Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Apr 1.
Published in final edited form as: J Comp Neurol. 2020 Aug 31;529(5):1066–1080. doi: 10.1002/cne.25004

The SNARE regulator Complexin3 is a target of the cone circadian clock

Jacob D Bhoi 1,2,#, Zhijing Zhang 3,#, Roger Janz 4,5,6,7, Yanan You 8,9, Haichao Wei 8,9, Jiaqian Wu 5,6,7,8,9, Christophe P Ribelayga 2,3,5,6,7,10,11
PMCID: PMC8190822  NIHMSID: NIHMS1618931  PMID: 32783205

Abstract

BMAL1 is a core component of the mammalian circadian clockwork. Removal of BMAL1 from the retina significantly affects visual information processing in both rod and cone pathways. To identify potential pathways and/or molecules through which BMAL1 alters signal transmission at the cone pedicle, we performed an RNA-seq differential expression analysis between cone-specific Bmal1 knockout cones (cone-Bmal1−/−) and wild-type (WT) cones. We found 88 genes differentially expressed. Among these, Complexin3 (Cplx3), a SNARE regulator at ribbon synapses, was downregulated fivefold in the mutant cones. The purpose of this work was to determine whether BMAL1 and/or the cone clock controls CPLX3 protein expression at cone pedicles. We found that CPLX3 expression level was decreased twofold in cone-Bmal1−/− cones. Furthermore, CPLX3 expression was downregulated at night compared to the day in WT cones but remained constitutively low in mutant cones both day and night. The transcript and protein expression levels of Cplx4—the other complexin expressed in cones—were similar in WT and mutant cones; in WT cones, CPLX4 protein level did not change with the time of day. In silico analysis revealed four potential BMAL1:CLOCK binding sites upstream from exon one of Cplx3 and none upstream of exon one of Cplx4. Our results suggest that CPLX3 expression is regulated at the transcriptional level by the cone clock. The modulation of CPLX3 may be a mechanism by which the clock controls the cone synaptic transfer function to second-order cells and thereby impacts retinal signal processing during the day/night cycle.

Keywords: Bmal1, circadian clock, cones, Cplx3, Cplx4, retina, ribbon synapses, SNARE proteins

1 ∣. INTRODUCTION

Most living organisms have daily rhythms of behavior and physiology, a result of the interplay between the light–dark cycle and endogenous circadian clocks, whose oscillations persist under constant environmental conditions, such as constant darkness (Dunlap, 1999). In the retina, circadian clocks are present in most cell types and together regulate rod and cone function with the time of day (Besharse & McMahon, 2016; McMahon, Iuvone, & Tosini, 2014). BMAL1 is an essential, nonredundant component of the mammalian circadian clockwork that partners with CLOCK to modulate large fractions of the transcriptome (Buhr & Takahashi, 2013; Takahashi, 2017). Retina-specific removal of BMAL1 abolishes circadian variations in gene expression and of most rhythms in retinal physiology, and compromises retinal cell viability (Baba, Piano, et al., 2018; Baba, Ribelayga, Michael Iuvone, & Tosini, 2018; Storch et al., 2007). However, despite the clear demonstration of the fundamental role of retinal clocks in the integrity and performance of the retinal tissue, much of the clock pathways that link retinal cell clocks, gene expression, and maintenance of retinal cell health and function remain to be identified.

Cone synapses are arguably some of the more complex synapses in the brain. Remarkably, cone synapses are ribbon synapses, a rare type characterized by the presence of a presynaptic plate-like organelle called a ribbon. Ribbons are anchored to the plasma membrane in close proximity of voltage-gated calcium channels. They have been proposed to act as conveyer belts to guide glutamate containing synaptic vesicles toward the active zone—a key process required to maintain glutamate release at a high and constant rate in the dark (DeVries & Baylor, 1993; Dick et al., 2003; Matthews & Fuchs, 2010; Parsons & Sterling, 2003). Many of the proteins forming conventional synapses and ribbon synapses are conserved, but there are a few critical differences. Notably, ribbon synapses contain specific members of the complexin family, a SNARE regulating protein family (Trimbuch & Rosenmund, 2016). The complexin proteins are thought to act by stabilizing the SNARE complex to maintain the vesicles in a highly releasable state (Chang et al., 2015; Reim et al., 2001; Trimbuch & Rosenmund, 2016). Four different types of complexin have been identified in mammals. Complexin1 (CPLX1) and CPLX2 are mainly present at conventional synapses whereas ribbon synapses primarily use CPLX3 and/or CPLX4 (Trimbuch & Rosenmund, 2016). In the retina, CPLX3 and CPLX4 have been found in both plexiform layers, at photoreceptor presynaptic terminals, and at presynaptic terminals of amacrine cells and rod bipolar cells (Reim et al., 2005). Mice knocked out for Cplx3 and Cplx4 show structural and functional aberrations of the retina accompanied of visual deficits (Babai et al., 2016; Landgraf et al., 2012; Mortensen et al., 2016; Reim et al., 2009). Thus, CPLX3 and CPLX4 play important functional roles at retinal ribbon synapses.

Previous studies demonstrated the importance of BMAL1 in cone function and maintenance (Baba, Piano, et al., 2018; Baba, Ribelayga, et al., 2018; Sawant et al., 2017). In the present study, we used an RNA-seq differential expression analysis to investigate the importance of BMAL1 in the cone transcriptome with the hope to identify genes involved circadian oscillation of cone function. We found that the removal of BMAL1 in cones is associated with a decrease in Cplx3 mRNA expression whereas Cplx4 expression remains unchanged. Decrease in Cplx3 transcript in mutant cones translates into a decrease in CPLX3 protein at cone pedicles. We also show that CPLX3 is downregulated at night compared to the day in wild-type (WT) cone pedicles. We posit that CPLX3 is expressed rhythmically in cones and is regulated by the cone circadian clock at the transcriptional level. CPLX3 could be an effector of the cone clock to modulate signaling to second-order neurons and thereby downstream circuit function.

2 ∣. METHODS

2.1 ∣. Animals

Male and female adult mice were used in the experiments. Animals were raised with a 12:12 hr light–dark cycle (lights on at 7:00) in the UT Health Science Center at Houston Center for Laboratory Animal Medicine and Care. Circadian conditions were created by keeping the mice in the dark for up to 36 hr, with dark adaptation starting at the end of the light phase (7.00 p.m.). We refer to the subjective day as the period between Circadian Time (CT) 0 and CT 12, between 12 and 24 hr after the beginning of dark adaptation, and the subjective night as the period between CT 12 and CT 24, between 24 and 36 hr after the beginning of dark adaptation. We typically collected retinal tissue in the middle of the subjective day (CT 05-07) or in the middle of the subjective night (CT 17–19). Under circadian conditions, deep dark-adapted conditions were preserved until after the fixation step. When indicated, some retinas were collected during the daytime under room lights that were ON since the beginning of the day (7.00 a.m.) or for 1 hr before experiment, at Zeitgeber Time (ZT) 05-07. All procedures on animals were in accordance with federal, local, and institutional guidelines, and reviewed and approved by the Institutional Animal Care and Use Committee of the University of Texas Health Science Center Houston.

Cone-Bmal1−/− and rod-Bmal1−/− mice were generated by crossing human red/green photopigment (HRGP)Cre (Le et al., 2004) or rhodopsin (Rho)i75Cre (Li et al., 2005) mice in which the Z/EG transgene (Novak, Guo, Yang, Nagy, & Lobe, 2000) was introduced, with Bmal1flox mice, causing cone or rod-specific Bmal1 knockouts. We used littermate mice that either lacked cre expression or had a single Bmal1 floxed allele as control mice. Specifically, we used HRGPcre;Bmal1f/+ or HRGP0;Bmal1f/f as controls for the cone-Bmal1−/− line, and Rhoi75cre;Bmal1f/+ or Rho0;Bmal1f/f as controls for the rod-Bmal1−/− line. All mice were anesthetized with a mixture of ketamine and xylazine 100/10 mg/kg (IM), and euthanized using cervical dislocation.

2.2 ∣. Fluorescent-assisted cell sorting and RNA-seq library preparation

Cone-Bmal1−/− and rod-Bmal1−/− mutant mice or WT control littermates were euthanized at one timepoint in the middle of the day in regular light–dark cycle, defined as ZT 05-07. These animals express the CRE reporter eGFP in cones and rods, respectively. Neural retinas were isolated and digested with 0.25% trypsin for up to 15 min at 37°C. Following digestion, they were dissociated to a single cell suspension in DNase I buffered with Ames' medium. Fluorescent assisted cell sorting was performed with an Aria II cell sorter (BD Biosciences), based on fluorescence and forward scatter. Collected cells were lysed and stored in a Nucleospin RNA XS kit (Macherey-Nagel).

Cells were sorted into TRIzol Reagent (Invitrogen) followed by RNA extraction according to manufacturer's protocol. Subsequently, RNA quantity and quality were determined by Nanodrop 1000 (Thermo Fisher Scientific). Total of 100 ng RNA per sample was used for RNA-seq library preparation, constructed using NEBNext Ultra Directional RNA Library Prep Kit for Ilumina with NEBNext Poly(A) mRNA Magnetic Isolation Module (New England Biolabs) according to manufacturer's protocol. The libraries were quantified using Agilent 2100 Bioanalyzer (Agilent) and Quibit quantification. RNA-seq libraries were sequenced using NextSeq550 PE150 (Illumina).

2.3 ∣. RNA-seq and data analysis

We combined GENCODE M14 annotation file with NCBI (GCF_000001635.25_GRCm38.p5, only gene labeled as “lncRNA”) (the official NCBI ftp repository: ftp://ftp.ncbi.nih.gov/genomes/refseq/vertebrate_mammalian/Mus_musculus/all_assembly_versions/GCF_000001635.25_GRCm38.p5/) as the final annotation file for mouse genome. Read mapping, transcript assembly, and expression estimation were performed as described in previous publication (Duran et al., 2017; Zhang et al., 2014). The 150-bp paired-end reads were aligned to the reference genome (mm10) using TopHat v2.1.0 with default parameters (Trapnell, Pachter, & Salzberg, 2009). Cufflinks v2.2.1 was used to analyze FPKM (Fragments Per Kilobase of transcript per Million mapped reads) values for genes and transcripts annotated (Trapnell et al., 2012). Any FPKM <0.1 was set to 0.1 to avoid ratio inflation (Quackenbush, 2002). Read counts for annotated genes and transcripts were acquired using HTSeq-count (Anders, Pyl, & Huber, 2015). DESeq2 software was used to analyze the FPKM to identify differentially expressed genes (DEGs) (Love, Huber, & Anders, 2014). The following thresholds were used to classify genes as DEGS: (a) at least one sample's FPKM >1, (b) ∣log2 (fold-change)∣ > 1, and (c) false discovery rate (FDR) < 0.05.

To conduct the pathway analysis, DEGs were uploaded to the MSigDB website to compute overlaps between DEGs and known gene sets (Liberzon et al., 2011; Subramanian et al., 2005). FDR (q-value) shows the FDR analog of hypergeometric p-value after correction for multiple hypothesis testing according to Benjamini and Hochberg method. Gene sets with an FDR < 5% were considered significant and reported with the corresponding biological processes.

2.4 ∣. Immunocytochemistry and imaging

Mice were euthanized and eyes were removed and rapidly fixed for 45 min in phosphate buffer solution (PBS) containing 4% paraformaldehyde at room temperature. Eyeballs were rinsed in PBS and then placed in 10% and then 20% sucrose solution for 2 hr and then 30% sucrose solution overnight. Eyeballs were embedded in OCT compound, solidified with liquid nitrogen, and vertically sectioned (20-μm-thick sections) using a cryostat. The retina was blocked with PBS containing 2.5% Normal Donkey Serum and 0.2% Triton X-100 for 2 hr at room temperature.

Sections were incubated with the following primary antibodies for 48 hr: anti-CPLX3 polyclonal rabbit antibody (1:1,000, made by Dr Nils Brose, Max Plank Institute, Göttingen, Germany); anti-CPLX4 polyclonal rabbit antibody (1:1000, made by Dr Nils Brose, Max Plank Institute, Göttingen, Germany); anti-PSD95 monoclonal mouse antibody (1:500, MAB1598, EMD Millipore, Billerica, MD); and anti-cone-arrestin (cArr) polyclonal rabbit antibody (1:500, AB15282, EMD Millipore). The samples were rinsed in PBS and treated with a secondary antibody conjugated with Dylight549 (1:600, Donkey anti-mouse antibody, 715-505-150, Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) or Dylight488 (1:600, Donkey anti-rabbit antibody, 711-485-152, Jackson ImmunoResearch) for 4 hr. Sections were mounted in mounting medium containing DAPI.

Images were taken at ×40 or ×63 magnification using Zeiss LSM 800 with Airyscan operated with Zeiss's Zen software (Carl Zeiss Microscopy, Thornwood, NY). For each experiment, the microscope settings were optimized for the most intensely stained sample and then remained the same for all of the trials in order to standardize between samples. For each animal, a randomly chosen part of the retina was selected and a z-stack of images of the section was taken from one side of the section to the other. The middle image of the stack was used for analysis to control for any differences in staining based on the depth in the section.

2.5 ∣. Immunocytochemistry data analysis

The confocal images were analyzed using the NIH ImageJ software. For all quantification, the background CPLX3/4 immuno-signal was recorded by taking the mean CPLX3/4 immuno-signal of the entire ONL, where there is minimal CPLX3/4 protein expression. This background was then subtracted from the mean CPLX3/4 immuno-signal to get the adjusted mean immuno-signal.

To quantify CPLX3 expression at the cone pedicles, all of the cone pedicles in the central image of the z-stack were selected using the freehand selection function in ImageJ (5–16 cones pedicles per animal). Cone terminals were discriminated from rod terminals as previously described (Li et al., 2013), and cone terminals were excluded if the cross-sectional area was less than 8 μm2. The mean CPLX3-immuno-signal of each cone pedicle recorded and the ONL background was subtracted. To control for any differences between immunocytochemistry (ICC) staining and imaging, sections from WT and mutant retinas were present on each slide, and the adjusted mean CPLX3 immuno-signal range was normalized between 0 and 1 for each slide. This was done by subtracting the minimum adjusted mean CPLX3-immuno-signal value representing the weakest cone pedicle and then dividing by the maximum adjusted mean CPLX3-immuno-signal representing the brightest cone pedicle. This controls for any differences in staining or imaging between slides and trials. The median cone pedicle immuno-signal was calculated for each animal. Median values from different animals were averaged and used to compare levels of expression between conditions or genotypes.

To quantify CPLX4 in OPL, the OPL was selected using the freehand selection tool based on the DAPI stain, and then the mean intensity of CPLX4 throughout the whole OPL was recorded. The ONL background was subtracted from the OPL mean, and then the mean adjusted immuno-signal for each slide was normalized between 0 and 1, as described in the CPLX3 cone quantification. The normalized OPL intensities from different animals were averaged and used to compare levels of expression between conditions or genotypes.

To conduct cluster analysis for CPLX3/4 in rods and cones, the area and CPLX3/4 intensity of rods and cones were measured in ImageJ based on PSD95 staining, and the mean ONL background was subtracted from the intensity. Then each point was clustered using k-means clustering algorithm (Hartigan & Wong, 1979) with two centers, for the two cell types, in R (R-project, v. 3.6.0.). The mean CPLX3/4 intensity for each cluster was calculated and reported.

All image analysis was conducted in unprocessed images. This data were analyzed using two-tailed unpaired Student's t tests in Excel and R.

2.6 ∣. RNAscope in situ hybridization

Mice were anesthetized with a mixture of ketamine and xylazine, euthanized using cervical dislocation, and eyes were removed and rapidly fixed overnight in 10% neutral buffered formalin. Eyeballs were rinsed with PBS and dehydrated using serial ethanol. The eyeballs were placed in xylene and then embedded in paraffin. Then, 9-μm vertical sections of retina were made, and RNAscope in situ hybridization (ISH) (Wang et al., 2012) was conducted using the RNAscope Fluorescent Multiplex kit (320850, Advanced Cell Diagnostics, Newark, CA) according to manufacturer protocol, with slight modifications as described by Kiyama and Mao (2020). Following RNAscope ISH, sections were immunostained with anti-cArr as described above, resulting in Cplx3 mRNA labeling using RNAscope ISH and immunolabeling of cones. Images were taken using the Zeiss LSM 800 confocal microscope at ×40 and ×63 magnification. Each fluorescent dot is one hybridized mRNA molecule. The probe used was the mouse Cplx3 (Probe-Mm-Cplx3, 467821, Advanced Cell Diagnostics).

2.7 ∣. Expression analysis

Cell-type specific expression of individual genes in rod bipolar cells, rod photoreceptors, and cone photoreceptors where analyzed using the data set from a recent massive single cell RNA sequencing study of 25,000 individual mouse retinal cells (Shekhar et al., 2016). Expression profiles for 7,784 rod bipolar cells, 32 rods, and 14 cones were extracted from the master expression matrix using the cluster analysis matrix to assign the cell types (https://singlecell.broadinstitute.org/single_cell/study/SCP3/retinal-bipolar-neuron-drop-seq#study-summary). The transcripts per million were calculated for each cell sequenced and averaged.

2.8 ∣. Promotor region analysis

The sequences for mouse Cplx3 and Cplx4 were downloaded from the Ensembl Project database, release 98 (Cunningham et al., 2019). The 600 base pairs upstream of Exon 1 were analyzed. The transcription factor binding sites and the corresponding sequences probed for are described in Table 1.

TABLE 1.

Circadian associated DNA binding factors and corresponding binding sequences

Binding site Description DNA binding sequence
E-box E-boxes are CLOCK:BMAL1 binding sites central to molecular circadian clock function (Buhr & Takahashi, 2013; Lowrey & Takahashi, 2011). 5′-CACGTG-3′
5′-CACGTT-3′
(Lowrey & Takahashi, 2011)
D-box D-boxes are binding sites for DBP, a transcription which oscillates with the circadian clock (Ueda et al., 2005) 5′-TTAYGTAA-3′
(Ueda et al., 2005)
RORE element RORE elements are Rev-erbα binding sites (Harding & Lazar, 1993) 5′-RGGTCA-3′
(Harding & Lazar, 1993).
CRE element Cyclic AMP response elements are binding sites for cAMP, which is commonly associated with the circadian clock (Carlezon, Duman, & Nestler, 2005) 5′-TGACGTCA-3′
(Carlezon et al., 2005)
Open in a new tab

3 ∣. RESULTS

3.1 ∣. Identification of target genes, biological processes, and signaling pathways in cones from RNA-seq differential expression analysis

To identify potential pathways/molecules in the cones through which BMAL1 may alter signal transfer at the cone pedicle, we performed an RNA-seq followed by differential expression analysis in cones in mice that lack Bmal1 specifically in cones (cone-Bmal1−/−, n = 3 mice, 3 × 2 retinas) and in WT mice (n = 3 mice, 3 × 2 retinas). Cone-Bmal1−/− mice were generated by crossing HRGPCre (Le et al., 2004) mice with Bmal1flox mice, causing cone-specific removal of BMAL1 (Figure 1a,b). Comparison of the cone transcriptome between cone-Bmal1−/− and WT littermates at a single time point (middle of the light phase, noon) revealed 88 genes differentially expressed (p < .05, fold change >2) (Figure 1c). As expected, these 88 genes included known clock components but also genes known to be involved in a wide range of functions: gene regulation, neuron development and structure, protein binding, and transport (Figure 1d). These data are consistent with the view that the cone clock controls many aspects of the development, maintenance, and function of the cones through its control of the transcriptome (Baba, Piano, et al., 2018; Baba, Ribelayga, et al., 2018; Sawant et al., 2017, 2019). Among the genes that were differentially expressed, we found only one gene associated with the core synaptic machinery: Cplx3. This result prompted us to further evaluate the expression of complexins in cones and rods.

FIGURE 1.

FIGURE 1

Open in a new tab

RNA-seq and analysis in cone-Bmal1−/− cones. (a) Wild-type retina section reacted with an antibody against BMAL1. Note the presence of BMAL1 expression in the inner retina (INL and GCL) and cone somas (white arrows). (b) Cone-Bmal1−/− retina section reacted with an antibody against BMAL1. Note the normal expression of BMAL1 in the inner retina and the absence of BMAL1 immunoreactivity in the cones (white arrows). Cones were labeled with an antibody against cone-arrestin [cARR]). GCL, ganglion cell layer; INL, inner nuclear layer; IPL, inner plexiform layer; ONL, outer nuclear layer; OPL, outer plexiform layer. DAPI (cyan) stains the cell nuclei. Bar is 50 μm (applies to all). (c) Heat map showing the expression pattern of 88 protein-coding gene transcripts between wild-type mice (C1, C2, C3) and cone-Bmal1−/− mutant mice (M1, M2, M3) and within triplicates. Red indicates upregulation and blue downregulation. Tissue was collected in the middle of the light phase (ZT 05-07). One triplicate = 3 × 1 animals (two retinas/animal). (d) Biological processes overrepresented in transcripts upregulated or downregulated in cones from cone-Bmal1−/− retinas (x fold increase in mutant cones)

3.2 ∣. Complexin gene expression in photoreceptors

Previous studies reported CPLX3 and CPLX4 protein expression in both plexiform layers in mouse retina, and CPLX1 and CPLX2 expression restricted to the inner plexiform layer (Landgraf et al., 2012; Reim et al., 2005, 2009). Double-ICC further suggested coexpression of CPLX3 and CPLX4 at cone pedicles and expression of CPLX4 in rod spherules (Reim et al., 2009). We estimated the level of expression of the four complexin genes using our RNA-seq and analysis data in WT cones and WT rods. We found very low levels of Cplx1 and Cplx2 mRNA in cones (Figure 2a) or in rods (Figure 2b). Furthermore, we found that both Cplx3 and Cplx4 genes are transcribed in cones, with a much larger relative Cplx4 mRNA abundance (Figure 2a,b). In the rods, Cplx3 mRNA levels were found to be about half those in cones while Cplx4 transcript levels were about twice those in cones. We were able to confirm the relatively higher level of Cplx3 transcript in the cones using RNAscope ISH (Figure 2c). Thus, our data are in agreement with previous IHC studies (Landgraf et al., 2012; Reim et al., 2005, 2009): Cplx1 and Cplx2 expression in cones and rods is very low; cones and rods express Cplx3 and Cplx4. We also examined the expression pattern of the different complexin isoforms in retinal neurons by analyzing the data set from a recent massive single-cell RNA-seq study of 25,000 individual mouse retinal cells (Shekhar et al., 2016). This analysis confirmed that Cplx3 is highly expressed in cones and rod bipolar cells whereas Cplx4 is expressed in both rods and cones but not in rod bipolar cells (Figure 2e).

FIGURE 2.

FIGURE 2

Open in a new tab

Cplx mRNA levels in wild-type (WT) and cone and rod specific BMAL1 knockouts. (a) Cplx3 is the only member of the Cplx family which is differentially expressed in WT and cone-Bmal1−/− cones. Cones also show high Cplx4 expression in both WT and cone-Bmal1−/− cones. (b) Cplx3 is the only member of the Cplx family which is differentially expressed in WT and rod-Bmal1−/−. A small difference was found between WT and mutant rods for Cplx2 but the levels were overall very low. Rods show a very high expression of Cplx4, consistent with previous research indicating that it is the primary Cplx found in rods. Error bars are SEM. Dots show individual data points. n = 1 animal (two retinas)/point, three animals/group. Tissue was collected in the middle of the light phase (ZT05-07). p-Values: unpaired t-test, two-tailed. Note the difference in y-axis scale for Cplx4 in rods and cones. (c) RNAscope in situ hybridization (ISH) followed by immunocytochemistry (ICC) revealed that Cplx3 mRNA (green dots) is found primarily in the outer layer of the outer nuclear layer (ONL), where the somas of cones primarily stratify, in WT retinas. Cplx3 mRNA is primarily surrounding cone somas (white), designated by white arrows. In cone-Bmal1−/− retinas, there is less Cplx3 mRNA in the outer ONL. DAPI (cyan) stains the cell nuclei. Scale bar: 25 μm (applies to all). (d) Vertical intensity plot of Cplx3 mRNA in WT (green) and cone-Bmal1−/− (purple). Cplx3 mRNA was higher in the upper ONL where the somas of cones stratify. (e) Expression of Cplx mRNA in cone photoreceptors, rod bipolar cells, and rod photoreceptors based on published single-cell RNA-seq data (Shekhar et al., 2016). Cplx1 and Cplx2 are not highly expressed in any of these cell types. Cplx3 mRNA is primarily in cones and rod bipolar cells. Cplx4 is primarily expressed in rods and cones. Cones (n = 14); rod bipolar cells (n = 7,784); and rods (n = 32). Error bars show SEM

3.3 ∣. Cplx3 transcript is downregulated in cones of cone-Bmal1−/− mice

We analyzed our RNA-seq data to compare mRNA transcript level of the complexin genes in cone-Bmal1−/− animals and in WT animals (littermates) (Figure 2a). In addition, we used a similar approach in animals that lack BMAL1 specifically in the rods (rod-Bmal1−/−) (Figure 2b). We found similarly low levels of Cplx1 and Cplx2 transcripts in mutant cones and mutant rods, as compared to their respective WT controls. Notably, Cplx3 was downregulated ~5-fold in cone-Bmal1−/− cones and approximately threefold in rod-Bmal1−/− rods. Cplx4 expression levels were similar in wild type and mutant animals in both cones and rods. The data indicate that removal of BMAL1 specifically affects the levels of Cplx3 in both cones and rods.

Next, we assessed the expression of CPLX3 and CPLX4 by ICC. As expected (Landgraf et al., 2012; Reim et al., 2005, 2009), expression of CPLX3 was observed in both plexiform layers (Figure 3a) at the middle of the subjective day (CT 05-07). We used an antibody against cARR to label the cones including their pedicles. In the OPL, CPLX3 labeling almost exclusively colocalized with cARR at cone pedicles (Figure 3b,c). In contrast, CPLX3 signal was weak or absent, in the direct vicinity of cone pedicles were rod spherules are located.

FIGURE 3.

FIGURE 3

Open in a new tab

CPLX3 is expressed in cone pedicles. (a) CPLX3 labeling (green) is found in both plexiform layers of mouse retina. Note intensity plot of CPLX3 and DAPI stain vertically through retinal section to the right. DAPI (cyan) stains the cell nuclei. Scale bar: 25 μm. (b) Labeling cones with cArr (red) reveals strong colocalization of CPLX3 at cone pedicles (asterisks). Also note the CPLX3 hotspots at the axon initial segment of some cones (arrowheads). DAPI (cyan) stains the cell nuclei. Scale bar: 25 μm. (c) Detail of (b). CPLX3 (green) fills the cone pedicles. (d) PSD95 (white) labels rod and cone terminals which can be distinguished based on their size. CPLX3 (green) strongly labels cone pedicles (asterisks) and there is slight CPLX3 in rod spherules (arrows). All images from retinas collected in the middle of the day subjective day (CT 05-07). Scale bar: 10 μm. (e) Rods (n = 94 rods from three animals) and cones (n = 34 cones from three animals) form two clusters based on area and CPLX3 intensity (Dunn index = 0.007) (f) mean CPLX3 intensity for rods (n = 94 rods from three animals) and cones (n = 34 cones from three animals). p-Value: unpaired t-test, two-tailed

Using an antibody against postsynaptic density protein 95 (PSD95), which labels both the rod spherules and cone pedicles, we discriminated between rod and cone terminals based on cross-sectional area, with rod spherules being less than 6 μm2 and cone pedicles less than 8 μm2 (Figure 3d). Using a k-means clustering algorithm (Hartigan & Wong, 1979), rods and cones could be distinguished with 99.2% accuracy based on CPLX3 intensity and size with a Dunn index of 0.007 (Figure 3e). We calculated the ratio of CPLX3 fluorescence intensity cones/rods to be 4.01 (Figure 3f,n = 10–12 cones and 30–32 rods from three animals) in the middle of the subjective day in WT animals. Thus, even though the RNA-seq data established a ratio of Cplx3 RNA expression ~2 between cones and rods, the difference in CPLX3 expression appears much greater and results from a low level of protein expression in the rods, suggesting posttranscriptional regulation is different in rods and cones.

We found CPLX4 in both plexiform layers, in agreement with previous reports (Landgraf et al., 2012; Reim et al., 2005, 2009) (Figure 4a). Using an antibody against PSD95 to label rod and cone terminals, we observed strong CPLX4 expression in both terminals (Figure 4b). We discriminated the two types of terminal based on their cross-sectional area as previously described. We found that the rod to cone ratio of CPLX4 intensity was 2.65 (Figure 4d, n = 45 cones and 144 rods from 5 animals) in the middle of the subjective day (CT 05-07). Using a k-means clustering algorithm (Hartigan & Wong, 1979), rods and cones could be distinguished with 100% accuracy based on CPLX4 intensity and size with a Dunn index of 0.18 (Figure 4c). The difference in CPLX4 expression between rods and cones compares well with the difference in Cplx4 transcript expression (~threefold) between the two cell types reported by the RNA-seq data (Figure 2a,b).

FIGURE 4.

FIGURE 4

Open in a new tab

CPLX4 is expressed in both rods and cones. (a) CPLX4 (red) is expressed in the outer plexiform layer (OPL) and the inner plexiform layer (IPL). Note the intensity plot of CPLX4 and DAPI stain vertically through retinal section to the right. DAPI (cyan) stains the cell nuclei. Scale bar: 25 μm. (b) An antibody against PSD95 labels rod and cone terminals (white), which can be differentiated based on their relative sizes (i.e., big ones: cone pedicles and small one: rod spherules). CPLX4 is expressed in both synaptic terminals but to a lower degree in cones (asterisks). White arrows indicate rod spherules. All images from retinas collected in the middle of the subjective day (CT 05-07). Scale bar: 5 μm. (c) Rods (n = 144 rods from five animals) and cones (n = 45 cones from five animals) form two clusters based on area and CPLX4 intensity (Dunn index = 0.18). (e) Mean CPLX4 intensity for rods (n = 144 rods from five animals) and cones (n = 45 cones from five animals). p-Values: unpaired t-test, two-tailed

Altogether, the data suggest that cones express both CPLX3 and CPLX4 whereas rods primarily express CPLX4. The reason for low level of expression of CPLX3 in rod spherules remains unclear. We did not pursue the study of the modulation of CPLX3 in the rods due to its low level of expression.

3.4 ∣. CPLX3 expression in cones is decreased during the day in cone-Bmal1−/− mice and at night in WT mice

Quantification of CPLX3 immunofluorescence in the middle of the day (ZT/CT 05-07) in the cone pedicles in WT (n = 5–18 cones per animal from 14 animals) and mutant animals (n = 5–18 cones per animal from six animals) revealed a ~ 2-fold decrease in the mutant cones (Figures 5a,b,d,e and 6a). This decrease is consistent with the decrease in Cplx3 mRNA (Figure 2) and suggests that CPLX3 might be regulated by the cone clock and/or the lighting conditions. We therefore tested whether CPLX3 expression levels were different between the middle of the subjective day (CT 05-07) and the middle of the subjective night (CT 17–19). We found that under circadian conditions, CPLX3 immuno-signal in the cone pedicles was stronger during the day (n = 5–18 cones per animal from nine animals) compared to the night (n = 5–18 cones per animal from five animals) in WT mice (Figures 5a,c and 6b). Furthermore, we found no significant difference between subjective day (n = 5–18 cones per animal from three animals) and subjective night (n = 5–18 cones per animal from two animals) in cone-Bmal1−/− mice (Figure 6b), an observation consistent with expectations as the circadian clock is nonfunctional in cone-Bmal1−/− cones. Notably, the night levels of CPLX3 in WT mice were about the same as the cone-Bmal1−/− mice, as the Bmal1-deficient clock lacks the positive loop and the retina is therefore expected to be stuck in a nighttime-like state (Storch et al., 2007). Additionally, a 1-hr light adaptation in the middle of the day had no effect on the daytime levels of CPLX3 (Figure 6a). Finally, we found no difference in the expression of CPLX4 in the outer plexiform layer due to Bmal1 knock out, time of day, or lighting (Figure 6c,d). When analyzing CPLX4 expression in solely cone pedicles, there was also no difference in expression due to Bmal1 knock out (n = 3 WT and 3 cone-Bmal1−/− animals, p = .52) or time of day (n = 3 subjective day and three subjective night WT animals, p = .68) (two-tailed unpaired Student's t test). Altogether the data indicate that CPLX3 is modulated by the cone clock whereas CPLX4 is constitutively expressed at photoreceptor synapses in both rods and cones.

FIGURE 5.

FIGURE 5

Open in a new tab

CPLX3 expression is controlled by the cone clock. (a–c) Wild-type (WT) retinal section collected during the subjective day (a). Cone-Bmal1−/− retinal section collected during the subjective day (b). Wild-type retinal section collected during the subjective night (c). Note that CPLX3 immunolabeling is dramatically decreased in the outer plexiform layer (OPL) in cone-Bmal1−/− retina but is normal in the IPL, compared to the wild-type. In wild-type, CPLX3 is decrease in both OPL and IPL during subjective night, compared to subjective daytime. Scale bar: 25 μm, applies to (a–c). (d–e) High magnification micrographs showing that during the subjective day, cone pedicles (white asterisk) of WT animals (d) have significant higher CPLX3 immunoreactivity than cone pedicles of cone-Bmal1−/− animals (e). PSD95 (white) labels photoreceptor terminals. DAPI (cyan) stains the cell nuclei. Arrows point to rod spherules. Scale bar: 5 μm

FIGURE 6.

FIGURE 6

Open in a new tab

Quantification of CPLX3 and CPLX4. (a) Quantification of CPLX3 in cone pedicles in the wild-type (WT) and mutant cones in the middle of the day under room light (ZT 05-07, yellow) or subjective day (CT 05-07, gray) showed that CPLX3 expression is decreased in cone-Bmal1−/− cones and there is no effect of light adaptation. (b) Quantification of CPLX3 in cone pedicles during the subjective day (CT 05-07, light gray) or during the subjective night (CT 07–19, dark gray) shows that CPLX3 is downregulated at night in WT cones and is constitutively low in cone-Bmal1−/− cones. (c,d) Quantification of CPLX4 in the outer plexiform layer (OPL) showed that there were no differences in CPLX4 expression in the OPL under the conditions depicted in (a,b). n is shown on the figures and represents number of animals (5–18 cones per animal). Error bars are SEM. Dots show individual data points. p-Values: unpaired t-test, two-tailed

3.5 ∣. CLOCK:BMAL1 binding sites in the promotor region of Cplx3

To understand how the circadian clock might be modulating Cplx3 and not Cplx4 expression, we conducted an in silico analysis of the promotor regions of the Cplx3 and Cplx4 genes. We searched for binding sites of circadian transcription factors such as E-boxes, D-boxes, ROR Elements, and CRE elements (see Table 1 for description and sequences). In the promotor region of Cplx3 we found three E-boxes, which are CLOCK/BMAL1 binding sites (Takahashi, 2017), and one ROR element, which are Rev-erbα binding site (Takahashi, 2017) (Figure 7a). All three E-boxes had the sequence 5′-CACGTG-3′ and the ROR element had the sequence 5′-GGGTCA-3′. However, in the promotor region of Cplx4 we found no such binding sites (Figure 7b). The presence of E-boxes in the promotor region of Cplx3 may provide a mechanism through which the circadian clock controls Cplx3 expression. The absence of such elements in the promotor region of Cplx4 is consistent with the lack of regulation of Cplx4 by BMAL1, the circadian cycle, and light/dark adaptation.

FIGURE 7.

FIGURE 7

Open in a new tab

Cplx3 and Cplx4 DNA promotor region analysis. (a) The Cplx3 DNA promotor region contains three E-boxes and one ROR element. (b) The Cplx4 DNA promotor region contains no E- or D-boxes and no ROR or CRE elements. The promotor region was designated as the 600 bp upstream of Exon 1. Gene sequences were downloaded from ENSEMBL

4 ∣. DISCUSSION

Vision starts in retinal photoreceptors—rods and cones—where light energy is converted into changes in membrane potential that modulate tonic synaptic transmission to second-order neurons. Second-order neurons process signals simultaneously in parallel circuits that converge onto ganglion cells-the output neurons of the retina. The cone pedicle plays a key role in parallel processing, extracting and distributing relevant information from the multitude of qualities that are present in the visual environment to at least a dozen parallel pathways that eventually project to the brain for further processing (Dowling, 2012; Wässle, 2004). Modulation of cone synaptic activity is therefore expected to critically affect retinal circuit function and eventually vision. Our data suggest that the circadian modulation of Cplx3/CPLX3 at the cone synapse may be an important process through which the cone clock modulates retinal function and visual behavior with the time of day.

CPLX3 at cone pedicles is ideally positioned to contribute to the day/night variation in retinal processing of visual information. Previous research demonstrated that CPLX3 is a critical protein in regulating vesicle priming and fusion at ribbon synapses (Babai et al., 2016; Chang et al., 2015; Mortensen et al., 2016; Reim et al., 2009; Vaithianathan, Henry, Akmentin, & Matthews, 2015). It has been suggested that CPLX3 has a dual role: it acts as a break on the SNARE complex to prevent spontaneous fusion in the absence of calcium influx, and as facilitator of transmitter release evoked by depolarization (Babai et al., 2016; Mortensen et al., 2016; Vaithianathan et al., 2015). Thus, the removal of CPLX3 in cones would be expected to (a) decrease glutamate release in the dark, when cones are depolarized, and (b) mitigate the decrease in glutamate release when light is present, that is when cones hyperpolarize. This assumption has been directly tested in Cplx3/4 double knockout mice. By recording from horizontal cells, Babai et al. concluded that (a) CPLX3/4 suppresses tonic activity in horizontal cells, and (b) facilitates horizontal cell responses elicited by electrical stimulation (depolarization) of the cones (Babai et al., 2016). Thus, removal of CPLX3 would be expected to have little effect on the cone photoresponse, which relies on cGMP-gated channels located in the outer segment, but affect glutamate release at cone terminals in a way that would result in a slowdown and/or decrease of the amplitude of the light responses of second order neurons.

In agreement with this, the cone photoresponse, as assessed by the a-wave of the electroretinogram (ERG) is normal whereas the implicit time of the b-wave, which reflects the response of the second-order neurons, is increased in Cplx3 knockout mice (Reim et al., 2009). Cplx4 knockout mice have a very mild phenotype, but the b-wave is both slowed down and of reduced amplitude in Cplx3/4 double knockout mice, revealing cooperative perturbing effect of Cplx3/4 deletion on the b-wave amplitude (Reim et al., 2009). Interestingly, changes in b-wave kinetics in Cplx3 knockout mice occur at bright light intensities (≥0.1 cds/m2) that is, when cones contribute to the ERG. A slower b-wave is also characteristic of the light-adapted ERG recorded at night (Cameron et al., 2008). Pan- or retina-specific Bmal1 knockout animals also show reduced and/or slower b-waves (Baba, Piano, et al., 2018; Storch et al., 2007). Although a clear limitation of our study is that we cannot exclude the possibility that other elements of the cone synaptic machinery may be regulated by the clock, the close similarity of the ERG waveform between Cplx3 knockout mice and clock mutant mice or nighttime recordings from WT mice suggests that CPLX3 may be a significant player in the circadian modulation of the cone synaptic activity.

Previous studies, including ours, have established the modulation of electrical coupling between photoreceptors by retinal circadian clocks (reviewed in Ribelayga & O'Brien, 2017). The circadian modulation of rod/cone coupling is expected to add to that of the cone synaptic machinery to shape the transfer of information to downstream neurons. By comparing the light responses of cones and horizontal cells, we recently provided electrophysiological evidence that the strength of rod/cone coupling cannot account entirely for rhythmic changes in the light response properties of horizontal cells (Ribelayga & Mangel, 2019). Rather, our study offered strong support to the view that retinal clocks shape the light responses of horizontal cells in part by modulating cone-to-horizontal cell synaptic transmission. The present study supports a control of the synaptic machinery by the cone clock. Future studies will be needed to isolate the specific contribution of CPLX3 in the circadian modulation of cone synaptic transfer and its impact on retinal processing. The daytime increase in CPLX3 in the cone pedicle may represent an adaptive advantage to optimize cone-driven vision under (expected) daylight conditions when contrast sensitivity, acuity, and color vision depend on reliable signal transmission at the cone synapse.

BMAL1 could be serving directly as a transcription factor for Cplx3. We identified four E-boxes, CLOCK:BMAL1 binding sites (Takahashi, 2015; Ye et al., 2014), in the promotor region of Cplx3. Thereby, the simplest explanation is that Cplx3 transcription is enhanced by the circadian clock through CLOCK:BMAL1 binding to these E-boxes. However, the in silico data cannot exclude the possibility that there may be additional steps along this pathway which could contribute to Cplx3 rhythmic expression. It is interesting to note though that the increase in Cplx3 gene expression occurs during the day that is when BMAL1 protein expression is maximal in mouse cones (Liu, Zhang, & Ribelayga, 2012).

Synaptic proteins are generally thought to have a relative long half-life (i.e., several days) (Cohen et al., 2013). The daily rhythm in CPLX3 expression we report here suggests that CPLX3 in cones may have an unusually fast turnover. Although surprising, this possibility would not be unprecedented in the retina since some retinal proteins have been shown to have extraordinarily high turnover rates. For example, Connexin36, a gap junction protein, has a half-life of about 3.1 hr (Wang, Lin, Mitchell, Ram, & O'Brien, 2015). Similarly, synaptic ribbons exhibit diurnal turnover with an average lifespan of less than 12 hr (Vollrath & Spiwoks-Becker, 1996). Therefore, CPLX3 could have a relatively fast turnover in cones to support daily rhythmic expression and function, although this remains to be demonstrated.

ACKNOWLEDGMENTS

This work was supported by grants from the National Institutes of Health (grant numbers R21-EY028647, R01-EY029408, P30-EY028102), a UTHealth BRAIN Initiative/CTSA grant (TR000371), a grant from the University of Texas System Neuroscience and Neurotechnology Research Institute (362469), and the Herman Eye Fund. The authors thank Eduardo Silveyra (UT Houston) for help with some of the initial experiments. The authors also thank Dr Nils Brose (Max Plank Institute, Göttingen, Germany) for providing the antibodies against CPLX3 and CPLX4. Finally, the authors thank Dr Takae Kiyama (UT Houston) and Dr Chai-An Mao (UT Houston) for help with the RNAscope experiments.

Funding information

National Eye Institute, Grant/Award Numbers: EY028102, EY028647, EY029408; University of Texas Medical School at Houston, Grant/Award Number: 362469; UTHealth Brain Initiative/CTSA, Grant/Award Number: TR000371

Footnotes

CONFLICT OF INTEREST

The authors declare no conflict of interest.

PEER REVIEW

The peer review history for this article is available at https://publons.com/publon/10.1002/cne.25004.

DATA AVAILABILITY STATEMENT

The dataset on RNAseq has been deposited in NCBI's Gene Expression Omnibus (GEO) and is accessible through GEO Series accession number GSE155550. All other data needed to evaluate the conclusions in the paper are present in the paper. Protocols related to the paper may be requested from the authors. The rod-Bmal1−/− and cone-Bmal1−/− mutant lines can be provided by C.P.R. pending scientific review and a completed material transfer agreement.

REFERENCES

  1. Anders S, Pyl PT, & Huber W (2015). HTSeq—A Python framework to work with high-throughput sequencing data. Bioinformatics, 31(2), 166–169. 10.1093/bioinformatics/btu638 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Baba K, Piano I, Lyuboslavsky P, Chrenek MA, Sellers JT, Zhang S, … Iuvone PM (2018). Removal of clock gene Bmal1 from the retina affects retinal development and accelerates cone photoreceptor degeneration during aging. Proceedings of the National Academy of Sciences of the United States of America, 115(51), 13099–13104. 10.1073/pnas.1808137115 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Baba K, Ribelayga CP, Michael Iuvone P, & Tosini G (2018). The retinal circadian clock and photoreceptor viability. In Ash JD, Anderson RE, LaVail MM, Bowes Rickman C, Hollyfield JG, & Grimm C (Eds.), Retinal degenerative diseases (Vol. 1074, pp. 345–350). New York: Springer International Publishing. 10.1007/978-3-319-75402-4_42 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Babai N, Sendelbeck A, Regus-Leidig H, Fuchs M, Mertins J, Reim K, … Brandstätter JH (2016). Functional roles of Complexin 3 and Complexin 4 at mouse photoreceptor ribbon synapses. The Journal of Neuroscience, 36(25), 6651–6667. 10.1523/JNEUROSCI.4335-15.2016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Besharse JC, & McMahon DG (2016). The retina and other light-sensitive ocular clocks. Journal of Biological Rhythms, 31(3), 223–243. 10.1177/0748730416642657 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Buhr ED, & Takahashi JS (2013). Molecular components of the mammalian circadian clock. In Handbook in experimental pharmacology (Vol. 217, pp. 3–27). Berlin, Heidelberg: 10.1007/978-3-642-25950-0_1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cameron MA, Barnard AR, Hut RA, Bonnefont X, van der Horst GTJ, Hankins MW, & Lucas RJ (2008). Electroretinography of wild-type and Cry mutant mice reveals circadian tuning of photopic and mesopic retinal responses. Journal of Biological Rhythms, 23(6), 489–501. 10.1177/0748730408325874 [DOI] [PubMed] [Google Scholar]
  8. Carlezon W, Duman R, & Nestler E (2005). The many faces of CREB. Trends in Neurosciences, 28(8), 436–445. 10.1016/j.tins.2005.06.005 [DOI] [PubMed] [Google Scholar]
  9. Chang S, Reim K, Pedersen M, Neher E, Brose N, & Taschenberger H (2015). Complexin stabilizes newly primed synaptic vesicles and prevents their premature fusion at the mouse calyx of held synapse. Journal of Neuroscience, 35(21), 8272–8290. 10.1523/JNEUROSCI.4841-14.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cohen LD, Zuchman R, Sorokina O, Müller A, Dieterich DC, Armstrong JD, … Ziv NE (2013). Metabolic turnover of synaptic proteins: Kinetics, interdependencies and implications for synaptic maintenance. PLoS One, 8(5), e63191. 10.1371/journal.pone.0063191 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cunningham F, Achuthan P, Akanni W, Allen J, Amode MR, Armean IM, … Flicek P (2019). Ensembl 2019. Nucleic Acids Research, 47(D1), D745–D751. 10.1093/nar/gky1113 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. DeVries SH, & Baylor DA (1993). Synaptic circuitry of the retina and olfactory bulb. Cell, 72, 139–149. 10.1016/S0092-8674(05)80033-9 [DOI] [PubMed] [Google Scholar]
  13. Dick O, tom Dieck S, Altrock WD, Ammermüller J, Weiler R, Garner CC, … Brandstätter JH (2003). The presynaptic active zone protein bassoon is essential for photoreceptor ribbon synapse formation in the retina. Neuron, 37(5), 775–786. 10.1016/S0896-6273(03)00086-2 [DOI] [PubMed] [Google Scholar]
  14. Dowling JE (2012). The retina: An approachable part of the brain (revised edition. Cambridge, MA: Harvard University Press. 10.1002/ajhb.22305 [DOI] [Google Scholar]
  15. Dunlap JC (1999). Molecular bases for circadian clocks. Cell, 96(2), 271–290. 10.1016/S0092-8674(00)80566-8 [DOI] [PubMed] [Google Scholar]
  16. Duran RC, Yan H, Zheng Y, Huang X, Grill R, Kim DH, … Wu JQ (2017). The systematic analysis of coding and long non-coding RNAs in the sub-chronic and chronic stages of spinal cord injury. Scientific Reports, 7, 41008. 10.1038/srep41008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Harding HP, & Lazar MA (1993). The orphan receptor Rev-ErbA alpha activates transcription via a novel response element. Molecular and Cellular Biology, 13(5), 3113–3121. 10.1128/MCB.13.5.3113 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hartigan JA, & Wong MA (1979). A k-means clustering algorithm. Journal of the Royal Statistical Society, 28(1), 100–108 Retrieved from https://www.jstor.org/stable/2346830 [Google Scholar]
  19. Kiyama T, & Mao C-A (2020). Ultrasensitive RNAscope in situ hybridization system on embryonic and adult mouse retinas. In Retinal development. Methods in molecular biology (Vol. 2092). New York, NY: Humana. 10.1007/978-1-0716-0175-4_11 [DOI] [PubMed] [Google Scholar]
  20. Landgraf I, Mühlhans J, Dedek K, Reim K, Brandstätter JH, & Ammermüller J (2012). The absence of Complexin 3 and Complexin 4 differentially impacts the ON and OFF pathways in mouse retina: Impact of Cplx 3 and Cplx 4 on ON and OFF pathways. European Journal of Neuroscience, 36(4), 2470–2481. 10.1111/j.1460-9568.2012.08149.x [DOI] [PubMed] [Google Scholar]
  21. Le Y-Z, Ash JD, Al-Ubaidi MR, Chen Y, Ma J-X, & Anderson RE (2004). Targeted expression of Cre recombinase to cone photoreceptors in transgenic mice. Molecular Vision, 10, 1011–1018. [PubMed] [Google Scholar]
  22. Li H, Zhang Z, Blackburn MR, Wang SW, Ribelayga CP, & O'Brien J (2013). Adenosine and dopamine receptors coregulate photoreceptor coupling via gap junction phosphorylation in mouse retina. Journal of Neuroscience, 33(7), 3135–3150. 10.1523/JNEUROSCI.2807-12.2013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Li S, Chen D, Sauve Y, McCandless J, Chen YJ, & Chen CK (2005). Rhodopsin-iCre transgenic mouse line for Cre-mediated rod-specific gene targeting. Genesis, 41(2), 73–80. 10.1002/gene.20097 [DOI] [PubMed] [Google Scholar]
  24. Liberzon A, Subramanian A, Pinchback R, Thorvaldsdottir H, Tamayo P, & Mesirov JP (2011). Molecular signatures database (MSigDB) 3.0. Bioinformatics, 27(12), 1739–1740. 10.1093/bioinformatics/btr260 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Liu X, Zhang Z, & Ribelayga CP (2012). Heterogeneous expression of the core circadian clock proteins among neuronal cell types in mouse retina. PLoS One, 7(11), e50602. 10.1371/journal.pone.0050602 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Love MI, Huber W, & Anders S (2014). Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biology, 15(12), 550. 10.1186/s13059-014-0550-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Lowrey PL, & Takahashi JS (2011). Genetics of circadian rhythms in mammalian model organisms. In Advances in genetics (Vol. 74, pp. 175–230). MA, USA: Elsevier. 10.1016/B978-0-12-387690-4.00006-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Matthews G, & Fuchs P (2010). The diverse roles of ribbon synapses in sensory neurotransmission. Nature Reviews Neuroscience, 11(12), 812–822. 10.1038/nrn2924 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. McMahon DG, Iuvone PM, & Tosini G (2014). Circadian organization of the mammalian retina: From gene regulation to physiology and diseases. Progress in Retinal and Eye Research, 39, 58–76. 10.1016/j.preteyeres.2013.12.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Mortensen LS, Park SJH, Ke J, Cooper BH, Zhang L, Imig C, … Singer JH (2016). Complexin 3 increases the fidelity of signaling in a retinal circuit by regulating exocytosis at ribbon synapses. Cell Reports, 15(10), 2239–2250. 10.1016/j.celrep.2016.05.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Novak A, Guo C, Yang W, Nagy A, & Lobe CG (2000). Z/EG, a double reporter mouse line that expresses enhanced green fluorescent protein upon Cre-mediated excision. Genesis, 28(3–4), 147–155. [PubMed] [Google Scholar]
  32. Parsons TD, & Sterling P (2003). Synaptic ribbon: Conveyor belt or safety belt. Neuron, 37, 379–382. 10.1016/S0896-6273(03)00062-X [DOI] [PubMed] [Google Scholar]
  33. Quackenbush J (2002). Microarray data normalization and transformation. Nature Genetics, 32, 496–501. 10.1038/ng1032 [DOI] [PubMed] [Google Scholar]
  34. Reim K, Mansour M, Varoqueaux F, McMahon HT, Südhof TC, Brose N, & Rosenmund C (2001). Complexins regulate a late step in Ca2+-dependent neurotransmitter release. Cell, 104(1), 71–81. 10.1016/s0092-8674(01)00192-1 [DOI] [PubMed] [Google Scholar]
  35. Reim K, Regus-Leidig H, Ammermuller J, El-Kordi A, Radyushkin K, Ehrenreich H, … Brose N (2009). Aberrant function and structure of retinal ribbon synapses in the absence of complexin 3 and complexin 4. Journal of Cell Science, 122(9), 1352–1361. 10.1242/jcs.045401 [DOI] [PubMed] [Google Scholar]
  36. Reim K, Wegmeyer H, Brandstätter JH, Xue M, Rosenmund C, Dresbach T, … Brose N (2005). Structurally and functionally unique complexins at retinal ribbon synapses. The Journal of Cell Biology, 169(4), 669–680. 10.1083/jcb.200502115 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Ribelayga C, & Mangel SC (2019). Circadian clock regulation of cone to horizontal cell synaptic transfer in the goldfish retina. PLoS One, 14(8), e0218818. 10.1371/journal.pone.0218818 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Ribelayga C, & O'Brien J (2017). Circadian and light-adaptive control of electrical synapse plasticity in the vertebrate retina. In Jing J (Ed.), Network functions and plasticity: Perspectives from studying electrical coupling in microcircuits (pp. 209–241). Cambridge, MA: Elsevier. 10.1016/B978-0-12-803471-2.00010-2 [DOI] [Google Scholar]
  39. Sawant OB, Horton AM, Zucaro OF, Chan R, Bonilha VL, Samuels IS, & Rao S (2017). The circadian clock gene Bmal1 controls thyroid hormone-mediated spectral identity and cone photoreceptor function. Cell Reports, 21(3), 692–706. 10.1016/j.celrep.2017.09.069 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Sawant OB, Jidiman VK, Fuller RD, Zucaro OF, Kpegba C, Yu M, … Rao S (2019). The circadian clock gene Bmal1 is required to control the timing of retinal neurogenesis and lamination of Müller glia in the mouse retina. FASEB Journal, 33(8), 8745–8758. 10.1096/fj.201801832RR [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Shekhar K, Lapan SW, Whitney IE, Tran NM, Macosko EZ, Kowalczyk M, … Sanes JR (2016). Comprehensive classification of retinal bipolar neurons by single-cell transcriptomics. Cell, 166(5), 1308–1323.e30. 10.1016/j.cell.2016.07.054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Storch K-F, Paz C, Signorovitch J, Raviola E, Pawlyk B, Li T, & Weitz CJ (2007). Intrinsic circadian clock of the mammalian retina: Importance for retinal processing of visual information. Cell, 130(4), 730–741. 10.1016/j.cell.2007.06.045 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, … Mesirov JP (2005). Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proceedings of the National Academy of Sciences of the United States of America, 102(43), 15545–15550. 10.1073/pnas.0506580102 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Takahashi JS (2015). Molecular components of the circadian clock in mammals. Diabetes, Obesity and Metabolism, 17, 6–11. 10.1111/dom.12514 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Takahashi JS (2017). Transcriptional architecture of the mammalian circadian clock. Nature Reviews Genetics, 18(3), 164–179. 10.1038/nrg.2016.150 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Trapnell C, Pachter L, & Salzberg SL (2009). TopHat: Discovering splice junctions with RNA-Seq. Bioinformatics, 25(9), 1105–1111. 10.1093/bioinformatics/btp120 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, … Pachter L (2012). Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks. Nature Protocols, 7(3), 562–578. 10.1038/nprot.2012.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Trimbuch T, & Rosenmund C (2016). Should I stop or should I go? The role of complexin in neurotransmitter release. Nature Reviews Neuroscience, 17(2), 118–125. 10.1038/nrn.2015.16 [DOI] [PubMed] [Google Scholar]
  49. Ueda HR, Hayashi S, Chen W, Sano M, Machida M, Shigeyoshi Y, … Hashimoto S (2005). System-level identification of transcriptional circuits underlying mammalian circadian clocks. Nature Genetics, 37(2), 187–192. 10.1038/ng1504 [DOI] [PubMed] [Google Scholar]
  50. Vaithianathan T, Henry D, Akmentin W, & Matthews G (2015). Functional roles of complexin in neurotransmitter release at ribbon synapses of mouse retinal bipolar neurons. Journal of Neuroscience, 35(9), 4065–4070. 10.1523/JNEUROSCI.2703-14.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Vollrath L, & Spiwoks-Becker I (1996). Plasticity of retinal ribbon synapses. Microscopy Research and Technique, 35, 472–487. [DOI] [PubMed] [Google Scholar]
  52. Wang F, Flanagan J, Su N, Wang L-C, Bui S, Nielson A, … Luo Y (2012). RNAscope. The Journal of Molecular Diagnostics, 14(1), 22–29. 10.1016/j.jmoldx.2011.08.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wang HY, Lin Y-P, Mitchell CK, Ram S, & O'Brien J (2015). Two-color fluorescent analysis of connexin 36 turnover: Relationship to functional plasticity. Journal of Cell Science, 128(21), 3888–3897. 10.1242/jcs.162586 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Wässle H (2004). Parallel processing in the mammalian retina. Nature Reviews Neuroscience, 5(10), 747–757. 10.1038/nrn1497 [DOI] [PubMed] [Google Scholar]
  55. Ye R, Selby CP, Chiou Y-Y, Ozkan-Dagliyan I, Gaddameedhi S, & Sancar A (2014). Dual modes of CLOCK:BMAL1 inhibition mediated by cryptochrome and period proteins in the mammalian circadian clock. Genes & Development, 28(18), 1989–1998. 10.1101/gad.249417.114 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Zhang Y, Chen K, Sloan SA, Bennett ML, Scholze AR, O'Keeffe S, … Wu JQ (2014). An RNA-sequencing transcriptome and splicing database of glia, neurons, and vascular cells of the cerebral cortex. Journal of Neuroscience, 34(36), 11929–11947. 10.1523/JNEUROSCI.1860-14.2014 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The dataset on RNAseq has been deposited in NCBI's Gene Expression Omnibus (GEO) and is accessible through GEO Series accession number GSE155550. All other data needed to evaluate the conclusions in the paper are present in the paper. Protocols related to the paper may be requested from the authors. The rod-Bmal1−/− and cone-Bmal1−/− mutant lines can be provided by C.P.R. pending scientific review and a completed material transfer agreement.

RESOURCES