We, along with others, have previously reported that a circadian rhythm in melatonin and dopamine levels is still present in 60-day-old −rdy
rats with severe photoreceptor degeneration (Doyle et al., 2002
; Sakamoto et al., 2004
). We have also shown that the circadian clock driving the melatonin rhythm in the dystrophic retina is sensitive to KA lesion, while the circadian clock driving Aanat
rhythmicity in intact retina is insensitive to KA (Sakamoto et al., 2004
). Such a result suggests that at least two different circadian-clock-generating mechanisms are present in the rat retina.
To gather more information about the functioning of the retinal circadian clock system, we decided to investigate the expression of clock genes in the rat retina under different lightning conditions and after photoreceptor degeneration. Our data demonstrate that many clock genes showed significant variation in the retina of −rdy+ under LD and DD, while under LL only Per1, Per2, Clock and RevErbα showed a significant variation over the sampling period (See Figures , and .) In the dystrophic retina (−rdy), Per1 and Per2 showed a significant variation in LD, DD and LL, while the other clock genes did not any variation in their expression levels. Moreover, in constant light, most of the clock genes were down-regulated (Figures , and ), while Rev-erbα and Naps2 were up-regulated (Figures , and ).
A few studies have investigated the expression of circadian clock genes in the rat retina. Namihira et al. (1999
reported that Per2
mRNA showed a significant circadian rhythm, whereas Per1
did not show any significant variation in their mRNA levels. More recently, Kamphuis et al. (2005)
investigated the daily rhythms of many of the clock genes in the rat retina and reported that only Per2, Per3, Cry2
were rhythmic. Finally, Numano et al. (2006)
and Sakamoto et al. (2006)
reported that Per1
mRNA levels were rhythmic in the rat retina. Overall, our data in rdy+
agree with the data reported by Namihira et al. [1999
, Numano et al. (2006)
and Sakamoto et al. (2006)
, although some discrepancies are present with respect to the data presented by Kamphuis et al.(2005)
. Recent studies have also reported the expression pattern of several clock genes in the intact mouse retina and in the retina of mice (rd/rd
) lacking photoreceptors [Ruan et al., 2006
; Dinet et al., 2006]. In one study, the authors reported that Per1, Per2, Cry1, Cry 2 and Bmal1
showed a significant rhythm in LD and DD in the retina of C3H rd/rd
mice (Ruan et al., 2006
), while another investigation reported a circadian rhythm in Per1
mRNA in C57BL and C3H rd/rd
, but not in Cry 2
mRNA (Dinet et al., 2007
). Therefore, it is clear that there are significant differences among the results obtained in different species or even within the same species among different strains.
The distribution of clock genes in the rat and mouse retinas has been also investigated by some studies. In the rat, Per1
transcripts are present, but at low levels, in the photoreceptors, and are more abundantly present in the inner nuclear layer, whereas Per2
transcripts are present in the inner nuclear layer and the ganglion cell layer (Namihira et al., 1999
), but not in the photoreceptor layer. In the mouse, Per1, Bmal1
transcripts are expressed in the photoreceptors (Gekakis et al., 1998
, Yujnovsky et al., 2006
; Dinet et al., 2007
transcripts seem to be absent from the photoreceptors (Miyamoto and Sancar, 1997), whereas a low level of expression for Cry2
transcripts and immunoreactivity have been detected in the outer nuclear layer of C57BL mice (Miyamoto and Sancar, 1997: Dinet et al., 2007
Therefore, we anticipated that degeneration of photoreceptors will only marginally affect the expression of Per1 and Cry2, and Per2 will not be affected. Indeed, Per1 RNA levels were slightly reduced by the degeneration of the photoreceptors (Figures -), while Cry2 mRNA levels were dramatically affected, either in their pattern of expression or in their levels, by photoreceptor degeneration (Figures -). Per2 mRNA levels and expression patterns were only slightly modified by the loss of the photoreceptors. This observation indicates that a re-arrangement of the clock gene expression within the different retinal layers must occur during the progression of the photoreceptor degeneration.
We have previously shown that, in −rdy
transcriptions are up-regulated in the inner nuclear layer where they showed a high-amplitude circadian rhythm (Sakamoto et al., 2004
). As we already mentioned, Aanat
transcription in the rat retina is under the direct control of the circadian clock via the action of CLOCK:BMAL1 on the E-box present on the promoter region of this gene (Chen and Baler, 2000
; Tosini and Fukuhara, 2002
). Therefore, we expected that the expression of clock genes in −rdy
retinae would have shed some light on the working of this inner retinal clock and, thus, on the mechanisms that control Aanat
transcription in the −rdy.
Unfortunately, our data do not provide any useful insight on this mechanism.
In conclusion, our data indicated that a group of four clock genes is rhythmic in LD, DD and LL in −rdy+
retina, whereas only Per1
are rhythmic in the retina of −rdy
. The expression pattern of these clock genes was greatly affected by degeneration of the photoreceptor cells. Recent evidence suggests that the mammalian retina is composed of a network of circadian clocks that are located in different types of retinal cells or layers (Sakamoto et al., 2004
; Ruan et al., 2006
), where they can present a different phase and, thus, affect the overall levels of expression observed in the entire retina.
Therefore, our work suggests that investigating the expression pattern of clock genes using the whole retina or animals with photoreceptor degeneration may not provide any definitive answers about the working of the retinal circadian clock system.