Download Synaptic Regulation of the Light-Dependent Oscillatory Currents in

J Neurophysiol 100: 993–1006, 2008.
First published May 21, 2008; doi:10.1152/jn.01399.2007.
Synaptic Regulation of the Light-Dependent Oscillatory Currents in Starburst
Amacrine Cells of the Mouse Retina
Jerome Petit-Jacques and Stewart A. Bloomfield
Departments of Ophthalmology, Physiology, and Neuroscience, New York University School of Medicine, New York City, New York
Submitted 27 December 2007; accepted in final form 14 May 2008
INTRODUCTION
driven oscillatory activity has been described for amacrine
cells in the fish retina (Sakai and Naka 1992). Consistent with
these findings, the presynaptic bipolar cells show calciumdependent oscillations of their membrane potential that leads to
pulsatile release of transmitter and oscillatory activity of
postsynaptic targets (Burrone and Lagnado 1997; Ma and Pan
2003). Interestingly, oscillations have also been reported in
other amacrine cell subtypes in fish and mouse that survive cell
isolation and are thus independent of synaptic drive (Feigenspan et al. 1998; Solessio et al. 2002). Thus both synaptically
mediated and intrinsically driven oscillatory activity occurs in
the retina.
Recently we described spontaneous, subthreshold oscillatory
activity in starburst amacrine cells, a unique subtype that
releases both acetylcholine and GABA and thereby subserves
both excitatory and inhibitory circuits within the proximal
retina (Petit-Jacques et al. 2005). Our results indicated that this
spontaneous rhythmic activity is synaptically driven, derived
from pulsatile, calcium-dependent glutamate release from presynaptic bipolar cells. This mechanism resides in the proximal
retina and is independent of light as evidenced by its experimental induction in the absence of photoreceptor signaling.
Here we report that starburst amacrine cells also show
prominent light-dependent oscillatory activity. The lightevoked responses of starburst cells consist of two components:
an initial transient peak inward current that relaxes during the
presentation of a light stimulus and oscillatory potentials that
ride atop this relaxation phase. Our results indicate that both
components result from glutamate release from presynaptic
bipolar cell axon terminals. However, they are affected differentially by a number of pharmacological agents that act on
inhibitory synaptic innervation of bipolar cell terminals or
glutamate reuptake transporters. Taken together, these results
suggest that the two response components result from the
sequential release of glutamate from a single pool or discrete
pools within presynaptic bipolar cell endings.
Oscillatory activity among neuronal ensembles has been
reported throughout the CNS (Leznick et al. 2002; Llinas et al.
1994). In the retina, rhythmic discharges, in the form of
spontaneous propagating waves, first appear prenatally and are
crucial to the proper development of synaptic circuitry within
both the retina and lateral geniculate nucleus (Meister et al.
1991; Wong 1993, 1999). In the adult, the oscillatory potentials
(OPs) are a prominent component of the electroretinogram,
indicating that widespread rhythmic activity exists across the
retina. There is now strong evidence that the OPs reflect
postsynaptic activity of amacrine and ganglion cells (Zhou
et al. 2007). Indeed adult ganglion cells display oscillatory
activity that is both light dependent and independent (Neuenschwander et al. 1999). The light-dependent rhythms show a
wide range of frequencies that can be altered by changes in
stimulus size and contrast (Stephens et al. 2006). Synaptically
METHODS
Address for reprint requests and other correspondence: J. Petit-Jacques,
Dept. of Physiology and Neuroscience, NYU School of Medicine, 550 First
Ave., New York, NY 10016 (E-mail: [email protected]).
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
www.jn.org
Mouse retina-eyecup preparation
All animal procedures complied with National Institutes of Health
guidelines for the ethical use of animals. C57BL6 wild-type (25– 60
days old) mice were deeply anesthetized with an intraperitoneal
injection of pentobarbital (0.08g/kg body wt). Lidocaine hydrochloride (20 mg/ml) was applied locally to the eyelids and surrounding
tissue. A flattened retinal-scleral eyecup preparation developed for
0022-3077/08 $8.00 Copyright © 2008 The American Physiological Society
993
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
Petit-Jacques J, Bloomfield SA. Synaptic regulation of the lightdependent oscillatory currents in starburst amacrine cells of the mouse
retina. J Neurophysiol 100: 993–1006, 2008. First published May 21,
2008; doi:10.1152/jn.01399.2007. Responses of on-center starburst
amacrine cells to steady light stimuli were recorded in the darkadapted mouse retina. The response to spots of dim white light appear
to show two components, an initial peak that correspond to the onset
of the light stimulus and a series of oscillations that ride on top of the
initial peak relaxation. The frequency of oscillations during light
stimulation was three time higher than the frequency of spontaneous
oscillations recorded in the dark. The light-evoked responses in
starburst cells were exclusively dependent on the release of glutamate
likely from presynaptic bipolar axon terminals and the binding of
glutamate to AMPA/kainate receptors because they were blocked by
6-cyano-7-nitroquinoxalene-2,3-dione. The synaptic pathway responsible for the light responses was blocked by AP4, an agonist of
metabotropic glutamate receptors that hyperpolarize on-center bipolar
cells on activation. Light responses were inhibited by the calcium
channel blockers cadmium ions and nifedipine, suggesting that the
release of glutamate was calcium dependent. The oscillatory component of the response was specifically inhibited by blocking the
glutamate transporter with D-threo-␤-benzyloxyaspartic acid, suggesting that glutamate reuptake is necessary for the oscillatory release.
GABAergic antagonists bicuculline, SR 95531, and picrotoxin increased
the amplitude of the initial peak while they inhibit the frequency of
oscillations. TTX had a similar effect. Strychnine, the blocker of glycine
receptors did not affect the initial peak but strongly decreased the
oscillations frequency. These inhibitory inputs onto the bipolar axon
terminals shape and synchronize the oscillatory component.
994
J. PETIT-JACQUES AND S. A. BLOOMFIELD
rabbit by Hu et al. (2000) was adopted and modified for the mouse.
Briefly, the eye was removed under dim red illumination and hemisected anterior to the ora serrata. Animals were killed immediately
after enucleation by cervical dislocation. The lens and vitreous humor
were removed, and the resultant eyecup preparation was placed on the
base of a submersion-type recording chamber. Several radial incisions
were made peripherally and the retina was flattened in the chamber
vitreal side up. The chamber was mounted on a microscope stage
within a Faraday cage and superfused (1–2 ml/min) with an oxygenated mammalian Ringer solution composed of (in mM) 120 NaCl, 5
KCl, 25 NaHCO3, 0.8 Na2HPO4, 0.1 NaH2PO4, 1 MgSO4, 2 CaCl2,
10 D-glucose. A pH of 7.4 was maintained by bubbling with 95%
O2-5% CO2 at room temperature of 20 –22°C.
Electrophysiological recordings
Light stimulation
The light stimuli were generated by the Vision Works software
Neurophysiology, outputted through a video projector onto a coherent
fiber optic and delivered to the retina through the microscope objective. The stimulus intensity was maintained in the mesopic illuminance range; for example, a 200-␮m-diam spot of white light stimulus
had an intensity of 0.7 ␮W/cm2 as measured with a radiometer/
photometer (Ealing Electro-Optics). Spot stimuli of various diameters
were used and were always visually centered over the starburst cell
soma under study.
Biocytin labeling
Neurons were labeled by allowing biocytin to diffuse from the
micropipette during patch recordings. After electrophysiological experiments were completed, retinas were fixed in a cold (4°C) solution
of 4% paraformaldehyde in 0.1 M phosphate buffer (pH ⫽ 7.3)
overnight. Retinas were then washed in phosphate buffer and soaked
in a solution of 0.18% hydrogen peroxide in methyl alcohol for 1 h.
This treatment completely abolished the endogenous peroxidase activity. Retinas were then washed in phosphate buffer and reacted with
the Elite ABC kit (Vector Laboratories) and 1% Triton X-100 in
sodium phosphate-buffered saline (9% saline, pH ⫽ 7.5). Retinas
were subsequently processed for peroxidase histochemistry using
J Neurophysiol • VOL
Statistical analyses
Data were analyzed using Student’s t-test statistic. Presentation of
data are in the form means ⫾ SE throughout.
RESULTS
Basic electrophysiological properties of starburst amacrine
cells in the mouse retina
Recordings were made from on-center starburst amacrine
cells, which have somata displaced to the ganglion cell layer
(GCL) and dendritic arbors stratifying within sublamina-b of
the inner plexiform layer (IPL). By specifically targeting small,
round somata in the GCL, we achieved a success rate of ⬎60%
in identifying starburst cells. After electrophysiological experiments, each recorded cell was injected with biocytin to confirm their identity by post hoc histology. Starburst amacrine
cells in the mouse showed the typical morphology described
previously in this (Ozaita et al. 2004; Petit-Jacques et al. 2005)
and other species (Bloomfield and Miller 1986; Famiglietti
1983; Tauchi and Masland 1984). This consisted of five to
seven main dendritic branches that formed a proximal zone,
which then divided into thinner intermediate segments that
divided further into a plexus of distal branches showing numerous varicosities (Fig. 1A).
As we have described previously, mouse starburst cells
displayed a number of basic and stereotypic electrophysiological properties (Petit-Jacques et al. 2005). One characteristic
feature of starburst cell was their membrane voltage response
to the injection of current steps. Membrane depolarization
triggered by pulses larger than ⫹50 pA tended to saturate, and
it was therefore not possible to depolarize the membrane to
potentials more positive than ⫺20 mV even with injections of
large current steps of ⫹450 pA (Fig. 1B). Under our recording
conditions, starburst cells never showed any spontaneous or
evoked spiking, consistent with our previous data from mouse
(Petit-Jacques et al. 2005) and studies of the rabbit retina using
the whole cell recording technique (Peters and Masland 1996;
Taylor and Wässle 1995). Large depolarizing current pulses
did evoke a small transient component, but it never reached
potentials ⬎0 mV, and it was insensitive to TTX. The absence
of spiking activity was consistent with voltage-clamp recordings that showed a total absence of inward currents for a full
range of membrane depolarization of ⫺70 to ⫹50 mV (Fig.
1C). In contrast to the absence of inward currents, membrane
depolarization above ⫺20 mV did trigger very large outward
currents comprised of two components: a transient current
immediately followed by a delayed rectifier component that did
not inactivate. On repolarization, the delayed rectifier component displayed very fast deactivation tail currents. These properties are consistent with those of voltage-gated Kv3 channels,
which have been shown to carry most of the outward current in
starburst cells of the mouse retina (Ozaita et al. 2004). Whereas
all starburst cells showed the delayed rectifier current, only a
subset of these cells showed the transient outward component
on depolarization (Fig. 1C). These results are consistent with
our previous finding that some starburst cells lack the transient
outward current (Ozaita et al. 2004) and support the recent
100 • AUGUST 2008 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
Recordings were made in the whole cell patch mode with an
Axopatch 200B amplifier (Axon Instruments, Burlingame, CA). Cells
were visualized with near infrared light (⬎775 nm) at 80⫻ magnification with a Nuvicon tube camera (Dage-MTI, Michigan City, IN)
and differential interference optics (DIC) on a fixed-stage microscope
(BX51WI; Olympus, Tokyo, Japan). Currents were recorded under
voltage clamp, filtered at 1 kHz, sampled at 20 kHz, and stored
directly on the computer’s hard drive using a Digidata 1322A A/D
interface (Axon Instruments). For the characterization of voltage
responses, neurons were recorded in the fast current-clamp mode of
the amplifier. The resting potential of neurons was adjusted to –70 mV
with small injections of DC. pCLAMP (v. 9.0; Axon Instruments)
software was used for data acquisition with subsequent data analysis
performed off-line using Minianalysis (v. 6.0.1; Synaptosoft, Decatur,
GA) and Origin (v. 6.1; OriginLab, Northampton, MA) software
packages.
Patch electrodes (3–5 M⍀) were pulled from standard wall borosilicate glass tubing (World Precision Instruments, Sarasota, FL) with
a Flaming/Brown type micropipette puller (Sutter Instruments, Novato, CA). Pipettes were filled with a K-gluconate internal solution
composed of (in mM) 144 K-gluconate, 3 MgCl2, 0.2 EGTA, 10
HEPES, 4 ATP-Mg, 0.5 GTP-Tris, pH 7.3 with KOH, and biocytin
(0.2% wt/vol, Sigma, St. Louis, MO). All recordings were made in
dark-adapted retinas.
3,3⬘-diaminobenzidine (DAB) as the chromagen, dehydrated and
flat-mounted for light microscopy.
LIGHT-DEPENDENT RESPONSE OF STARBURST AMACRINE CELLS
995
A
D
40 pA
200 ms
B
E
-20 mV
10 mV
10 mV
200 ms
200 ms
F
C
40 pA
1000 pA
200 ms
50 ms
0 pA
FIG. 1. Characteristics of ON starburst amacrine cells in the mouse retina. A: a photomicrograph of a starburst amacrine cell in the mouse retina labeled with
biocytin shows the characteristic dendritic arborization. The bar in the top right corner is 50 ␮m long. B: representative current-clamp recording from a starburst
amacrine cell. Steps of currents were injected in the cell for 600 ms, and the membrane voltage responses were recorded under whole cell patch clamp. Between
pulses, the cell was maintained at a voltage of ⫺75 mV by constant injection of a small amount of current (indicated by the arrow at the left of the traces). The
voltage traces are in response to injection of 50-pA current steps from ⫺100 to ⫹450 pA. Note the saturation of the membrane depolarization at ⫺20 mV for
current pulses greater than ⫹50 pA. C: representative voltage-clamp recording from the same starburst cell. The cell was maintained at a holding potential of
⫺70 mV, and the membrane was depolarized by 10-mV steps from ⫺70 to ⫹50 mV during 150 ms. On return from depolarization, the cell was maintained at
⫺40 mV for 70 ms to visualize the deactivation tail currents. Note the total absence of inward currents, but the presence of large outward currents with a fast
transient outward component and a slower delayed rectifier component. The holding current was ⫹25 pA. D: spontaneous oscillatory currents were recorded in
voltage-clamp at ⫺70 mV in another starburst cell. In absence of light, oscillations of variable amplitude and shape emerged from the baseline current and the
miniature events. Holding current was ⫺10 pA. E: in current-clamp mode, voltage membrane variations were recorded at ⫺70 mV in the same cell as in D.
During the application of a 70-␮m-diam spot of light (represented by the horizontal line below the voltage trace), the cell membrane potential displayed a series
of outward oscillations with a large initial peak followed by events with decreasing amplitude. After the light stimulation was cut off, the oscillations disappeared
and the membrane voltage returned to the baseline. Holding voltage was ⫺72 mV. F: in voltage clamp at a holding potential of ⫺70 mV, the light stimulation
triggered a series of oscillatory inward currents with a large initial peak followed by events of decreasing amplitude. After the light cutoff, the oscillatory currents
gradually disappeared into the baseline current. Same cell as in D and E. Holding current was ⫺18 pA.
finding of two types of murine starburst cells with different
physiological properties (Kaneda et al. 2007). However, we
saw no differences in the light-evoked responses of these two
subsets of starburst cells and so we do not differentiate them in
the results described in the following text.
Light-evoked responses of starburst amacrine cells
In a previous study, we showed that starburst amacrine cells
in the mouse retina display spontaneous current oscillations
(Petit-Jacques et al. 2005). In the dark-adapted retina, starburst
cells held at ⫺70 mV, exhibit random, spontaneous inward
currents of varying shape and amplitude (Fig. 1D) (PetitJacques et al. 2005). To further investigate the characteristics
of these oscillatory currents, we examined how they are afJ Neurophysiol • VOL
fected by presentation of light stimuli. At the resting potential
(approximately ⫺70 mV) in current-clamp mode, the presentation of a small spot (70 ␮m diameter) of light centered on the
starburst cell soma triggered a postsynaptic potential (PSP)
consisting of an initial peak followed by smaller-amplitude
oscillations of the membrane potential that lasted for the
duration of the light stimulus. At light offset, the final oscillatory wave gradually returned to the resting membrane potential
(Fig. 1E). Likewise, light stimulation of cells voltage clamped
at ⫺70 mV resulted in a large, initial inward current peak
followed by a series of current oscillations that lasted for the
duration of the stimulus (Fig. 1F). Similar light-evoked responses have been observed for on-center starburst amacrine
cells in the rabbit retina (Bloomfield 1992; Peters and Masland
100 • AUGUST 2008 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
-75 mV
996
J. PETIT-JACQUES AND S. A. BLOOMFIELD
A
40 pA
200 ms
Light-evoked oscillations in starburst amacrine cell
responses are mediated by glutamate
The spontaneous oscillations in starburst cells are synaptically mediated and are dependent on the excitatory drive
from presynaptic bipolar cells (Petit-Jacques et al. 2005).
Bipolar cells form glutamatergic synapses onto starburst
cells that involve AMPA/kainate ionotropic receptors
(Brandstätter et al. 1998; Firth et al. 2003; Thoreson and
Witkovsky 1999). To determine whether glutamate release
was responsible for both of the light-evoked response components of starburst cells, cells were stimulated in the
presence of 6-cyano-7-nitroquinoxalene-2,3-dione (CNQX),
a specific blocker of AMPA/kainate receptors. Application
of CNQX (10 ␮M) almost totally blocked all response
components to a small spot of light (Fig. 3, A and B). On
washout, the actions of CNQX were totally reversed and
both phases of the light response were recovered (Fig. 3C).
On average, CNQX reduced the maximal amplitude of the
initial peak by 87% (Fig. 3D; n ⫽ 27 stimulations in 3 cells,
P ⬍ 0.0001). Our results thus indicate that both the peak and
oscillatory response components of starburst cells are dependent on the glutamate release from presynaptic bipolar
cells.
Light responses of starburst cells are derived from
the ON pathway
On-center bipolar cells receive glutamate input from photoreceptors via the metabotropic glutamate receptors
mGluR6 (Nomura et al. 1994; Ueda et al. 1997). The
activation of these receptors leads to a hyperpolarization of
on-center bipolar cells and a reduction of their excitatory
inputs to more proximal neurons (Nakajima et al. 1993; Tian
and Slaughter 2003). Application of AP4, an agonist of
these receptors, totally blocked the light response of starburst cells to small spots of light. Only the spontaneous,
light-independent random activity was visible in the presence of the drug (Fig. 3, E and F). At its steady-state effect,
AP4 reduced the maximal amplitude of the initial peak by
95% and eliminated all light-dependent current oscillations
(Fig. 3G; P ⬍ 0.0001).
Average Frequency (Hz)
B
in 10 cells, not shown). The synaptic delay for excitatory
responses appears to be shorter in the rabbit retina in which
values near 60 ms have been reported (Lee and Zhou 2006;
Peters and Masland 1996).
Glutamate release from bipolar cell terminals is dependent
on calcium channels
Spontaneous Light-Evoked
FIG. 2. Characteristics of light-evoked responses in starburst amacrine
cells. A: a representative light-evoked response recorded in voltage clamp at
⫺70 mV. The light stimulation triggered oscillatory inward currents with the
initial “on” peak followed by a group of oscillations. ■, points of maximal
relaxation for each oscillation. The relationship between these points could be
described by a 1st-order exponential decay (- - -) and totally superposed with
the fitting of the relaxation of the initial on peak. The duration of light
stimulation is represented by — below the current trace. Holding current was
⫺10 pA. B: comparison between average frequency of spontaneous and
light-evoked current oscillations recorded at ⫺70 mV for 10 different cells. For
the light-evoked responses, the initial on peak was not counted as an oscillation, and the light stimulation was a 70-␮m-diam spot of light. Note that the
frequency of light-evoked oscillations is ⬎3 times higher than the frequency of
spontaneous oscillations.
J Neurophysiol • VOL
The glutamate release from bipolar cells is modulated by the
activity of different types of calcium channels (Berntson et al.
2003; Pan 2000, 2001), which co-localize with the vesicle
docking sites at ribbon synapses (Sterling and Matthews 2005).
We showed previously that the spontaneous oscillatory currents of starburst amacrine cells in mouse retina are completely
inhibited by calcium channel blockers (Petit-Jacques et al.
2005). Here we examined the effect of these blockers on the
light-evoked responses of starburst cells. In the presence of
cadmium ions, a nonspecific blocker of calcium channels, the
light response to a small spot of light was totally blocked, with
100 • AUGUST 2008 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
1996). These results indicate that current oscillations not only
appear spontaneously but are also an active component of the
light response of starburst cells in the mouse retina. Interestingly, the frequency of the light-evoked oscillatory currents
was more than three times that of the spontaneous oscillations
for individual cells (3.79 ⫾ 0.26 Hz for spontaneous and
11.97 ⫾ 0.98 Hz for light-evoked oscillations, n ⫽ 10 cells,
Fig. 2B).
An analysis of the light-evoked current response indicated
that it consisted of two discrete components: an initial transient
peak at light onset followed by a burst of oscillations riding
atop the relaxation phase of the initial peak. Figure 2A illustrates a typical light response of a starburst amacrine cell to a
70-␮m-diam spot of light. The maximal point of relaxation for
each current oscillation was measured (■). The relationship
between these points could be described by a first-order exponential decay that tightly matched the relaxation kinetics of the
initial peak.
On average, starburst cells in the mouse retina responded to
light stimulation with a synaptic delay of 108.5 ⫾ 2.4 ms.
Interestingly, the delay between the light offset and the disappearance of the light response was twice as long as the synaptic
delay at light onset (average of 233.9 ⫾ 2.9 ms; n ⫽ 88 tests
LIGHT-DEPENDENT RESPONSE OF STARBURST AMACRINE CELLS
997
A
E
40 pA
200 ms
100 pA
200 ms
F
C
G
n=18
D
n=27
Control
Control
AP4
CNQX
FIG.
3. Glutamate binding onto AMPA/kainate receptors and the on pathway define the light-evoked responses in starburst amacrine cells. A–C: the response
to a 70-␮m-diam spot of light was recorded at ⫺70 mV in control condition (A), in the presence of 10 ␮M 6-cyano-7-nitroquinoxalene-2,3-dione (CNQX; B),
and after return in control condition (C). Note the almost total disappearance of light response in the presence of CNQX, a blocker of AMPA/kainate receptors.
The — below the current traces represents the light stimulation durations. Holding current was ⫺10, ⫺5, and 0 pA. D: the maximum amplitude of current is
shown as current density in control condition and in the presence of CNQX. The blocker inhibited 87% of the maximum amplitude. E and F: the response to
a 70-␮m-diam spot of light was recorded at ⫺70 mV in control condition (E) and in the presence of 50 ␮M AP4 (F). The light response totally disappeared in
the presence of AP4. Note that the peak appearing after the light cutoff is not light dependent but just a random spontaneous activity. —, light stimulation
durations. Holding current was ⫺30 and ⫺40 pA. G: the maximum amplitude of current expressed as current density is shown in control condition and in the
presence of AP4. The amplitude was inhibited 95% by AP4.
loss of both the initial peak and the oscillatory components
(Fig. 4, A and B). For the example illustrated in Fig. 4, the
average maximal amplitude of the initial peak was 5.47 ⫾ 0.44
pA/pF (n ⫽ 9), and the average frequency of oscillations was
11.25 ⫾ 0.29 Hz (n ⫽ 8) in control, both of which were
eliminated by cadmium (n ⫽ 8). The only events detectable
under cadmium superfusion were light-independent spontaneJ Neurophysiol • VOL
ous miniature events. The effect of cadmium was extremely
fast with a maximal inhibition apparent after only 2 min, and,
although the full light response returned on wash out, the
recovery was slow. Nifedipine, a specific blocker of L-type
calcium channels, also had a strong inhibitory effect on the
light response of starburst cells, although less effective than
that of cadmium. Nifedipine blocked only 54% of the initial
100 • AUGUST 2008 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
B
998
J. PETIT-JACQUES AND S. A. BLOOMFIELD
Blockade of glutamate transporters differentially affects
the light-evoked response components of starburst
amacrine cells
A
40 pA
200 ms
C
Effects of GABA blockers on the light-evoked responses of
starburst cells
40 pA
200 ms
D
FIG. 4. The presynaptic release of glutamate occurring during the light
response is calcium dependent. A and B: the response to a 70-␮m-diam spot of
light was recorded at ⫺70 mV in control condition (A) and in the presence of
200 ␮M CdCl2, a nonspecific blocker of voltage-dependent calcium channels
(B). Note the total disappearance of the light response in the presence of
cadmium ions. Holding current was – 4 and 0 pA. C and D: the light response
to a 200-␮m-diam spot was recorded at ⫺70 mV in control condition (C) and
in the presence of 30 ␮M nifedipine, a specific blocker of L-type calcium
channels (D). With nifedipine, the initial on peak amplitude was largely
reduced and its kinetics slowed, and the current oscillations disappeared. —,
the light stimulation durations. Holding current was – 4 and ⫺10 pA.
peak and decreased the current oscillations frequency by 74%
(n ⫽ 9 tests) for the cell illustrated in Fig. 4, C and D. Similar
results were found for five additional starburst cells. These data
suggest that both components of the light response of starburst
cell are dependent on the calcium-dependent release of glutamate from the bipolar cell synaptic terminals although cadmium and nifedipine could also affect more distally-located
calcium channels.
J Neurophysiol • VOL
The excitatory release of glutamate from bipolar cell synaptic terminals can be modulated by GABAergic feedback inhibition from postsynaptic amacrine cells (Freed et al. 2003;
Matsui et al. 2001; Shen and Slaughter 2001; Wässle et al.
1998). We have reported previously that GABA receptors
antagonists have a strong stimulatory effect on the spontaneous
current oscillations of mouse starburst cells (Petit-Jacques et al.
2005). Here we extended the study of the different GABA
receptors blockers by examining their effects on the lightevoked responses of starburst cells. The GABAA receptor
blockers bicuculline (BMI, 10 ␮M) and SR 95531 (10 ␮M)
strongly enhanced the amplitude of the initial peak response
yet slightly decreased the frequency of current oscillations
(Fig. 6, A–D). Picrotoxin (PTX, 50 ␮M), an antagonist of Aand C-type GABA receptors, had an effect similar to that of the
GABAA blockers, triggering a comparable increase in the
amplitude of the initial peak response. However, PTX had a
stronger inhibitory effect than the GABAA antagonists on the
frequency of the oscillations (Fig. 6, E and F). On average, the
amplitude of the initial peak showed a 35% increase following
BMI (P ⬍ 0.05, n ⫽ 17 and 18 tests in 2 cells), a 22% increase
following SR 95531 (P ⬍ 0.01, n ⫽ 18 and 16 tests in 2 cells),
and 26% increase following PTX application (P ⬍ 0.001, n ⫽
18 tests in 2 cells) when compared with control conditions
(Fig. 5G). In contrast, the average frequency of oscillations
showed an 11% decrease following BMI (P ⬍ 0.05), a 20%
decrease following SR 95531 (P ⬍ 0.001), and a 43% decrease
following PTX application (P ⬍ 0.001) from control levels
(Fig. 5H). Interestingly, studies of murine rho subunits have
suggested that the native GABAC receptors in mouse to be
insensitive to PTX (Greka et al. 1998 2000). However, our
finding that PTX produce a significantly larger decrease in the
oscillatory frequency than either BMI or SR 95531 argues
against this. Taken together, these data suggest that the
100 • AUGUST 2008 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
B
Different types of glutamate transporters are present in the
mammalian retina, localized to presynaptic terminals of photoreceptors and bipolar cells (Hasegawa et al. 2006; Palmer
et al. 2003). The glutamate transporters in the axon terminals of
rod bipolar cells are effectively blocked by D-threo-␤-benzyloxyaspartic acid (TBOA), a nonselective nontransported
blocker (Veruki et al. 2006). Interestingly, application of
TBOA (20 ␮M) differentially affected the peak and oscillatory
response components of mouse starburst cells (Fig. 5, A and B).
TBOA had no significant effect on the amplitude of the initial
peak component (Fig. 5C; 7.10 ⫾ 0.64 pA/pF in control,
6.97 ⫾ 0.70 pA/pF in TBOA, n ⫽ 26 and 27 tests in 3 cells),
but it reduced the average frequency of the current oscillations
by 83% (Fig. 5D; 10.99 ⫾ 0.18 Hz in control compared with
1.85 ⫾ 0.16 in TBOA, P ⬍ 0.0001). The different pharmacology of these two components suggests that the peak and
oscillatory components are generated by distinct mechanisms
related to the release of glutamate from bipolar cell axon
terminals (see DISCUSSION).
LIGHT-DEPENDENT RESPONSE OF STARBURST AMACRINE CELLS
999
A
40 pA
200 ms
C
D
GABAergic feedback inhibition of bipolar cell axon terminals
plays an important role in the modulation of the glutamate
release.
In the inner retina, spike-dependent feedforward and -back
inhibition from amacrine cells are thought to play an important
role in the spatial tuning of ganglion cells (Cooks and McReynolds 1998; Cooks et al. 1998; Shields and Lukasiewicz 2003;
Taylor 1999). To test the possible involvement of spikedependent inhibition in the regulation of starburst cell responses, we applied tetrodotoxin (TTX), a specific inhibitor of
voltage-dependent sodium channels, after application of PTX.
In the presence of PTX, TTX further increased the amplitude
of the initial peak due to a presumed blockade of nonGABAergic spiking amacrine cells. Application of TTX also
strongly inhibited the current oscillations (Fig. 7, A and B). On
wash out, the amplitude of the initial peak was slightly increased and the current oscillations fully recovered (Fig. 7C).
J Neurophysiol • VOL
On average, TTX slightly increased the amplitude of the initial
peak by 7%, although the difference was not significant when
compared with PTX alone (Fig. 7D). The major effect of TTX
was a 40% inhibition of the oscillations maximal amplitude
when compared with PTX (Fig. 7E, P ⬍ 0.0005) and a 34%
inhibition of the current oscillations frequency (Fig. 7F, P ⬍
0.05). Application of TTX alone showed similar effects on the
two response components. These results suggest that spikedependent inhibition, at least partly non-GABAergic, modulates the oscillatory release of glutamate from bipolar cells.
Effect of strychnine of starburst cell light responses
The results with TTX suggest a role for glycinergic inhibition in modulating glutamate release from bipolar cells and
thereby affecting starburst cell responses (Cui et al. 2003; Du
and Yang 2002; Eggers and Lukasiewicz 2006; Ivanova et al.
100 • AUGUST 2008 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
FIG. 5.
Presynaptic glutamate reuptake through
a glutamate transporter is necessary for the oscillatory
release of glutamate during light stimulation. A and
B: the response to a 70-␮m-diam spot of light was
recorded at ⫺70 mV in control condition (A) and in
the presence of 20 ␮M D-threo-␤-benzyloxyaspartic
acid (TBOA, B). Note the disappearance of current
oscillations in the presence of TBOA. —, the light
stimulation durations. Holding current was ⫺50 and
⫺57 pA. C: the maximum amplitude of current
expressed as current density is shown in control
condition and in the presence of TBOA. The amplitude of the initial on peak was not modified by
TBOA. D: the average frequency of current oscillations is shown for control condition and for the steady
state effect of TBOA. The frequency was decreased
by 83% in the presence of the glutamate transporter
blocker.
B
1000
J. PETIT-JACQUES AND S. A. BLOOMFIELD
A
E
40 pA
40 pA
200 ms
200 ms
B
F
50 pA
200 ms
D
Max Amplitude (pA/pF)
G
n=16
12
n=18
n=18
9
6
n=18
n=18
n=17
3
0
Average Frequency (Hz)
H
15
12
9
6
3
0
FIG. 6. Effect of GABAA and GABAC receptors blockers on the light-evoked responses in starburst amacrine cells. A and B: the response to a 70-␮m-diam
spot of light was recorded at ⫺70 mV in control condition (A) and in the presence of 10 ␮M bicuculline (BMI), a GABAA receptor antagonist (B). BMI largely
increased the amplitude of the initial on peak without significantly affecting the current oscillations. Holding current was ⫹9 and ⫺4 pA. C and D: the response
to a 70-␮m-diam spot of light was recorded at ⫺70 mV in control condition (C) and in the presence of 10 ␮M SR 95531, another GABAA receptor antagonist
(D). SR 95531 increased the amplitude of the initial peak and slightly reduced the current oscillations frequency. Holding current was ⫺10 and ⫺18 pA. —,
the light stimulation durations. E and F: the response to a 70-␮m-diam spot of light was recorded at ⫺70 mV in control condition (E) and in the presence of
50 ␮M picrotoxin (PTX), a GABAA/C receptor antagonist (F). PTX increased the amplitude of the initial peak and strongly reduced the number of current
oscillations. —, the light stimulation durations. Holding current was ⫹16 and ⫹20 pA. G: the maximum amplitude of the initial peak is shown as current density
for control condition and for BMI (left), SR 95531 (middle), and PTX (right). The 3 drugs significantly increased the amplitude of the initial peak compared with
control condition. H: the average frequency of current oscillations is shown for control condition and for BMI (left), SR 95531 (middle), and PTX (right). The
3 GABA receptors antagonists reduced the frequency of oscillations with PTX showing the most potent effect.
2006). We therefore examined the effect of the glycine receptor blocker, strychnine, on the light-evoked response of starburst cells. Application of strychnine (10 ␮M) almost totally
abolished the oscillatory currents, whereas the initial peak
component was largely unaffected (Fig. 8, A and B). Strychnine did produced a slight decrease in the amplitude of the
J Neurophysiol • VOL
initial peak response, but this was not significant (Fig. 8C, from
8.9 ⫾ 0.56 to 7.49 ⫾ 0.43 pA/pF, n ⫽ 18 tests in 2 cells). In
contrast, the oscillatory component frequency was reduced by
69% (Fig. 8D, from 10.1 ⫾ 0.22 to 3.1 ⫾ 0.28 Hz, P ⬍
0.0001). These data suggest that glycinergic inhibition of
bipolar cell axon terminals specifically controls the releasable
100 • AUGUST 2008 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
C
LIGHT-DEPENDENT RESPONSE OF STARBURST AMACRINE CELLS
1001
A
40 pA
200 ms
B
D
Initial Peak
n=18
return to PTX
E
Oscillations
F
n=18
pool of glutamate that underlies the starburst cell oscillatory
responses.
DISCUSSION
Taken together with our previous report (Petit-Jacques et al.
2005), the present results indicate that starburst amacrine cells
J Neurophysiol • VOL
Oscillations
n=18
in mouse retina show spontaneous and light-evoked oscillatory
activity. Both apparently arise from a common mechanism, a
pulsatile release of glutamate from presynaptic bipolar cell
axon terminals, which can be induced in darkness or by light.
The light-evoked responses of dark-adapted starburst cells
were composed of two distinct components: a prominent peak
100 • AUGUST 2008 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
C
FIG. 7. The presynaptic oscillatory release of glutamate is regulated by a spiking
inhibitory input. A–C: the response to a 70␮m-diam spot of light was recorded at ⫺70
mV in the presence of PTX (A), in the
presence of PTX and 0.3 ␮M tetrodotoxin
(TTX) a specific blocker of voltage-dependent sodium channels (B), and after return to
PTX (C). Note that in the presence of TTX, most
of the current oscillations disappeared. —, the
light stimulation durations. Holding current
was ⫹20, ⫹28, and ⫹40 pA. D–F: the
average effect of TTX is shown for the
maximum amplitude of the initial on peak
(D), for the maximum amplitude of current
oscillations (E), and for the oscillations frequency (F). When compared with PTX,
TTX did not significantly increase the amplitude of the initial peak, but it strongly
reduced the oscillations maximum amplitude and frequency.
1002
J. PETIT-JACQUES AND S. A. BLOOMFIELD
A
40 pA
200 ms
C
D
of inward current at stimulus onset and oscillations that rode
along the relaxation phase of the initial peak.
Our results indicate that these response components both
result from the temporal properties of glutamate release from
presynaptic bipolar cells. However, our finding that the transient and oscillatory response components were differentially
affected by a number of pharmacological agents suggests that
they result from two distinct mechanisms related to the release of
glutamate from presynaptic bipolar cell axon terminals (Fig. 9).
Basic electrophysiological characteristics of starburst cells
There is a controversy as to whether adult starburst amacrine
cells support Na⫹-dependent spike activity. In the rabbit, some
studies reported light-evoked spiking of starburst cells (Bloomfield 1992; Cohen 2001; Gavrikov et al. 2003), whereas others
found them to be totally absent (Peters and Masland 1996;
Taylor and Wässle 1995; Zhou and Fain 1996). Under our
experimental conditions, we never recorded spiking behavior
in starburst amacrine cells of the mouse retina (Ozaita et al.
2004; Petit-Jacques et al. 2005) (see also Fig. 1). Likewise we
J Neurophysiol • VOL
did not record any inward currents in response to the depolarization of the cell membrane (Ozaita et al. 2004) (see also Fig.
1). Consistent with our findings, Kaneda et al. (2007) recently
reported the absence of voltage-gated Na⫹ currents in murine
starburst cells. However, they did report two types of voltagegated Ca2⫹ currents in starburst cells. This discrepancy between their data and ours is difficult to explain, but it may
relate to differences in experimental conditions. Kaneda et al.
(2007) recorded from starburst cells in retinal slices and dissociated in culture, whereas our experiments were performed
on intact retinas. Nevertheless we are confident in our observation that starburst cells lacked voltage-gated Na⫹ and Ca2⫹
currents under our experimental conditions because we could
record robust voltage-gated inward currents and associated
spiking activity from ganglion cells in the same retinas. The
absence of voltage-gated inward currents in our starburst cells
is consistent with our conclusion that the effects of TTX,
cadmium, and nifedipine on light-evoked responses do not
reflect direct actions on starburst amacrine cells but rather the
glutamate release from presynaptic bipolar cells. Consistent
100 • AUGUST 2008 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
FIG. 8. The presynaptic oscillatory release of
glutamate is regulated by a glycinergic input. A and
B: the response to a 70-␮m-diam spot of light was
recorded at ⫺70 mV in control condition (A) and in
the presence of 10 ␮M strychnine, a glycine receptors antagonist (B). Almost all current oscillations
disappeared during the application of strychnine. —,
the light stimulation durations. Holding current was
⫺30 and ⫺50 pA. C: the maximum amplitude of the
initial peak expressed as current density is shown in
control condition and in the presence of strychnine.
The antagonist did not significantly affect the initial
peak amplitude. D: the average frequency of current
oscillations is shown for control condition and for
the steady-state effect of strychnine. Strychnine
strongly reduced the current oscillations frequency.
B
LIGHT-DEPENDENT RESPONSE OF STARBURST AMACRINE CELLS
Bipolar Cell
GABAergic Amacrine Cell
1003
Glycinergic Amacrine Cell
GLY
GABA
GLY?
GLUT/Cl
Spiking Amacrine Cell
2+
Ca
GLUT
Starburst Amacrine Cell
FIG. 9. Schematic of the synapse between bipolar cell and starburst amacrine cell showing the different mechanisms regulating the glutamate release from
the bipolar cell presynaptic terminal. In response to a light stimulus, the ribbon synapse of the bipolar cell terminal release glutamate into the synaptic cleft. The
light response in starburst cells shows 2 components, a large initial peak in response to the onset of light followed by a series of oscillations that last the entire
duration of the light stimulus. The oscillations depend on an oscillatory release of glutamate that is dependent on the reuptake of glutamate from the synaptic
cleft by the glutamate transporter (labeled T, it cotransport glutamate and Cl⫺ ions, GLUT/Cl⫺). The initial peak release and the oscillatory release (symbolized
by the double arrow labeled GLUT for glutamate) are both dependent on the activity of presynaptic calcium channels that localize with the ribbon synapse. Three
types of presynaptic inputs onto the bipolar cell terminal participate in the regulation of the glutamate release. A GABAergic input that could be a feedback from
starburst amacrine cell or a direct input from a nonspiking amacrine cell affects both components of the release, the initial peak and the oscillations. Three
different GABA receptors blockers, bicuculline, SR 95531, and picrotoxin, increase the amplitude of the initial peak and decrease the frequency of the
oscillations. The 2 other inputs seem to affect predominantly the oscillatory component of the glutamate release. A spiking-dependent inhibitory input is blocked
by a specific blocker of voltage-dependent sodium channels, TTX, it might be glycinergic in nature. A glycinergic input is blocked by the glycine receptors
specific antagonist, strychnine. Both TTX and strychnine significantly decrease the frequency of oscillations.
with our previous report (Ozaita et al. 2004), we found that
starburst cells displayed prominent K⫹-mediated outward currents. Kaneda et al. (2007) reported recently that a starburst cell
in the mouse can show one of two types of outward current
activity. Some cells showed both transient and delayed outward currents, whereas others showed only the delayed current.
Here we illustrated starburst amacrine cell responses with both
the transient and delayed outward currents (Fig. 1C), and we
described previously starburst cells in the mouse retina with
only delayed outward currents (Ozaita et al. 2004). Therefore
we agree with the findings of Kaneda et al. (2007) that two
types of starburst cell responses exist in the mouse retina with
regard to their voltage-gated outward currents. The delayed
outward current in starburst cells is mediated mainly by Kv3.1/
3.2 channels (Ozaita et al. 2004).
Light-evoked responses in starburst cells are carried
by glutamate
Both the transient and oscillatory light-evoked response
components were blocked by CNQX, indicating that they are
dependent on the binding of glutamate to AMPA/kainate receptors located on starburst cell postsynaptic membranes. Similar results have been reported previously for starburst amacrine cells in the rabbit retina (Peters and Masland 1996). The
light-evoked response components were also blocked by the
mGluR6 agonist, AP4, indicating that they are derived from the ON
pathway. In addition, our results indicate that the glutamate
J Neurophysiol • VOL
release from bipolar cell axon terminals that gives rise to the
two starburst cell response components is calcium-activated
and largely dependent on the activity of L-type calcium channels sensitive to nifedipine. In bipolar cells, calcium channels
organized in clusters around the ribbon synapse control the
exocytosis of glutamate vesicles through a calcium-induced
calcium release process, which is triggered by a light-dependent depolarization (Burrone et al. 2002; Llobet et al. 2003;
Sterling and Matthews 2005).
Although both light-evoked response components are dependent on glutamate release, we found that they were differentially affected by pharmacological agents. The agents
TBOA, bicuculline, SR95531, TTX, and strychnine all affected
either one of the response components or affected the two in
opposite directions. These observations indicate the later oscillatory component is not the result of the initial peak component activating a mechanism intrinsic to the starburst cell
such as voltage-gated ionic channels. Rather they support our
conclusion that these components both result synaptically from
the glutamate release from bipolar cell terminals.
The differential effects of TBOA, a blocker of the glutamate
transporter, were most striking. Application of TBOA had an
insignificant effect on the amplitude of the initial peak component, but it suppressed almost all of the current oscillations.
The inability of TBOA to affect the initial peak amplitude,
even after lengthy application, indicates that this component is
not readily dependent on the recycling of glutamate from the
100 • AUGUST 2008 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
AMPA-Kainate R
1004
J. PETIT-JACQUES AND S. A. BLOOMFIELD
Regulation of glutamate release by presynaptic receptors
Bipolar cell axon terminals receive a variety of synaptic
inputs that can modulate the release of glutamate. GABAergic
inhibition derived from amacrine cell processes feedback onto
bipolar cell terminals to limit and synchronize the release
(Euler and Masland 2000; Freed et al. 2003; Shields et al.
2000). Activation of GABAA, GABAC, and glycine receptors
J Neurophysiol • VOL
all differentially affect light-evoked signaling in mouse retinal
bipolar cells (Cui et al. 2003; Eggers and Lukasiewicz 2006;
Eggers et al. 2007; Frech and Backus 2004). For example,
some GABAergic and glycinergic inputs onto bipolar cell
terminals participate in the lateral inhibition of ganglion cells
and are spike-dependent (Cook et al. 1998; Shields and
Lukasiewicz 2003). Our results suggest that inhibitory circuits
are involved in the regulation of glutamate release from bipolar
cells to starburst cells during light stimulation (Fig. 9).
GABAergic and glycinergic inhibitory inputs, the latter which
is likely spike-dependent in part, influence the response of
starburst cells to light. Activation of both GABAA and GABAC
receptors is involved in the regulation of the light responses,
which likely includes both feedforward and -back circuitry.
Because glycinergic synapses appear not to occur onto cone
bipolar cells, but do occur on rod bipolar cells in the murine
retina (Cui et al. 2003; Ivanova et al. 2006), the strychnine
effects likely reflect suppression of glycinergic receptors on
rod bipolar cell axon terminals.
The increased amplitude of the initial peak component
following application of the GABA blockers and TTX appears
to be the logical result of reducing feedback inhibition to
bipolar cell terminals, thereby increasing glutamate release. In
contrast, the mechanism of the reduced frequency of the
oscillatory component by these agents is less clear. However,
our results are consistent with the idea that feedback inhibition
plays a crucial role in the synchronization of the oscillatory
release of glutamate from bipolar cell terminals (Euler and
Masland 2000; Freed et al. 2003). Our TTX data suggest that,
under control conditions, repetitive spike-dependent inhibition
repolarizes the bipolar cell terminal membrane at regular intervals, reinforcing an oscillatory release of glutamate. Overall
our results indicate that inhibition derived from amacrine cells,
some of which is spike-dependent, act to shape the oscillatory
release of glutamate from bipolar cell terminals.
Physiological role of light-evoked oscillations of
starburst cells
We showed previously that oscillations form an important
mechanism of spontaneous transmitter release from bipolar cell
axon terminals (Petit-Jacques et al. 2005). Our present results
indicate that oscillatory release from bipolar cells is also
triggered by light and thus likely plays a role in propagating
visual signals. While our pharmacological data indicate that the
mechanisms that modulate the transient peak and oscillatory
components of starburst cell response are different, those for
the latter are similar to those we reported previously for
spontaneous oscillatory activity (Petit-Jacques et al. 2005).
These results suggest that the oscillatory glutamate release
from bipolar cells is modulated both dependently and independently of light. One idea is that spontaneous oscillatory release
corresponds to a basal synaptic noise. Thus changes in the
frequency of oscillatory events during stimulation by light
could provide a mechanism for postsynaptic starburst cells to
distinguish light-evoked signals from synaptic noise (Singer
et al. 2004).
In the visual cortex, oscillatory potentials may play a role in
the binding of separate neuronal aggregates into sensory units.
Oscillatory responses form a time/frequency coding mechanism for neurons to detect the physical properties of a stimulus
100 • AUGUST 2008 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
synaptic cleft. In contrast, the light-evoked oscillations were
largely inhibited during the first minutes of the TBOA application, indicating that they arise from glutamate stores dependent on the transporter for their recycling process. The dependence of the oscillatory component on glutamate recycling may
simply reflect the fact that it occurred sequentially after the
transient peak component and a possible corresponding depletion of transmitter. However, these data also raise the possibility that the two starburst cell response components reflect
the existence of two distinct releasable pools of glutamate from
bipolar axon terminals with different depletion kinetics. Consistent with this notion, the vesicular release at bipolar cell
synaptic terminals exhibits two distinct components, a fast pool
released within a few milliseconds and a sustained pool that is
released over the next several hundred milliseconds (Mennerick and Matthews 1996; Singer and Diamond 2003; Von
Gersdorff et al. 1998). The fast pool corresponds to the vesicles
docked at the base of the ribbon synapse, whereas the sustained
pool is dependent on vesicles tethered to the ribbon in higher
rows more distant from the plasma membrane (Sterling and
Matthews 2005). It has been suggested that filaments connecting vesicles to the ribbon could constitute a molecular motor
that transports primed vesicles in higher rows of the ribbon in
successive waves to the base where they fuse with the plasma
membrane. Such a structural organization of the ribbon may
underlie successive waves of release that could give rise to the
oscillations we observed.
The glutamate transporter located at the bipolar cell axon
terminal is a co-transporter of chloride ions inside the presynaptic terminal (Kugler and Beyer 2003; Palmer et al. 2003;
Rauen et al. 1996). This coupled anion current can counteract
the stimulus-evoked depolarization of the bipolar axon terminal and thereby suppress transmitter release (Veruki et al.
2006). Such a mechanism may also be involved in the development of an oscillatory release of glutamate. In this scheme,
each vesicle or group of vesicles that fuse into the plasma
membrane to release a pool of glutamate is followed by the
co-transport of chloride ions that will repolarize the synaptic
terminal membrane and suppress further release. Thus the
cascade of calcium channel-induced depolarization, glutamate
release, activation of transporter/chloride current, hyperpolarization, and suppression of release, could trigger successive
oscillatory waves of glutamate release (Fig. 9). Furthermore
the oscillatory release would stimulate the release of GABA
and glycine from amacrine cells, including starburst cells. The
timing of these inhibitory inputs arriving on the bipolar cell
terminal would follow each wave of the oscillatory release with
a slight delay, participating in the suppression of the release
and the return to the current baseline between each wave.
Further, the interventions of these different inhibitory inputs
would likely be synchronized with the system of glutamate
release-activity of the glutamate transporter.
LIGHT-DEPENDENT RESPONSE OF STARBURST AMACRINE CELLS
and are thereby involved in sensory information processing
relevant for perceptual grouping (Neuenschwander and Singer
1996; Sannita 2000). We propose that the oscillatory component of light-evoked responses in the retina constitute an
important early step in the coding of visual clues transmitted to
the brain. The specificity of the pharmacology of the oscillatory component provides compelling evidence of its importance for encoding visual signals. Considering the major role
played by starburst amacrine cells in the computation of
direction selectivity in the retina (Fried et al. 2005; Hausselt
et al. 2007; Taylor and Vaney 2003), it will be important in
future studies to determine how the oscillatory component is
modulated in response to moving light stimuli.
GRANTS
REFERENCES
Berntson A, Taylor WR, Morgans CW. Molecular identity, synaptic localization, and physiology of calcium channels in retinal bipolar cells. J Neurosci Res 71: 146 –151, 2003.
Bloomfield SA. Relationship between receptive and dendritic field size of
amacrine cells in the rabbit retina. J Neurophysiol 68: 711–725, 1992.
Bloomfield SA, Miller RF. A functional organization of ON and OFF
pathways in the rabbit retina. J Neurosci 6: 1–13, 1986.
Brandstätter JH, Koulen P, Wässle H. Diversity of glutamate receptors in
the mammalian retina. Vision Res 38: 1385–1397, 1998.
Burrone J, Lagnado L. Electrical resonance and Ca2⫹ influx in the synaptic
terminal of depolarizing bipolar cells from the goldfish retina. J Physiol 505:
571–584, 1997.
Burrone J, Neves G, Gomis A, Cooke A, Lagnado L. Endogenous calcium
buffers regulate fast exocytosis in the synaptic terminal of retinal bipolar
cells. Neuron 33: 101–112, 2002.
Cohen ED. Voltage-gated calcium and sodium currents of starburst amacrine
cells in the rabbit retina. Visual Neurosci 18: 799 – 809, 2001.
Cook PB, Lukasiewicz PD, McReynolds JS. Action potentials are required
for the lateral transmission of glycinergic transient inhibition in the amphibian retina. J. Neurosci 18: 2301–2308, 1998.
Cook PB, McReynolds JS. Lateral inhibition in the inner retina is important
for spatial tuning of ganglion cells. Nat Neurosci 1: 714 –719, 1998.
Cui J, Ma YP, Lipton SA, Pan ZH. Glycine receptors and glycinergic input
at the axon terminals of mammalian retinal rod bipolar cells. J Physiol 553:
895–909, 2003.
Du JL, Yang XL. Glycinergic synaptic transmission to bullfrog retinal bipolar
cells is input specific. Neuroscience 113: 779 –784, 2002.
Eggers ED, Lukasiewicz PD. GABA(A), GABA(C) and glycine receptormediated inhibition differentially affects light-evoked signaling from mouse
retinal rod bipolar cells. J Physiol 572: 215–225, 2006.
Eggers ED, McCall MA, Lukasiewicz PD. Presynaptic inhibition differentially shapes transmission in distinct circuits in the mouse retina. J Physiol
582: 569 –582, 2007.
Euler T, Masland RH. Light-evoked responses of bipolar cells in a mammalian retina. J Neurophysiol 83: 1817–1829, 2000.
Famiglietti EV. On and off pathways through amacrine cells in mammalian
retina: the synaptic connections of “starburst” amacrine cells. Vision Res 23:
1265–1279, 1983.
Feigenspan A, Gustincich S, Bean BP, Raviola E. Spontaneous activity of
solitary dopaminergic cells of the retina. J Neurosci 18: 6776 – 6789, 1998.
Firth SI, Li W, Massey SC, Marshak DW. AMPA receptors mediate
acetylcholine release from starburst amacrine cells in the rabbit retina.
J Comp Neurol 466: 80 –90, 2003.
Frech MJ, Backus KH. Characterization of inhibitory postsynaptic currents in
rod bipolar cells of the mouse retina. Vis Neurosci 21: 645– 652, 2004.
Freed MA, Smith RG, Sterling P. Timing of quantal release from the retinal
bipolar terminal is regulated by a feedback circuit. Neuron 38: 89 –101,
2003.
Fried SI, Münch TA, Werblin FS. Directional selectivity is formed at
multiple levels by laterally offset inhibition in the rabbit retina. Neuron 46:
117–127, 2005.
J Neurophysiol • VOL
Gavrikov KE, Dmitriev AV, Keyser KT, Mangel SC. Cation-chloride
cotransporters mediate neural computation in the retina. Proc Natl Acad Sci
USA 100: 16047–16052, 2003.
Greka A, Koolen JA, Lipton SA, Zhang D. Cloning and characterization of
mouse GABAC receptor subunits. Mol Neurosci 9: 229 –232, 1998.
Greka A, Lipton SA, Zhang D. Expression of GABAC receptor rho1 and rho2
subunits during development of the mouse retina. Eur J Neurosci 12:
3575–3582, 2000.
Hasegawa J, Obara T, Tanaka K, Tachibana M. High-density presynaptic
transporters are required for glutamate removal from the first visual synapse.
Neuron 50: 63–74, 2006.
Hausselt SE, Euler T, Detwiler PB, Denk W. A dendrite-autonomous
mechanism for direction selectivity in retinal starburst amacrine cells. PLoS
Biol 5: 1474 –1493, 2007.
Hu EH, Dacheux RF, Bloomfield SA. A flattened retina-eyecup preparation
suitable for electrophysiological studies of neurons visualized with transscleral infrared illumination. J Neurosci Methods 103: 209 –216, 2000.
Ivanova E, Müller U, Wässle H. Characterization of the glycinergic input to
bipolar cells of the mouse retina. Eur J Neurosci 23: 350 –364, 2006.
Kaneda M, Ito K, Morishima Y, Shigematsu Y, Shimode Y. Characterization of voltage-gated ionic channels in cholinergic amacrine cells in the
mouse retina. J Neurophysiol 97: 4225– 4234, 2007.
Kugler P, Beyer A. Expression of glutamate transporters in human and rat
retina and rat optic nerve. Histochem Cell Biol 120: 199 –212, 2003.
Lee SE, Zhou ZJ. The synaptic mechanism of direction selectivity in distal
processes of starburst amacrine cells. Neuron 51: 787–799, 2006.
Leznik E, Makarenko V, Llinas R. Electronically mediated oscillatory
patterns in neuronal ensembles: an in vitro voltage-dependent dye-imaging
study in the inferior olive. J Neurosci 22: 2804 –2815, 2002.
Llinas R, Ribary U, Joliot M, Wang XJ. In: Temporal Coding in the Brain,
edited by Buzsaki G, Llinas R, Singer W, Berthoz A, Christen Y. Berlin:
Springer, 1994, p. 251–272.
Llobet A, Cooke A, Lagnado L. Exocytosis at the ribbon synapse of retinal
bipolar cells studied in patches of presynaptic membrane. J Neurosci 23:
2706 –2714, 2003.
Ma YP, Pan ZH. Spontaneous regenerative activity in mammalian retinal
bipolar cells: roles of multiple subtypes of voltage-dependent Ca2⫹ channels. Vis Neurosci 20: 131–139, 2003.
Matsui K, Hasegawa J, Tachibana M. Modulation of excitatory synaptic
transmission by GABA(C) receptor-mediated feedback in the mouse inner
retina. J Neurophysiol 86: 2285–2298, 2001.
Meister M, Wong RO, Baylor DA, Shatz CJ. Synchronous bursts of action
potentials in ganglion cells of the developing mammalian retina. Science
252: 939 –943, 1991.
Mennerick S, Matthews G. Ultrafast exocytosis elicited by calcium current in
synaptic terminals of retinal bipolar neurons. Neuron 17: 1241–1249, 1996.
Nakajima Y, Iwakabe H, Akazawa C, Nawa H, Shigemoto R, Mizuno N,
Nakanishi S. Molecular characterization of a novel retinal metabotropic
glutamate receptor mGluR6 with a high agonist selectivity for L-2-amino4-phosphonobutyrate. J Biol Chem 268: 11868 –11873, 1993.
Neuenschwander S, Castelo-Branco M, Singer W. Synchronous oscillations
in the cat retina. Vision Res 39: 2485–2497, 1999.
Neuenschwander S, Singer W. Long-range synchronization of oscillatory
light responses in the cat retina and lateral geniculate nucleus. Nature 379:
728 –732, 1996.
Nomura A, Shigemoto R, Nakamura Y, Okamoto N, Mizuno N, Nakanishi
S. Developmentally regulated postsynaptic localization of a metabotropic
glutamate receptor in rat rod bipolar cells. Cell 77: 361–369, 1994.
Ozaita A, Petit-Jacques J, Volgyi B, Ho CS, Joho RH, Bloomfield SA,
Rudy B. A unique role for Kv3 voltage-gated potassium channels in
starburst amacrine cell signaling in mouse retina. J Neurosci 24: 7335–7343,
2004.
Palmer MJ, Taschenberger H, Hull C, Tremere L, von Gersdorff H.
Synaptic activation of presynaptic glutamate transporter currents in nerve
terminals. J Neurosci 23: 4831– 4841, 2003.
Pan ZH. Differential expression of high- and two types of low-voltageactivated calcium currents in rod and cone bipolar cells of the rat retina.
J Neurophysiol 83: 513–527, 2000.
Pan ZH. Voltage-activated Ca2⫹ channels and ionotropic GABA receptors
localized at axon terminals of mammalian retinal bipolar cells. Vis Neurosci
18: 279 –288, 2001.
Peters BN, Masland RH. Responses to light of starburst amacrine cells.
J Neurophysiol 75: 469 – 480, 1996.
100 • AUGUST 2008 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
This work was supported by Nationel Eye Institute Grant EY-07360 to S. A.
Bloomfield.
1005
1006
J. PETIT-JACQUES AND S. A. BLOOMFIELD
J Neurophysiol • VOL
Taylor WR, Vaney DI. New directions in retinal research. Trends Neurosci
26: 379 –385, 2003.
Taylor WR, Wässle H. Receptive field properties of starburst cholinergic
amacrine cells in the rabbit retina. Eur J Neurosci 7: 2308 –2321, 1995.
Tian N, Slaughter MM. Structure of glutamate analogs that activate the ON
bipolar cell metabotropic glutamate receptor in vertebrate retina. Vis Neurosci 20: 231–240, 2003.
Thoreson WB, Witkovsky P. Glutamate receptors and circuits in the vertebrate retina. Prog Retin Eye Res 6: 765– 810, 1999.
Ueda Y, Iwakabe H, Masu M, Suzuki M, Nakanishi S. The mGluR6 5⬘
upstream transgene sequence directs a cell-specific and developmentally
regulated expression in retinal rod and ON-type cone bipolar cells. J Neurosci 17: 3014 –23, 1997.
Veruki ML, Mørkve SH, Harveit E. Activation of a presynaptic glutamate
transporter regulates synaptic transmission through electrical signaling. Nat
Neurosci9: 1388 –1396, 2006.
Von Gersdorff H, Sakaba T, Berglund K, Tachibana M. Submillisecond
kinetics of glutamate release from a sensory synapse. Neuron 21: 1177–
1188, 1998.
Wässle H, Koulen P, Brandstatter JH, Fletcher EL, Becker CM. Glycine
and GABA receptors in the mammalian retina. Vision Res 38: 1411–1430,
1998.
Wong RO. The role of spatio-temporal firing patterns in neuronal development
of sensory systems. Curr Opin Neurobiol 3: 595– 601, 1993.
Wong RO. Retinal waves and visual system development. Annu Rev Neurosci
22: 29 – 47, 1999.
Zhou W, Rangaswamy N, Ktonas P, Frishman LJ. Oscillatory potentials of
the slow-sequence multifocal ERG in primates extracted using the Matching
Pursuit method. Vision Res 15: 2021–2036, 2007.
Zhou ZJ, Fain GL. Starburst amacrine cells change from spiking to nonspiking neurons during retinal development. Proc Natl Acad Sci USA 93:
8057– 8062, 1996.
100 • AUGUST 2008 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.33.3 on August 1, 2017
Petit-Jacques J, Völgyi B, Rudy B, Blooomfield SA. Spontaneous oscillatory
activity of starburst amacrine cells in the mouse retina. J Neurophysiol 94:
1770 –1780, 2005.
Rauen T, Rothstein JD, Wässle H. Differential expression of three glutamate
transporter subtypes in the rat retina. Cell Tiss Res 286: 325–336, 1996.
Sakai HM, Naka K. Response dynamics and receptive-field organization of
catfish amacrine cells. J Neurophysiol 67: 430 – 42, 1992.
Sannita WG. Stimulus-specific oscillatory responses of the brain: a time/
frequency- related coding process. Clin Neurophysiol 111: 565–583, 2000.
Shen W, Slaughter MM. Multireceptor GABAergic regulation of synaptic
communication in amphibian retina. J Physiol 530: 55– 67, 2001.
Shields CR, Lukasiewicz PD. Spike-dependent GABA inputs to bipolar cell
axon terminals contribute to lateral inhibition of retinal ganglion cells.
J Neurophysiol 89: 2449 –2458, 2003.
Shields CR, Tran MN, Wong RO, Lukasiewicz PD. Distinct ionotropic
GABA receptors mediate presynaptic and postsynaptic inhibition in retinal
bipolar cells. J Neurosci 20: 2673–2682, 2000.
Singer JH, Diamond JS. Sustained Ca2⫹ entry elicits transient postsynaptic
currents at a retinal ribbon synapse. J. Neurosci 23: 10923–10933, 2003.
Singer JH, Lassova L, Vardi N, Diamond JS. Coordinated multivesicular
release at a mammalian ribbon synapse. Nat Neurosci 7: 826 – 833, 2004.
Solessio E, Vigh J, Cuenca N, Rapp K, Lasater EM. Membrane properties
of an unusual intrinsically oscillating, wide-field teleost retinal amacrine
cell. J Physiol 544: 831– 847, 2002.
Sterling P, Matthews G. Structure and function of ribbon synapses. Trends
Neurosci 28: 20 –29, 2005.
Stephens GJ, Neuenschwander S, George JS, Singer W, Kenyon GT. See
globally, spike locally: oscillations in a retinal model encode large visual
features. Biol Cybern 95: 327–348, 2006.
Tauchi M, Masland RH. The shape and arrangement of the cholinergic
neurons in the rabbit retina. Proc R Soc Lond B Biol Sci 223: 101–119, 1984.
Taylor WR. TTX attenuates surround inhibition in rabbit retinal ganglion
cells. Vis Neurosci 16: 285–290, 1999.