Hostname: page-component-74656f9884-2dmjx Total loading time: 0 Render date: 2026-11-06T21:28:10.170Z Has data issue: false hasContentIssue false
  • English
  • Français

Recurrent axon collaterals of intrinsically photosensitive retinal ganglion cells

Published online by Cambridge University Press:  09 July 2013

HANNAH R. JOO ,
BETH B. PETERSON ,
DENNIS M. DACEY ,
SAMER HATTAR  and
SHIH-KUO CHEN
HANNAH R. JOO
Affiliation:
Department of Biology, Johns Hopkins University, Baltimore, Maryland Department of Biological Structure and the National Primate Research Center, University of Washington, Seattle, Washington
BETH B. PETERSON
Affiliation:
Department of Biological Structure and the National Primate Research Center, University of Washington, Seattle, Washington
DENNIS M. DACEY
Affiliation:
Department of Biological Structure and the National Primate Research Center, University of Washington, Seattle, Washington
SAMER HATTAR*
Affiliation:
Department of Biology, Johns Hopkins University, Baltimore, Maryland Department of Neuroscience, Johns Hopkins University, Baltimore, Maryland
SHIH-KUO CHEN*
Affiliation:
Department of Biology, Johns Hopkins University, Baltimore, Maryland Department of Zoology, National Taiwan University, Taipei, Taiwan
*
*Address correspondence to: Dr. Shih-Kuo Chen, Department of Zoology, National Taiwan University, Taipei, Taiwan 106. E-mail: alenskchen@ntu.edu.tw and Dr. Samer Hattar, Departments of Biology and Neuroscience, Johns Hopkins University, Baltimore, MD 21218. E-mail: shattar@jhu.edu
*Address correspondence to: Dr. Shih-Kuo Chen, Department of Zoology, National Taiwan University, Taipei, Taiwan 106. E-mail: alenskchen@ntu.edu.tw and Dr. Samer Hattar, Departments of Biology and Neuroscience, Johns Hopkins University, Baltimore, MD 21218. E-mail: shattar@jhu.edu
Rights & Permissions [Opens in a new window]

Abstract

Retinal ganglion cells (RGCs), the output neurons of the retina, have axons that project via the optic nerve to diverse targets in the brain. Typically, RGC axons do not branch before exiting the retina and thus do not provide it with synaptic feedback. Although a small subset of RGCs with intraretinal axon collaterals has been previously observed in human, monkey, cat, and turtle, their function remains unknown. A small, more recently identified population of RGCs expresses the photopigment melanopsin. These intrinsically photosensitive retinal ganglion cells (ipRGCs) transmit an irradiance-coding signal to visual nuclei in the brain, contributing both to image-forming vision and to several nonimage-forming functions, including circadian photoentrainment and the pupillary light reflex. In this study, using melanopsin immunolabeling in monkey and a genetic method to sparsely label the melanopsin cells in mouse, we show that a subgroup of ipRGCs have axons that branch en route to the optic disc, forming intraretinal axon collaterals that terminate in the inner plexiform layer of the retina. The previously described collateral-bearing population identified by intracellular dye injection is anatomically indistinguishable from the collateral-bearing melanopsin cells identified here, suggesting they are a subset of the melanopsin-expressing RGC type and may therefore share its functional properties. Identification of an anatomically distinct subpopulation in mouse, monkey, and human suggests this pathway may be conserved in these and other species (turtle and cat) with intraretinal axon collaterals. We speculate that ipRGC axon collaterals constitute a likely synaptic pathway for feedback of an irradiance signal to modulate retinal light responses.

Keywords

MelanopsinCollateralsRetinaPrimateMouse

Information

Type
Brief Communication
Information
Visual Neuroscience , Volume 30 , Issue 4 , July 2013 , pp. 175 - 182
Copyright
Copyright © Cambridge University Press 2013 

Introduction

Visual processing in the retina proceeds via excitatory synaptic transmission from photoreceptors to bipolar cells to retinal ganglion cells (RGCs). Ganglion cells, the output neurons of the retina, often branch to terminate in multiple central targets. They are distinct from other long projection neurons in that they typically lack recurrent collaterals terminating in the vicinity of the cell of origin. However, a small number of ganglion cells with recurrent axon collaterals have been rarely but consistently reported in human, monkey, cat, and turtle retina (Peterson & Dacey, Reference Peterson and Dacey1998). In addition to a primary axon that exits the eye via the optic disk, these cells have axon collateral branches terminating within the inner plexiform layer (IPL) of the retina itself. These collateral-bearing cells have very large, sparsely branching dendritic trees stratified in the inner half of the IPL and thin, branching collateral arbors terminating in the outer IPL. Although their consistent anatomy suggests a single cell type, it remains unclear whether the collaterals themselves reflect a functional specialization of a low-density type or a rare, nonfunctional aberration to which this cell type is uniquely vulnerable.

Recently, physiological evidence has emerged that a small population of atypical ganglion cells provide a positive feedback signal to the retinal circuitry (Barnard et al., Reference Barnard, Hattar, Hankins and Lucas2006; Zhang et al., Reference Zhang, Wong, Sollars, Berson, Pickard and McMahon2008; Zhang et al., Reference Zhang, Belenky, Sollars, Pickard and McMahon2012). These cells express the photopigment melanopsin and are intrinsically photosensitive (Berson et al., Reference Berson, Dunn and Takao2002; Hattar et al., Reference Hattar, Liao, Takao, Berson and Yau2002; Panda et al., Reference Panda, Sato, Castrucci, Rollag, DeGrip, Hogenesch, Provencio and Kay2002; Provencio et al., Reference Provencio, Rollag and Castrucci2002; Lucas et al., Reference Lucas, Hattar, Takao, Berson, Foster and Yau2003), contributing to image-forming vision (Brown et al., Reference Brown, Gias, Hatori, Keding, Semo, Coffey, Gigg, Piggins, Panda and Lucas2010; Ecker et al., Reference Ecker, Dumitrescu, Wong, Alam, Chen, Legates, Renna, Prusky, Berson and Hattar2010; Renna et al., Reference Renna, Weng and Berson2011) but serving primarily as irradiance detectors for nonimage-forming functions, such as circadian photoentrainment, the pupillary light reflex, and mood (Lucas et al., Reference Lucas, Hattar, Takao, Berson, Foster and Yau2003; Panda et al., Reference Panda, Provencio, Tu, Pires, Rollag, Castrucci, Pletcher, Sato, Wiltshire, Andahazy, Kay, Van Gelder and Hogenesch2003; Gamlin et al., Reference Gamlin, McDougal, Pokorny, Smith, Yau and Dacey2007; Goz et al., Reference Goz, Studholme, Lappi, Rollag, Provencio and Morin2008; Guler et al., Reference Guler, Ecker, Lall, Haq, Altimus, Liao, Barnard, Cahill, Badea, Zhao, Hankins, Berson, Lucas, Yau and Hattar2008; Hatori et al., Reference Hatori, Le, Vollmers, Keding, Tanaka, Schmedt, Jegla and Panda2008; Chen et al., Reference Chen, Badea and Hattar2011; LeGates et al., Reference LeGates, Altimus, Wang, Lee, Yang, Zhao, Kirkwood, Weber and Hattar2012). A melanopsin-based feedback signal appears to entrain the retinal circadian clock, modulating sensitivity of cone circuitry according to time of day (Barnard et al., Reference Barnard, Hattar, Hankins and Lucas2006). In addition, a melanopsin-dependent intrinsic light response has been discovered in dopaminergic amacrine cells (Zhang et al., Reference Zhang, Wong, Sollars, Berson, Pickard and McMahon2008; Zhang et al., Reference Zhang, Belenky, Sollars, Pickard and McMahon2012) via an excitatory synapse either directly or serially from melanopsin ganglion cells. However, no direct anatomical evidence of synaptic input from melanopsin-expressing cells to other retinal cell types has been found.

Using genetic and immunohistochemical methods, we describe a novel subpopulation of melanopsin cells with intraretinal axon collaterals that terminate in the IPL of both mouse and macaque monkey retina. We present anatomical evidence suggesting that previously described collateral-bearing cells belong to the melanopsin cell population, indicating a functional role consistent with that of melanopsin cells. We hypothesize that the axon collaterals of melanopsin cells constitute at least one pathway for the positive feedback of irradiance information to the retina, possibly via synaptic contact with the dopaminergic amacrine cell.

Materials and methods

Mouse

All mice were of mixed background (BL/6; 129SvJ) and sacrificed at ∼2–4 months. Mice were housed and treated in accordance with National Institutes of Health (NIH) and Institutional Animal Care and Use Committee guidelines. Animal care and use protocols were approved by the Johns Hopkins University Animal Care and Use Committee. To sparsely label melanopsin-expressing cells, 0.5–2.0 mg of tamoxifen was injected intraperitoneally into Opn4CreERT2; Z/AP mice 2 weeks prior to alkaline phosphatase (AP) staining as previously described (Lobe et al., Reference Lobe, Koop, Kreppner, Lomeli, Gertsenstein and Nagy1999; Ecker et al., Reference Ecker, Dumitrescu, Wong, Alam, Chen, Legates, Renna, Prusky, Berson and Hattar2010; Chen et al., Reference Chen, Badea and Hattar2011; Percival et al., Reference Percival, Martin and Grunert2011). In brief, mice were perfused with 45 ml of 4% paraformaldehyde (PFA) and retinas were dissected out. After staining, whole mount retinas were postfixed in 4% PFA overnight and washed with ethanol series overnight before mounting. Stack images were taken with Zeiss Axio Imager 1 microscope with 1 or 2 μm intervals and imported into Adobe Photoshop for detailed tracing using a Wacom tracing tablet or traced by hand from the printed images of composited stacks. Total retinal area, soma size, and dendritic and axonal field areas (defined as the area of a convex polygon enclosing the outermost processes) were measured in ImageJ (NIH). When a collateral branched into two large arbors, axonal field area was determined for each arbor separately. Total collateral length was measured in ImageJ from print images of individual cells. Cells were traced by hand from print images.

Primate

Macaque tissue was obtained from the Regional Primate Center at the University of Washington. Human eyes were obtained from donors to the Lions Eye Bank at the University of Washington. The anti-melanopsin immunolabeling has been described elsewhere (Dacey et al., Reference Dacey, Liao, Peterson, Robinson, Smith, Pokorny, Yau and Gamlin2005). In brief, macaque ganglion cells were labeled in whole mount retina using a rabbit polyclonal antibody derived from cloned full-length c-DNA for human melanopsin (hNA; provided by Dr. King-Wai Yau) and biotinylated goat anti-rabbit (BA-1000; Vector Labs, Burlingame, CA), and processed for horseradish peroxidase histochemistry (ABC Elite kit, Vector Labs). In separate experiments, macaque and human giant cells with axon collaterals were intracellularly injected with Neurobiotin (SP-1120; Vector Labs, Burlingame, CA) in the in vitro retina and processed for HRP histochemistry as previously described (Peterson & Dacey, Reference Peterson and Dacey1998). Cells were traced by hand using a camera lucida.

Results

To study the axonal targeting of melanopsin cells in mouse, we crossed the Opn4-driven tamoxifen-inducible Cre mouse line (Opn4CreERT2) with the Z/AP reporter line (Indra et al., Reference Indra, Warot, Brocard, Bornert, Xiao, Chambon and Metzger1999; Lobe et al., Reference Lobe, Koop, Kreppner, Lomeli, Gertsenstein and Nagy1999; Chen et al., Reference Chen, Badea and Hattar2011; Percival et al., Reference Percival, Martin and Grunert2011). This allowed controlled expression of AP on the plasma membrane of melanopsin-expressing cells. The number of cells labeled was roughly proportional to the volume of tamoxifen injected. High-volume tamoxifen injection densely labeled melanopsin-expressing cells (Fig. 1A, left panel). Both inner-stratifying M2 (Fig. 1A, right panel, top) and outer-stratifying M1 cells (Fig. 1A, right panel, bottom) were identified. Lower volume tamoxifen injection (Fig. 1B, left panel) resulted in much sparser labeling, primarily of M1 cells, making it possible to detect fine-caliber collateral processes that branched from M1 cell axons intraretinally but would be obscured by overlapping processes in densely labeled retinas (Fig. 1B, right panel). Two M1 cells are shown at high magnification in Fig. 2A and 2B. The cell in Fig. 2A had a highly branched axon collateral stratified in the outer IPL. The axon of the M1 cell in Fig. 2B gave rise to a sparsely branched collateral stratified in the inner IPL.

Fig. 1.

Titrated tamoxifen injection controls the density of labeled cells in knock-in mice (Opn4Cre ERT2; Z/AP) to reveal intraretinal collaterals. (A) Micrograph of a high-volume tamoxifen-injected retina with dense labeling of melanopsin-expressing cells. Area indicated by white box is shown at right at higher magnification. Top panel: focus is on the dendrites of M2 cells (open arrows) in the inner IPL. Bottom panel: focus is shifted to the dendrites of M1 cells in the outer IPL (open arrows). (B) Low tamoxifen injection resulted in a sparsely labeled retina with seven nonoverlapping M1 cells (asterisks). Area within dashed lines is shown at right at higher magnification. The axon (closed arrow) of an M1 cell (open arrow) gives off a single collateral process (arrowhead).

Fig. 2.

Photomicrographs of intraretinal axon collateral-bearing M1 cells in mouse. (A) Photomicrographs of an M1 cell with outer-stratifying collateral. Left panel: focus is on the soma (open arrow) and axon (closed arrows). A single collateral (arrowhead) branches from the axon. Right panel: focus is shifted to the dendrites (open arrowhead) and axon collateral (closed arrowhead) in the outer IPL. (B) Photomicrographs of an M1 cell with inner-stratifying collateral. Left panel: focus is on the soma (open arrow) and axon (closed arrow) that gave rise to a single collateral (closed arrowhead) that terminated in the inner IPL. Right panel: focus is on the dendritic arbor in the outer IPL.

Axon collaterals varied in arbor size and extent of branching (Fig. 3). The cells in Fig. 3A–3C had relatively small collateral arbors, but the arbors in Fig. 3A and 3C were highly branched, while the arbor in Fig. 3B had only two branches. The cell in Fig. 3D had a very large, sparsely branched collateral arbor. Collaterals extended a variable distance, terminating up to as much as 800 μm away from their cell bodies.

Fig. 3.

Mouse axon collateral arbors show variability in size and extent of branching. (A-D) Tracings of 4 M1 cells with axon collaterals. Cell bodies indicated by open arrows; primary axons and collaterals are shown in red. Primary axons (arrows) extended to the optic disk and gave rise to collateral processes (arrowheads) that terminated in either the inner (A, B) or the outer (C, D) IPL.

To estimate the percentage of M1 cells with axon collaterals, we counted the total number of labeled M1 cells and the number with collaterals in retinas (n = 23) in which overlap between cells was minimal enough for individual cells to be distinguished. We estimate the percentage of cells with axon collaterals to be ∼7% (7 outer, 4 inner, 11 total cells with collaterals; 165 labeled M1 cells). About 4% of the cells had outer-stratifying collaterals and ∼3% had inner-stratifying collaterals.

To estimate the total retinal area covered by collaterals, we calculated area based on the mean collateral field diameter (127.8 ± 16.4 μm, n = 9) and multiplied it by the number of collateral-bearing M1 cells (7% of ∼700 total M1 cells in adult mouse retina, n = 49). Dividing by mean retinal area (11.0 ± 2.4 mm2, n = 3), we estimated that inner and outer-stratifying collaterals together cover ∼5.7% of the retina. Inner-stratifying collaterals cover ∼2.4%, outer collaterals ∼3.3%. In these estimates, the collateral field diameter was taken as the diameter of a circle with area equal to that covered by a convex polygon enclosing the outermost tips of the collateral arbor. Because collateral arbors were often very sparse (Fig. 3D), the field diameter and thus the retinal percentage covered are both likely to be overestimates.

The giant melanopsin-expressing ganglion cells of macaque retina also exhibit axon collaterals. Collaterals were found on both outer- (Fig. 4A) and inner- (not shown) stratifying melanopsin cells and, in turn, branched in either the inner or outer IPL. In two samples of neighboring melanopsin-expressing cells (21 inner, 82 outer, 103 cells total), ∼10.7 % (2 inner, 9 outer, 11 total) had intraretinal axon collaterals, similar to the frequency in mouse retina. The two inner cells had inner-stratifying collaterals and the nine outer cells had outer-stratifying collaterals. A previous study of monkey and human ganglion cells using intracellular dye injection reported inner-stratifying cells with outer-stratifying intraretinal axon collaterals (Peterson & Dacey, Reference Peterson and Dacey1998). These cells, re-examined in the present study, had large somas and very large, sparsely branched dendritic trees (Fig. 4B and 4C), consistent with the anatomy of monkey giant melanopsin-expressing cells. As in mouse, axon collaterals of monkey and human cells often showed secondary and tertiary branching (Fig. 4B and 4C) and extended up to ∼800 µm from the primary axon. It is likely that all the large, axon collateral-bearing cells in both monkey and human are associated with the same low-density ganglion cell population that labels for melanopsin and has been shown to be intrinsically photosensitive (Dacey et al., Reference Dacey, Liao, Peterson, Robinson, Smith, Pokorny, Yau and Gamlin2005).

Fig. 4.

Primate giant melanopsin ganglion cells exhibit axon collaterals. (A) Photomicrograph of an outer stratifying melanopsin immunolabeled giant cell in macaque retina. Soma (open arrow) was displaced to the INL. The axon (closed arrows) gave rise to a thin collateral (arrowheads) studded with varicosities that could be followed for ∼800 µm in the outer IPL before fading. (B-C) Tracings of inner stratifying giant cells (open arrows) intracellularly filled with Neurobiotin from macaque (B) and human (C) retina. Axons (closed arrows) gave off several collateral branches (arrowheads) en route to the optic disk. Collaterals terminated in the outer IPL with the exception of the cell on left in (B), where two short collaterals terminated in the inner IPL (shown in red in inset). Insets show magnified views of collaterals indicated by the dashed lines.

Discussion

In both monkey and mouse, axon collaterals were associated with a relatively small percentage, ∼10%, of melanopsin-expressing ganglion cells. However, in macaque, the fine caliber processes and their light immunostaining made the collaterals very difficult to observe, and our finding that 11% of monkey melanopsin-expressing cells form collaterals must be considered a minimum estimate. In mouse, the collaterals were much easier to observe due to the clear, sparse labeling. However, we observed intraretinal axon collaterals only for the outer-stratifying M1 cells. This is possibly because our method preferentially labels M1 cells at low-volume tamoxifen injection, possibly due to their relatively high melanopsin expression (Hattar et al., Reference Hattar, Kumar, Park, Tong, Tung, Yau and Berson2006; Baver et al., Reference Baver, Pickard and Sollars2008; Schmidt & Kofuji, Reference Schmidt and Kofuji2009; Berson et al., Reference Berson, Castrucci and Provencio2010). If, as in primate, mouse inner-stratifying melanopsin cells also form collaterals, the actual percentage of cells with collaterals would be higher in mouse than the estimated 7%.

Regardless of the difficulties in observing collaterals, the relatively low frequency and the variable size and density of the associated terminal arbors suggest that the collaterals do not tile the retina in a regular spatial array (Wässle & Boycott, Reference Wässle and Boycott1991) and are therefore not associated with a separate and distinctive type of melanopsin-expressing ganglion cell. It is unlikely that the collateral-bearing cell is an artifact, however, as it has been seen using different methods in different species. The observation that melanopsin-expressing cells give rise to axon collaterals in both mouse and primate suggests that collaterals are a conserved, functional specialization associated specifically with melanopsin-expressing cells.

The dopaminergic amacrine cell, which costratifies with the dendrites of melanopsin cells in the outer IPL, signals by volume transmission to all retinal cell classes, modulating synaptic input for optimal visual function at varying light levels (Dowling & Ehinger, Reference Dowling and Ehinger1975; Lasater & Dowling, Reference Lasater and Dowling1985; Nir et al., Reference Nir, Harrison, Haque, Low, Grandy, Rubinstein and Iuvone2002; Hayashida & Ishida, Reference Hayashida and Ishida2004; Ichinose & Lukasiewicz, Reference Ichinose and Lukasiewicz2007). An estimated 21% of DA cells exhibit sustained melanopsin-dependent light responses consistent with glutamatergic input from melanopsin cells (Zhang et al., Reference Zhang, Zhou and McMahon2007; Zhang et al., Reference Zhang, Wong, Sollars, Berson, Pickard and McMahon2008; Zhang et al., Reference Zhang, Belenky, Sollars, Pickard and McMahon2012). Zhang et al. (Reference Zhang, Belenky, Sollars, Pickard and McMahon2012) propose two potential routes for retrograde transmission from melanopsin cells to DA cells: via dendrodendritic synapses, as have been reported in catfish retina (Sakai et al., Reference Sakai, Naka and Dowling1986), or by recurrent axon collaterals of ganglion cells, as previously described in cat and human retina (Dacey, Reference Dacey1985; Peterson & Dacey, Reference Peterson and Dacey1998). Although melanopsin and amacrine cell dendrites do cofasciculate extensively and this route for retrograde transmission of the melanopsin signal cannot be ruled out, there is as yet no evidence of synaptic input from melanopsin dendrites to dopaminergic amacrine cell dendrites (Belenky et al., Reference Belenky, Smeraski, Provencio, Sollars and Pickard2003; Dacey et al., Reference Dacey, Peterson, Liao and Yau2006; Jusuf et al., Reference Jusuf, Lee, Hannibal and Grunert2007; Viney et al., Reference Viney, Balint, Hillier, Siegert, Boldogkoi, Enquist, Meister, Cepko and Roska2007). The data shown here support the alternative hypothesis that the axon collaterals of melanopsin cells provide the route for transmission of light signals to the DA cells.

In double- (melanopsin, TH) and triple-labeled (melanopsin, TH, presynaptic SNARE complex protein SNAP25) rat and human retina, rare examples of potential collateral input to DA cell dendrites have been observed (Vugler et al., Reference Vugler, Redgrave, Semo, Lawrence, Greenwood and Coffey2007, initial study; personal correspondence suggesting collateral input). Despite forming a relatively sparse plexus, if collaterals selectively target DA cell dendrites, assuming that the entire length of each collateral made contact with DA cell dendrites, we estimate that, as an upper bound, a DA cell could receive input at up to ∼12% of its total length, or 300 µm. This estimate is based on dividing the mean collateral length per mouse retina (24,956 µm, our measurements) by the mean DA dendrite length per mouse retina, calculated as the product of the mean DA cell dendritic length and the number of DA cells known to receive melanopsin input (21%; Raven et al., Reference Raven, Eglen, Ohab and Reese2003; Badea et al., Reference Badea, Hua, Smallwood, Williams, Rotolo, Ye and Nathans2009).

Over a third of the collaterals we observed are stratified at the inner border of the IPL and would therefore not be positioned to contact the dendrites of dopaminergic amacrine cells at the outer border of the IPL. This suggests collateral-bearing cells may have at least two functional roles. In the developing retina, melanopsin cell input has been demonstrated to play a role in the segregation of retinogeniculate projections by modulating the waves of synchronous depolarization in neighboring ganglion cells driven by the starburst amacrine cell network (Renna et al., Reference Renna, Weng and Berson2011). Melanopsin cell input to this circuit could be to starburst amacrines or to spike-generating RGCs themselves (Renna et al., Reference Renna, Weng and Berson2011). Although axon collaterals do not costratify with the ON or OFF starburst amacrine plexus, located at ∼30% and ∼70% depth, respectively (Famiglietti, Reference Famiglietti1983; Mariani & Hersh, Reference Mariani and Hersh1988; Rodieck, Reference Rodieck1989), or with known RGC types other than the melanopsin cells themselves, it is possible that melanopsin cells influence retinal waves by collateral-mediated synapses with an intermediate cell type. This does, however, seem less likely than signaling to other ganglion cells by dendrodendritic gap junctions (Sekaran et al., Reference Sekaran, Foster, Lucas and Hankins2003).

Ganglion cells with branching axons have been described in calf, dog, lizard, cat, turtle, mouse, monkey, and human (Gallego & Cruz, Reference Gallego and Cruz1965; Honrubia & Elliott, Reference Honrubia and Elliott1968; Dacey, Reference Dacey1985; Usai et al., Reference Usai, Ratto and Bisti1991; Peterson & Dacey, Reference Peterson and Dacey1998). Our results in primate indicate that the previously observed collateral-bearing cells of monkey and human are, in fact, melanopsin cells, and suggest the hypothesis that all previously observed collateral-bearing cells of other species express melanopsin. Further investigation of the development, connectivity, and physiology of the recurrent collaterals will be required to determine the precise role that this unique circuit plays in the regulation of retinal function.

Acknowledgments

This work was supported by the National Institutes of Health grants GM076430 (S.H.), EY06678 (D.M.D.), RR00166 (the National Primate Research Center at the University of Washington), the David and Lucile Packard Foundation (S.H.), the Alfred P. Sloan Foundation (S.H.), and Vision Research Core grants EY01730 (D.M.D.). Conflict of Interest: The authors declare no competing financial interests.

References

Badea, T., Hua, Z., Smallwood, P., Williams, J., Rotolo, T., Ye, X. & Nathans, J. (2009). New mouse lines for the analysis of neuronal morphology using CreER(T)/loxP-directed sparse labeling. PLoS One 4, e7859.CrossRefGoogle ScholarPubMed
Barnard, A., Hattar, S., Hankins, M. & Lucas, R. (2006). Melanopsin regulates visual processing in the mouse retina. Current Biology: CB 16, 389–395.CrossRefGoogle ScholarPubMed
Baver, S.B., Pickard, G.E. & Sollars, P.J. (2008). Two types of melanopsin retinal ganglion cell differentially innervate the hypothalamic suprachiasmatic nucleus and the olivary pretectal nucleus. The European Journal of Neuroscience 27, 1763–1770.CrossRefGoogle ScholarPubMed
Belenky, M.A., Smeraski, C.A., Provencio, I., Sollars, P.J. & Pickard, G.E. (2003). Melanopsin retinal ganglion cells receive bipolar and amacrine cell synapses. The Journal of Comparative Neurology 460, 380–393.CrossRefGoogle ScholarPubMed
Berson, D., Castrucci, A. & Provencio, I. (2010). Morphology and mosaics of melanopsin-expressing retinal ganglion cell types in mice. The Journal of Comparative Neurology 518, 2405–2422.CrossRefGoogle ScholarPubMed
Berson, D.M., Dunn, F.A. & Takao, M. (2002). Phototransduction by retinal ganglion cells that set the circadian clock. Science 295, 1070–1073.CrossRefGoogle ScholarPubMed
Brown, T.M., Gias, C., Hatori, M., Keding, S.R., Semo, M., Coffey, P.J., Gigg, J., Piggins, H.D., Panda, S. & Lucas, R.J. (2010). Melanopsin contributions to irradiance coding in the thalamo-cortical visual system. PLoS Biol 8, e1000558.CrossRefGoogle ScholarPubMed
Chen, S.K., Badea, T.C. & Hattar, S. (2011). Photoentrainment and pupillary light reflex are mediated by distinct populations of ipRGCs. Nature 476, 92–95.CrossRefGoogle ScholarPubMed
Dacey, D.M. (1985). Wide-spreading terminal axons in the inner plexiform layer of the cat’s retina: Evidence for intrinsic axon collaterals of ganglion cells. The Journal of Comparative Neurology 242, 247–262.CrossRefGoogle ScholarPubMed
Dacey, D.M., Liao, H-W., Peterson, B.B., Robinson, F.R., Smith, V.C., Pokorny, J., Yau, K-W. & Gamlin, P.D. (2005). Melanopsin-expressing ganglion cells in primate retina signal colour and irradiance and project to the LGN. Nature 433, 749–754.CrossRefGoogle Scholar
Dacey, D.M., Peterson, B.B., Liao, H-W. & Yau, K-W. (2006). Two types of melanopsin-containing ganglion cell in the primate retina: Links to dopaminergic amacrine and DB6 cone bipolar cells. Investigative Ophthalmology & Visual Science 47, ARVO E-Abstract 3111.Google Scholar
Dowling, J.E. & Ehinger, B. (1975). Synaptic organization of the amine-containing interplexiform cells of the goldfish and Cebus monkey retina. Science 188, 270–273.CrossRefGoogle Scholar
Ecker, J.L., Dumitrescu, O.N., Wong, K.Y., Alam, N.M., Chen, S.K., Legates, T., Renna, J.M., Prusky, G.T., Berson, D.M. & Hattar, S. (2010). Melanopsin-expressing retinal ganglion-cell photoreceptors: Cellular diversity and role in pattern vision. Neuron 67, 49–60.CrossRefGoogle ScholarPubMed
Famiglietti, E.V. Jr. (1983). ‘Starburst’ amacrine cells and cholinergic neurons: Mirror-symmetric on and off amacrine cells of rabbit retina. Brain Research 261, 138–144.CrossRefGoogle ScholarPubMed
Gallego, A. & Cruz, J. (1965). Mammalian retina: Associational nerve cells in ganglion cell layer. Science 150, 1313–1314.CrossRefGoogle ScholarPubMed
Gamlin, P.D., McDougal, D.H., Pokorny, J., Smith, V.C., Yau, K-W. & Dacey, D.M. (2007). Human and macaque pupil responses driven by melanopsin-containing retinal ganglion cells. Vision Research 47, 946–954.CrossRefGoogle ScholarPubMed
Goz, D., Studholme, K., Lappi, D.A., Rollag, M.D., Provencio, I. & Morin, L.P. (2008). Targeted destruction of photosensitive retinal ganglion cells with a saporin conjugate alters the effects of light on mouse circadian rhythms. PLoS One 3, e3153.CrossRefGoogle ScholarPubMed
Guler, A.D., Ecker, J.L., Lall, G.S., Haq, S., Altimus, C.M., Liao, H.W., Barnard, A.R., Cahill, H., Badea, T.C., Zhao, H., Hankins, M.W., Berson, D.M., Lucas, R.J., Yau, K.W. & Hattar, S. (2008). Melanopsin cells are the principal conduits for rod-cone input to non-image-forming vision. Nature 453, 102–105.CrossRefGoogle ScholarPubMed
Hatori, M., Le, H., Vollmers, C., Keding, S.R., Tanaka, N., Schmedt, C., Jegla, T. & Panda, S. (2008). Inducible ablation of melanopsin-expressing retinal ganglion cells reveals their central role in non-image forming visual responses. PLoS One 3, e2451.CrossRefGoogle ScholarPubMed
Hattar, S., Kumar, M., Park, A., Tong, P., Tung, J., Yau, K. & Berson, D. (2006). Central projections of melanopsin-expressing retinal ganglion cells in the mouse. The Journal of Comparative Neurology 497, 326–349.CrossRefGoogle ScholarPubMed
Hattar, S., Liao, H.W., Takao, M., Berson, D.M. & Yau, K.W. (2002). Melanopsin-containing retinal ganglion cells: Architecture, projections, and intrinsic photosensitivity. Science 295, 1065–1070.CrossRefGoogle ScholarPubMed
Hayashida, Y. & Ishida, A.T. (2004). Dopamine receptor activation can reduce voltage-gated Na+ current by modulating both entry into and recovery from inactivation. Journal of Neurophysiology 92, 3134–3141.CrossRefGoogle ScholarPubMed
Honrubia, F.M. & Elliott, J.H. (1968). Efferent innervation of the retina I. Morphologic study of the human retina. Archives of Ophthalmology 80, 98–103.CrossRefGoogle ScholarPubMed
Ichinose, T. & Lukasiewicz, P.D. (2007). Ambient light regulates sodium channel activity to dynamically control retinal signaling. Journal of Neuroscience 27 17, 4756–4764.CrossRefGoogle ScholarPubMed
Indra, A.K., Warot, X., Brocard, J., Bornert, J.M., Xiao, J.H., Chambon, P. & Metzger, D. (1999). Temporally-controlled site-specific mutagenesis in the basal layer of the epidermis: Comparison of the recombinase activity of the tamoxifen-inducible Cre-ER(T) and Cre-ER(T2) recombinases. Nucleic Acids Research 27, 4324–4327.CrossRefGoogle ScholarPubMed
Jusuf, P.R., Lee, S.C., Hannibal, J. & Grunert, U. (2007). Characterization and synaptic connectivity of melanopsin-containing ganglion cells in the primate retina. The European Journal of Neuroscience 26, 2906–2921.CrossRefGoogle ScholarPubMed
Lasater, E.M. & Dowling, J.E. (1985). Dopamine decreases conductance of the electrical junctions between cultured retinal horizontal cells. Proceedings of the National Academy of Sciences of the United States of America 82, 3025–3029.CrossRefGoogle ScholarPubMed
LeGates, T.A., Altimus, C.M., Wang, H., Lee, H.K., Yang, S., Zhao, H., Kirkwood, A., Weber, E.T. & Hattar, S. (2012). Aberrant light directly impairs mood and learning through melanopsin-expressing neurons. Nature 491, 594–598.CrossRefGoogle ScholarPubMed
Lobe, C.G., Koop, K.E., Kreppner, W., Lomeli, H., Gertsenstein, M. & Nagy, A. (1999). Z/AP, a double reporter for cre-mediated recombination. Developmental Biology 208, 281–292.CrossRefGoogle ScholarPubMed
Lucas, R.J., Hattar, S., Takao, M., Berson, D.M., Foster, R.G. & Yau, K.W. (2003). Diminished pupillary light reflex at high irradiances in melanopsin-knockout mice. Science 299, 245–247.CrossRefGoogle ScholarPubMed
Mariani, A.P. & Hersh, L.B. (1988). Synaptic organization of cholinergic amacrine cells in the rhesus monkey retina. The Journal of Comparative Neurology 267, 269–280.CrossRefGoogle ScholarPubMed
Nir, I., Harrison, J., Haque, R., Low, M., Grandy, D., Rubinstein, M. & Iuvone, P. (2002). Dysfunctional light-evoked regulation of cAMP in photoreceptors and abnormal retinal adaptation in mice lacking dopamine D4 receptors. Journal of Neuroscience 22, 2063–2073.CrossRefGoogle ScholarPubMed
Panda, S., Provencio, I., Tu, D.C., Pires, S.S., Rollag, M.D., Castrucci, A.M., Pletcher, M.T., Sato, T.K., Wiltshire, T., Andahazy, M., Kay, S.A., Van Gelder, R.N. & Hogenesch, J.B. (2003). Melanopsin is required for non-image-forming photic responses in blind mice. Science 301, 525–527.CrossRefGoogle ScholarPubMed
Panda, S., Sato, T.K., Castrucci, A.M., Rollag, M.D., DeGrip, W.J., Hogenesch, J.B., Provencio, I. & Kay, S.A. (2002). Melanopsin (Opn4) requirement for normal light-induced circadian phase shifting. Science 298, 2213–2216.CrossRefGoogle ScholarPubMed
Percival, K.A., Martin, P.R. & Grunert, U. (2011). Synaptic inputs to two types of koniocellular pathway ganglion cells in marmoset retina. The Journal of Comparative Neurology 519, 2135–2153.CrossRefGoogle ScholarPubMed
Peterson, B.B. & Dacey, D.M. (1998). Morphology of human retinal ganglion cells with intraretinal axon collaterals. Visual Neuroscience 15, 377–387.CrossRefGoogle ScholarPubMed
Provencio, I., Rollag, M.D. & Castrucci, A.M. (2002). Photoreceptive net in the mammalian retina. Nature 415, 493–494.CrossRefGoogle ScholarPubMed
Raven, M.A., Eglen, S.J., Ohab, J.J. & Reese, B.E. (2003). Determinants of the exclusion zone in dopaminergic amacrine cell mosaics. The Journal of Comparative Neurology 461, 123–136.CrossRefGoogle ScholarPubMed
Renna, J.M., Weng, S. & Berson, D.M. (2011). Light acts through melanopsin to alter retinal waves and segregation of retinogeniculate afferents. Nature Neuroscience 14, 827–829.CrossRefGoogle ScholarPubMed
Rodieck, R.W. (1989). Starburst amacrine cells of the primate retina. The Journal of Comparative Neurology 285, 18–37.CrossRefGoogle ScholarPubMed
Sakai, H.M., Naka, K-I. & Dowling, J.E. (1986). Ganglion cell dendrites are presynaptic in catfish retina. Nature 319, 495–497.CrossRefGoogle ScholarPubMed
Schmidt, T.M. & Kofuji, P. (2009). Functional and morphological differences among intrinsically photosensitive retinal ganglion cells. Journal of Neuroscience 29, 476–482.CrossRefGoogle ScholarPubMed
Sekaran, S., Foster, R.G., Lucas, R.J. & Hankins, M.W. (2003). Calcium imaging reveals a network of intrinsically light-sensitive inner-retinal neurons. Current Biology: CB 13, 1290–1298.CrossRefGoogle ScholarPubMed
Usai, C., Ratto, G.M. & Bisti, S. (1991). Two systems of branching axons in monkey’s retina. The Journal of Comparative Neurology 308, 149–161.CrossRefGoogle ScholarPubMed
Viney, T.J., Balint, K., Hillier, D., Siegert, S., Boldogkoi, Z., Enquist, L.W., Meister, M., Cepko, C.L. & Roska, B. (2007). Local retinal circuits of melanopsin-containing ganglion cells identified by transsynaptic viral tracing. Current Biology: CB 17, 981–988.CrossRefGoogle ScholarPubMed
Vugler, A., Redgrave, P., Semo, M., Lawrence, J., Greenwood, J. & Coffey, P. (2007). Dopamine neurones form a discrete plexus with melanopsin cells in normal and degenerating retina. Experimental Neurology 205, 26–35.CrossRefGoogle Scholar
Wässle, H. & Boycott, B.B. (1991). Functional architecture of the mammalian retina. Physiological Reviews 71, 447–480.CrossRefGoogle ScholarPubMed
Zhang, D.Q., Belenky, M.A., Sollars, P.J., Pickard, G.E. & McMahon, D.G. (2012). Melanopsin mediates retrograde visual signaling in the retina. PLoS One 7, e42647.CrossRefGoogle ScholarPubMed
Zhang, D.Q., Wong, K.Y., Sollars, P.J., Berson, D.M., Pickard, G.E. & McMahon, D.G. (2008). Intraretinal signaling by ganglion cell photoreceptors to dopaminergic amacrine neurons. Proceedings of the National Academy of Sciences of the United States of America 105, 14181–14186.CrossRefGoogle ScholarPubMed
Zhang, D.Q., Zhou, T.R. & McMahon, D.G. (2007). Functional heterogeneity of retinal dopaminergic neurons underlying their multiple roles in vision. Journal of Neuroscience 27, 692–699.CrossRefGoogle ScholarPubMed
Figure 0

Fig. 1. Titrated tamoxifen injection controls the density of labeled cells in knock-in mice (Opn4CreERT2; Z/AP) to reveal intraretinal collaterals. (A) Micrograph of a high-volume tamoxifen-injected retina with dense labeling of melanopsin-expressing cells. Area indicated by white box is shown at right at higher magnification. Top panel: focus is on the dendrites of M2 cells (open arrows) in the inner IPL. Bottom panel: focus is shifted to the dendrites of M1 cells in the outer IPL (open arrows). (B) Low tamoxifen injection resulted in a sparsely labeled retina with seven nonoverlapping M1 cells (asterisks). Area within dashed lines is shown at right at higher magnification. The axon (closed arrow) of an M1 cell (open arrow) gives off a single collateral process (arrowhead).

Figure 1

Fig. 2. Photomicrographs of intraretinal axon collateral-bearing M1 cells in mouse. (A) Photomicrographs of an M1 cell with outer-stratifying collateral. Left panel: focus is on the soma (open arrow) and axon (closed arrows). A single collateral (arrowhead) branches from the axon. Right panel: focus is shifted to the dendrites (open arrowhead) and axon collateral (closed arrowhead) in the outer IPL. (B) Photomicrographs of an M1 cell with inner-stratifying collateral. Left panel: focus is on the soma (open arrow) and axon (closed arrow) that gave rise to a single collateral (closed arrowhead) that terminated in the inner IPL. Right panel: focus is on the dendritic arbor in the outer IPL.

Figure 2

Fig. 3. Mouse axon collateral arbors show variability in size and extent of branching. (A-D) Tracings of 4 M1 cells with axon collaterals. Cell bodies indicated by open arrows; primary axons and collaterals are shown in red. Primary axons (arrows) extended to the optic disk and gave rise to collateral processes (arrowheads) that terminated in either the inner (A, B) or the outer (C, D) IPL.

Figure 3

Fig. 4. Primate giant melanopsin ganglion cells exhibit axon collaterals. (A) Photomicrograph of an outer stratifying melanopsin immunolabeled giant cell in macaque retina. Soma (open arrow) was displaced to the INL. The axon (closed arrows) gave rise to a thin collateral (arrowheads) studded with varicosities that could be followed for ∼800 µm in the outer IPL before fading. (B-C) Tracings of inner stratifying giant cells (open arrows) intracellularly filled with Neurobiotin from macaque (B) and human (C) retina. Axons (closed arrows) gave off several collateral branches (arrowheads) en route to the optic disk. Collaterals terminated in the outer IPL with the exception of the cell on left in (B), where two short collaterals terminated in the inner IPL (shown in red in inset). Insets show magnified views of collaterals indicated by the dashed lines.