Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Apr 7.
Published in final edited form as: J Comp Neurol. 2020 Nov 23;529(8):1863–1875. doi: 10.1002/cne.25061

The projections of ipRGCs and conventional RGCs to retinorecipient brain nuclei

Corinne Beier 1, Ze Zhang 1, Maria Yurgel 1, Samer Hattar 1,*
PMCID: PMC10081000  NIHMSID: NIHMS1887896  PMID: 33104235

Abstract

Retinal ganglion cells (RGCs), the output neurons of the retina, allow us to perceive our visual environment. RGCs respond to rod/cone input through the retinal circuitry, however, a small population of RGCs are in addition intrinsically photosensitive (ipRGCs) and project to unique targets in the brain to modulate a broad range of subconscious visual behaviors such as pupil constriction and circadian photoentrainment. Despite the discovery of ipRGCs nearly two decades ago, there is still little information about how or if conventional RGCs (non-ipRGCs) target ipRGC-recipient nuclei to influence subconscious visual behavior. Using a dual recombinase fluorescent reporter strategy, we showed that conventional RGCs innervate many subconscious ipRGC-recipient nuclei, apart from the suprachiasmatic nucleus. We revealed previously unrecognized stratification patterns of retinal innervation from ipRGCs and conventional RGCs in the ventral portion of the lateral geniculate nucleus. Further, we found that the percent innervation of ipRGCs and conventional RGCs across ipsi- and contralateral nuclei differ. Our data provide a blueprint to understand how conventional RGCs and ipRGCs innervate different brain regions to influence subconscious visual behaviors.

Graphical Abstract

graphic file with name nihms-1887896-f0012.jpg

INTRODUCTION

Retinal ganglion cells (RGCs) are the output neurons that connect the eye to the brain and are responsible for all visual perception in mammals. Light information conveyed by RGCs contributes to pattern vision (image forming), as well as subconscious visual behaviors (non-image forming) such as pupil constriction, circadian photoentrainment, sleep and mood. Intrinsically photosensitive RGCs (ipRGCs) are a small population of RGCs that express melanopsin and project to unique targets in the brain that modulate a broad range of non-image forming visual behaviors (Altimus et al., 2008; Chen et al., 2011; Ecker et al., 2010; Hattar et al., 2006; LeGates et al., 2013; Lupi et al., 2008; Rupp et al., 2019). There is still, however, little information about how or if conventional RGCs (non-ipRGCs) can influence subconscious visual behaviors. To address this knowledge gap, we must first delineate whether conventional RGCs innervate non-image forming nuclei.

To determine if conventional RGCs innervate non-image forming nuclei (Ecker et al., 2010; Hattar et al., 2002, 2006; Li and Schmidt, 2018; Morin and Studholme, 2014), we utilized a dual recombinase two color fluorescent reporter strategy to delineate, in the same animal, how ipRGCs and conventional RGCs converge and diverge in ipRGC-recipient brain regions. We found that the master clock in the suprachiasmatic nucleus (SCN) is almost exclusively innervated by ipRGCs, as was previously implicated (Hattar et al., 2006). However, the SCN was the exception to the rule, as most ipRGC-recipient areas received substantial projections from conventional RGCs. The dual color strategy allowed us to differentiate areas where conventional RGCs and ipRGCs converge and intermingle, from brain regions where they form completely distinct subdivisions. Specifically, we found that conventional RGCs and ipRGCs innervate distinct subdivisions of the ventral lateral geniculate nucleus (vLGN). As we found in the SCN, the vLGN core region receives exclusive ipRGC input. Furthermore, we found differences in the percent innervation between the ipsi- and contralateral patterns of conventional RGCs and ipRGCs. Our results provide a basis for understanding how ipRGCs and conventional RGC modulate non-image forming behaviors for normal function.

RESULTS

The SCN receives nearly exclusive input from ipRGCs

To investigate the relative innervations by conventional RGCs and ipRGCs to retinorecipient nuclei, we developed a method to selectively label both cell populations in a single mouse (Figure 1a). We used the Opn4Cre mouse line that expresses Cre in ipRGCs, in combination with a Flp-Cre dual recombinase reporter mouse line, RC::FLTG (Rosa26-CAG-Frt-Stop-Frt-loxP-tdTomato-loxP-eGFP, shorthand Rosa26CAG-tdTomato-eGFP) (Ecker et al., 2010; Plummer et al., 2015). In the RC::FLTG mouse line, Flp recombinase excises the Frt flanked stop signal to label Flp-positive cells with tdTomato. The tdTomato transcription termination signal prevents eGFP expression. Cells that are only Cre-positive excise the loxP sites but stay unlabeled while the Frt flanked stop signal remains (Figure 1b). Only cells that express Cre and Flp recombinases are labeled with eGFP (Figure 1c).

Figure 1. Experiment design and verification.

Figure 1.

Open in a new tab

(a) Schematic detailing experiment strategy. AAV2-hSyn-Flpo is injected intravitreally into one eye of adult Opn4Cre/+;Rosa26CAG-tdTomato-eGFP/+ mice. Conventional RGC (magenta) and ipRGC (green) projections can be visualized in retinorecipient nuclei. (b) Immunostaining of GFP and RFP in the retina of the eye without intravitreal injection (left eye) in Opn4Cre/+;Rosa26CAG-tdTomato-eGFP/+ mice. No cells are labeled. Scale bar 1000 μm. (c) Immunostaining of GFP (green) and RFP (magenta) in the retina of the eye that received intravitreal viral injection in Opn4Cre/+;Rosa26CAG-tdTomato-eGFP/+ mice. Scale bar 1000 μm. (d) Detail view from location indicated in (c). Melanopsin (Opn4) antibody stains a subset of cells (cyan, arrows). All melanopsin-positive cells are GFP-positive (green). RFP-positive cells (magenta) do not overlap with melanopsin- or GFP-positive cells. Scale bar 100 μm. (e) Most ganglion cells (86.5 ± 6.5 %) are labeled after an intravitreal injection. RBPMS antibody stains all retinal ganglion cells (cyan). Most RBPMS positive cells are also positive for either GFP (green) or RFP (magenta). Some ganglion cells are not labeled by the intravitreal injection (dashed circles).

We generated Opn4Cre/+;Rosa26CAG-tdTomato-eGFP/+ mice and injected AAV2-hSyn-Flpo intravitreally into one eye of adult mice to infect all RGCs with Flp (Figure 1A). AAV2-hSyn-Flpo expresses the Flp recombinase under the neuronal cell specific synapsin promoter. In the retina of Opn4Cre/+;Rosa26CAG-tdTomato-eGFP/+ mice infected with AAV2-hSyn-Flpo, ipRGCs express Flp and Cre, whereas conventional RGCs only express Flp. As expected, this method specifically labels ipRGCs with eGFP and conventional RGCs with tdTomato (Figure 1c). The dual recombinase strategy worked as expected, since no cells expressed both tdTomato and eGFP (Figure 1D). To test whether our strategy is selective to ipRGCs, we stained the retina of Opn4Cre/+;Rosa26CAG-tdTomato-eGFP/+ mice infected with AAV2-hSyn-Flpo with a melanopsin antibody, which labels a subset of ipRGCs (Ecker et al., 2010). We found complete overlap between melanopsin expressing cells and eGFP (Figure 1d), confirming that we can selectively label ipRGCs from conventional RGCs. Furthermore, we co-stained for RBPMS to label all RGCs and found a high degree of RGC infection with our injections (86.5 ± 6.5% of all RGCs are labeled, n = 2 retinas, Figure 1e).

After confirming our labeling strategy, we traced axonal fibers that project to ipRGC-recipient brain regions to answer the following questions: (1) Do ipRGC-recipient brain regions receive exclusive input from ipRGCs? (2) If not, do the inputs of conventional RGCs and ipRGCs converge or diverge? (3) What are the projection patterns of conventional RGCs and ipRGCs in their ipsilateral and contralateral regions? We focused our analysis on retinorecipient nuclei known to receive robust ipRGC innervation: the suprachiasmatic nucleus (SCN), dorsal lateral geniculate nucleus (dLGN), ventral LGN (vLGN), intergeniculate leaflet (IGL), olivary pretectal nucleus (OPN), and superior colliculus (SC). We also observed modest ipRGC innervation to the zona incerta (ZI), peri habenula (PHb), and supraoptic nucleus (SON). Additional retinorecipient areas that are known to receive minor ipRGC innervation were labeled inconsistently by the dual recombinase mouse and hence were not included in this study (Delwig et al., 2016; Ecker et al., 2010; Hattar et al., 2002, 2006; Li and Schmidt, 2018).

Testifying to the specificity of the dual recombinase strategy, we first examined the SCN, thought to receive exclusive input from ipRGCs (Baver et al., 2008; Chew et al., 2017; Güler et al., 2008; Hattar et al., 2006). We show indeed that the SCN barely receives any input from conventional RGCs (Figure 2, inset in middle panel). From rostral to caudal, the SCN receives only ipRGC innervations that are uniform across the ipsilateral and contralateral regions (Figure 2). The ipsilateral SCN nucleus received 48% of ipRGC innervation and this innervation was equivalent across the ipsi- and contralateral regions (Student’s paired t test, p = 0.31).

Figure 2. ipRGC and conventional RGC projections in the SCN.

Figure 2.

Open in a new tab

ipRGC (green) and conventional RGC (magenta, inset) innervation to the suprachiasmatic nucleus (SCN). ipRGC input is uniform across the contra (left) and ipsilateral (right) lobes (dashed outlines). Scale bars 100 μm.

ipRGC projections overlap with conventional RGCs in the SC and dLGN

A subset of ipRGCs play a role in image-forming vision and project to the image-forming retinorecipient targets, the SC and dLGN (Ecker et al., 2010; Estevez et al., 2012; Quattrochi et al., 2018; Schmidt et al., 2014; Stabio et al., 2017). In the contralateral SC, conventional RGC innervations overlapped in the retinorecipient layer with ipRGC innervation and appeared patchy (Figure 3a). However, ipRGC innervation was confined to the deeper areas of the SC while conventional RGC innervation covered the entire region (Figure 3a,b). The ipsilateral SC received minimal, but completely overlapping ipRGC and conventional RGC input (Figure 3a,c).

Figure 3. ipRGC and conventional RGC projections in the SC.

Figure 3.

Open in a new tab

(a) ipRGC (green) and conventional RGC (magenta) innervation to the contralateral (left) and ipsilateral (right) superior colliculus (SC). Conventional RGC and ipRGC innervation to the ipsilateral SC overlap (inset). Scale bar 500 μm. (b-c) Quantification of conventional RGC and ipRGC innervation overlap in the contralateral (b) and ipsilateral (c) SC. (See Methods for overlap calculation, Mean ± SEM quantified from n = 2–4 coronal brain slices from n = 4 mice). ipRGC innervation overlaps with conventional RGC innervation 97±3% of the time in the contralateral SC and 78±8% of the time in the ipsilateral SC.

In the dLGN, we also observed overlapping ipRGC and conventional RGC innervation (Figure 4). As we have shown previously (Ecker et al., 2010), ipRGC innervation is concentrated towards the medial border of the dLGN nucleus and this pattern becomes more apparent in the caudal dLGN (Figure 4 and Figure 7b–d). However, this area was not exclusive to ipRGCs as conventional RGCs innervated the entirety of the dLGN, including the medial border (Figure 4 and Figure 7b–d). ipRGCs stratify with conventional RGCs 88±3%, 80±2%, 92±2% in the rostral, medial, caudal dLGN (Figure 7b–d). We did not find any correlation in the change of stratification overlap along the rostral-caudal axis (Pearson’s correlation, R = 0.3, p = 0.11) In the ipsilateral dLGN, there was considerable overlap between the ipRGCs and conventional RGCs (Figure 4), with ipRGCs contributing substantially to the ipsilateral region despite being a minority population in the retina (Figure 5).

Figure 4. ipRGC and conventional RGC projections to the geniculate complex.

Figure 4.

Open in a new tab

Conventional RGC (magenta) and ipRGC (green) innervation to the contralateral and ipsilateral geniculate. The dorsal lateral geniculate nucleus (dLGN) and the ventral LGN (vLGN) are separated by a thin region, the intergeniculate leaflet (IGL). Retinal innervations to the contralateral dLGN and vLGN are outlined (dashed lines). Scale bars 500 μm. Sections are displayed top to bottom, from rostral to caudal, with distances from Bregma listed far right.

Figure 7. Stratification of ipRGCs and conventional RGCs in the dLGN and vLGN.

Figure 7.

Open in a new tab

(a) Schematic summarizing quantification of the stratification of ipRGCs (green) and conventional RGCs (cRGCs, magenta) in the dorsal and ventral lateral geniculate nucleus (dLGN and vLGN). (b-d) Quantification of the innervation of ipRGCs (green) and conventional RGCs (magenta) in the rostral, medial, and caudal dLGN as distance from the lateral edge of the optic tract towards the medial edge of the dLGN. ipRGC innervation is concentrated towards the medial edge of the dLGN but overlaps with conventional RGC innervation. (e-g) Same as in (b-d) except innervation of the vLGN is shown. Conventional RGCs overlap with ipRGC innervation in the rostral vLGN (e) but segregate in the medial (f) and caudal vLGN (g).

Figure 5. Ipsilateral ipRGC and conventional RGC projections to the geniculate complex.

Figure 5.

Open in a new tab

(a-c) The projection in percent of ipRGC (green) and conventional RGC (cRGCs, magenta) innervation to the dorsal lateral geniculate nucleus (dLGN), the intergeniculate leaflet (IGL), and the ventral LGN (vLGN). Ipsilateral innervation between ipRGCs and conventional RGCs was not significantly different except in the caudal IGL. See Methods for ipsilateral projection calculation. Individual points show the ipsilateral projection calculated from a single coronal brain slice (n = 1–2 slices from n = 4 mice per group, Anova one way, post hoc Bonferroni, p < 0.05 indicated by * and p > 0.05 shown as n.s.). (d-e) Detailed view of conventional RGC (magenta) and ipRGC (green) innervation to the rostral (d) and caudal (e) ipsilateral vLGN and IGL, as seen in Figure 4. The approximate area of retinal innervation from the contralateral eye to the vLGN is outlined (dashed lines). The IGL is indicated by arrows between the vLGN and dLGN. Scale bars 100 μm.

The caudal ipsilateral IGL receives conventional RGC innervation

The IGL has been implicated in circadian photoentrainment, and, more recently, in nonphotic behaviors such as sleep and feeding (Edelstein and Amir, 1999; Fernandez et al., 2020; Huang et al., 2019; Shi et al., 2020). Consistent with its role in circadian photoentrainment, it is not surprising that the contralateral IGL was predominately innervated by ipRGCs and received little to no conventional RGC innervation (Figure 4 and Figure 5e,f). The ipsilateral IGL also received substantial ipRGC input (Figure 5b). Surprisingly, in contrast to the dLGN, there was more ipsilateral input from conventional RGCs at the far caudal IGL (Figure 5b). The conventional RGC projections overlapped, however, completely with ipRGC input (Figure 5f).

ipRGC and conventional RGC innervations stratify in the vLGN

The vLGN is a retinorecipient region associated with non-image forming behaviors, although its function is not well understood (Davidowa and Albrecht, 1992; Harrington, 1997; Huang et al., 2019; Monavarfeshani et al., 2017; Shi et al., 2020). The vLGN also shows retinotopic responses to visual stimuli necessary for gaze control, suggesting a role in image-forming vision as well (Nagata and Hayashi, 1984). We observed a remarkable separation of ipRGC and conventional RGC projections within the caudal vLGN (Figure 4 and Figure 6). In the rostral vLGN, ipRGC and conventional RGC projections overlapped entirely (Figure 6 and Figure 7e–g). In the caudal vLGN, conventional RGCs innervated only the lateral side of the vLGN, leaving the medial (core) side to ipRGC projections without any overlap (Figure 6 and Figure 7e–g). Indeed, conventional RGCs reduced their overlap with ipRGCs along the rostral-caudal axis, overlapping 85±4%, 46±3%, and 23±2% in the rostral, medial, and caudal vLGN, respectively (Pearson’s correlation, R = −0.9, p = 1e-11). In the ipsilateral vLGN, stratification was still visible with areas of overlap (Figure 5e,f) and it has the characteristics of both the dLGN and IGL. Namely, in the rostral portion, more conventional RGCs innervate the ipsilateral vLGN, whereas, more ipRGCs innervate the ipsilateral vLGN in the caudal region (Figure 5e,f). Together, these data show that there are clear stratification patterns in the lateral geniculate complex between ipRGCs and conventional RGCs.

Figure 6. ipRGC and conventional RGC projections to the contralateral vLGN and IGL.

Figure 6.

Open in a new tab

Conventional RGC (magenta) and ipRGC (green) innervation to the contralateral ventral lateral geniculate nucleus (vLGN) and intergeniculate leaflet (IGL). Retinal innervation to the vLGN is outlined (dashed lines). The IGL is indicated by arrows between the vLGN and dorsal LGN (dLGN). Scale bars 100 μm.

Conventional RGC and ipRGC projections overlap to innervate the OPN

The OPN shell is known to receive innervation from ipRGCs that are required for normal pupil constriction (Chen et al., 2011; Hattar et al., 2002, 2006). The OPN core has also been shown to receive ipRGC innervation (Chen et al., 2011; Delwig et al., 2016; Ecker et al., 2010; Quattrochi et al., 2018), although its function remains unknown. We found that conventional RGCs substantially innervated the core but avoided the shell of the contralateral OPN (Figure 8). In the ipsilateral OPN, projections from conventional RGCs and ipRGCs overlap (Figure 8).

Figure 8. ipRGC and conventional RGC projections to the OPN.

Figure 8.

Open in a new tab

ipRGC (green) and conventional RGC (magenta) innervation to the contralateral and ipsilateral olivary pretectal nucleus (OPN). Retinal innervation to the OPN is outlined (dashed lines). Scale bars 100 μm.

Retinorecipient targets with modest ipRGC innervation receive little input from conventional RGCs and weak ipsilateral innervation

The PHb mediates the direct effect of light on mood and is thought to be innervated primarily by ipRGCs (Fernandez et al., 2018). Similar to the projection patterns we observed in the SCN and vLGN core, we found that ipRGC innervations are the predominate retinal afferents to the PHb, with only a few conventional RGC puncta, indicative of minimal innervation (Figure 9). No ipsilateral innervation was visible in this region.

Figure 9. ipRGC and conventional RGC projections to the PHb.

Figure 9.

Open in a new tab

(a) ipRGC (green) and conventional RGC (magenta, inset) innervation to the contralateral peri habenula (PHb). Dashed lines indicate the border between the hippocampus and the lateral habenula (LHb). (b) Another example of PHb innervation. Scale bars 100 μm.

The ZI responds to some visual stimuli but its role in visual behaviors is not well understood (Wang et al., 2020; Zhao et al., 2019). At the border of the optic tract and ZI lies the peripeduncular nucleus, which we found is innervated exclusively by conventional RGCs (Figure 10). The significance of the retinal input to the peripeduncular nucleus is completely unknown, but it is likely innervated by a direction-selective RGC subtype (Rivlin-Etzion et al., 2011). The ZI, on the other hand, is innervated entirely by ipRGCs, except for a few conventional RGC fibers (Figure 10, inset). We did not observe retinal afferents in the ipsilateral ZI, as expected (Morin and Studholme, 2014).

Figure 10. ipRGC and conventional RGC projections to the ZI.

Figure 10.

Open in a new tab

(a) ipRGC (green) and conventional RGC (magenta, inset) innervation to the contralateral zona incerta (ZI). Arrowhead indicates retinal innervation to the peripeduncular nucleus. Retinal innervation to the zona incerta lies between the arrows. Dashed line outlines the cerebral peduncle along the optic tract (ot) and the ZI. (b) Another example of ZI innervation. Scale bars 100 μm.

The SON is a neurotransmitter-rich nucleus whose visual function is unclear, although it has been implicated in metabolism, hydration, general anesthetics, sleep, as well as lactation in nursing females (Cunningham et al., 2004; Jiang-Xie et al., 2019). We found both conventional RGC and ipRGC innervation to the SON (Figure 11). Conventional RGC innervation was sparser than ipRGC innervation but covered the region almost as completely.

Figure 11. ipRGC and conventional RGC projections to the SON.

Figure 11.

Open in a new tab

(a) ipRGC (green) and conventional RGC (magenta) innervation to the contralateral supraoptic nucleus (SON). Above the optic tract, retinal innervation to the SON is outlined (dashed line). (b) Another example of SON innervation. Scale bars 100 μm.

DISCUSSION

The role of ipRGCs in many non-image forming visual behaviors has been well-established, and their dense innervation to non-image forming brain nuclei is well known, but it remains unclear if ipRGCs are the sole conveyers of non-image forming visual information. Previous work has described ipRGC innervation patterns in comparison to all RGCs, but these studies could not readily delineate conventional RGC innervation to targets with dense ipRGC innervation due to the overlap in their labeling (Hattar et al., 2006; Morin and Studholme, 2014). Here, we used a dual recombinase labeling strategy to compare, in the same mouse, the innervation patterns of ipRGCs and conventional RGCs in a non-overlapping manner. We found that many non-image forming retinorecipient nuclei are innervated by conventional RGCs. The overlap and innervation patterns of ipRGC and conventional RGC projections vary drastically across retinorecipient nuclei and between ipsi and contralateral targets. This provides the intriguing possibility that conventional RGCs may modulate a subset of, but not all, non-image forming behaviors.

We were most surprised to find clear stratification and separation of ipRGC and conventional RGC inputs to the vLGN. Well-segregated cellular layers have been recognized in the dLGN and vLGN-like regions in primates and other diurnal mammals, but mice were not thought to have the same rigid organization in the geniculate complex (Kerschensteiner and Guido, 2017; Livingston and Mustari, 2000; Monavarfeshani et al., 2017). Some stratification of retinal subtype innervation to the mouse geniculate has been noted, but it is unclear if the stratification is indicative of non-overlapping retinal inputs (Kerschensteiner and Guido, 2017; Monavarfeshani et al., 2017). Our findings suggest that there is previously unrecognized organization in the vLGN that receives non-overlapping retinal innervation, consistent with recent data on cell type laminae within the vLGN (Sabbagh et al., 2020).

The pupillary light response relies on ipRGC innervation to the OPN shell majorly from M1 ipRGCs (Chen et al., 2011; Hattar et al., 2002, 2006). We found that the OPN receives overlapping ipRGC and conventional RGC innervation, except in the contralateral shell of the OPN, which receives only ipRGC input. The function of the OPN core is not well-understood (Dhande et al., 2018), but it is likely that ipRGCs and conventional RGCs both play a role in its visual response.

The IGL is another non-image forming nucleus that receives conventional RGC input. Excitingly, the innervation from conventional RGCs increases significantly to the caudal ipsilateral IGL (Anova one way, post hoc Bonferroni comparing rostral and caudal ipsilateral conventional RGC innervation to the IGL, p = 0.0001). The IGL has been implicated in numerous non-image forming behaviors, making its conventional RGC innervation somewhat surprising (Edelstein and Amir, 1999; Fernandez et al., 2020; Huang et al., 2019; Shi et al., 2020). Furthermore, the increase in conventional RGC innervation to the IGL along the rostral-caudal and contra-ipsi axes fits well with previous studies suggesting underlying complexity within the cellular organization of the IGL (Moore and Card, 1994; Morin, 2013). It remains to be seen what role conventional RGCs will play in the ipsilateral IGL.

Removing ipRGC innervation to the SCN and PHb results in the loss of their light-modulated behavioral output (Fernandez et al., 2018; Güler et al., 2008; Hatori et al., 2008). Therefore, it is not surprising that ipRGCs were the predominant afferents to the SCN and PHb and it is unlikely that the few conventional RGC fibers we observed play any substantial functional role in these regions. Similarly, while the function of the ZI visual response remains to be elucidated, it is unlikely that conventional RGCs play a substantial role in its modulation due to their minimal innervation. The SON has been implicated in a myriad of behaviors (sleep and anesthesia, inflammatory pain processing, and thirst) but its role in vision and dependence on light modulation is unknown (Eliava et al., 2016; Jiang-Xie et al., 2019; Zimmerman et al., 2017). The SON received moderate conventional RGC input, especially when compared to the few puncta we see in the SCN, PHb and ZI. Our findings suggest that conventional RGC influence on the SON should not be ruled out.

The Opn4Cre mouse labels all ipRGCs which are a heterogeneous population made up of at least 6 known subtypes. The Brn3b-negative M1 ipRGCs are known to exclusively innervate the SCN core (Chen et al., 2011). It is possible, or even likely, that retinal innervation specificity exists amongst other ipRGC and conventional RGC subtypes. As genetic mouse lines and tools to differentiate individual RGC subtypes become ever more available, it will be important to understand if innervations by all RGC subtypes segregate or overlap to contribute to visual function.

Taken together, we show that non-image forming retinorecipient brain regions are often not exclusive to ipRGC innervation. Our findings provide a basis for future experiments that will reveal how both ipRGCs and conventional RGC together modulate non-image forming behaviors for normal function.

METHODS

Animals

All mice were handled in accordance with guidelines of the Animal Care and Use Committees of the National Institute of Mental Health (NIMH). RC::FLTG mice (Jackson Laboratory, Stock # 026932, (Plummer et al., 2015)) were crossed with Opn4Cre mice (Ecker et al., 2010) to produce Opn4Cre/+;Rosa26CAG-tdTomato-eGFP/+ mice for experiments. Mice were group housed in a temperature and humidity-controlled room under a 12-hour light-dark cycle. Food and water were available ad libitum.

Viral intravitreal injections

Adult Opn4Cre/+;Rosa26CAG-tdTomato-eGFP/+ mice (8–10 weeks) were anesthetized with isoflurane. The virus (Addgene plasmid pAAV-hSyn-Flpo, # 60663, (Xue et al., 2014), packaged into AAV2 at the NINDS Viral Production Core Facility) was placed on a piece of Parafilm and loaded into a glass microcapillary (Fisher Scientific, Cat. No. 11714) pulled needle (Sutter Instruments, Model P-2000). The glass needle was used to puncture the sclera of the right eye and 1.5μL of virus was injected into the vitreous chamber (Drummond Science Nanoinject II, Cat. No. 3–000-204). After 410 the injection, mice were recovered from anesthesia and placed back into their home cages.

Immunohistochemistry

Mice were deeply anesthetized 4 weeks after intravitreal injection and perfused with phosphate-buffered saline (PBS) followed by 4% paraformaldehyde. Eyecups and brains were removed for post fixation in 4% paraformaldehyde (1 hour and overnight, respectively). Retinas were isolated and washed in 0.5% Triton X-100 in PBS for 15 minutes three times. Blocking solution, with 3% donkey serum (Vector Labs), 0.5% Triton X-100, in PBS was applied to retinas overnight with gentle agitation at 4° C. Retinas were washed and incubated with primary antibodies in 1% donkey serum, 0.5% Triton X-100, in PBS solution for 3 days at 4° C. Primary antibodies used were chicken anti-GFP (1:2000, Abcam, Cat. No. ab13970), goat anti-tdTomato (1:1000, LSBio, LS-C340696) with rabbit anti-melanopsin (1:1000, Advanced Target Systems, Cat. No. AB-N38) or rabbit anti-RBPMS (1:200, Abcam, Cat. No. ab194213). Retinas were washed and incubated with secondary antibodies overnight at 4° C. Secondary antibodies used were donkey anti-chicken 488nm (1:1000, Jackson ImmunoResearch, Cat. No. 703–545-155), donkey anti-goat 555nm (1:1000, Invitrogen, Cat. No. A-21432), and donkey anti-rabbit 647nm (1:1000, Invitrogen, Cat. No. A-31573). Retinas were washed and then mounted flat in Vectashield (Vector Labs, H-1000). Brains were sunk in 30% sucrose, frozen, and sliced into 40 μm think coronal sections on a cryostat (Leica). Brain sections were treated as retinas with the following exceptions. Sections were blocked in 3% goat serum for 2 hours at room temperature. Sections were incubated in primary antibodies overnight at 4° C. Primary antibodies used were chicken anti-GFP (1:2000, Abcam, Cat. No. ab13970) and rabbit anti-RFP (1:1000, MBL International, Cat. No. PM005). Sections were incubated in secondaries for 2 hours at room temperature. Secondary antibodies used were goat anti-chicken 488nm (Invitrogen, Cat. No. A-11039) and goat anti-rabbit 546nm (Invitrogen, Cat. No. A-11035). Sections were mounted in Flouromount-G with DAPI (ThermoFisher, Cat. No. 00–4959-52).

Imaging and analysis

Fluorescent images of retinas and brain sections were taken on a confocal microscope (Nikon, Eclipse Ti2). Images in the manuscript represent the maximal projections of a Z-stack (3μm step size) taken through an entire 40 μm thick coronal brain section. Any contrast or brightness adjustments were applied to the entire image in Fiji.

To calculate the percent of RGCs infected after an intravitreal injection, RBPMS-positive cells in two 400μm2 regions from n = 2 retinas were counted. The number of RBPMS- and GFP/RFP-positive cells were divided by the total number of RBPMS-positive cells.

To determine the percent ipsilateral projections, first, the maximal projection images were binarized using the Otsu method in Fiji. Regions of interest were created based on DAPI staining around the SCN, dLGN, IGL and vLGN. The total area of the contralateral and ipsilateral ipRGC or conventional RGC innervation to each region of interest was measured in the same coronal brain slice. For the geniculate measurements, coronal brain slices were grouped into rostral, medial, caudal groups (rostral distance from Bregma −2.00 to −2.18, medial distance from Bregma −2.30 to −2.46, caudal distance from Bregma −2.54 to −2.70, Bregma based on (Franklin and Paxinos, 2007), n = 4–6 brain slices per group). The percent ipsilateral innervation was calculated by ipsilateral area over total retinal innervation multiplied by 100 to give percentage.

To quantify the stratification, overlap between ipRGCs and conventional RGCs in the SC, dLGN and vLGN, first, the maximal projection images were binarized as before except that the Yen binary method was used for the SC images. The regions were then straightened by manually tracing the optic tract for the dLGN and vLGN or the dorsal edge of SC (See Figure 7a for an example). The intensity profile of the binarized image was taken along the lateral-medial axis for the dLGN and vLGN and along the dorsal-ventral axis for the SC. The amount of overlap (between the conventional RGCs and ipRGCs) is stated with respect to the smaller population for each nucleus (ipRGCs in the SC and dLGN and conventional retinal ganglion cells in the vLGN).

Acknowledgements

This project was supported by funding from the National Institute of Mental Health project number MH002964 (S.H.). We would like to thank Hui Wang in the Hattar lab for her management of the Opn4Cre mouse line. This work was supported, in part, by the NIMH IRP Rodent Behavioral Core (MH002952).

REFERENCES

  1. Altimus CM, Güler AD, Villa KL, McNeill DS, Legates TA, and Hattar S. (2008). Rods-cones and melanopsin detect light and dark to modulate sleep independent of image formation. Proc. Natl. Acad. Sci. U. S. A 105, 19998–20003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Baver SB, Pickard GE, Sollars PJ, and Pickard GE (2008). Two types of melanopsin retinal ganglion cell differentially innervate the hypothalamic suprachiasmatic nucleus and the olivary pretectal nucleus. Eur. J. Neurosci 27, 1763–1770. [DOI] [PubMed] [Google Scholar]
  3. Chen S-K, Badea TC, and Hattar S. (2011). Photoentrainment and pupillary light reflex are mediated by distinct populations of ipRGCs. Nature 476, 92–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chew KS, Renna JM, McNeill DS, Fernandez DC, Keenan WT, Thomsen MB, Ecker JL, Loevinsohn GS, VanDunk C, Vicarel DC, et al. (2017). A subset of ipRGCs regulates both maturation of the circadian clock and segregation of retinogeniculate projections in mice. Elife 6, e22861. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cunningham JT, Penny ML, and Murphy D. (2004). Cardiovascular regulation of supraoptic neurons in the rat: Synaptic inputs and cellular signals. Prog. Biophys. Mol. Biol 84, 183–196. [DOI] [PubMed] [Google Scholar]
  6. Davidowa H, and Albrecht D. (1992). Modulation of visually evoked responses in units of the ventral lateral geniculate nucleus of the rat by somatic stimuli. Behav. Brain Res 50, 127–133. [DOI] [PubMed] [Google Scholar]
  7. Delwig A, Larsen DD, Yasumura D, Yang CF, Shah NM, and Copenhagen DR (2016). Retinofugal projections from melanopsin-expressing retinal ganglion cells revealed by intraocular injections of Cre-dependent virus. PLoS One 11, 1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dhande OS, Seabrook1 TA, Phan AH, Salaly LD, Ishiko N, Nguyen PL, Wang JT, Evans SM, and Huberman AD (2018). Assembly of functionally antagonistic visual circuits for controlling pupil dynamics. BioRxiv. [Google Scholar]
  9. Ecker JL, Dumitrescu ON, Wong KY, Alam NM, Chen S-K, LeGates T, Renna JM, Prusky GT, Berson DM, and Hattar S. (2010). Melanopsin-Expressing Retinal Ganglion-Cell Photoreceptors: Cellular Diversity and Role in Pattern Vision. Neuron 67, 49–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Edelstein K, and Amir S. (1999). The role of the intergeniculate leaflet in entrainment of 300 circadian rhythms to a skeleton photoperiod. J. Neurosci 19, 372–380. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Eliava M, Melchior M, Knobloch-Bollmann HS, Wahis J, da Silva Gouveia M, Tang Y, Ciobanu AC, Triana del Rio R, Roth LC, Althammer F, et al. (2016). A New Population of Parvocellular Oxytocin Neurons Controlling Magnocellular Neuron Activity and Inflammatory Pain Processing. Neuron 89, 1291–1304. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Estevez ME, Fogerson PM, Ilardi MC, Borghuis BG, Chan E, Weng S, Auferkorte ON, Demb JB, and Berson DM (2012). Form and function of the M4 cell, an intrinsically photosensitive retinal ganglion cell type contributing to geniculocortical vision. J. Neurosci 32, 13608–13620. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Fernandez DC, Fogerson PM, Lazzerini Ospri L, Thomsen MB, Layne RM, Severin D, Zhan J, Singer JH, Kirkwood A, Zhao H, et al. (2018). Light Affects Mood and Learning through Distinct Retina-Brain Pathways. Cell 175, 71–84.e18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Fernandez DC, Komal R, Langel J, Ma J, Duy PQ, Penzo MA, Zhao H, and Hattar S. (2020). Retinal innervation tunes circuits that drive nonphotic entrainment to food. Nature 1–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Franklin K, and Paxinos G. (2007). The Mouse Brain in Stereotaxic Coordinates. [Google Scholar]
  16. Güler AD, Ecker JL, Lall GS, Haq S, Altimus CM, Liao H-W, Barnard AR, Cahill H, Badea TC, Zhao H, et al. (2008). Melanopsin cells are the principal conduits for rod–cone input to non-image-forming vision. Nature 453, 102–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Harrington ME (1997). The Ventral Lateral Geniculate Nucleus and the Intergeniculate Leaflet: Interrelated Structures in the Visual and Circadian Systems. Neurosci. Biobehav. Rev 21, 705–727. [DOI] [PubMed] [Google Scholar]
  18. Hatori M, Le H, Vollmers C, Keding SR, Tanaka N, Schmedt C, Jegla T, and Panda S. (2008). Inducible ablation of melanopsin-expressing retinal ganglion cells reveals their central role in non-image forming visual responses. PLoS One 3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hattar S, Liao H-W, Takao M, Berson DM, and Yau K-W (2002). Melanopsin-Containing Retinal Ganglion Cells: Architecture, Projections, and Intrinsic Photosensitivity. Science (80-. ). 295, 1065–1070. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hattar S, Kumar M, Park A, Tong P, Tung J, Yau K-W, and Berson DM (2006). Central Projections of Melanopsin- Expressing Retinal Ganglion Cells in the Mouse. J. Comp. Neurol 502, 275–290. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Huang L, Xi Y, Peng Y, Yang Y, Huang X, Fu Y, Tao Q, Xiao J, Yuan T, An K, et al. (2019). A Visual Circuit Related to Habenula Underlies the Antidepressive Effects of Light Therapy. Neuron 102, 1–15. [DOI] [PubMed] [Google Scholar]
  22. Jiang-Xie LF, Yin L, Zhao S, Prevosto V, Han BX, Dzirasa K, and Wang F. (2019). A Common Neuroendocrine Substrate for Diverse General Anesthetics and Sleep. Neuron 102, 1053–1065.e4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kerschensteiner D, and Guido W. (2017). Organization of the dorsal lateral geniculate nucleus in the mouse. Vis. Neurosci 34, E008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. LeGates TA, Altimus CM, Wang H, Lee H-K, Yang S, Zhao H, Kirkwood A, Weber ET, and Hattar S. (2013). Aberrant light directly impairs mood and learning through melanopsin-expressing neurons. Nature 491, 594–598. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Li JY, and Schmidt TM (2018). Divergent projection patterns of M1 ipRGC subtypes. J. Comp. Neurol 526, 2010–2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Livingston CA, and Mustari MJ (2000). The anatomical organization of the macaque pregeniculate complex. Brain Res. 876, 166–179. [DOI] [PubMed] [Google Scholar]
  27. Lupi D, Oster H, Thompson S, and Foster RG (2008). The acute light-induction of sleep is mediated by OPN4-based photoreception. Nat. Neurosci 11. [DOI] [PubMed] [Google Scholar]
  28. Monavarfeshani A, Sabbagh U, and Fox MA (2017). Not a one-trick pony: Diverse connectivity and functions of the rodent lateral geniculate complex. Vis. Neurosci 34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Moore RY, and Card JP (1994). Intergeniculate leaflet: An anatomically and functionally distinct subdivision of the lateral geniculate complex. J. Comp. Neurol 344, 403–430. [DOI] [PubMed] [Google Scholar]
  30. Morin LP (2013). Neuroanatomy of the extended circadian rhythm system. Exp. Neurol 243, 4–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Morin LP, and Studholme KM (2014). Retinofugal Projections in the Mouse. J. Comp. Neurol 522, 3733–3753. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Nagata T, and Hayashi Y. (1984). The visual field representation of the rat ventral lateral geniculate nucleus. J. Comp. Neurol 227, 582–588. [DOI] [PubMed] [Google Scholar]
  33. Plummer NW, Evsyukova IY, Robertson SD, de Marchena J, Tucker CJ, and Jensen P. (2015). Expanding the power of recombinase-based labeling to uncover cellular diversity. Dev. 142, 4385–4393. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Quattrochi LE, Stabio ME, Kim I, Ilardi MC, Fogerson PM, Leyrer ML, and Berson DM (2018). The M6 cell: A small-field bistratified photosensitive retinal ganglion cell. J. Comp. Neurol 1–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Rivlin-Etzion M, Zhou K, Wei W, Elstrott J, Nguyen PL, Barres BA, Huberman AD, and Feller MB (2011). Transgenic mice reveal unexpected diversity of on-off direction-selective retinal ganglion cell subtypes and brain structures involved in motion processing. J. Neurosci 31, 8760–8769. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Rupp AC, Ren M, Altimus CM, Fernandez DC, Richardson M, Turek F, Hattar S, and Schmidt T. (2019). Distinct ipRGC subpopulations mediate light’s acute and circadian 370 effects on body temperature and sleep. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Sabbagh U, Govindaiah G, Somaiya RD, Ha RV, Wei JC, Guido W, and Fox MA (2020). Diverse GABAergic neurons organize into subtype-specific sublaminae in the ventral lateral geniculate nucleus. J. Neurochem 1–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Schmidt TM, Alam NM, Chen S, Kofuji P, Li W, Prusky GT, and Hattar S. (2014). A Role for Melanopsin in Alpha Retinal Ganglion Cells and Contrast Detection. Neuron 82, 781–788. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Shi HY, Xu W, Guo H, Dong H, Qu WM, and Huang ZL (2020). Lesion of intergeniculate leaflet GABAergic neurons attenuates sleep in mice exposed to light. Sleep 43, 1–12. [DOI] [PubMed] [Google Scholar]
  40. Stabio ME, Sabbah S, Quattrochi LE, Ilardi MC, Fogerson PM, Leyrer ML, Renna JM, Kim MT, Kim I, Schiel M, et al. (2017). The M5 Cell: A Color-Opponent Intrinsically Photosensitive Retinal Ganglion Cell. Neuron 97, 150–163.e4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Wang X, X lin Chou, Zhang LI, and Tao HW (2020). Zona Incerta: An Integrative Node for Global Behavioral Modulation. Trends Neurosci. 43, 82–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Xue M, Atallah BV, and Scanziani M. (2014). Equalizing excitation-inhibition ratios across visual cortical neurons. Nature 511, 596–600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Z dong Zhao, Chen Z, Xiang X, Hu M, Xie H, Jia X, Cai F, Cui Y, Chen Z, Qian L, et al. (2019). Zona incerta GABAergic neurons integrate prey-related sensory signals and induce an appetitive drive to promote hunting. Nat. Neurosci 22, 921–932. [DOI] [PubMed] [Google Scholar]
  44. Zimmerman CA, Leib DE, and Knight ZA (2017). Neural circuits underlying thirst and fluid homeostasis. Nat. Rev. Neurosci 18, 459–469. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES