| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Ophthalmology and Visual Science, University of Texas Medical School, Houston, Texas
Submitted 10 June 2004; accepted in final form 28 October 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
). Therefore rod signals must first enter cone bipolars, and there are three well-defined anatomical pathways by which this occurs (for review, see Bloomfield and Dacheux 2001; Sharpe and Stockman 1999). The first utilizes synapses in the order of rod-rod bipolar-AII amacrine cells (Dacheux and Raviola 1986; Famiglietti and Kolb 1975; Kolb and Famiglietti 1974; Sterling et al. 1988; Strettoi et al. 1992). AII amacrines then relay the rod bipolar signals to ON cone bipolars via gap junctions (Veruki and Hartveit 2002b) and to OFF cone bipolars via inhibitory glycinergic synapses (Muller et al. 1988; Strettoi et al. 1992, 1994). In the second pathway, signals flow from rods directly to cones via gap junctions between the two photoreceptor types (DeVries and Baylor 1995; Nelson 1977; Raviola and Gilula 1973; Schneeweis and Schnapf 1995, 1999; Smith et al. 1986). The third pathway involves direct glutamate release from rods onto OFF cone bipolars (Hack et al. 1999; Li et al. 2004; Soucy et al. 1998; Tsukamoto et al. 2001).
In the first two pathways, gap junctions are a necessary conduit for rod signals. Cx36 is a neuronal connexin that has been localized in the retina to AII amacrine cells (Feigenspan et al. 2001; Mills et al. 2001) and to photoreceptors (Feigenspan et al. 2004). Recently, an elegant study from Deans et al. (2002) demonstrated by recording from ON center ganglion cells in Cx36 knockout mice that both the pathways described were abolished. That is, coupling between rods and cones and between AIIs and ON cone bipolars is necessary for transmission of rod signals to ON ganglion cells. These data confirm that there are multiple, perhaps redundant, pathways for rod signals, yet how these signals are routed depending on light intensity is not known. To address this question, we have examined the AII amacrine light response and the relative contributions of rod bipolar and ON cone bipolar cells that synapse with AIIs. Evidence from both physiological and tracer coupling indicates that the AII/AII and AII/ON cone bipolar gap junctions are open and allow for bidirectional communication over a large range of light intensities (Trexler et al. 2001; Veruki and Hartveit 2002b; Xin and Bloomfield 1999). In this manuscript, we take advantage of the fact that in the absence of the rod bipolar inputs to AIIs, the only other rod input to ON cone bipolars is via rod-cone coupling. Thus rod bipolar and ON cone bipolar inputs to AIIs represent the operation of two different rod pathways. We demonstrate that both pathways operate simultaneously over greater than a 3 log unit range, both substantially contributing to AII and (more importantly) cone bipolar light responses.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Described in detail previously (Massey and Mills 1999), the isolation of the rabbit retina is briefly summarized as follows. Under a protocol approved by the Institutional Animal Welfare Committee, adult New Zealand Albino white rabbits of either sex (1.5–3 kg) were deeply anesthetized with urethan (loading dose, 1.5 g/kg ip), and the orbit was infused with 2% lidocaine hydrochloride before enucleation. The eye was then removed and hemisected.
The inferior portion of the eyecup was cut into strips and attached to filter paper, vitreal side down. The schlera and choroid were removed and the retinas were bathed in a modified Ames medium (see following text). Retinas were then stored at 10°C for later recording. Slices were cut on a vertical slicer to varying thickness (120–200 µm) and transferred to the recording chamber. Experiments measuring light responses were performed on retinas from rabbits that were dark adapted for 1–2 h prior to enucleation; surgery, isolation of the retina, and preparation of slices were done under dim red light.
Electrophysiology
A modified Ames solution was used for storing retinas, bath perfusion of slices, and puffer application of drugs. The core salts, which are shared by the three solutions mentioned, consisted of (in mM) 115 NaCl, 3.1 KCl, 1.24 MgCl2, 2 CaCl2, 6 glucose, 2 succinate, 1 malate, and 1 lactate. For the storage solution, 10 mM HEPES, 12 mM NaHCO3, and 1 mM pyruvate were added and pH was adjusted to 7.4 with NaOH. For bath perfusion, 24 mM NaHCO3 and 1 mM pyruvate were added and the solution was bubbled with 95% O2-5% CO2. The perfusate was heated to 37°C with an inline heater (Warner Instruments, Hamden, CT).
Patch pipette filling solution consisted of (in mM) 110 potassium gluconate, 10 CsCl, 10 NaCl, 10 HEPES, 10 EGTA, 2 MgCl2, 5 K+ATP, and 0.5 Na+GTP. The pH was adjusted to 7.2 with KOH. Lucifer yellow (0.5 mg/ml) or 0.1 mM Alexa-568 (Molecular Probes, Eugene, OR) were included in the patch solutions to confirm cell identity by epifluorescent visualization after recording. All chemicals and pharmacological agents were obtained from Sigma-Aldrich (St. Louis, MO) unless otherwise noted. Drugs were dissolved in core salts buffered with 20 mM HEPES to pH 7.4 and were delivered by single- or triple-barrel puffer pipettes. A single barrel would contain two or more drugs for testing the effects of simultaneous application rather than perfusion from multiple barrels.
Patch electrodes were pulled on a Flaming-Brown horizontal puller from Sutter Instruments (Novato, CA) and fire polished on a Narishige MF-83 microforge (East Meadow, NY) to a resistance of 7–9 M
(in KCl pipette solutions, pipette resistances were 5–7 M due to the higher mobility of chloride relative to gluconate). Seals in the cell attached configuration ranged from 8 to 20 G. On obtaining a whole cell patch, series resistance was estimated from the peak of the capacitive transients due to a square wave voltage pulse and ranged from 20 to 30 M. Series resistance was not compensated but was monitored periodically by the method of capacitive transients. Rod bipolar cell input resistances ranged from 500 M to 2.0 G depending on the holding potential (–30 to –70 mV); thus the worst case series resistance would decrease the command voltage by <6%. Cone bipolar and AII input resistances were lower, ranging from 300 M to 1.0 G. Thus series resistance was more of a factor for these cells and measurement of gap junction conductance must be considered an estimate.
Cell currents and voltages were amplified by Axopatch 200B patch-clamp amplifiers (Axon Instruments, Foster City, CA) in resistive feedback mode. Current and voltage signals were filtered on the amplifiers at 2 kHz and sampled at 10 kHz using a PCI-MIO16XE-10 data-acquisition board from National Instruments (Austin, TX). Custom software was used for data acquisition and analysis.
Light stimuli
Light responses were recorded from slices visualized using infrared videomicroscopy with DIC optics. Light stimuli were delivered to the slice through the microscope objective from a source consisting of 20 light-emitting diodes (LEDs; 525 nm peak, ±17 nm at half-width) mounted in a diffusing plastic sphere to provide a uniform field of illumination. The LEDs were driven by digital counter/timers on the data-acquisition board, and light intensity was varied by pulse width modulation of a 1-kHz square wave as well as neutral density filters moved into the light path. Maximum output was 5 x 105 photons µm–2 s–1, measured at the focal plane of the slice using an IL-1700 radiometer (International Light, Newburyport, MA) and apertures of 100 or 400 µm (Lenox Laser, Glen Arm, MD). Photon density was converted to an estimate of R* · rod–1 · s–1 using the effective collecting area for rods, taken from
 |
), where d = 1.9 µm, l = 19 µm, the dimensions of rod outer segments (Nakatani et al. 1991), Qisom = 0.67, and 
= 0.016 µm–1. The factor F = 0.5 is used to account for the absorption of unpolarized light orthogonal to the outer segment's longitudinal axis. Thus the collecting area was calculated as 0.8 µm2 (1.6 µm2 for polarized light). Further accounting for the 50% absorption efficiency of 525 nm photons by rhodopsin yields a collecting area of 0.4 µm2. |
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Rods differ from cones in their activation threshold and response kinetics, features commonly used to distinguish between rod and cone inputs to a given neuron. Individual rods respond to single photon absorptions, whereas the threshold for cone responses is much higher. We made estimates of the threshold for cone responses in our preparation based on studies performed in macaque. Cones in macaque respond to photon fluxes corresponding to photoisomerization rates as low as 50 s–1 (Schnapf et al. 1990; Schneeweis and Schnapf 1999). Taking the cone collecting area as 0.6 µm2 (Schneeweis and Schnapf 1999) and the near maximal absorption of 525 nm photons by green cones in rabbit (Nuboer 1971), a photon flux of >80 photons µm–2 s–1 would be necessary to excite cones in our rabbit retina slices. This corresponds to a rod photoisomerization rate of 
30 R* · rod–1 · s–1, in agreement with the cone threshold intensity determined for ON-center ganglion cells in mouse retina (Deans et al. 2002). In addition, a signature of rod photoreceptors is a long-lasting afterpotential following exposure to bright stimuli (Euler and Masland 2000; Nelson 1977; Svaetichin 1956).
Examples of rod driven light responses from three different cell types are displayed in Fig. 1. In Fig. 1A, a family of averaged light responses is shown for an AII amacrine cell. The flash duration was 100 ms, and the hallmarks of rod input are evident. The AII responds to flashes as dim as 0.4 R* · rod–1 · s–1, much lower than the calculated cone threshold of 30 R* · rod–1 · s–1. Rise times quicken with increasing flash intensity, and the AII reaches its maximum response amplitude with a flash intensity of 10 R* · rod–1 · s–1. Above this value, the peak of the response actually decreases slightly, but its duration increases. Later portions of the depolarizing response are of lower amplitude than the peak, and for the two brightest flashes, there is a secondary depolarization that appears to rebound from the decay of the first. A hyperpolarizing OFF response follows the depolarizing portion of the waveform and increases in magnitude and duration with increasing intensity.
|
In Fig. 1C, responses to 100-ms flashes were recorded for a rod bipolar. The nine traces are averages of nine or more flashes. Responses were obtained for flashes as dim as 0.4 R* · rod–1 · s–1. Although its origin is unknown, there is a small "shoulder" that occurs with intermediate intensities in rodents (Euler and Masland 2000; Field and Rieke 2002) that also appears in rabbit rod bipolar responses. The maximum response amplitude was reached with 10 R* · rod–1 · s–1 flashes, and the shoulder is evident for flash intensities of 10, 20, and 40 R* · rod–1 · s–1 in Fig. 1C (*). With intensities greater than those that produce the shoulder (
100 R* · rod–1 · s–1), the rod bipolar responses reached a plateau, and the duration of this plateau increased with further increases in flash intensity.
In Fig. 1D, the responses of the rod bipolar in Fig. 1C, the AII in Fig. 1A, and an ON cone bipolar, different from that in Fig. 1B, are aligned. The flash durations are all 100 ms at 200 R* · rod–1 · s–1 intensity to for comparison of response kinetics. All three cell types exhibit a light response that lasts much longer than the stimulus. Furthermore, the AII and ON cone bipolar reach their peak within 40 ms, whereas the rod bipolar requires 120 ms to rise to its maximum. These results confirm previous findings (Dacheux and Raviola 1986; Nelson 1982), showing that AII responses are quicker than rod bipolars, and extend those findings to show that bipolars downstream of the AII are also quickened.
These experiments demonstrate that all three cell types exhibit the hallmarks of rod input. This conclusion is based on the criteria that the lowest intensity flash responses are below cone threshold (Schnapf et al. 1990; Schneeweis and Schnapf 1999), and the highest intensity flash responses are influenced by rod afterpotentials. The next experiments aim to determine the routes rod signals take on their way to AIIs.
AII light responses consist of two distinct components
The AII functions as a relay for rod bipolar signals, communicating with ON cone bipolar cells via gap junctions (Strettoi et al. 1992; Veruki and Hartveit 2002b). However, there are two possible pathways for rod signals to reach ON cone bipolar cells. In addition to input from AIIs, ON cone bipolars can receive rod signals directly from cones because of rod-cone gap junctional coupling (Raviola and Gilula 1973; Smith et al. 1986; Zhang and Wu 2004). One would expect that as ON bipolar cells are activated via synaptic pathways in the outer plexiform layer, an ON cone bipolar contribution to the AII light response would arise. Thus two components of the AII light response should be evident: rod bipolar input via glutamate receptors and ON cone bipolar input via gap junctions. Recording light responses from AIIs in the presence of AMPA antagonists would eliminate rod bipolar input (Cohen and Miller 1999; Ghosh et al. 2001; Li et al. 2002; Morkve et al. 2002; Qin and Pourcho 1999; Singer and Diamond 2003), effectively isolating the ON cone bipolar component, if it exists.
We recorded the current responses of AII amacrine cells to full field monochromatic (525 nm) stimuli in the absence and presence of AMPA antagonists. Figure 2 depicts responses to flash intensities of 0.5, 1, 100, and 1,000 R* · rod–1 · s–1. AIIs were held at –50 mV, the reversal potential for Cl– conductances, to isolate excitatory input. For these intensities, application of 20 µM 1-(4-aminophenyl)-3-methylcarbamyl-4-methyl-3,4-dihydro-7,8-methylenedioxy-5H-2,3-benzodiaz-epine (GYKI, Fig. 2, A and B) or 20 µm 1,2,3,4-tetrahydro-6-nitro-2,3-dioxo-benzo[f]quinoxaline-7-sulfonamide (NBQX, C and D) resulted in a significant decrease in the amplitude of the light response. In addition, AMPA antagonists blocked all spontaneous excitatory postsynaptic currents (EPSCs) in the dark. However, a small current remained that was not blocked by GYKI or NBQX. This residual current was evident for flash intensities below cone threshold (Fig. 2, A and B) as well as for brighter flashes that elicited rod afterpotentials (D). Thus it appears that the residual current is rod driven. Peak normalizing the current remaining in AMPA antagonists to the current in control conditions (Fig. 2B, inset) highlights the difference in waveform of the two currents. The rising phases overlap, but the residual current is more sustained with a distinct inward component at light off. The difference in waveform of the AMPA and non-AMPA components suggests that the two currents arise from two different sources.
|
|
) and Ca2+ channels in photoreceptors (Vessey et al. 2004) as well as its lack of reversibility. However, the carbenoxolone blockade taken together with the lack of reversal of the residual current suggests that the residual current arises due to coupling with ON cone bipolars and that the AII light response is the sum of inputs from rod bipolars and ON cone bipolars. Because AIIs are coupled to each other and some or all of the neighboring ON cone bipolars, the NBQX-insensitive component represents the average response of the AII/ON cone bipolar network in the absence of rod bipolar input. In the absence of rod bipolar input to AIIs, the remaining rod input to ON cone bipolars most likely comes from rod-cone coupling. Thus the residual current allows for an estimation of the magnitude of the contribution of rod-cone coupling to inner retina responses (see DISCUSSION). The experiments described in the preceding text support the conclusion that the light response of AIIs is generated by two different rod pathways via two different bipolar cells. In the following sections, we examine each bipolar cell's input to the AII in greater detail.
Synaptic transmission between rod bipolars and AIIs
Rod bipolar cells and AII amacrine cells in slices of rabbit retina were identified by morphology, and synaptically connected pairs were found with a high level of success. AII amacrine cells were identified by a characteristically small soma and stout primary dendrite that descends into the inner plexiform layer (IPL) (Kolb and Nelson 1983; Massey and Mills 1999; Strettoi et al. 1992) (Fig. 4A). In some cases, lobular appendages in sublaminas 2 and 3 of the IPL, as well as fine dendritic processes in sublaminas 4 and 5, could be visualized. Rod bipolar cells had large somas (10–12 µm) near the top of the inner nuclear layer (INL) with axons that branched into three or four bulbous terminals in sublamina 5 of the IPL. Rod bipolar cells could be distinguished from other bipolar cells by their larger soma and the depth and size of the axon terminals. Patch pipettes included fluorescent dyes, and after recording, cells were imaged using epifluorescence to confirm their identity (Fig. 4B).
|
; Veruki and Hartveit 2002a; Veruki et al. 2003), isolating the glutamatergic EPSCs. Analysis of averaged EPSCs revealed an inward current with a large transient and a much smaller sustained component (Fig. 1C) (see also Singer and Diamond 2003). We determined pertinent kinetic parameters of the AII postsynaptic response from averages of 9 to >20 repeated rod bipolar stimulations in 10 separate pairs. The average EPSCs followed the onset of rod bipolar depolarization after a delay of 1.65 ± 0.33 (SE) ms (n = 10 pairs) and rose to a peak within 1.28 ± 0.43 ms (10–90%). They decayed with an exponential time course (
= 2.06 ± 0.34 ms) to 4 ± 2.2% of the peak current. Peak synaptic currents ranged from 42 to 1.1 nA at a holding potential of –50 mV (373 ± 340 pA, n = 18 pairs).
The transient component of the averaged EPSC decay slowed to = 9 ms in cyclothiazide from = 2 ms in control conditions (Fig. 4D) (Singer and Diamond 2003), and the sustained component was slightly increased in cyclothiazide, likely due to the slowing of the decay of individual exocytotic events (Fig. 4D, inset) (Veruki et al. 2003). The amplitude of the sustained relative to the peak amplitude in control (4%) or cyclothiazide (7%) conditions suggests that the transient component of the EPSC is not due solely to receptor desensitization but a decrease in vesicle release rate.
We compared the effects of the specific antagonists NBQX and GYKI-53655. NBQX blocks AMPA and kainate receptor subtypes, whereas GYKI is specific for AMPA receptors alone (Lukasiewicz et al. 1997). Puffer application of 20 µM NBQX reversibly abolished the evoked EPSCs as well as all spontaneous EPSCs in 11 of 11 pairs (not shown). In separate experiments (n = 5), 10–20 µM GYKI was applied (Fig. 4E). We saw complete blockade of evoked and spontaneous EPSCs with GYKI. These data taken together with the effects of cyclothiazide indicate that AMPA receptors are the sole current carriers of rod bipolar to AII synaptic transmission at –50 mV.
Electrical synapses between AIIs and ON cone bipolars
Next we examined synaptic transmission between AIIs and ON cone bipolar cells. Anatomical studies have shown that AIIs and ON cone bipolars are connected by gap junctions (Strettoi et al. 1994), and tracer injection studies in whole-mount retina have shown that the two heterologous cell types form an extensively coupled network (Hampson et al. 1992; Mills and Massey 1995, 1998). ON cone bipolars could be identified and distinguished from rod bipolars due to their IPL axon projection depth, which was not as deep as that of rod bipolar axons (Ghosh et al. 2004; MacNeil et al. 2004), and terminal branching pattern, which involved a larger number of processes with smaller endings. In Fig. 5 A, a fluorescence image of an AII –ON cone bipolar pair is overlaid on a dim DIC image of a slice. The bipolar cell ramified in the middle of sublamina b of the IPL, and its axon terminals spread out near the AII dendrites. Gap junction plaques labeled with an antibody specific for Cx36 have been found at this level (Feigenspan et al. 2001; Mills et al. 2001), suggesting that pairs with proper overlap should exhibit electrical coupling.
|
). Furthermore, the synaptic currents persisted in the presence of antagonists of chemical synaptic transmission (Fig. 3B2).
Altogether we successfully recorded from 18 AII–ON cone bipolar pairs. Two of the pairs exhibited no coupling, and no clear contacts were visible between the two cells in those pairs. Electrical conductance was measured in 14 pairs and ranged from 130 pS to 1 nS [510 ± 246 (SE) pS]. Coupling strengths were determined by chord conductances for positive and negative transjunctional voltages, as in Fig. 5B, or from slope conductances (Fig. 5D). Gap junctional currents were linear and exhibited no voltage dependence over a large range (Fig. 5D), consistent with the voltage insensitivity of Cx36 gap junctions in exogenous expression systems (Al-Ubaidi et al. 2000; Srinivas et al. 1999; Teubner et al. 2000). These experiments confirm that in our slice preparation AIIs and ON cone bipolars are coupled by ohmic electrical synapses.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Rod bipolar-AII amacrine synapse
Using simultaneous patch-clamp recordings from both the pre- and postsynaptic cells, we were able to demonstrate that AMPA receptor antagonists totally blocked spontaneous and evoked glutamatergic input to AIIs (Veruki et al. 2003). Rod bipolar input to AIIs was completely blocked by puffer applied 20 µM NBQX or 20 µM GYKI-53655. Singer and Diamond (2003) found incomplete block of rod bipolar input to AIIs in rat using 25 µM GYKI-54266 to block AMPA receptors. The differences in our findings might be explained by the eightfold higher efficacy of GYKI-53655 relative to -54266 in cultured rat hippocampal neurons (Donevan et al. 1994) and retinal ganglion cells (Lukasiewicz et al. 1997). The IC50 of GYKI-53655 is 1 µM.
The present study, as well as Singer and Diamond (2003), provide evidence that contradicts previous reports in the literature regarding transmission at the rod bipolar-AII synapse. Here we have shown that in addition to total block of synaptic transmission in rod bipolar-AII pairs, the light response of AIIs in dark-adapted retina are substantially attenuated by AMPA antagonists. However, in a study of AII amacrines with intracellular recordings, AMPA antagonists were unable to block AII light responses. In fact, application of 100 µM CNQX depolarized the AII and increased the amplitude of depolarizing light responses (Bloomfield and Xin 2000). Although, at present, we can find no explanation to reconcile our differences, we feel that enough immunological (Ghosh et al. 2001; Li et al. 2002) and pharmacological (Morkve et al. 2002; Singer and Diamond 2003; Veruki et al. 2003) evidence has accumulated to say with certainty that rod bipolar to AII synaptic transmission is mediated solely by AMPA receptors.
In addition to the pharmacology, the kinetics of this synapse and EPSC waveform are of interest as well. It is widely held that for low scotopic vision, all rod signals must cross the rod bipolar-AII synapse. Nelson (1982) showed that the response rise time of the AII was faster than that of the presynaptic rod bipolar, and we have shown that cone bipolars downstream of AIIs are also faster than rod bipolars. Perhaps the initial transient component of vesicle release from rod bipolars is related to the speeding of responses, or the large initial volley of release might be required to faithfully transmit small depolarizations in a rod bipolar that result from a single photon absorption by one of its presynaptic rods (Field and Rieke 2002). However, the slow rate of rod bipolar depolarization by light might not evoke the same transient release as a voltage step. In fact, the output of a rod bipolar might be linear with respect to its slow light response, based on the following argument.
Rods themselves respond sluggishly to photon absorption and likely account for a majority of the rod bipolar's sluggishness at low intensities. The sluggishness of the rods is also visible in the ON cone bipolar component of the AII light response (Fig. 2), and, by analogy, this current likely reflects the voltage waveform of the rod bipolar (Fig. 2B, inset, gray trace). Therefore in a single AII recording, we can visualize the light responses of ON cone bipolars rod bipolars in addition to measuring the glutamate release from the rod bipolar. In a comparison of the bipolar waveform to the glutamate release (Fig. 2B, inset), the rising phase of the two overlap, suggesting that the vesicle release machinery of the rod bipolar is linearly dependent on the membrane voltage, at least initially. The glutamate release does decline over time, whereas the rod bipolar voltage remains steady. Perhaps the decline in release rate is due to the significant GABA input that rod bipolar terminals receive from S1 and S2 amacrine cells (Zhang et al. 2002). The decrease in vesicle release rate might represent desensitization of the vesicle fusion machinery to a sustained increase in [Ca2+]i (Hsu et al. 1996), or it might reflect a decrease in the number of vesicles available for fusion (Heidelberger et al. 1994; Heinemann et al. 1994; Neher and Zucker 1993; Thomas et al. 1993). Nevertheless, because the rod bipolar waveform and the glutamate current on the AII have the same time course initially, the speeding of responses of cells downstream from the rod bipolar (Fig. 1D) must arise from elsewhere than the rod bipolar-AII synapse. It is more likely that the quickening of voltage responses in AIIs and ON and OFF cone bipolars is the result of Na+ spikes in AII amacrines (Boos et al. 1993; Veruki and Hartveit 2002a). Further work is necessary to address this issue.
ON cone bipolar-AII synapse
In our rabbit retina slices, we have shown that AIIs and ON cone bipolars are coupled by gap junctions with an average conductance of 500 pS. This value is lower than the 1.2-nS average reported for paired recordings of AII–ON cone bipolar synapses in rat (Veruki and Hartveit 2002b). Pipette solutions and series resistance values are very similar for recordings in that study and those reported here. Perhaps the disparity in the average values obtained for coupling conductance represent a true species difference. A species related difference in coupling strengths is supported by comparison of recordings of AII–AII pairs. We have determined the average conductance to be 260 ± 107 pS (n = 6 pairs), which is smaller than the 700-pS value reported for rat (Veruki and Hartveit 2002a). For both gap junctions, AII–AII and AII–ON cone bipolar, the conductance values for rabbit are 40% of those found in rat.
Although our mean gap junction conductance values differ, there are similarities in our findings regarding the large range of conductances around the mean. The large range may reflect differences in the number of AII dendritic processes retained as a result of slicing, or the difference in coupling strengths between AIIs and ON cone bipolars of different classes (Veruki and Hartveit 2002b). We did not classify our ON cone bipolars from paired recordings and therefore cannot speak to differences between types. However, we did notice a substantial variability in the ON cone bipolar component of the AII light response, which we attribute to different numbers of contacts between the recorded AII and neighboring ON cone bipolars. Although not tested exhaustively, we expect that for a given AII, increasing flash intensities should lead to larger ON cone bipolar contributions to the AII light response (see Fig. 2).
Rod contributions to ON cone bipolar light responses
In the absence of rod bipolar input, the AII receives rod input from ON cone bipolars, and the most likely route for the rod component of the ON cone bipolar light response involves rod-cone coupling. However, we must consider the possibility of direct contacts between rods and ON cone bipolars, as these connections have been demonstrated for OFF cone bipolars (Li et al. 2004; Soucy et al. 1998; Tsukamoto et al. 2001). Direct rod-ON cone bipolar contacts are unlikely considering the results from Deans et al. (2002). The complete loss of rod input to ON center ganglion cells in the Cx36 knockout mouse indicates that gap junctions are required for transmission of rod signals in the ON pathways. Our interpretation that rod signals in ON cone bipolars results from rod-cone coupling depends on the lack of direct ON cone bipolar to rod contacts in the rabbit as was shown in the mouse. Although unlikely, the influence of direct contacts cannot be eliminated at this time.
One of the most interesting findings in the study by Deans et al. (2002) relates to the separate populations of ganglion cells with different thresholds and different intensity response functions. One explanation for these results is that the different ganglion cell populations receive disproportionate input from the different rod pathways. Concomitantly, different classes of ON cone bipolar cells must receive disproportionate input as well. The results presented here suggest that the rod-rod bipolar-AII pathway and the rod-cone coupling pathway operate at similar intensities, down to 0.5 R* · rod–1 · s–1, and that ON cone bipolars get input from both pathways. However, there might be a population of ON cone bipolars that are not coupled to AIIs. These cells would not contribute to the AII light response.
We can make predictions about the relative contributions of the rod bipolar-AII and rod-cone coupling pathways to ON cone bipolar responses based on data presented here. The values for the AII current with and without NBQX or GYKI can be translated to the values for coupled ON cone bipolars, using published values for input resistances and coupling coefficients (Veruki and Hartveit 2002b). In Fig. 6, a diagram of a cell pair is displayed together with arrows representing the flow of current injected into one cell. This simple circuit yields a system of equations
 | (1) |
 | (2) |
 | (3) |
 | (4) |
 | (5) |
, respectively, CCAII-ONCB was 0.6, and CCONCB-AII was 0.3 (Veruki and Hartveit 2002b). For the case when cell 1 is the AII, IT is 100 pA due to rod bipolar input (Fig. 2B, black trace). From Eq. 5, IJ1
2 is then 20 pA, which represents the rod bipolar input to ON cone bipolars via gap junctions with AIIs. For the case when cell 1 is the ON cone bipolar, the value for IJ12 is 3 pA (Fig. 2B, gray trace), which gives IT as 7.3 pA. This IT value represents the rod-cone coupling derived current in ON cone bipolars that gives rise to the NBQX resistant component of the AII light response. In the AII, the ratio of the rod bipolar to rod-cone coupling component is >30 (100:3), whereas for ON cone bipolars, this ratio is reduced to <3 (20:7.3). Thus the tiny currents we recorded from AIIs in the presence of AMPA antagonists represent proportionally larger currents in ON cone bipolars. We found an ON cone bipolar component of the AII light response at 0.5 R* · rod–1 · s–1, the lowest intensity attempted. These results suggest that rod cone coupling contributions to retina function occur at intensities lower than the predicted 1 R* · rod–1 · s–1, (Smith et al. 1986). Further investigations would require recording directly from identified cone bipolar cells to determine the contributions each rod pathway makes to their light responses and the intensity ranges over which each pathway operates.
|
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Present address and address for reprint requests and other correspondence: E. B. Trexler, Departments of Ophthalmology and Neuroscience, Mount Sinai School of Medicine, 1 Gustave L. Levy Place, Box 1183, New York, NY 10029 (E-mail: brady.trexler{at}mssm.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Baylor DA, Lamb TD, and Yau KW. The membrane current of single rod outer segments. J Physiol 288: 589–611, 1979.
Bennett MVL. Electrical transmission: a functional analysis and comparison to chemical transmission. In: The Handbook of Physiology. The Nervous System. Cellular Biology of Neurons. Washington, DC: Am. Physiol. Soc., 1977, p. 357–416, 1977.
Bloomfield SA and Dacheux RF. Rod vision: pathways and processing in the mammalian retina. Prog Retinal Eye Res 20: 351–384, 2001.[CrossRef][ISI][Medline]
Bloomfield SA and Xin D. Surround inhibition of mammalian AII amacrine cells is generated in the proximal retina. J Physiol 523: 771–783, 2000.
Boos R, Schneider H, and Wassle H. Voltage- and transmitter-gated currents of all-amacrine cells in a slice preparation of the rat retina. J Neurosci 13: 2874–2888, 1993.[Abstract]
Cohen ED and Miller RF. The network-selective actions of quinoxalines on the neurocircuitry operations of the rabbit retina. Brain Res 831: 206–228, 1999.[CrossRef][ISI][Medline]
Dacheux RF and Raviola E. The rod pathway in the rabbit retina: a depolarizing bipolar and amacrine cell. J Neurosci 6: 331–345, 1986.[Abstract]
Deans MR, Volgyi B, Goodenough DA, Bloomfield SA, and Paul DL. Connexin36 is essential for transmission of rod-mediated visual signals in the mammalian retina.[see comment]. Neuron 36: 703–712, 2002.[CrossRef][ISI][Medline]
DeVries SH and Baylor DA. An alternative pathway for signal flow from rod photoreceptors to ganglion cells in mammalian retina. Proc Natl Acad Sci USA 92: 10658–10662, 1995.
Donevan SD, Yamaguchi S, and Rogawski MA. Non-N-methyl-D-aspartate receptor antagonism by 3-N-substituted 2,3-benzodiazepines: relationship to anticonvulsant activity. J Pharmacol Exp Therap 271: 25–29, 1994.
Euler T and Masland RH. Light-evoked responses of bipolar cells in a mammalian retina. J Neurophysiol 83: 1817–1829, 2000.
Famiglietti EV Jr and Kolb H. A bistratified amacrine cell and synaptic cirucitry in the inner plexiform layer of the retina. Brain Res 84: 293–300, 1975.[CrossRef][ISI][Medline]
Feigenspan A, Janssen-Bienhold U, Hormuzdi S, Monyer H, Degen J, Sohl G, Willecke K, Ammermuller J, and Weiler R. Expression of connexin36 in cone pedicles and OFF-cone bipolar cells of the mouse retina. J Neurosci 24: 3325–3334, 2004.
Feigenspan A, Teubner B, Willecke K, and Weiler R. Expression of neuronal connexin36 in AII amacrine cells of the mammalian retina. J Neurosci 21: 230–239, 2001.
Field GD and Rieke F. Nonlinear signal transfer from mouse rods to bipolar cells and implications for visual sensitivity. [see comment]. Neuron 34: 773–785, 2002.[CrossRef][ISI][Medline]
Ghosh KK, Bujan S, Haverkamp S, Feigenspan A, and Wassle H. Types of bipolar cells in the mouse retina. J Comp Neurol 469: 70–82, 2004.[CrossRef][ISI][Medline]
Ghosh KK, Haverkamp S, and Wassle H. Glutamate receptors in the rod pathway of the mammalian retina. J Neurosci 21: 8636–8647, 2001.
Hack I, Peichl L, and Brandstatter JH. An alternative pathway for rod signals in the rodent retina: rod photoreceptors, cone bipolar cells, and the localization of glutamate receptors. Proc Natl Acad Sci USA 96: 14130–14135, 1999.
Hampson EC, Vaney DI, and Weiler R. Dopaminergic modulation of gap junction permeability between amacrine cells in mammalian retina. J Neurosci 12: 4911–4922, 1992.[Abstract]
Heidelberger R, Heinemann C, Neher E, and Matthews G. Calcium dependence of the rate of exocytosis in a synaptic terminal. Nature 371: 513–515, 1994.[CrossRef][Medline]
Heinemann C, Chow RH, Neher E, and Zucker RS. Kinetics of the secretory response in bovine chromaffin cells following flash photolysis of caged Ca2+. Biophys J 67: 2546–2557, 1994.
Hsu SF, Augustine GJ, and Jackson MB. Adaptation of Ca(2+)-triggered exocytosis in presynaptic terminals. Neuron 17: 501–512, 1996.[CrossRef][ISI][Medline]
Kolb H and Famiglietti EV. Rod and cone pathways in the inner plexiform layer of cat retina. Science 186: 47–49, 1974.
Kolb H and Nelson R. Rod pathways in the retina of the cat. Vision Res 23: 301–312, 1983.[CrossRef][ISI][Medline]
Li W, Keung JW, and Massey SC. Direct synaptic connections between rods and OFF cone bipolar cells in the rabbit retina. J Comp Neurol 474: 1–12, 2004.[CrossRef][ISI][Medline]
Li W, Trexler EB, and Massey SC. Glutamate receptors at rod bipolar ribbon synapses in the rabbit retina. J Comp Neurol 448: 230–248, 2002.[CrossRef][ISI][Medline]
Lukasiewicz PD, Wilson JA, and Lawrence JE. AMPA-preferring receptors mediate excitatory synaptic inputs to retinal ganglion cells. J Neurophysiol 77: 57–64, 1997.
MacNeil MA, Heussy JK, Dacheux RF, Raviola E, and Masland RH. The population of bipolar cells in the rabbit retina. J Comp Neurol 472: 73–86, 2004.[CrossRef][ISI][Medline]
Massey SC and Mills SL. Antibody to calretinin stains AII amacrine cells in the rabbit retina: double-label and confocal analyses. J Comp Neurol 411: 3–18, 1999.[CrossRef][ISI][Medline]
Mills SL and Massey SC. Differential properties of two gap junctional pathways made by AII amacrine cells.[see comment]. Nature 377: 734–737, 1995.[CrossRef][Medline]
Mills SL and Massey SC. The kinetics of tracer movement through homologous gap junctions in the rabbit retina. Visual Neurosci 15: 765–777, 1998.[CrossRef][ISI][Medline]
Mills SL, O'Brien JJ, Li W, O'Brien J, and Massey SC. Rod pathways in the mammalian retina use connexin 36. J Comp Neurol 436: 336–350, 2001.[CrossRef][ISI][Medline]
Morkve SH, Veruki ML, and Hartveit E. Functional characteristics of non-NMDA-type ionotropic glutamate receptor channels in AII amacrine cells in rat retina. J Physiol 542: 147–165, 2002.
Muller F, Wassle H, and Voigt T. Pharmacological modulation of the rod pathway in the cat retina. J Neurophysiol 59: 1657–1672, 1988.
Nakatani K, Tamura T, and Yau KW. Light adaptation in retinal rods of the rabbit and two other nonprimate mammals. J Gen Physiol 97: 413–435, 1991.
Neher E and Zucker RS. Multiple calcium-dependent processes related to secretion in bovine chromaffin cells. Neuron 10: 21–30, 1993.[CrossRef][ISI][Medline]
Nelson R. Cat cones have rod input: a comparison of the response properties of cones and horizontal cell bodies in the retina of the cat. J Comp Neurol 172: 109–135, 1977.[CrossRef][ISI][Medline]
Nelson R. AII amacrine cells quicken time course of rod signals in the cat retina. J Neurophysiol 47: 928–947, 1982.
Nuboer JF. Spectral discrimination in a rabbit. Document Ophthalmol 30: 279–298, 1971.[CrossRef]
Qin P and Pourcho RG. Localization of AMPA-selective glutamate receptor subunits in the cat retina: a light- and electron-microscopic study. Vis Neurosci 16: 169–177, 1999.
