Neuron
Previews €, T., Zinkowski, R., Buerger, K., Ewers, M., Pirttila Alafuzoff, I., Teipel, S.J., DeBernardis, J., Kerkman, D., McCulloch, C., Soininen, H., and Hampel, H. (2006). CSF phosphorylated tau protein correlates with neocortical neurofibrillary pathology in Alzheimer’s disease. Brain 129, 3035–3041. Grundke-Iqbal, I., Iqbal, K., Tung, Y.C., Quinlan, M., Wisniewski, H.M., and Binder, L.I. (1986). Abnormal phosphorylation of the microtubuleassociated protein tau (tau) in Alzheimer cytoskeletal pathology. Proc. Natl. Acad. Sci. USA 83, 4913–4917. Mattsson, N., Schӧll, M., Strandberg, O., Smith, R., €gerstrӧm, D., Ohlsson, Palmqvist, S., Insel, P.S., Ha
T., Zetterberg, H., Jӧgi, J., et al. (2017). 18F-AV1451 and CSF T-tau and P-tau as biomarkers in Alzheimer’s disease. EMBO Mol. Med. 9, 1212– 1223. Meredith, J.E., Jr., Sankaranarayanan, S., Guss, V., Lanzetti, A.J., Berisha, F., Neely, R.J., Slemmon, J.R., Portelius, E., Zetterberg, H., Blennow, K., et al. (2013). Characterization of novel CSF tau and ptau biomarkers for Alzheimer’s disease. PLoS ONE 8, e76523. Olsson, B., Lautner, R., Andreasson, U., O¨hrfelt, A., €, M., Rose´n, C., Portelius, E., Bjerke, M., Hӧltta Olsson, C., Strobel, G., et al. (2016). CSF and blood biomarkers for the diagnosis of Alzheimer’s
disease: a systematic review and meta-analysis. Lancet Neurol. 15, 673–684. Sato, C., Barthe´lemy, N.R., Mawuenyega, K.G., Patterson, B.W., Gordon, B.A., Jockel-Balsarotti, J., Sullivan, M., Crisp, M.J., Kasten, T., Kirmess, K.M., et al. (2018). Tau kinetics in neurons and the human central nervous system. Neuron 97, this issue, 1284–1298. Vandermeeren, M., Mercken, M., Vanmechelen, E., Six, J., van de Voorde, A., Martin, J.J., and Cras, P. (1993). Detection of tau proteins in normal and Alzheimer’s disease cerebrospinal fluid with a sensitive sandwich enzyme-linked immunosorbent assay. J. Neurochem. 61, 1828–1834.
An Old Neuron Learns New Tricks: Redefining Motion Processing in the Primate Retina Ben L. Murphy-Baum1 and Gautam B. Awatramani1,* 1Department of Biology, University of Victoria, Victoria, BC V8W 3N5, Canada *Correspondence:
[email protected] https://doi.org/10.1016/j.neuron.2018.03.007
Motion sensitivity requires the comparison of neural responses activated by nearby points in visual space. In this issue of Neuron, Manookin et al. (2018) find that in the primate retina, such comparisons are already manifest in second-order retinal bipolar cells, relying on lateral excitation mediated by gap junctions. The vertebrate retina is composed of a complex array of microcircuits that serve to extract salient features from the visual scene, such as object motion, direction, orientation, color, etc. These features are then relayed to higher visual centers via 20–30 functionally and morphologically distinct types of ganglion cells, whose axons leave the eye to form the optic nerve. Unlike most other vertebrates, the primate retina is unique in that the vast majority of the output is carried by the midget and parasol ganglion cells, which provide dominant inputs to the parvocellular and magnocellular pathways, respectively. Most other ganglion cell types have a numerically sparser representation and are functionally less well characterized (reviewed by Field and Chichilnisky, 2007). For this reason, dominant models of primate vision generally focus on the midget and parasol pathways, which are traditionally thought to act as simple relays, each adapted for conveying complementary spatiotemporal aspects of
the visual scene. Parasol cells have high temporal resolution and large receptive fields, making them ideally suited for conveying information about object motion. Midget cells have much smaller receptive field sizes and poor temporal resolution and are better suited for conveying information about object form and finer textures (Field and Chichilnisky, 2007). Importantly, the projections of all ganglion cells are organized in a precise retino-topographical manner, allowing some of the more complex features to be extracted at higher visual centers. For example, motion information can be extracted from the waves of activity relayed by a population of parasol ganglion cells as objects sweep across the retina (Frechette et al., 2005). Data such as these have led to the common perception that the primate visual system has moved some complex visual processing back to higher brain centers so it can dedicate more retinal circuitry toward attaining very high visual acuity. However, in this issue of
Neuron, Manookin et al. (2018) show that individual parasol ganglion cells deliver complex information about object motion to higher visual centers, refuting the long-standing notion that they are simple relays of spatiotemporal information. Manookin et al. (2018) recorded synaptic responses from parasol and midget ganglion cells in a macaque monkey whole-mount retinal preparation. To study motion sensitivity, they compared how the two ganglion cell types respond to rectangular light stimuli presented in either a ‘‘random’’ or a ‘‘moving’’ condition. In the moving condition, the rectangle drifts smoothly across the ganglion cell’s dendritic arbor. In the random condition, the rectangle is presented over the same regions of space but in a randomized temporal sequence. If the ganglion cells behaved like simple spatiotemporal filters, the response to the random and moving stimuli would be similar. However, parasol, but not midget, ganglion cells responded to moving stimuli
Neuron 97, March 21, 2018 ª 2018 Elsevier Inc. 1205
Neuron
Previews
with nearly twice the number only does the current through A of action potentials compared glutamate receptors sum with to the random condition, a gap junction current arising stark deviation from the linear from a pre-junctional cell, prediction. Motion sensitivity but it also does not flow was observed over a variback into the upstream ety of stimulus conditions ‘‘donor’’ cells because the (speeds and contrasts), indivoltage gradient across the cating that it was a robust gap junction opposes it, and general property of the thus further augmenting recell. How do parasol ganglion sponses (Figure 1B). Consecells acquire their sensitivity quently, neurons that are to motion? positioned ‘‘in front’’ of the B Voltage-clamp recordings stimulus experience a larger revealed larger excitatory curjunctional depolarization than rents in response to moving those positioned ‘‘behind’’ stimuli compared to random the stimulus. sequences, indicating that Sequential activation of gap motion sensitivity arises from junction-coupled cells also the presynaptic circuitry. gives rise to a distinct ‘‘popuBlocking GABAergic and glylation effect’’ (Figure 1B), cinergic receptors reduced which could be especially the motion sensitivity of the important in processing weak excitatory input. However, signals. When a given bipolar motion sensitivity was still cell response is strengthened present even during inhibitory by electrical signals, it will in blockade, suggesting that turn produce a stronger gap inhibition may contribute to junction signal in its downmotion sensitivity but is stream post-junctional partnot necessarily critical. These ners. In this way, augmendata led Manookin et al. tation of electrical signals (2018) to conclude that mobuilds up over several neuFigure 1. Motion Sensitivity Is an Intrinsic Property of Electrically tion sensitivity originates in rons, ultimately limited by the Coupled Networks the network of bipolar cells strength of coupling between (A) Schematic illustrating that the photoreceptor (PR)-evoked depolarization that provide the glutamatergic cells (Trenholm et al., 2013). of an isolated bipolar cell (BP) is weakened in the presence of gap junctions. (B) A simple model demonstrating how depolarization builds over a population excitatory input to parasol All of these factors, when of five coupled bipolar cells during a moving stimulus. The only model ganglion cells. The question combined with the nonlinear parameters are the coupling strength (how far the depolarizations spread now becomes how do relationship between voltage laterally) and the kinetics of the bipolar cell response (how long dethe presynaptic bipolar cells and synaptic release, can polarizations last). Current through the gap junctions was made to flow according to the voltage gradient between neighboring cells. Each column detect object motion? cause extra glutamate to be represents the membrane potential (Vm) of a given neuron at five different Recently, Rieke and colreleased in response to a times (T1–T5), during which the stimulus bar steadily moves from BP1 to BP5. leagues demonstrated a moving stimulus. Since biThe diagonal (dotted line) represents the response of the BP at a time when they are being directly stimulated by PRs. The pure chemical synaptic pivotal role for gap junctionpolar cells driving parasol component (blue) sums with the gap junction-mediated response (red). Note coupled bipolar cells in generganglion cells exhibit strong that the response grows over the population. Arrows indicate that current will ating motion sensitivity in the electrical coupling, Manookin flow across gap junctions in the direction of motion (see Preview text for details). It is important to note that these effects of gap junctions are strongly mouse retina, findings that et al., (2018) rationalized that amplified by postsynaptic nonlinearities. they substantiated using gap motion sensitivity in primate might also arise via similar junction knockouts (Kuo mechanisms. et al., 2016). Below, we The gap junction model for motion discuss the principle considerations that response to light due to a reduction in make gap junctions particularly useful in input resistance (Figure 1A). However, sensitivity requires coincident gap junction this context. When a bipolar cell in a when bipolar cells are depolarized in and synaptic input at the level of bipolar coupled network is stimulated by light in quick succession, as they would be in cells and puts experimentally verifiable relative isolation, the current driven by response to a moving edge, gap junction constraints on the stimuli that would glutamatergic input will flow laterally inputs can sum with photoreceptor inputs be considered moving. To systematically across gap junctions into the cell’s electri- and strengthen bipolar cell voltage re- test these predictions, Manookin et al. cally coupled neighbors. In this situation, sponses (Figure 1B). During the rising (2018) presented ganglion cells with a gap junctions weaken the bipolar cell’s phase of the bipolar cell response, not ‘‘paired bar’’ stimulus, varying the space 1206 Neuron 97, March 21, 2018
Neuron
Previews and time intervals between the bar presentations. They found that when the two bars were presented close together in space and time, they evoked much larger excitatory currents in parasol ganglion cells than what is predicted by the linear summation of the responses to each bar presented separately. Importantly, the supralinearity was not present if the bars were separated by more than the distance of two bipolar cell receptive fields (64 mm) and was absent if the bars were presented more than 30 ms apart. The spatial and temporal dependence of the supralinear summation matches expectations from models where motion sensitivity is mediated by electrically coupled bipolar cells. One potentially confusing result is that the bipolar cells that provide input to midget ganglion cells are also gap junction coupled, yet midget ganglion cells are not motion sensitive. Manookin et al. (2018) argue that lateral inhibition, which is stronger in the midget circuit, offsets the augmentation provided by gap junctions. However, this reasoning may be problematic in the central retina and the fovea, where lateral inhibition in the inner retina is significantly weaker (Sinha et al., 2017). Other possibilities, which are not mutually exclusive, are that coupling between bipolar cells is stronger in the parasol pathway or, alternatively, that junctional depolarizations are processed differently in different bipolar cells. Indeed, diffuse bipolar cells driving parasol ganglion cells express voltagegated sodium channels, whereas as midget bipolar cells do not (Puthussery et al., 2013). It seems likely that having this additional nonlinearity in the parasol excitatory pathway would accentuate the effects of gap junction coupling. This study, together with work done in mouse (Kuo et al., 2016), raises a new and exciting idea that motion sensitivity originates from electrically coupled bipolar cells and opens doors to new lines of future investigations. One particularly interesting question is whether other ele-
ments in the parasol circuit that exhibit gap junction coupling also contribute to motion selectivity. For example, in addition to bipolar cell coupling, parasol ganglion cells (but not midgets) may also be coupled to each other and/or to amacrine cell networks (Dacey and Brace, 1992; Trong and Rieke, 2008). Indeed, a recent study demonstrates that amacrine cells mediate lateral excitation via gap junctions in the mouse retina (Roy et al., 2017). In addition, direct evidence for the essential factors required for motion sensitivity, including the lateral flow of electrical currents, nonlinear interactions between electrical/chemical input, and the ‘‘population effect,’’ was obtained using paired recordings from the electrically coupled direction-selective ganglion cells in the mouse retina (Trenholm et al., 2013). Here, too, the effects of gap junctions are relatively local, as electrical inputs are strongly attenuated by ganglion cell dendrites. However, in the study by Manookin et al. (2018), the relative contribution of electrical inputs and bipolar cell glutamate inputs cannot be easily distinguished based on voltage-clamp recordings of the motion-evoked excitatory currents in parasol ganglion cells. Thus, addressing the contribution of other coupled elements that may collaborate to enhance motion sensitivity would require alternate approaches. Finally, motion sensitivity seems to be emerging as a general property of gap junctioncoupled networks (Figure 1). Whether such properties manifest in other electrically coupled networks in monkey retina or in higher motion processing centers in the brain remain to be determined. In summary, work from Manookin et al. (2018) elegantly demonstrating motion sensitivity in the parasol circuit extends the conclusions drawn from work primarily done in mice to monkeys. The common circuit mechanism for processing motion information observed between primates and rodents suggests that, despite their differences in retinal architecture and circuitry, there may be more common design
principles than previously appreciated. Moreover, the rationale for such a design is clear: motion detection early in the visual pathway would enable a faster response to movement than would be permitted if information processing only began in higher cortical pathways. This would be extremely valuable for behavioral outcomes. Future work that could uncover more examples of complex computations in the primate retina (e.g., direction selectivity) is an exciting prospect. REFERENCES Dacey, D.M., and Brace, S. (1992). A coupled network for parasol but not midget ganglion cells in the primate retina. Vis. Neurosci. 9, 279–290. Field, G.D., and Chichilnisky, E.J. (2007). Information processing in the primate retina: circuitry and coding. Annu. Rev. Neurosci. 30, 1–30. Frechette, E.S., Sher, A., Grivich, M.I., Petrusca, D., Litke, A.M., and Chichilnisky, E.J. (2005). Fidelity of the ensemble code for visual motion in primate retina. J. Neurophysiol. 94, 119–135. Kuo, S.P., Schwartz, G.W., and Rieke, F. (2016). Nonlinear spatiotemporal integration by electrical and chemical synapses in the retina. Neuron 90, 320–332. Manookin, M.B., Patterson, S.S., and Linehan, C.M. (2018). Neural mechanisms mediating motion sensitivity in parasol ganglion cells of the primate retina. Neuron 97, this issue, 1327–1340. Puthussery, T., Venkataramani, S., Gayet-Primo, J., Smith, R.G., and Taylor, W.R. (2013). NaV1.1 channels in axon initial segments of bipolar cells augment input to magnocellular visual pathways in the primate retina. J. Neurosci. 33, 16045– 16059. Roy, K., Kumar, S., and Bloomfield, S.A. (2017). Gap junctional coupling between retinal amacrine and ganglion cells underlies coherent activity integral to global object perception. Proc. Natl. Acad. Sci. USA 114, E10484–E10493. Sinha, R., Hoon, M., Baudin, J., Okawa, H., Wong, R.O.L., and Rieke, F. (2017). Cellular and circuit mechanisms shaping the perceptual properties of the primate fovea. Cell 168, 413–426.e12. Trenholm, S., Schwab, D.J., Balasubramanian, V., and Awatramani, G.B. (2013). Lag normalization in an electrically coupled neural network. Nat. Neurosci. 16, 154–156. Trong, P.K., and Rieke, F. (2008). Origin of correlated activity between parasol retinal ganglion cells. Nat. Neurosci. 11, 1343–1351.
Neuron 97, March 21, 2018 1207
