Modeling the impact of common noise inputs on the network activity of retinal ganglion cells

Synchronized spontaneous firing among retinal ganglion cells (RGCs), on timescales faster than visual responses, has been reported in many studies. Two candidate mechanisms of synchronized firing include direct coupling and shared noisy inputs. In neighboring parasol cells of primate retina, which exhibit rapid synchronized firing that has been studied extensively, recent experimental work indicates that direct electrical or synaptic coupling is weak, but shared synaptic input in the absence of modulated stimuli is strong. However, previous modeling efforts have not accounted for this aspect of firing in the parasol cell population. Here we develop a new model that incorporates the effects of common noise, and apply it to analyze the light responses and synchronized firing of a large, densely-sampled network of over 250 simultaneously recorded parasol cells. We use a generalized linear model in which the spike rate in each cell is determined by the linear combination of the spatio-temporally filtered visual input, the temporally filtered prior spikes of that cell, and unobserved sources representing common noise. The model accurately captures the statistical structure of the spike trains and the encoding of the visual stimulus, without the direct coupling assumption present in previous modeling work. Finally, we examined the problem of decoding the visual stimulus from the spike train given the estimated parameters. The common-noise model produces Bayesian decoding performance as accurate as that of a model with direct coupling, but with significantly more robustness to spike timing perturbations.     

Carbenoxolone blocks the light-evoked rise in intracellular calcium in isolated melanopsin ganglion cell photoreceptors

Background

Retinal ganglion cells expressing the photopigment melanopsin are intrinsically photosensitive (ipRGCs). These ganglion cell photoreceptors send axons to several central targets involved in a variety of functions. Within the retina ipRGCs provide excitatory drive to dopaminergic amacrine cells via glutamatergic signals and ipRGCs are coupled to wide-field GABAergic amacrine cells via gap junctions. However, the extent to which ipRGCs are coupled to other retinal neurons in the ganglion cell layer via gap junctions is unclear. Carbenoxolone, a widely employed gap junction inhibitor, greatly reduces the number of retinal neurons exhibiting non-rod, non-cone mediated light-evoked Ca²⁺ signals suggesting extensive intercellular coupling between ipRGCs and non-ipRGCs in the ganglion cell layer. However, carbenoxolone may directly inhibit light-evoked Ca²⁺ signals in ipRGCs independent of gap junction blockade.

Methodology/Principal Findings

To test the possibility that carbenoxolone directly inhibits light-evoked Ca²⁺ responses in ipRGCs, the light-evoked rise in intracellular Ca²⁺ ([Ca²⁺]_i) was examined using fura-2 imaging in isolated rat ipRGCs maintained in short-term culture in the absence and presence of carbenoxolone. Carbenoxolone at 50 and 100 µM concentrations completely abolished the light-evoked rise in [Ca²⁺]_i in isolated ipRGCs. Recovery from carbenoxolone inhibition was variable.

Conclusions/Significance

We demonstrate that the light-evoked rise in [Ca²⁺]_i in isolated mammalian ganglion cell photoreceptors is inhibited by carbenoxolone. Since the light-evoked increase in [Ca²⁺]_i in isolated ipRGCs is almost entirely due to Ca²⁺ entry via L-type voltage-gated calcium channels and carbenoxolone does not inhibit light-evoked action potential firing in ipRGCs in situ, carbenoxolone may block the light-evoked increase in [Ca²⁺]_i in ipRGCs by blocking L-type voltage-gated Ca²⁺ channels. The ability of carbenoxolone to block evoked Ca²⁺ responses must be taken into account when interpreting the effects of this pharmacological agent on retinal or other neuronal circuits, particularly if a change in [Ca²⁺]_i is the output being measured.     

Dynamic Range of Vertebrate Retina Ganglion Cells: Importance of Active Dendrites and Coupling by Electrical Synapses

The vertebrate retina has a very high dynamic range. This is due to the concerted action of its diverse cell types. Ganglion cells, which are the output cells of the retina, have to preserve this high dynamic range to convey it to higher brain areas. Experimental evidence shows that the firing response of ganglion cells is strongly correlated with their total dendritic area and only weakly correlated with their dendritic branching complexity. On the other hand, theoretical studies with simple neuron models claim that active and large dendritic trees enhance the dynamic range of single neurons. Theoretical models also claim that electrical coupling between ganglion cells via gap junctions enhances their collective dynamic range. In this work we use morphologically reconstructed multi-compartmental ganglion cell models to perform two studies. In the first study we investigate the relationship between single ganglion cell dynamic range and number of dendritic branches/total dendritic area for both active and passive dendrites. Our results support the claim that large and active dendrites enhance the dynamic range of a single ganglion cell and show that total dendritic area has stronger correlation with dynamic range than with number of dendritic branches. In the second study we investigate the dynamic range of a square array of ganglion cells with passive or active dendritic trees coupled with each other via dendrodendritic gap junctions. Our results suggest that electrical coupling between active dendritic trees enhances the dynamic range of the ganglion cell array in comparison with both the uncoupled case and the coupled case with cells with passive dendrites. The results from our detailed computational modeling studies suggest that the key properties of the ganglion cells that endow them with a large dynamic range are large and active dendritic trees and electrical coupling via gap junctions.     

Synaptic connections of cone bipolar cells that express the neurokinin 1 receptor in the rabbit retina

We have investigated and further characterized, in the rabbit retina, the synaptic connectivity of the ON-type cone bipolar cells that are immunoreactive for an antibody against the neurokinin-1 receptor (NK1R). NK1R-immunoreactive bipolar cell axons terminate in stratum 4 of the inner plexiform layer. The axons of NK1R-positive bipolar cells receive synaptic inputs from amacrine cells through conventional synapses and from putative AII amacrine cells via gap junctions. The major outputs from NK1R-positive bipolar cells make contacts with amacrine cell processes. The most frequent postsynaptic dyads comprise two amacrine cell processes. Double-labeling experiments with antibodies against NK1R and either calretinin or glycine have demonstrated that NK1R-immunoreactive bipolar cells form gap junctions with AII amacrine cells. Thus, NK1R-positive cone bipolar cells, together with calbindin-positive cone bipolar cells, may play an important role in transferring rod signals to the ON-type ganglion cells of the cone pathway in the rabbit retina.I.-B. Kim and M.R. Park contributed equally to this work.This work was supported by the Ministry of Science and Technology of Korea (grant no. M1-0108-00-0059; Neurobiology Support Grant).     

Synaptic connections of calbindin-immunoreactive cone bipolar cells in the inner plexiform layer of rabbit retina

In the mammalian retina, information concerning various aspects of an image is transferred in parallel, and cone bipolar cells are thought to play a major role in this parallel processing. We have examined the synaptic connections of calbindin-immunoreactive (IR) ON cone bipolar cells in the inner plexiform layer (IPL) of rabbit retina and have compared these synaptic connections with those that we have previously described for neurokinin 1 (NK1) receptor-IR cone bipolar cells. A total of 325 synapses made by calbindin-IR bipolar axon terminals have been identified in sublamina b of the IPL. The axons of calbindin-IR bipolar cells receive synaptic inputs from amacrine cells through conventional synapses and are coupled to putative AII amacrine cells via gap junctions. The major output from calbindin-IR bipolar cells is to amacrine cell processes. These data resemble our findings for NK1 receptor-IR bipolar cells. However, the incidences of output synapses to ganglion cell dendrites of calbindin-IR bipolar cells are higher compared with the NK1-receptor-IR bipolar cells. On the basis of stratification level and synaptic connections, calbindin-IR ON cone bipolar cells might thus play an important role in the processing of various visual aspects, such as contrast, orientation, and approach sensing, and in transferring rod signals to the ON cone pathway.     

Synchronized firing in the retina

Synchronized firing in neural populations has been proposed to constitute an elementary aspect of the neural code, but a complete understanding of its origins and significance has been elusive. Synchronized firing has been extensively documented in retinal ganglion cells, the output neurons of the retina. However, differences in synchronized firing across species and cell types have led to varied conclusions about its mechanisms and role in visual signaling. Recent work on two identified cell populations in the primate retina, the ON-parasol and OFF-parasol cells, permits a more unified understanding. Intracellular recordings reveal that synchronized firing in these cell types arises primarily from common synaptic input to adjacent pairs of cells. Statistical analysis indicates that local pairwise interactions can explain the pattern of synchronized firing in the entire parasol cell population. Computational analysis reveals that the aggregate impact of synchronized firing on the visual signal is substantial. Thus, in the parasol cells, the origin and impact of synchronized firing on the neural code may be understood as locally shared input which influences the visual signals transmitted from eye to brain.     

Mouse Ganglion-Cell Photoreceptors Are Driven by the Most Sensitive Rod Pathway and by Both Types of Cones

Intrinsically photosensitive retinal ganglion cells (ipRGCs) are depolarized by light by two mechanisms: directly, through activation of their photopigment melanopsin; and indirectly through synaptic circuits driven by rods and cones. To learn more about the rod and cone circuits driving ipRGCs, we made multielectrode array (MEA) and patch-clamp recordings in wildtype and genetically modified mice. Rod-driven ON inputs to ipRGCs proved to be as sensitive as any reaching the conventional ganglion cells. These signals presumably pass in part through the primary rod pathway, involving rod bipolar cells and AII amacrine cells coupled to ON cone bipolar cells through gap junctions. Consistent with this interpretation, the sensitive rod ON input to ipRGCs was eliminated by pharmacological or genetic disruption of gap junctions, as previously reported for conventional ganglion cells. A presumptive cone input was also detectable as a brisk, synaptically mediated ON response that persisted after disruption of rod ON pathways. This was roughly three log units less sensitive than the rod input. Spectral analysis revealed that both types of cones, the M- and S-cones, contribute to this response and that both cone types drive ON responses. This contrasts with the blue-OFF, yellow-ON chromatic opponency reported in primate ipRGCs. The cone-mediated response was surprisingly persistent during steady illumination, echoing the tonic nature of both the rod input to ipRGCs and their intrinsic, melanopsin-based phototransduction. These synaptic inputs greatly expand the dynamic range and spectral bandpass of the non-image-forming visual functions for which ipRGCs provide the principal retinal input.     

A key role of starburst amacrine cells in originating retinal directional selectivity and optokinetic eye movement.

The directional selectivity of retinal ganglion cell responses represents a primitive pattern recognition that operates within a retinal neural circuit. The cellular origin and mechanism of directional selectivity were investigated by selectively eliminating retinal starburst amacrine cells, using immunotoxin-mediated cell targeting techniques. Starburst cell ablation in the adult retina abolished not only directional selectivity of ganglion cell responses but also an optokinetic eye reflex derived by stimulus movement. Starburst cells therefore serve as the key element that discriminates the direction of stimulus movement through integrative synaptic transmission and play a pivotal role in information processing that stabilizes image motion.     

Expression of alpha 7 nicotinic acetylcholine receptors by bipolar, amacrine, and ganglion cells of the rabbit retina.

Cholinergic agents affect the light responses of many ganglion cells (GCs) in the mammalian retina by activating nicotinic acetylcholine receptors (nAChRs). Whereas retinal neurons that express beta2 subunit-containing nAChRs have been characterized in the rabbit retina, expression patterns of other nAChR subtypes remain unclear. Therefore, we evaluated the expression of alpha7 nAChRs in retinal neurons by means of single-, double-, and triple-label immunohistochemistry. Our data demonstrate that, in the rabbit retina, several types of bipolar cells, amacrine cells, and cells in the GC layer express alpha7 nAChRs. At least three different populations of cone bipolar cells exhibited alpha7 labeling, whereas glycine-immunoreactive amacrine cells comprised the majority of alpha7-positive amacrine cells. Some GABAergic amacrine cells also displayed alpha7 immunoreactivity; alpha7 labeling was never detected in rod bipolar cells or rod amacrine cells (AII amacrine cells). Our data suggest that activation of alpha7 nAChRs by acetylcholine (ACh) or choline may affect glutamate release from several types of cone bipolar cells, modulating GC responses. ACh-induced excitation of inhibitory amacrine cells might cause either inhibition or disinhibition of other amacrine and GC circuits. Finally, ACh may act on alpha7 nAChRs expressed by GCs themselves.     

RdgB2 is required for dim-light input into intrinsically photosensitive retinal ganglion cells

A subset of retinal ganglion cells is intrinsically photosensitive (ipRGCs) and contributes directly to the pupillary light reflex and circadian photoentrainment under bright-light conditions. ipRGCs are also indirectly activated by light through cellular circuits initiated in rods and cones. A mammalian homologue (RdgB2) of a phosphoinositide transfer/exchange protein that functions in Drosophila phototransduction is expressed in the retinal ganglion cell layer. This raised the possibility that RdgB2 might function in the intrinsic light response in ipRGCs, which depends on a cascade reminiscent of Drosophila phototransduction. Here we found that under high light intensities, RdgB2^−/− mutant mice showed normal pupillary light responses and circadian photoentrainment. Consistent with this behavioral phenotype, the intrinsic light responses of ipRGCs in RdgB2^−/− were indistinguishable from wild-type. In contrast, under low-light conditions, RdgB2^−/− mutants displayed defects in both circadian photoentrainment and the pupillary light response. The RdgB2 protein was not expressed in ipRGCs but was in GABAergic amacrine cells, which provided inhibitory feedback onto bipolar cells. We propose that RdgB2 is required in a cellular circuit that transduces light input from rods to bipolar cells that are coupled to GABAergic amacrine cells and ultimately to ipRGCs, thereby enabling ipRGCs to respond to dim light.     

How voltage-gated ion channels alter the functional properties of ganglion and amacrine cell dendrites

The present study compares the structure and function of retinal ganglion and amacrine cell dendrites. Although a superficial similarity exists between amacrine and ganglion cell dendrites, a comparison between the branching pattern of the two cell types reveals differences which can only be appreciated at the microscopic level. Whereas decremental branching is found in ganglion cells, a form of non-decremental or "trunk branching" is observed in amacrine cell dendrites. Physiological differences are also observed in amacrine vs ganglion cells in which many amacrine cells generate dendritic impulses which can be readily distinguished from those of the soma, while separate dendritic impulses in ganglion cell dendrites have not been reported. Despite these differences, both amacrine and ganglion cell dendrites appear to contain voltage-gated ion channels, including TTX-sensitive sodium channels. One way to account for separate dendritic impulses in amacrine cells is to have a higher density of sodium channels and we generally find in modeling studies that a dendritic sodium channel density that is more than about 50% of that in the soma is required for excitatory, synaptic currents to give rise to local dendritic spike activity. Under these conditions, impulses can be generated in the dendrites and propagate for some distance along the dendritic tree. When the soma generates impulse activity in amacrine cells, it can activate, antidromically, the entire dendritic tree. Although ganglion cell dendrites do not appear to generate independent impulses, the presence of voltage-gated ion channels in these structures appears to be important for their function. Modeling studies demonstrate that when dendrites lack voltage-gated ion channels, impulse activity evoked by current applied to the cell body is generated at rates that are much higher than those observed physiologically. However, by placing ion channels in the dendrites at a reduced density compared to those of amacrine cells, the firing rate of ganglion cells becomes more physiological and the relationship between frequency and current (F/I relationship) can be precisely matched with physiological data. Recent studies have demonstrated the presence of T-type calcium channels in ganglion cells and our analysis suggests that they are found in higher density in the dendrites compared to the soma. This is the first voltage-gated ion channel which appears more localized to the dendrites than other cell copartments and this difference alone cries for an interpretation. The presence of a significant T-type calcium channel density in the dendrites can influence their integrative properties in several important ways. First, excitatory synaptic currents can be augmented by the activation of T-type calcium channels, although this is more likely to occur for transient rather than sustained synaptic currents because T-type currents show strong inactivation properties. In addition, T-type calcium channels may serve to limit the electrical load which dendrites impose on the spike initiation process and thus enhance the speed with which impulses can be triggered by the impulse generation site. This role whill enhance the safety factor for impulses traveling in the orthograde direction.     

Modelling the Effects of Electrical Coupling between Unmyelinated Axons of Brainstem Neurons Controlling Rhythmic Activity

Gap junctions between fine unmyelinated axons can electrically couple groups of brain neurons to synchronise firing and contribute to rhythmic activity. To explore the distribution and significance of electrical coupling, we modelled a well analysed, small population of brainstem neurons which drive swimming in young frog tadpoles. A passive network of 30 multicompartmental neurons with unmyelinated axons was used to infer that: axon-axon gap junctions close to the soma gave the best match to experimentally measured coupling coefficients; axon diameter had a strong influence on coupling; most neurons were coupled indirectly via the axons of other neurons. When active channels were added, gap junctions could make action potential propagation along the thin axons unreliable. Increased sodium and decreased potassium channel densities in the initial axon segment improved action potential propagation. Modelling suggested that the single spike firing to step current injection observed in whole-cell recordings is not a cellular property but a dynamic consequence of shunting resulting from electrical coupling. Without electrical coupling, firing of the population during depolarising current was unsynchronised; with coupling, the population showed synchronous recruitment and rhythmic firing. When activated instead by increasing levels of modelled sensory pathway input, the population without electrical coupling was recruited incrementally to unpatterned activity. However, when coupled, the population was recruited all-or-none at threshold into a rhythmic swimming pattern: the tadpole “decided” to swim. Modelling emphasises uncertainties about fine unmyelinated axon physiology but, when informed by biological data, makes general predictions about gap junctions: locations close to the soma; relatively small numbers; many indirect connections between neurons; cause of action potential propagation failure in fine axons; misleading alteration of intrinsic firing properties. Modelling also indicates that electrical coupling within a population can synchronize recruitment of neurons and their pacemaker firing during rhythmic activity.     

Antibodies against neurofilament subunits label retinal ganglion cells but not displaced amacrine cells of hamsters.

Although neurofilament (NF) antibodies have been used to visualize ganglion cells and their axons in the retina, it is not known, however, how many ganglion cells contain NF, and how the various NF subunits are distributed in the ganglion cells. Moreover, it is not known whether displaced amacrine cells in the ganglion cell layer are also labelled. In order to see whether NF antibodies can be used as a specific marker for ganglion cells, antibodies raised against the low (NF-L), middle (NF-M) and high (NF-H) molecular weight subunits of NF were employed to stain retinal whole-mounts of adult hamsters after pre-labelling the ganglion cells with Granular Blue. It was found that NF-L and NF-H antibodies labelled 38,777 and 17,750 cells in the ganglion cell layer respectively. By co-localization with GB-labelled cells, 88% of NF-L positive cells and 91% of NF-H positive cells were found to be ganglion cells. In contrast, the NF-M antibody labelled only very few ganglion cells (418 per retina) although robust staining of axonal bundles was observed. Thus, NF antibodies may prove useful in studying this population of ganglion cells.     

Electrophysiological evidence for push-pull interactions in the inner retina of turtle

The responses of the inner retinal neurons of turtle to light spots of sizes were studied in an attempt to reveal characteristics that may reflect possible interactions of the neural circuits underlying the center and surround responses. For the ON-OFF cells, the responses were also analyzed to observe whether interference or augmentation of these responses occur. The intracellular recordings revealed several such interactions, observed either in the form of altered spike activity or as changes in the transiency of the light responses. The ON-responding amacrine cell presented in this study became more sustained, while for the ON-OFF amacrine cells larger light spots tended to make the responses more transient and both the ON and OFF components became more pronounced. The spiking activity of the OFF-type ganglion cell shifted in relation to the light stimulus and the number of spikes observed upon presentation of larger spots increased. We suggest that the surround circuits activated by increasing light spots may substantially influence and reorganize not only the overall center-surround balance, but also the center response of the cells. Although it cannot be excluded that intrinsic membrane properties also influence these processes to some extent, it is more likely that lateral inhibition and disinhibitory mechanisms play the leading role in this process.     

Gap junction permeability modulated by dopamine exerts effects on spatial and temporal correlation of retinal ganglion cells’ firing activities

Synchronized activities among retinal ganglion cells (RGCs) via gap junctions can be increased by exogenous dopamine (DA). During DA application, single neurons’ firing activities become more synchronized with its adjacent neighbors. One intriguing question is how the enhanced spatial synchronization alters the temporal firing structure of single neurons. In the present study, firing activities of bullfrog’s dimming detectors in response to binary pseudo-random checker-board flickering were recorded via a multi-channel recording system. DA was applied in the retina to modulate synchronized activities between RGCs, and the effect of DA on firing activities of single neurons was examined. It was found that, during application of DA, synchronized activities between single neuron and its neighboring neurons was enhanced. At the meantime, the temporal structures of single neuron spike train changed significantly, and the temporal correlation in single neuron’s response decreased. The pharmacological study results indicated that the activation of D1 receptor might have effects on gap junction permeability between RGCs. Our results suggested that the dopaminergic pathway participated in the modulation of spatial and temporal correlation of RGCs’ firing activities, and may exert critical effects on visual information processing in the retina.     

Growth and synapse formation among major classes of adult salamander retinal neurons in vitro

Adult neurons, isolated from the salamander retina, were maintained in low-density cell culture and examined for synapse formation by electrophysiological and electron microscopic techniques. Morphologically identifiable rod, cone, horizontal, bipolar, and amacrine/ganglion cells survived for many months, grew processes, and formed numerous cell contacts. Intracellular recordings showed the presence of a variety of voltage- and time-dependent conductances and both electrical and chemical transmission among these cells. At the ultrastructural level, gap junctions, monad ribbon synapses, and conventional synapses, like those present in the intact retina, were observed in sibling cultures. Thus, all major classes of adult retinal neurons, in addition to ganglion cells, are able to regenerate processes and reform synapses. The regenerated synaptic contacts are functional and structurally diverse.     

Intraretinal axon diameter: a single cell analysis in the marmoset (Callithrix jacchus)

We have labelled individual retinal ganglion cells of a New World primate, the common marmoset (Callithrix jacchus) with neurobiotin and then measured axon, soma and dendritic field diameter. A total of 111 cells were analysed (62 parasol cells, 22 midget cells, 16 hedge cells and 11 small bistratified cells). When all retinal ganglion cells were grouped together axon diameter was positively correlated to soma diameter. When analysed according to cell class only midget cells showed a positive correlation between soma size and mean axon diameter. Dendritic field diameter and mean axon diameter of both parasol and midget cells showed significant correlations. Axon diameter is not constant along the intraretinal length of the axon and the rate of change in diameter appears to be related to the cell class and the initial size of the axon. Midget cell axons showed a rapid increase of up to 20% over the first 200 μm in contrast to parasol cell axons which increased more slowly over this distance but then showed a marked increase in diameter of up to 40% over the next 450 μm. However, axon diameter did not remain at these increased diameters but decreased at greater distances from the soma. The degree to which an axon changes its diameter is related to retinal ganglion cell class and the initial size of the axon. We postulate that these variations in intraretinal axon diameter may have a direct influence on conduction velocity and reflect a compensatory mechanism to minimise spatiotemporal dispersion along the visual pathway.     

Features and functions of nonlinear spatial integration by retinal ganglion cells

Ganglion cells in the vertebrate retina integrate visual information over their receptive fields. They do so by pooling presynaptic excitatory inputs from typically many bipolar cells, which themselves collect inputs from several photoreceptors. In addition, inhibitory interactions mediated by horizontal cells and amacrine cells modulate the structure of the receptive field. In many models, this spatial integration is assumed to occur in a linear fashion. Yet, it has long been known that spatial integration by retinal ganglion cells also incurs nonlinear phenomena. Moreover, several recent examples have shown that nonlinear spatial integration is tightly connected to specific visual functions performed by different types of retinal ganglion cells. This work discusses these advances in understanding the role of nonlinear spatial integration and reviews recent efforts to quantitatively study the nature and mechanisms underlying spatial nonlinearities. These new insights point towards a critical role of nonlinearities within ganglion cell receptive fields for capturing responses of the cells to natural and behaviorally relevant visual stimuli. In the long run, nonlinear phenomena of spatial integration may also prove important for implementing the actual neural code of retinal neurons when designing visual prostheses for the eye.     

The morphology and coupling of Aplysia bag cells within the abdominal ganglion and in cell culture

The bag cells in the abdominal ganglion of Aplysia californica control egg-laying behavior by releasing a polypeptide (ELH) during an afterdischarge of synchronous action potentials. We have used intracellular injection of Lucifer Yellow to study the morphology and interconnections of the bag cells. These neurosecretory cells are typically multipolar and their processes extend in all directions out from the bag cell clusters into the surrounding connective tissue, where they branch in a complex manner. In some of the dye injection experiments, dye transfer from the injected cell to neighboring cells was observed. Freeze fracture of the bag cell clusters and their surrounding connective tissue revealed numerous gap junctions on bag cell processes within the clusters as well as on more distal processes. We have also examined the morphology and coupling between bag cells in primary culture. As in the intact ganglion, bag cells in culture were found to be multipolar. All pairs of bag cells whose somata or processes had formed contacts in culture were electrically coupled. The strongest coupling was observed between pairs of cells whose somata appeared closely apposed. In these cases transfer of Lucifer Yellow between cells could also be observed. It is therefore likely that the synchrony of bag cell action potentials during a bag cell afterdischarge is a result of coupling between individual cells in the bag cell cluster.