Most mammalian circadian rhythms are orchestrated by a biological clock located in the hypothalamic suprachiasmatic nuclei (SCN) (Stetson and Watson-Whitmyre 1976; Menaker et al. 2013). Circadian rhythms must be kept in phase (i.e., synchronized) with the 24-h light:dark (LD) cycle to maintain a stable temporal relationship with the environment (Golombek and Rosenstein 2010). Photic information from the LD cycle is the main synchronizing stimulus for the SCN clock, and consequently of its circadian outputs. In behavioral protocols, the circadian phase is modified when a single light pulse is delivered concomitant with nocturnal locomotor activity [i.e., the ‘subjective night’ under constant darkness (DD)]. Specifically, delaying or advancing the clock will occur if light is delivered at the beginning or at the end of the subjective night, respectively (Gillette and Mitchell 2002).
The ventrolateral ‘core’ SCN sub-region receives the major retinal afferences from a subset of ganglion cells, responding to light stimulation by glutamatergic transmission through metabotropic glutamate-N-methyl-d-aspartate receptor (Vileikyte et al. 2005). Calcium (Ca2+) influx through N-methyl-d-aspartate receptor is followed by activation of both intracellular Ca2+/calmodulin-dependent kinase II and neuronal nitric oxide synthase (nNOS), leading to an increase of nitric oxide (NO) synthesis (Golombek et al. 2000; Agostino et al. 2004). Pharmacological inhibition of nNOS blocks both light-induced phase-advancing and delaying mechanism (Watanabe et al. 1995; Melo et al. 1997; Golombek et al. 2004). However, it is suggested that downstream of nNOS there is a bifurcation of the pathway. Activation of the guanylyl cyclase-cGMP (GC) and cGMP-dependent protein kinase is involved in phase advances but not in delays (Golombek et al. 2004). However, the activation of the ryanodine receptor (RyR) at the endoplasmatic reticule is involved in light-induced phase delays (Ding et al. 1998; Pfeffer et al. 2009). Regardless of this difference, both signaling pathways (through GC or RyR) modify the expression of clock genes at the core of the molecular circadian oscillator (Akiyama et al. 1999; Albrecht et al. 2001).
Besides intracellular signal transduction, the SCN neuronal network coupling is also modulated in circadian synchronization (Shirakawa et al. 2001; Quintero et al. 2003). Light-induced-glutamatergic transmission sets the phase of the ventrolateral-retinorecipient neurons of the SCN (Antle and Silver 2005). In turn, these neurons release different neurochemical signals [e.g., vasoactive intestinal polypeptide (Ibata et al. 1989; Watanabe et al. 2000; Reed et al. 2001; Aida et al. 2002; Vasalou and Henson 2011), gastrin-releasing peptide (Aida et al. 2002; Dardente et al. 2002; Vasalou and Henson 2011)] toward the dorsomedial region of the SCN. We have previously demonstrated that NO can also participate as an extracellular messenger, by coordinating the ventrolateral-dorsomedial SCN communication necessary for photic synchronization (Plano et al. 2007, 2010).
In this work, we have employed a novel NO-donor, N-nitrosomelatonin (Turjanski et al. 2000b; Kirsch and de Groot 2009) to increase NO levels within the SCN tissue, in order to potentiate photic synchronization of locomotor activity rhythms. Previous studies have demonstrated that this drug can efficiently release NO in vitro (Blanchard-Fillion et al. 2001; De Biase et al. 2005; Peyrot et al. 2005; Berchner-Pfannschmidt et al. 2008), but there are still no in vivo studies assessing its neurochemical effects on circadian synchronization. In particular, we have studied the chronobiotic properties of N-nitrosomelatonin (NOMel) using both light-pulse and jet-lag behavioral protocols. These experiments led us to propose some mechanistic insights about NO signal transduction in the photic synchronization pathway.
- Top of page
- Materials and methods
- Supporting Information
Our behavioral experiments consistently show that NOMel potentiates light-induced phase advances of locomotor activity rhythms of hamsters, without affecting photic phase delays. This phase-dependent effect was observed in both LP (Fig. 1) and simulated jet-lag protocols (Fig. 7). Even when a 10x higher dose was evaluated for light-induced phase delays, no potentiation was obtained (see Figure S2). Taking these results into account, we only evaluated the lowest dose for the rest of the experiments. It should also be stated that all animals showed a normal behavior (e.g., stereotypical grooming, exploration, and gross motor activity) after the injection of the drug, even for the higher dose (data not shown).
Even though not directly correlated with the magnitude of the phase shifts (Trávnícková et al. 1996), light exposure during the night induces the expression of several immediate early genes, including c-fos (Kornhauser et al. 1996; Golombek et al. 2003; Porterfield and Mintz 2009), which is considered as a marker of neuronal activation in the circadian clock. Moreover, PER1 is a key component of the molecular clock (Buhr and Takahashi 2013), strongly induced by light and strictly correlated with photic phase shifts (Shigeyoshi et al. 1997). We have previously demonstrated that the blockade of extracellular NO communication within the SCN impeded both photic behavioral advances as well as the ventral-dorsal spreading of cFOS expression within the SCN, while it does not affect phase delays (Plano et al. 2007). In addition, the same treatment also inhibited both photic advances for the steady-state synchronization to LD cycles, and SCN-PER1 expression at CT18, without affecting the synchronization by phase delays (Plano et al. 2010). In agreement with our previous reports, here we show that NOMel-treated animals only exhibited an increased number of photically induced SCN cFOS-positive cells at CT18, and importantly, an increased number of PER1 immunoreactivity at the dorsomedial region of the SCN, demonstrating the same kind of neuronal activation as when modulating extracellular NO communication. Therefore, our results support the idea that NO, as an extracellular messenger, is involved only in the signaling of photic phase advances. On the other hand, NOMel did not potentiate phase advances induced by saturating light pulses (Fig. 2a) but did increase cFOS induction in response to such stimulation (Fig. 2b). As stated before, cFOS induction is not necessarily correlated with the magnitude of the phase shift. It is possible that the saturating light-induced phase changes in locomotor activity may be limited by the circadian pace making system, reaching a ceiling effect that cannot be further modulated by our pharmacological treatment. In contrast, NOMel increased even further the number of cFOS-expressing cells in response to bright light pulses. We assume that the increased levels of cFOS induction could be due to an enhanced extracellular NO communication; however, this effect does not correlate with behavioral phase shifts because of the ceiling effect mentioned before. In all cases, NOMel potentiates the effects of light on gene expression, but lacks any circadian effects per se, suggesting that NO is capable of modulating the signal transduction pathway of photic entrainment, but is not sufficient to drive changes in the circadian pacemaker.
We hypothesized that the NOMel potentiation observed in both behavioral and immunohistochemical results might be due to an increase of NO levels acting both as an intracellular as well as an extracellular messenger. Therefore, to evaluate if this compound is indeed working as a NO-donor, we decided to measure nitrate and nitrite levels (i.e., the main NO metabolites) in the SCN after the drug administration (Fig. 5). SCN nitrate and nitrite levels confirm that NOMel is working as an efficient NO-donor in vivo. In addition, taking into account the behavioral, the immunohistochemical and the nitrate-nitrite results, we can confirm that the increase of NO levels per se, without light stimulation, is not enough to generate a circadian phase advance. On the other hand, nitrate levels are used as an indirect way to measure NOS activity (Kleinbongard et al. 2003), and ex vivo measurement of brain tissue nitrate and nitrite levels accurately reflects NOS activity in vivo (Salter et al. 1996). However, in our experiments light stimulation did not generate a significant increase of SCN nitrate and nitrite levels as compared to controls. In order to evaluate NO synthesis, it would be necessary to take SCN samples several minutes after light stimulation. However, as this was not one of the objectives of the present work, where SCN tissue was immediately obtained after the LP, future experiments will be necessary to confirm this hypothesis. In addition, since SCN nitrite levels measured both at CT14 and CT18 after NOMel treatment were similar (see Figure S4), we can discard any circadian-dependent differences in the bioavailability of the drug, in relation to the timing of NOMel administration.
In order to determine the mechanism by which NOMel is modulating the photic pathway in the SCN, we co-administered l-NAME, a well-known NOS inhibitor (Weber et al. 1995), together with an i.p. injection of NOMel, followed by a light pulse. We hypothesized that the NO levels generated by the NOMel treatment, regardless of the photic NOS activation, would allow the photic signal transduction cascade to continue, by activating the required substrates downstream of NOS, thus generating the phase shift. The present results agreed with our hypothesis (Fig. 6). Therefore, NOMel should be delivering the appropriate NO levels for the continuation of the photic pathway, but it cannot activate the pathway by itself. Although not measured in this work, we assumed that NO levels in other tissues would be increased due to the systemic NOMel administration. However, the central inhibition of the NOS activity performed in the present experiments, together with the increased nitrate and nitrite levels found, allow us to suggest that the main effect of the NOMel is due to its action at the SCN.
We also evaluated the effect of the drug on the collectively named ‘non-photic’ phase shifts, using a cage and bedding change as stimulus [(Figure S3) (Mistlberger 1992; Yannielli and Harrington 2004)]. The administration of NOMel at CT6 did not affect non-photic phase advances induced by such stimulus, nor generated a phase advance per se. To our knowledge, there is no data about the participation of NOS in the transduction of these circadian phase shifts.
The photic-signal transduction pathway bifurcates downstream of NO: while GC is involved in circadian phase advances, the opening of the RyR is related to the phase-delaying mechanism. It has been shown that another type of NO-donor, S-Nitroso-N-acetylpenciallamine (SNAP), can potentiate both phase advances and delays, at the same CTs as we have studied in this work (Melo et al. 1997). However, our results with NOMel indicate that this drug can only enhance phase advances. Therefore, NOMel could only be potentiating the activation of the GC, but not that of the RyR. Indeed, SNAP can stimulate the opening of RyR in other tissues (Hart and Dulhunty 2000; Wang et al. 2010), and this mechanism involves the S-nitrosylation of the thiol group of cysteine residues, generating a S-NO group (Gonzalez et al. 2008; Kakizawa et al. 2012). S-nitrosilation by NOMel has only been demonstrated for the glyceraldehyde 3-phosphate dehydrogenase under in vitro conditions (Kirsch and de Groot 2008). We presume that under in vivo conditions, the NO levels released by NOMel had little thiol reactivity able to increase the S-nitrosylation of the RyR, which is likely involved in the aperture of the channel for enhancing the phase-delaying effects of light. Another important difference is that the activation of the GC is much more sensitive to NO activation, within the nanomolar range (Russwurm and Koesling 2004; Rodriguez-Juarez et al. 2007; Hall and Garthwaite 2009). Instead, S-nitrosylation is considered as a short-ranged mechanism in which a high concentration of NO is necessary (Evangelista et al. 2013; Martinez-Ruiz et al. 2013). This was taken into account for the inclusion of the experiment in which a 10-times higher dose of NOMel was evaluated at CT14. However, NOMel still failed to potentiate the phase-delaying effects of light. In addition, SNAP can actually be considered as a low molecular weight S-nitrosothiol (Oliveira et al. 2008), since the NO-group is linked to the rest of the molecule through a sulfur bond. Therefore, a trans-nitrosyaltion (Matsumoto and Gow 2011; Nakamura and Lipton 2013) of the RyR could also be hypothesized, since SNAP trans-nitrosylates other substrates by de-nitrosylating itself [similar to the glutathione or thioredoxin systems (Sengupta and Holmgren 2013)]. The NOMel compound does not have this ability, because the NO-group is linked to the rest of the molecule through a nitrogen-nitrogen bond [N-nitrosomelatonin (Turjanski et al. 2000a)]. Therefore, we hypothesize that it is because of the nature of the NO released, and/or of the mechanism by which it is released, that NOMel can only potentiate the light-induced phase advances.
In conclusion, we have demonstrated the chronobiotic properties of NOMel, enhancing photic synchronization of circadian rhythms. Importantly, these are the first in-vivo experiments showing the NO-donor property of this drug. We propose that this drug can eventually be tested in other mammalian models, in order to evaluate its ability for the treatment of different types of circadian disorders which affect human health, such as those occurring in shift-work or jet-lag. Melatonin by itself has been reported to be adequate for treating and preventing jet lag in human trials (Herxheimer and Petrie 2002; Caspi 2004). We are introducing here the nitroso version of this compound, which has a promising future for studies aimed to treat circadian alterations.