Development of melanopsin-based irradiance detecting circuitry

Abstract

Background: Most retinal ganglion cells (RGCs) convey contrast and motion information to visual brain centers. Approximately 2% of RGCs are intrinsically photosensitive (ipRGCs), express melanopsin and are necessary for light to modulate specific physiological processes in mice. The ipRGCs directly target the suprachiasmatic nucleus (SCN) to photoentrain circadian rhythms, and the olivary pretectal nucleus (OPN) to mediate the pupillary light response. How and when this ipRGC circuitry develops is unknown.

Results: Here, we show that some ipRGCs follow a delayed developmental time course relative to other image-forming RGCs. Specifically, ipRGC neurogenesis extends beyond that of other RGCs, and ipRGCs begin innervating the SCN at postnatal ages, unlike most RGCs, which innervate their image-forming targets embryonically. Moreover, the appearance of ipRGC axons in the OPN coincides precisely with the onset of the pupillary light response.

Conclusions: Some ipRGCs differ not only functionally but also developmentally from RGCs that mediate pattern-forming vision.

Figure 1

Birth and melanopsin expression of ipRGCs. (A,B) The birthdates of ipRGCs were compared to Brn3a-positive RGCs (representative images in (A), paradigm in (B)). (C) For each time point, we determined the proportion of ipRGCs labeled with EdU; asterisks indicate significant difference between ipRGCs and Brn3a (t-test, P < 0.05). Representative sections with double-labeled RGCs (yellow arrowheads) and EdU-negative RGCs (white arrows) from postnatal day 0 (P0) retinas pulsed with EdU at embryonic day 15 (E15) are shown in (A). Note that the ipRGC marker beta-galactosidase (β-gal), is cytoplasmic, while Brn3a and EdU are nuclear. For all EdU time points, n = 3 to 4 retinas per time point, mean ± standard error of the mean. (D-F) Melanopsin (Opn4) expression begins at E15 based on immunofluorescence (D), and the Melanopsin^Cre/+;Z/AP (E) and Melanopsin^tauLacZ/+ genetic labeling methods (F). The Melanopsin^Cre/+;Z/AP labels all ipRGC subtypes and does not depend on the melanopsin locus for signal strength. Arrows in D-F denote migrating cells. GCL; Ganglion Cell Layer and RPE; Retinal Pigmented Epithelium. (G) Coronal sections from Melanopsin^tauLacZ/+ mice show an initial lack of ipRGCs in the periphery at E15, which is entirely filled in by P0 (arrows; D, dorsal; V, ventral). Note the lack of X-gal staining within the central optic nerve at E18 (G, green arrowhead). Scale bars: 50 μm (A,D-F); 100 μm (G).

Figure 2

ipRGCs innervate the suprachiasmatic nucleus postnatally. Innervation of the suprachiasmatic nucleus (SCN) begins at P0 and continues throughout the first two postnatal weeks. (A) Labeling of all RGCs in coronal sections with fluorescently labeled cholera toxin B subunit (CTB), one color per eye. Retinal fibers fill the SCN by P7 and innervation becomes bilateral by P14. Note that for CTB labeling, a P0 time point was not possible due to the survival period required for the tracer to label distal axons. (B,C) Labeling of ipRGCs and their axons with both Melanopsin^Cre/+;Z/AP (B) and Melanopsin^tauLacZ/+ (C) reveals a similar innervation pattern. (D-F) In contrast, CTB labeling reveals RGC axons have already penetrated the LGN by P1 (D), and genetic labeling of Brn3b-positive RGCs (E) reveals that this subset of RGCs begins entering the pretectum as early as E17 (F). Arrows indicate fibers entering target area. n = 3 or more animals for each time point. OT, optic tract; dashed outlines demarcate the SCN prior to innervation in (A,B). Scale bars: 100 μm (A-C); 500 μm (F).

Figure 3

Spatial and temporal progression of SCN innervation. Serial 50-μm coronal sections from individual brains of Melanopsin^tauLacZ/+ mice at different time points reveal the spatial and temporal progression of ipRGC innervation of the SCN. Nuclei are counterstained with Vector Nuclear Fast Red. Scale bar: 100 μm.

Figure 4

Olivary pretectal nucleus innervation and onset of pupillary light responses. (A) The olivary pretectal nucleus (OPN) is innervated by CTB-labeled RGC axons at P0 and ipsilateral and contralateral axons have segregated by P7. (B) Melanopsin^Cre/+;Z/AP-labeled axons show a similar innervation pattern from birth. (C) In contrast, Melanopsin^tauLacZ/+-labeled axons are not detected until P7 and show adult-like innervation of the OPN shell by P14. (D,E) The paradigm used for all pupil measurements (D) and representative images (E). Note that the pupil appears white at P7 because at this early time point it was necessary to reflect infrared light off the back of the eye to make the iris visible to the camera. Carbachol was used to induce maximal constriction of the ciliary muscle. (F) Percent constriction of pupil was measured and the summary of data for PLR is shown. Significant light responses are detected from P7 onward (n = 3 or more mice per time point, one-way ANOVA with Tukey post hoc test, error bars are mean ± standard error of the mean). No baseline constriction was observed in adult mice. Scale bars: 100 μm (A-C).

Figure 5

ipRGC innervation of the LGN. Comparison of LGN innervation by Opn4^tauLacZ/+-labeled ipRGC axons and Opn4^Cre/+;Z/AP labeling of total ipRGC axons across postnatal development. Arrows denote the intergeniculate leaflet, which separates the dorsal and ventral LGN. Scale bar: 100 μm.