Expression patterns of zebrafish nocturnin genes and the transcriptional activity of the frog nocturnin promoter in zebrafish rod photoreceptors

Abstract

Purpose: Daily modulation of gene expression is critical for the circadian rhythms of many organisms. One of the modulating mechanisms is based on nocturnin, a deadenylase that degrades mRNA in a circadian fashion. The nocturnin genes are expressed broadly, but their tissue expression patterns differ between mice and the frog Xenopus laevis; this difference suggests that the extent of the regulation of nocturin gene expression varies among species. In this study, we set out to characterize the expression patterns of two zebrafish nocturnin genes; in addition, we asked whether a frog nocturnin promoter has transcriptional activity in zebrafish.

Methods: We used reverse transcription (RT)-PCR, quantitative real-time PCR (qRT-PCR), and rapid amplification of cDNA ends (RACE) analysis to determine whether the nocturnin-a and nocturnin-b genes are expressed in the eye, in situ hybridization to determine the cellular expression pattern of the nocturnin-b gene in the retina, and confocal microscopy to determine the protein expression pattern of the transgenic reporter green fluorescent protein (GFP) driven by the frog nocturnin promoter.

Results: We found that the amino acid sequences of zebrafish nocturnin-a and nocturnin-b are highly similar to those of frog, mouse, and human nocturnin homologs. Only nocturnin-b is expressed in the eye. Within the retina, nocturnin-b mRNA was expressed at higher levels in the retinal photoreceptors layer than in other cellular layers. This expression pattern echoes the restricted photoreceptor expression of nocturnin in the frog. We also found that the frog nocturnin promoter can be specifically activated in zebrafish rod photoreceptors.

Conclusions: The high level of similarities in amino acid sequences of human, mouse, frog, and zebrafish nocturnin homologs suggest these proteins maintain a conserved deadenylation function that is important for regulating retinal circadian rhythmicity. The rod-specific transcriptional activity of the frog nocturnin promoter makes it a useful tool to drive moderate and rod-specific transgenic expression in zebrafish. The results of this study lay the groundwork to study nocturnin-based circadian biology of the zebrafish retina.

Figure 1

Zebrafish nocturnin-a and nocturnin-b are similar to frog and mammalian nocturnin genes. A: An unrooted phylogenetic tree of nocturnin proteins suggests zebrafish nocturnin-b is more similar to frog nocturnin than nocturnin-a. The scale bar represents 5% estimated sequence divergence. B: An alignment of the amino acid sequences of the nocturnin homologs of human, mouse, Xenopus laevis, and zebrafish revealed high conservation: Identical amino acid residues between different nocturnin homologs are highlighted in gray. The deadenylation catalytic domain is boxed. C: An reverse transcription (RT)-PCR analysis with two primer pairs for each zebrafish nocturnin gene showed that nocturnin-b was expressed in 5-dpf and adult zebrafish eyes, as well as in 5-dpf whole larval fish, whereas nocturnin-a expression was not detectable in the eyes, although it was detectable in the 5-dpf whole larval fish. D: RT–PCR analysis of nocturnin-a and nocturnin-b expression in adult eyes at 8 AM, 12 PM, 4 PM, 8 PM, 12 AM, and 4 AM, noting prominent nocturnin-b expression but undetectable nocturnin-a expression. E: Quantitative real-time PCR (qRT-PCR) revealed the rhythmic changes in nocturnin-b expression in zebrafish adult eyes at the 8 AM, 12 PM, 4 PM, 8 PM, 12 AM, and 4 AM y-axis, fold differences between nocturnin-b and elf2a, normalizing against the level at 8 AM. Error bars (standard deviations) are based on three experiments.

Figure 2

In situ hybridization analysis showed that the nocturnin-b gene is expressed in multiple retinal cell types. A: The nocturnin-b mRNA signals, visualized with an anti-sense probe (blue), were stronger in the photoreceptor layer (PRL) than in the inner nuclear (INL) and the ganglion cell layer (GC). B: A local region of the photoreceptor layer in panel A (boxed region) is magnified to better illustrate the enrichment of the nocturnin-b mRNA signals at the inner segment regions. The cell nuclei were stained with YO-PRO (red). C: In situ hybridization with a sense probe for nocturnin-b did not stain the retina, supporting the staining specificity by the anti-sense probe in panels A and B. Scale bars, 20 μm.

Figure 3

Frog nocturnin promoter drives rod-specific transcription in the zebrafish retina. A: A schematic illustrates the configuration of the transgene cassette to express green fluorescent protein (GFP) in the zebrafish under the control of the frog nocturnin promoter, which covers 398 bp upstream of and 21 bp downstream of the translation start site of the nocturnin gene. This construct, thus, drives the expression of a fusion protein between GFP and the first seven amino acids of frog nocturnin. SV40 polyA stands for the 3′- untranslated region (UTR) of the SV40 gene and its polyadenylation signals. The transgene is flanked by two Tol2 transposon elements (Tol2). The diagram is not drawn to scale. A double-headed arrow indicates the region that was amplified when the fish was PCR genotyped. B–E: In Tg(nocturnin^frog:GFP) ^pt158, GFP immunostaining highlighted rods. GFP was enriched in the rod cell nuclei in the basal half of the outer nuclear layer (green; arrows). Double arrows indicate the GFP signals in the rod spherules. In contrast, no GFP expression was detected in the UV cone nuclei (arrowhead) and the RGB cones (red, green, and blue cones). The cell nuclei were stained with TO-PRO (red). Panel E is a magnification of the boxed region in panel D. F-I: In Tg(nocturnin^frog:GFP) ^pt158, GFP localized to cells positive for the rod marker Zpr3 at the outer segments (arrows). Panel I is a magnification of the boxed region in panel H. J–M: In Tg(nocturnin^frog:GFP) ^pt158, GFP signals did not colocalize with double cone marker Zpr1. Panel M is a magnification of the boxed region in panel L. Arrowheads indicate the double cone pedicle synaptic junctions, and arrows indicate the rod spherules. The GFP signals at the outer segments were stronger when GFP was visualized directly with its own florescence (panels F and J) than indirectly with immunostaining (panel B). This difference is believed to be due to the strong autofluorescence of the visual pigments at the outer segments: When GFP expression was as weak as in Tg(nocturnin^frog:GFP) ^pt158, to directly visualize GFP expression with its own florescence under confocal microscopy, we needed to use a much stronger laser power, which made the autofluorescence of the visual pigments more prominent. Thus, the GFP expression patterns in Tg(nocturnin^frog:GFP) ^pt158 could be more reliably determined by its presence in the synaptic junctions, cell body, and inner segments. N: Western blotting showed that even after a 1,000-fold dilution with wild-type eye extracts, the GFP level in the Tg(−3.7rho:EGFP)^kj2 transgenic fish eye was still significantly higher than that in Tg(nocturnin^frog:GFP) ^pt158 (compare lane 6 with lane 4). α-tubulin blotting served as a loading control. O: A histogram of the intensity ratio between Tg(nocturnin^frog:GFP) ^pt158 GFP and α-tubulin signals (N) normalized against the ratio at 8 AM, shows no apparent rhythmic fluctuation in GFP levels among the four time points. Scale bars, 20 μm.