeed confirmed this hypothesis [19,20].

The major advantage of VSD imaging compared to whole-cell recording is the availability of spatial information (Figure 6.4C). Deflection of the C2 whisker evokes a sensory signal that is initially (at 10 ms after whisker deflection) confined to the corresponding barrel column. Over the next milliseconds, the sensory response spreads across a large part of the barrel map [19,20,24]. Such propagation of the evoked activity may be critical for the integration of tactile information from different whiskers, other body parts, and even multimodal sensory integration. The spreading VSD signals evoked by deflection of single whiskers are in good agreement with the large subthreshold receptive fields measured in layer 2/3 barrel cortex [42–44]. The spatiotemporal properties of the cortical activation induced by sensory input have been further studied, in mice, by imaging larger cortical regions covering both the somatosensory and the motor cortex. Such recordings revealed that a single deflection of the C2 whisker evokes, in addition to the initial spreading somatosensory response, a secondary localized region of activation in the motor cortex, which also spreads. Intracortical microstimulation experiments demonstrated that this secondary activated region corresponds to the region of the motor cortex that controls whisker movements. VSD imaging of cortical responses induced by the stimulation of different whiskers in the same conditions revealed a somatotopic arrangement of the activated areas within the primary motor cortex [29]. These first images of motor cortex activation upon sensory stimulation indicate that the whisker-related motor cortex is likely to integrate sensory input with motor commands.

Because of the relatively good signal-to-noise ratio of the recordings obtained with RH1691, it has also been possible to optically record spontaneous cortical activity, which by necessity demands single trial resolution. Under anesthesia, the membrane potential of cortical neurons often exhibits slow and large-amplitude fluctuations, which have been termed UP and DOWN states [20,45–48]. These spontaneous membrane potential changes correlate closely with the simultaneously measured VSD signal (Figure 6.4D). Again, since the whole-cell recording reports single neuron activity, whereas the VSD signal arises from a population, these data would seem to indicate that many neurons are behaving electrically in a very similar manner. Dual whole-cell recordings from anesthetized animals confirm that membrane potential fluctuations indeed occur synchronously in neurons within a local cortical region [20,49,50]. Although the subthreshold membrane potential changes recorded in individual neurons correlate closely with the VSD signals, the same cannot be said of the suprathreshold activity. Action potentials recorded in single neurons occur during the depolarized UP state, but whereas the whole-cell recordings show a large electrical signal change during the action potential, there is no indication of this event in the VSD record. Because the VSD is sufficiently fast to follow action potential waveforms, we must conclude that only very few nearby neurons spike at the same time. This would indicate a sparse action potential coding in layer 2/3 barrel cortex, as also suggested by electrophysiological recordings [19,20,44,48,51].

The spatiotemporal dynamics of the spontaneous activity imaged by VSD reveal a great deal of complexity. In many cases the spontaneous depolarized active periods (UP states), appear as propagating waves of activity (Figure 6.4E). They can travel as planar waves, spirals, or even more complex patterns. In some cases, interesting correlations between spontaneously active regions have been observed. For example, in the mouse, we have observed synchronous spontaneous activity between the somatosensory and the motor cortex [29]. In the cat visual cortex, the spontaneous activity under anesthesia often takes the form of orientation maps [13,18]. In the rat parietal association area, which lies between the primary visual and the barrel cortex, imaging reveals a clear preference for propagation of the spontaneous waves within the cross-modal axis [36]. These observations from different species suggest that the spontaneous patterns of activity imaged under anesthesia are likely to relate intrinsic properties of the cortical functional connectivity, which do not depend upon the vigilance state of the brain [52].

The close correlation in vivo under anesthesia between the VSD signal from RH1691 and the membrane potential in layer 2/3 neurons, for both evoked and spontaneous activity, suggests that the optical signals are open to a simple and straightforward interpretation. Because action potentials in layer 2/3 neurons occur rarely under our experimental conditions, the VSD signal is almost entirely dominated by subthreshold membrane potential changes. We conclude that VSD imaging reports the spatiotemporal dynamics of subthreshold membrane potential changes of the supragranular cortical layers.


6.5.1. Experimental Strategies

The relative simplicity and robustness of the in vivo VSD imaging technique and our ability to interpret the signals in a simple manner has allowed us to investigate the spatiotemporal cortical dynamics of awake mice with the aim of obtaining direct correlations between behavior, sensory input, and cortical activity. In order to perform VSD imaging of awake mice, we developed two different experimental strategies. The first approach consisted in imaging the cortex of head-restrained mice [20,29], (Figure 6.5A) and the second method, based on the use of a flexible fiber optic bundle, allowed us to image cortical activity while the mouse was moving freely within a limited area [24] (Figure 6.5B). Both strategies yielded interesting and complementary results.

FIGURE 6.5. Fluorescence VSD imaging of awake mice with conventional epifluorescence and fiber optic imaging.


Fluorescence VSD imaging of awake mice with conventional epifluorescence and fiber optic imaging. (A) Mice readily accept head-fixation and conventional optics can therefore be applied for fluorescence imaging of awake mice. (B) Fiber optics can also (more...)

Mice accept head-restraint with surprising ease. Following implantation of a metal fixation post on the occipital bone, the mouse is trained to sit calmly in the experimental recording setup for a few days, on the basis of a couple of habituation sessions per day. At first the mouse is head-restrained for a few minutes and, as the mouse gradually adapts, the period of head-fixation can be increased up to a period of a couple of hours. On the day of the imaging experiment, the standard staining procedures are carried out using reversible gas anesthesia. After complete recovery from the anesthesia, it is then possible to image the cortical dynamics in the awake head-fixed mouse (Figure 6.5A). This approach has the great advantage that it should be possible to combine awake VSD imaging with other experimental approaches such as simultaneous electrophysiological recordings or targeted drug injections.

However, head-restraint necessarily reduces the behavioral repertoire of animals, which in some cases is helpful in reducing the complexity of the experimental environment but also raises concerns relating to the physiological relevance of the data collected. It is therefore also important to attempt to record cortical activity of freely moving animals that can exhibit more natural behaviors. In this aim, we have begun to develop a relatively simple fiber optic imaging approach, which we applied to freely moving mice [24] (Figure 6.5B). The fiber we used consisted of a well-ordered array of 300 × 300 individual fibers with 8-μm cores and high numerical aperture (0.6 NA). The fibers were rigidly arranged at each end to obtain reliable image transfer but, between the ends, they were loose, allowing great flexibility and rotation. This flexibility is crucial for the ability of the mouse to move and turn around relatively freely. Compared to direct imaging, we found large light losses using the fiber optic approach, with total collected fluorescence being only about a quarter of the normal. Nonetheless, there remained sufficient VSD signal to resolve cortical dynamics in single trials with direct correlation to the ongoing behavior.

There is, therefore, growing interest now in VSD imaging as we enter a new era, where—for the first time—we can resolve the real-time spatiotemporal dynamics of cortical activity in awake-behaving animals. In the final sections of this review, we present our first glimpses of cortical activity in awake mice, and we discuss our hopes for the future.

6.5.2. State-Dependent Sensory Processing

Sensory perception does not arise purely from the ongoing inflow of sensory stimuli arriving at any moment. Perception should rather be viewed as a creative process, where these sensory signals are interpreted and the percept constructed based in large part on our previous experiences. It is therefore not surprising that many studies have found profound differences in how sensory signals are processed in the brain, depending upon the behavior of the animal [41,53]. Such so-called top-down processing of sensory information is also strongly suggested by the intricate anatomical feedback loops that are present throughout the neocortex and other brain areas.

With respect to the whisker sensorimotor system, when the animal is awake, there are two obvious states. During some periods the whiskers are still, whereas during other periods, the whiskers move rapidly and rhythmically following forward and backward movements, in a behavior that has been termed “whisking.” Through VSD imaging of head-fixed awake mice [29], we found that evoked responses in the mouse sensorimotor cortex were very different, depending upon the ongoing behavior of the mouse (Figure 6.6A). These results agreed well with independent measurements made using whole-cell recordings in awake head-fixed mice [54]. During quiet periods, when the mouse was not whisking, a small brief deflection of a whisker evoked a sensory response that began in the barrel cortex after ~10 ms. Over the next milliseconds, the response spread over the somatosensory cortex. In addition, a secondary localized region of activation was observed in the whisker-related motor cortex, which also subsequently propagates. Clearly, even a single, brief whisker deflection evokes a highly distributed sensory response. The spreading responses observed under anesthesia (Figure 6.4C) therefore also appear to be relevant to physiological brain states. These large and distributed sensory responses might correspond to a “wake up” call for driving the cortex into an active state. Sensory-evoked responses in the whisker-related motor cortex correlated with the initiation of whisker movement. The motor cortex activity might therefore contribute to driving whisker movements, which could be directed to gather information relating to the immediate environment, and could be viewed as an attempt by the mouse to understand what caused the whisker deflection.

FIGURE 6.6. Imaging sensorimotor processing in awake-behaving mice.


Imaging sensorimotor processing in awake-behaving mice. (A) VSD imaging of sensorimotor cortex in a head-restrained mouse. A small metal particle is attached to the C2 whisker, and whisker deflections are evoked by brief 2 ms magnetic pulses. A craniotomy (more...)

The same stimulus during whisking evoked a very different cortical response (Figure 6.6A). The response began in a similar manner compared to during quiet wakefulness, with a localized response in the primary somatosensory barrel cortex. However, the amplitude was smaller, and it did not spread. During the active brain state, sensory processing can therefore remain confined to highly localized cortical regions. From these observations, it is clear that the cortical response to a simple whisker deflection depends strongly on the behavioral state. Future experiments must determine the mechanisms underlying this state-dependence and the relationships to perceptual thresholds. Furthermore, it would also be interesting to combine VSD imaging in behaving animals with electrophysiological measurements to reveal how the activity of single neurons relates to the population VSD signals and how the different signals are influenced by behavioral states.

6.5.3. Imaging the Cortical Response During Active Touch

A mouse whisks as it explores a novel environment, and it is therefore thought that this behavior is an active processing mode during which the mouse tries to acquire sensory information with its whiskers. Thus, in the same way that humans would reach out to touch and manipulate an object in order to gauge its size, shape, and texture, mice actively contact objects with their whiskers to obtain tactile information. Applying the VSD fiber imaging approach to freely moving mice [24], we obtained the first images of the cortical processing of touch as a mouse approached and contacted an object with a whisker (Figure 6.6B). We found that active touch of real objects evoked large spreading signals in the barrel cortex. They were again initiated in a small localized region, with a diameter similar to that of an individual barrel, and colocalized with the sensory-evoked response mapped under anesthesia. In the milliseconds that followed the first response, the VSD signal spread over a large cortical area. Future experiments must investigate how sensory responses to the touch of real objects (Figure 6.6B) compare quantitatively to the artificial deflections evoked by passive stimuli (Figure 6.6A). These experiments are just the beginnings of investigations into the cortex of awake animals, and there is obviously much further work to be done. We find it encouraging that the VSD imaging technique appears to provide useful information relating to cortical dynamics during behavior.


Sensory information is rapidly processed in highly distributed cortical networks. The spatiotemporal dynamics of cortical membrane potential changes can be resolved in vivo with millisecond temporal and columnar spatial resolution through VSD imaging. Application of this technique to awake-behaving mice reveals prominent state-dependent processing of whisker-related information within sensorimotor cortex. This VSD imaging technique opens the door to investigation of the distributed processing underlying active acquisition of tactile sensory information.

However, it is clear that there are major limitations to the current VSD imaging approach discussed in this review: (1) VSD signal amplitudes are small, (2) membranes are nonspecifically stained, and (3) the technique is restricted to the supragranular layers of the neocortex. These limitations mean that it is not currently possible to image the membrane potential of a network of neurons in vivo with cellular or subcellular resolution. Equally, right now it is not possible to differentiate between membrane potential changes in different subtypes of neurons, for example, GABAergic neurons versus excitatory neurons. Furthermore, it would clearly be very interesting to investigate the spatiotemporal dynamics of membrane potential changes in deep brain structures and to correlate them with the complex dynamics observed in the neocortex (Figure 6.6).

Fortunately, over the past years, there have been a number of technical advances that offer hope for improvements to in vivo imaging, which may allow in vivo imaging of subcellular compartments, specific neuronal subtypes and deep brain structures.


6.7.1. New Dyes and Nonlinear Optics

The development of RH1691, RH1692, and RH1838 operating at long wavelengths [12] has helped to increase the signal-to-noise ratio by reducing the contribution of heartbeat-related artifacts. However, the actual signal amplitude remains very small, usually well below 1% ΔF/F0. Calcium-sensitive dyes in comparison offer signals that are 100 times bigger. These large signals from calcium dyes, together with the development of methods for high-resolution two-photon imaging [55] and loading populations of cells [56], have allowed in vivo calcium imaging of networks with single cell resolution [48,57–61]. However, the calcium dye signals report almost exclusively action potential activity, and the subthreshold membrane potential dynamics remain hidden from this technique [27]. Increasing the signal amplitude for VSDs could therefore result in dramatic advances in our understanding of brain function.

The most obvious approach is simply to keep synthesizing new dyes, and by persistent screening one might find dyes with better characteristics than the currently used ones [62,63]. An equally promising route has been to excite the VSD with laser light, in order to optimize the excitation wavelength and therefore to obtain larger signal amplitudes [62]. So far, these approaches have yielded moderate but significant improvements for in vitro imaging that may in the future also contribute to improving in vivo imaging.

An exciting prospect is to use nonlinear optical phenomenon to enhance VSD signals. With the increasing availability of easy-to-use pulsed laser sources, multiphoton microscopy is becoming a widely used technology for in vivo imaging. The most important advantage offered by two-photon microscopy is the inherent optical sectioning resulting from the need for near-simultaneous absorption of two photons per excitation event, which in practice occurs primarily in the focal plane. In contrast to water soluble dyes, VSDs insert into the plasma membranes and are therefore oriented in a specific direction. All VSD molecules on a given patch of membrane are oriented in the same direction, and this can contribute to an interesting nonlinear phenomenon termed second harmonic generation (SHG). In the SHG process, two photons are near simultaneously absorbed as for two-photon fluorescence. However, whereas in fluorescence the excited molecule emits a longer wavelength photon, in SHG the two photons are converted into a single photon with twice the energy. The SHG signal depends upon the hyperpolarizability of the VSD, enhanced by resonance of the excitation electromagnetic field, in addition to the electric field experienced by the VSD molecule. Recent SHG measurements show promising results for high-resolution VSD imaging using the steryl dye FM4-64 [64,65].

6.7.2. Voltage-Sensitive Fluorescent Proteins

Increasing the signal amplitudes of fluorescent dyes will be extremely helpful for improving the resolution of VSD measurements; however, if membranes are non-specifically labeled, then we will still not know to which neurons the VSD signal relates. Neurons have very fine intermingled axonal and dendritic processes, and glial cells also have extremely intricate membranes. If all membranes are similarly labeled, then it will not be possible to distinguish nearby fluorescently labeled dendrites, axons, or glia. One interesting approach would be to label single neurons by dye intracellular injections, but this approach limits the possibilities for investigating network activity in many neurons. Labeling of specific cells types in the brain has been accomplished most elegantly by genetic methods. Through cell-type-specific promoters and advanced combinatorial genetic strategies of gene regulation, it has become possible to label different populations of neurons with fluorescent proteins. For example, in one mouse line, all GABAergic neurons express the green fluorescent protein (GFP) [66]. A great deal of hope for the future development of voltage imaging has therefore been placed in the development of genetically encoded voltage-sensitive fluorescent proteins.

The first genetically encoded indicators responded to changes in calcium concentration [67], and since then, a number of improved genetically encoded calcium-sensitive indicators (GECIs) have been developed and found to be functional in the mammalian brain [68–70]. Fluorescent proteins sensitive to changes in membrane potential have also been developed, most of which are based upon GFP attached to different voltage-gated ion channels [71–73]. So far these approaches have not been successful in the mammalian brain, mainly because the proteins are not trafficked efficiently to the plasma membrane [74]. However, renewed hope now comes from coupling a voltage-sensitive phosphatase to GFP [75].

6.7.3. Fiber Optic VSD Imaging

The ability to image through fiber optics [24,76,77] has two major advantages over conventional imaging. Firstly, the animal is free to move around, and such measurements are therefore inherently more physiological. However, this advantage is balanced by decreased experimental control. Perhaps the best compromise is the recently designed “freely moving head-restrained” system, where the mouse can run around on a floating ball while head-fixed [60]. In the future, one could combine this approach with virtual environments, providing enormous experimental potential. The second advantage offered by fiber optics is the possibility to image deep in the brain through microendoscopy. Fiber arrays attached to graded-index (GRIN) lenses can be manufactured to have diameters as small as a few hundred microns and can therefore be inserted deep into the brain with minimal damage. In combination with dye delivery systems, or genetically encoded voltage-sensitive fluorescent proteins, such fiber optic microendoscopes could provide the first glimpses of membrane potential spatiotemporal dynamics in deep brain areas like thalamus or superior colliculus.


F.M. is a recipient of a European Molecular Biology Organization (EMBO) long-term fellowship.


Nicolelis MA, Dimitrov D, Carmena JM, Crist R, Lehew G, Kralik JD, Wise SP. Chronic, multisite, multielectrode recordings in macaque monkeys. Proc Natl Acad Sci USA. 2003;100:11041–11046. [PMC free article: PMC196923] [PubMed: 12960378]
Buzsáki G. Large-scale recording of neuronal ensembles. Nat Neurosci. 2004;7:446–451. [PubMed: 15114356]
Grinvald A, Lieke E, Frostig RD, Gilbert CD, Wiesel TN. Functional architecture of cortex revealed by optical imaging of intrinsic signals. Nature. 1986;324:361–364. [PubMed: 3785405]
Frostig RD, Lieke EE, Ts’o DY, Grinvald A. Cortical functional architecture and local coupling between neuronal activity and the microcirculation revealed by in vivo high-resolution optical imaging of intrinsic signals. Proc Natl Acad Sci USA. 1990;87:6082–6086. [PMC free article: PMC54476] [PubMed: 2117272]
Polley DB, Chen-Bee CH, Frostig RD. Two directions of plasticity in the sensory-deprived adult cortex. Neuron. 1999;24:623–637. [PubMed: 10595514]
Polley DB, Kvasnak E, Frostig RD. Naturalistic experience transforms sensory maps in the adult cortex of caged animals. Nature. 2004;429:67–71. [PubMed: 15129281]
Tasaki I, Watanabe A, Sandlin R, Carnay L. Changes in fluorescence, turbidity and birefringence associated with nerve excitation. Proc. Natl. Acad. Sci. USA. 1968;61:883–888. [PMC free article: PMC305410] [PubMed: 4301149]
Salzberg BM, Davila HV, Cohen LB. Optical recording of impulses in individual neurons of an invertebrate central nervous system. Nature. 1973;246:508–509. [PubMed: 4357630]
Grinvald A, Anglister L, Freeman JA, Hildesheim R, Manker A. Real-time optical imaging of naturally evoked electrical activity in intact frog brain. Nature. 1984;308:848–850. [PubMed: 6717577]
Orbach HS, Cohen LB, Grinvald A. Optical mapping of electrical activity in rat somatosensory and visual cortex. J Neurosci. 1985;5:1886–1895. [PubMed: 4020423]
Shoham D, Glaser DE, Arieli A, Kenet T, Wijnbergen C, Toledo Y, Hildesheim R, Grinvald A. Imaging cortical dynamics at high spatial and temporal resolution with novel blue voltage-sensitive dyes. Neuron. 1999;24:791–802. [PubMed: 10624943]
Grinvald A, Hildesheim R. VSDI: A new era in functional imaging of cortical dynamics. Nat Rev Neurosci. 2004;5:874–885. [PubMed: 15496865]
Tsodyks M, Kenet T, Grinvald A, Arieli A. Linking spontaneous activity of single cortical neurons and the underlying functional architecture. Science. 1999;286:1943–1946. [PubMed: 10583955]
Seidemann E, Arieli A, Grinvald A, Slovin H. Dynamics of depolarization and hyperpolarization in the frontal cortex and saccade goal. Science. 2002;295:862–865. [PubMed: 11823644]
Slovin H, Arieli A, Hildesheim R, Grinvald A. Long-term voltage-sensitive dye imaging reveals cortical dynamics in behaving monkeys. J Neurophysiol. 2002;88:3421–3438. [PubMed: 12466458]
Spors H, Grinvald A. Spatiotemporal dynamics of odor representations in the mammalian olfactory bulb. Neuron. 2002;34:301–315. [PubMed: 11970871]
Derdikman D, Hildesheim R, Ahissar E, Arieli A, Grinvald A. Imaging spatiotemporal dynamics of surround inhibition in the barrels somatosensory cortex. J Neurosci. 2003;23:3100–3105. [PubMed: 12716915]
Kenet T, Bibitchkov D, Tsodyks M, Grinvald A, Arieli A. Spontaneously emerging cortical representations of visual attributes. Nature. 2003;425:954–956. [PubMed: 14586468]
Petersen CCH, Grinvald A, Sakmann B. Spatiotemporal dynamics of sensory responses in layer 2/3 of rat barrel cortex measured in vivo by voltage-sensitive dye imaging combined with whole-cell voltage recordings and anatomical reconstructions. J Neurosci. 2003;23:1298–1309. [PubMed: 12598618]
Petersen CCH, Hahn TTG, Mehta M, Grinvald A, Sakmann B. Interaction of sensory responses with spontaneous depolarization in layer 2/3 barrel cortex. Proc Natl Acad Sci USA. 2003;100:13638–13643. [PMC free article: PMC263866] [PubMed: 14595013]
Jancke D, Chavane F, Naaman S, Grinvald A. Imaging cortical correlates of illusion in early visual cortex. Nature. 2004;428:423–426. [PubMed: 15042090]
Chen Y, Geisler WS, Seidemann E. Optimal decoding of correlated neural population responses in the primate visual cortex. Nat Neurosci. 2006;9:1412–1420. [PMC free article: PMC1851689] [PubMed: 17057706]
Civillico EF, Contreras D. Integration of evoked responses in supragranular cortex studied with optical recordings in vivo. J Neurophysiol. 2006;96:336–351. [PubMed: 16571736]
Ferezou I, Bolea S, Petersen CCH. Visualizing the cortical representation of whisker touch: Voltage-sensitive dye imaging in freely moving mice. Neuron. 2006;50:617–629. [PubMed: 16701211]
Roland PE, Hanazawa A, Undeman C, Eriksson D, Tompa T, Nakamura H, Valentiniene S, Ahmed B. Cortical feedback depolarization waves: A mechanism of top-down influence on early visual areas. Proc Natl Acad Sci USA. 2006;103:12586–12591. [PMC free article: PMC1531644] [PubMed: 16891418]
Benucci A, Frazor RA, Carandini M. Standing waves and traveling waves distinguish two circuits in visual cortex. Neuron. 2007;55:103–117. [PMC free article: PMC2171365] [PubMed: 17610820]
Berger T, Borgdorff AJ, Crochet S, Neubauer FB, Lefort S, Fauvet B, Ferezou I, Carleton A, Luscher HR, Petersen CCH. Combined voltage and calcium epifluorescence imaging in vitro and in vivo reveals subthreshold and suprathreshold dynamics of mouse barrel cortex. J Neurophysiol. 2007;97:3751–3762. [PubMed: 17360827]
Borgdorff AJ, Poulet JFA, Petersen CCH. Facilitating sensory responses in developing mouse somatosensory barrel cortex. J Neurophysiol. 2007;97:2992–3003. [PubMed: 17287446]
Ferezou I, Haiss F, Gentet LJ, Aronoff R, Weber B, Petersen CCH. Spatiotemporal dynamics of cortical sensorimotor integration in behaving mice. Neuron. 2007;56:907–923. [PubMed: 18054865]
Lippert MT, Takagaki K, Xu W, Huang X, Wu JY. Methods for voltage-sensitive dye imaging of rat cortical activity with high signal-to-noise ratio. J Neurophysiol. 2007;98:502–512. [PMC free article: PMC2855339] [PubMed: 17493915]
Sharon D, Jancke D, Chavane F, Na’aman S, Grinvald A. Cortical response field dynamics in cat visual cortex. Cereb Cortex. 2007;17:2866–2877. [PubMed: 17395608]
Wallace DJ, Sakmann B. Plasticity of representational maps in somatosensory cortex observed by in vivo voltage-sensitive dye imaging. Cereb Cortex. 2008;18:1361–1373. [PubMed: 17921458]
Xu W, Huang X, Takagaki K, Wu JY. Compression and reflection of visually evoked cortical waves. Neuron. 2007;55:119–129. [PMC free article: PMC1988694] [PubMed: 17610821]
Yang Z, Heeger DJ, Seidemann E. Rapid and precise retinotopic mapping of the visual cortex obtained by voltage-sensitive dye imaging in the behaving monkey. J Neurophysiol. 2007;98:1002–1014. [PMC free article: PMC2214852] [PubMed: 17522170]
Chen Y, Geisler WS, Seidemann E. Optimal temporal decoding of neural population responses in a reaction-time visual detection task. J Neurophysiol. 2008;99:1366–1379. [PMC free article: PMC2667890] [PubMed: 18199810]
Takagaki K, Zhang C, Wu JY, Lippert MT. Crossmodal propagation of sensory-evoked and spontaneous activity in the rat neocortex. Neurosci Lett. 2008;431:191–196. [PMC free article: PMC2292672] [PubMed: 18178313]
Yoshida T, Sakagami M, Katura T, Yamazaki K, Tanaka S, Iwamoto M, Tanaka N. Anisotropic spatial coherence of ongoing and spontaneous activities in auditory cortex. Neurosci Res. 2008;61:49–55. [PubMed: 18314210]
Nauhaus I, Benucci A, Carandini M, Ringach DL. Neuronal selectivity and local map structure in visual cortex. Neuron. 2008;57:673–679. [PMC free article: PMC2322861] [PubMed: 18341988]
Zhang J, Davidson RM, Wei MD, Loew LM. Membrane electric properties by combined patch clamp and fluorescence ratio imaging in single neurons. Biophys J. 1998;74:48–53. [PMC free article: PMC1299360] [PubMed: 9449308]
Ratzlaff EH, Grinvald A. A tandem-lens epifluorescence macroscope: Hundred-fold brightness advantage for wide-field imaging. J Neurosci Methods. 1991;36:127–137. [PubMed: 1905769]
Petersen CCH. The functional organization of the barrel cortex. Neuron. 2007;56:339–355. [PubMed: 17964250]
Moore CI, Nelson SB. Spatio-temporal subthreshold receptive fields in the vibrissa representation of rat primary somatosensory cortex. J Neurophysiol. 1998;80:2882–2892. [PubMed: 9862892]
Zhu JJ, Connors BW. Intrinsic firing patterns and whiskers-evoked synaptic responses of neurons in the rat barrel cortex. J Neurophysiol. 1999;81:1171–1183. [PubMed: 10085344]
Brecht M, Roth A, Sakmann B. Dynamic receptive fields of reconstructed pyramidal cells in layers 3 and 2 of rat somatosensory barrel cortex. J Physiol. 2003;553:243–265. [PMC free article: PMC2343497] [PubMed: 12949232]
Steriade M, Nunez A, Amzica F. A novel slow (<1 Hz) oscillation of neo-cortical neurons in vivo: Depolarizing and hyperpolarizing components. J Neurosci. 1993;13:3252–3265. [PubMed: 8340806]
Cowan RL, Wilson CJ. Spontaneous firing patterns and axonal projections of single corticostriatal neurons in the rat medial agranular cortex. J Neurophysiol. 1994;71:17–32. [PubMed: 8158226]
Sachdev RN, Ebner FF, Wilson CJ. Effect of subthreshold up and down states on the whisker-evoked response in somatosensory cortex. J Neurophysiol. 2004;92:3511–3521. [PubMed: 15254074]
Kerr JN, Greenberg D, Helmchen F. Imaging input and output of neocortical networks in vivo. Proc Natl Acad Sci USA. 2005;102:14063–14068. [PMC free article: PMC1201343] [PubMed: 16157876]
Lampl I, Reichova I, Ferster D. Synchronous membrane potential fluctuations in neurons of the cat visual cortex. Neuron. 1999;22:361–374. [PubMed: 10069341]
Volgushev M, Chauvette S, Mukovski M, Timofeev I. Precise long-range synchronization of activity and silence in neocortical neurons during slow-wave oscillations. J Neurosci. 2006;26:5665–5672. [PubMed: 16723523]
de Kock CP, Bruno RM, Spors H, Sakmann B. Layer- and cell-type-specific suprathreshold stimulus representation in rat primary somatosensory cortex. J Physiol. 2007;581:139–154. [PMC free article: PMC2075227] [PubMed: 17317752]
Vincent JL, Patel GH, Fox MD, Snyder AZ, Baker JT, Van Essen DC, Zempel JM, Snyder LH, Corbetta M, Raichle ME. Intrinsic functional architecture in the anaesthetized monkey brain. Nature. 2007;447:83–86. [PubMed: 17476267]
Gilbert CD, Sigman M. Brain states: Top-down influences in sensory processing. Neuron. 2007;54:677–696. [PubMed: 17553419]
Crochet S, Petersen CCH. Correlating whisker behavior with membrane potential in barrel cortex of awake mice. Nat Neurosci. 2006;9:608–610. [PubMed: 16617340]
Denk W, Strickler JH, Webb WW. Two-photon laser scanning fluorescence microscopy. Science. 1990;248:73–76. [PubMed: 2321027]
Stosiek C, Garaschuk O, Holthoff K, Konnerth A. In vivo two-photon calcium imaging of neuronal networks. Proc Natl Acad Sci USA. 2003;100:7319–7324. [PMC free article: PMC165873] [PubMed: 12777621]
Ohki K, Chung S, Ch’ng YH, Kara P, Reid RC. Functional imaging with cellular resolution reveals precise micro-architecture in visual cortex. Nature. 2005;433:597–603. [PubMed: 15660108]
Ohki K, Chung S, Kara P, Hübener M, Bonhoeffer T, Reid RC. Highly ordered arrangement of single neurons in orientation pinwheels. Nature. 2006;442:925–928. [PubMed: 16906137]
Kerr JN, de Kock CP, Greenberg DS, Bruno RM, Sakmann B, Helmchen F. Spatial organization of neuronal population responses in layer 2/3 of rat barrel cortex. J Neurosci. 2007;27:13316–13328. [PubMed: 18045926]
Dombeck DA, Khabbaz AN, Collman F, Adelman TL, Tank DW. Imaging large scale neural activity with cellular resolution in awake mobile mice. Neuron. 2007;56:43–57. [PMC free article: PMC2268027] [PubMed: 17920014]
Sato TR, Gray NW, Mainen ZF, Svoboda K. The functional microarchitecture of the mouse barrel cortex. PLoS Biol. 2007:e189. 5. [PMC free article: PMC1914403] [PubMed: 17622195]
Kuhn B, Fromherz P, Denk W. High sensitivity of stark-shift voltage-sensing dyes by one- or two-photon excitation near the red spectral edge. Biophys. J. 2004;87:631–639. [PMC free article: PMC1304385] [PubMed: 15240496]
Zhou WL, Yan P, Wuskell JP, Loew LM, Antic SD. Intracellular long-wavelength voltage-sensitive dyes for studying the dynamics of action potentials in axons and thin dendrites. J Neurosci Methods. 2007;164:225–239. [PMC free article: PMC2001318] [PubMed: 17560661]
Dombeck DA, Sacconi L, Blanchard-Desce M, Webb WW. Optical recording of fast neuronal membrane potential transients in acute mammalian brain slices by second-harmonic generation microscopy. J Neurophysiol. 2005;94:3628–3636. [PubMed: 16093337]
Nuriya M, Jiang J, Nemet B, Eisenthal KB, Yuste R. Imaging membrane potential in dendritic spines. Proc Natl Acad Sci USA. 2006;103:786–790. [PMC free article: PMC1334676] [PubMed: 16407122]
Tamamaki N, Yanagawa Y, Tomioka R, Miyazaki J, Obata K, Kaneko T. Green fluorescent protein expression and colocalization with calretinin, parvalbumin, and somatostatin in the GAD67-GFP knock-in mouse. J Comp Neurol. 2003;467:60–79. [PubMed: 14574680]
Miyawaki A, Llopis J, Heim R, McCaffery JM, Adams JA, Ikura M, Tsien RY. Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin. Nature. 1997;388:882–887. [PubMed: 9278050]
Hasan MT, Friedrich RW, Euler T, Larkum ME, Giese G, Both M, Duebel J, Waters J, Bujard H, Griesbeck O, Tsien RY, Nagai T, Miyawaki A, Denk W. Functional fluorescent Ca2+ indicator proteins in transgenic mice under TET control. PLoS Biol. 2004;2:e163. [PMC free article: PMC423138] [PubMed: 15208716]
Díez-García J, Matsushita S, Mutoh H, Nakai J, Ohkura M, Yokoyama J, Dimitrov D, Knöpfel T. Activation of cerebellar parallel fibers monitored in transgenic mice expressing a fluorescent Ca2+ indicator protein. Eur J Neurosci. 2005;22:627–635. [PubMed: 16101744]
Heim N, Garaschuk O, Friedrich MW, Mank M, Milos RI, Kovalchuk Y, Konnerth A, Griesbeck O. Improved calcium imaging in transgenic mice expressing a troponin C-based biosensor. Nat Methods. 2007;4:127–129. [PubMed: 17259991]
Siegel MS, Isacoff EY. A genetically encoded optical probe of membrane voltage. Neuron. 1997;19:735–741. [PubMed: 9354320]
Sakai R, Repunte-Canonigo V, Raj CD, Knopfel T. Design and characterization of a DNA-encoded, voltage-sensitive fluorescent protein. Eur J Neurosci. 2001;13:2314–2318. [PubMed: 11454036]
Ataka K, Pieribone VA. A genetically targetable fluorescent probe of channel gating with rapid kinetics. Biophys J. 2002;82:509–516. [PMC free article: PMC1302490] [PubMed: 11751337]
Baker BJ, Lee H, Pieribone VA, Cohen LB, Isacoff EY, Knopfel T, Kosmidis EK. Three fluorescent protein voltage sensors exhibit low plasma membrane expression in mammalian cells. J Neurosci Methods. 2007;161:32–38. [PubMed: 17126911]
Dimitrov D, He Y, Mutoh H, Baker BJ, Cohen L, Akemann W, Knöpfel T. Engineering and characterization of an enhanced fluorescent protein voltage sensor. PLoS ONE. 2007;2:e440. [PMC free article: PMC1857823] [PubMed: 17487283]
Helmchen F, Fee MS, Tank DW, Denk W. A miniature head-mounted two-photon microscope: High resolution brain imaging in freely moving animals. Neuron. 2001;31:903–912. [PubMed: 11580892]
Flusberg BA, Cocker ED, Piyawattanametha W, Jung JC, Cheung EL, Schnitzer MJ. Fiber-optic fluorescence imaging. Nat Methods. 2005;2:941–950. [PMC free article: PMC2849801] [PubMed: 16299479]
Copyright © 2009, Taylor & Francis Group, LLC.
Bookshelf ID: NBK20229
PubReader format: click here to try


  • PubReader
  • Print View
  • Cite this Page

Other titles in this collection

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to pubmed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...