Optical sensing of neural activity in the retina



Over the past 40 years, electrophysiology has been the main experimental tool for understanding how neurons in the retina respond to light and communicate with one another to process visual information. However, recent advances in optical recording methods provide new opportunities for functional studies of the retina, with several potential advantages over purely electrophysiological methods. Optical recordings are relatively non-invasive, and depending on the type of indicator dye, can reveal different aspects of neuronal activity, including changes in intracellular ion concentrations, changes in membrane potential, and release of synaptic vesicles. Unlike single cell electrical measurements, optical recordings can be made from many neurons at once, revealing more of the “big picture” of how the retina responds to light.



Function of the synaptic ribbon in rods and cones

We are particularly interested in the synapses of rod and cone photoreceptors, which transmit information to other neurons through a specialized structure called the synaptic ribbon. Unlike most synapses, ribbon synapses continuously release neurotransmitter, with the sensory stimulus (light) causing a decrease in the rate of release. We have used a fluorescent indicator to study how vesicles move inside rod and cone terminals (Rea et al., Neuron 2004), how Ca2+ controls release (Sheng et al., Journal of Neuroscience 2007; Bartoletti et al., Journal of Neurophysiology 2011), and how the rate of synaptic vesicle fusion encodes light intensity (Choi et al., Neuron 2005, Choi et al., Visual Neuroscience 2008 ; Jackman et al., Nature Neuroscience 2009).

The function of the synaptic ribbon is a long-standing mystery and one of our goals is to directly observe its interaction with synaptic vesicles to understand its role in controlling their release. Most recently, we have found that Rab-3a, a synaptic vesicle-associated protein, binds to the synaptic ribbon and is necessary and sufficient for the delivery of vesicles to release sites (Tian et al., J Neurosci, 2012). Further studies are aimed at determining whether Rab-3a escorts vesicles to the ribbon and then leaves, or participates in shuttling vesicles along the ribbon to the plasma membrane. To really understand what the ribbon does and how it operates we will need to observe single vesicles moving along the ribbon. We have recently embarked on a project employing super-bright and super-small “quantum dots” to help us accomplish this goal.

Synaptic ribbons have been thought to accelerate the delivery of vesicles to release sites at the plasma membrane, but our recent studies on cones show that the ribbon actually constrains vesicle delivery to the membrane. Two-photon ratiometric Ca2+ imaging has given us the first view of the intra-terminal Ca2+ in light and dark (Choi et al., Neuron 2008), essential information for understanding how vesicle release is controlled. Electron microscopy shows that in dark, when cones are continuously releasing neurotransmitter, the synaptic ribbon is depleted of synaptic vesicles (Jackman et al., Nat. Nsci 2009). This indicates that vesicle resupply, not vesicle fusion, is the rate-limiting step that controls release in cones, a surprising result. Electrophysiological recordings of postsynaptic responses also revealed that the ribbon behaves like a capacitor, charging with vesicles in light and discharging in a phasic burst at light offset.



Cell and molecular mechanisms of lateral inhibition in the retina

Lateral inhibition is a key neural network phenomenon that enhances contrast sensitivity in sensory systems. The reciprocal synapse between cone photoreceptors and horizontal cells (HCs) underlies lateral inhibition in the retina and establishes the antagonistic center-surround organization of all subsequent neurons in the visual system. In darkness, cones release the neurotransmitter glutamate to depolarize HCs, which reciprocate by inhibiting cone voltage-gated Ca2+ channels and reducing glutamate release. However, despite 50 years of study by many scientists, the nature of the synaptic negative feedback signal has remained unclear.

We have employed a genetically-encoded fluorescent pH indicator to show that protons are the key messengers mediating negative feedback (Wang et al., Nature Neuroscience 2014). We generated a transgenic zebrafish line with cones expressing Cali-pHluorin, a pH-sensitive GFP spliced onto the extracellular side of the cone Ca2+ channel. Two-photon imaging showed that light stimulation increased Cali-pHluorin fluorescence, consistent with a change in pH in the synaptic cleft. In accordance with HCs being the source of the pH change, the signal was blocked by the glutamate receptor antagonist DNQX, which blocks receptors on HCs but not on cones. The Cali-pHluorin signal grew as the illuminated area was expanded far beyond the imaged region. An annulus of light surrounding the imaged region also generated a pH change, indicating mediation by laterally projecting neurons. The pH change was suppressed by amiloride, which blocks negative feedback but has no effect on synaptic transmission of the cone light response. The magnitude, direction, kinetics, and spatial dependency of the Cali-pHluorin signal are all consistent with the notion that the change in pH mediated by HCs is the key signal underlying lateral inhibition in the retina. These results are the first to indicate that protons serve as an unconventional synaptic neurotransmitter in the nervous system of a vertebrate animal.

Rows of cone terminals in a zebrafish retina expressing the genetically-encoded pH sensor Cali-pHluorin. Image obtained by 2-photon microscopy of the intact isolated retina. From Wang et al., (2014).



Lateral inhibition is enhanced by a positive feedback synapse

Lateral inhibition in the retina was described in invertebrates >60 years ago and in vertebrates more than 50 years ago. However, we recently found that HCs also transmit to cone terminals a positive feedback signal (Jackman et al. 2011). The horizontal cell signal elevates intracellular Ca2+ and accelerates neurotransmitter release from cones. Positive and negative feedback are both initiated by AMPA receptors on HCs, but positive feedback appears to be mediated by a change in HC Ca2+, whereas negative feedback is mediated by a change in HC membrane potential. Local uncaging of AMPA receptor agonists indicates that positive feedback is spatially constrained to active HC-cone synapses, whereas the negative feedback signal spreads through HCs to affect release from surrounding cones. By locally offsetting the effects of negative feedback, positive feedback amplifies photoreceptor synaptic release without sacrificing HC-mediated contrast enhancement. Hence in the retina, activation of one postsynaptic receptor (AMAP-R) leads two distinct signals (voltage and Ca2+) that operate over different spatial domains to trigger opposite feedback responses in presynaptic neuron. Counter-intuitively, we show that the combination of these opposing systems enhances contrast more than lateral inhibition could accomplish alone. Ongoing studies are aimed at discovering the nature of the positive feedback signal from HCs, and investigating whether similar phenomena occur elsewhere in the nervous system.

Feedback mechanisms in outer retina. The differential spread of positive and negative feedback signals within an HC. The top bar denotes illumination pattern. A cone in darkness will release glutamate, activating AMPA receptors causing depolarization and Ca2+ influx. The rise in Ca2+ is constrained to an individual dendrite, localizing positive feedback to an individual cone. The depolarization spreads through the HC, resulting in negative feedback from all of the dendrites. Adapted from Jackman et al., PLoS Biology 2011.



Optical sensing of neural activity in the retina

Electrophysiology used to be the only experimental tool for understanding how neurons in the retina respond to light and communicate with one another to process visual information. However, advances in optical recording methods provide new opportunities for understanding how the retina processes visual information, with many advantages over purely electrophysiological methods. Optical recordings are non-invasive and can reveal different aspects of neuronal activity, including changes in calcium, changes in voltage, and changes in the release of synaptic vesicles. Among our accomplishments: We engineered novel fluorescent probes to quantify synaptic events at the first synapse in the retina, between presynaptic rods and cones and postsynaptic horizontal and bipolar cells (Rea et al.,2004; Choi et al., 2005; Jackman et al., . We used these tools to understand the mechanism of signaling at this synapse during lateral inhibition. These studied spearheaded optical technology for understand the normal functioning of the retinal circuit, and new studies are revealing the effectiveness of vision restoration technology in the eyes of blind mice.