Studying the Evolution of the Vertebrate Circadian Clock: The Power of Fish as Comparative Models.

The utility of any model species cannot be judged solely in terms of the tools and approaches it provides for genetic analysis. A fundamental consideration is also how its biology has been shaped by the environment and the ecological niche which it occupies. By comparing different species occupying very different habitats we can learn how molecular and cellular mechanisms change during evolution in order to optimally adapt to their environment. Such knowledge is as important as understanding how these mechanisms work. This is illustrated by the use of fish models for studying the function and evolution of the circadian clock. In this review we outline our current understanding of how fish clocks sense and respond to light and explain how this differs fundamentally from the situation with mammalian clocks. In addition, we present results from comparative studies involving two species of blind cavefish, Astyanax mexicanus and Phreatichthys andruzzii. This work reveals the consequences of evolution in perpetual darkness for the circadian clock and its regulation by light as well as for other mechanisms such as DNA repair, sleep, and metabolism which directly or indirectly are affected by regular exposure to sunlight. Major differences in the cave habitats inhabited by these two cavefish species have a clear impact on shaping the molecular and cellular adaptations to life in complete darkness.

Relaxed selective constraints drove functional modifications in peripheral photoreception of the cavefish P. andruzzii and provide insight into the time of cave colonization.

The genetic basis of phenotypic changes in extreme environments is a key but rather unexplored topic in animal evolution. Here we provide an exemplar case of evolution by relaxed selection in the Somalian cavefish Phreatichthys andruzzii that has evolved in the complete absence of light for at least 2.8 million years. This has resulted in extreme degenerative phenotypes, including complete eye loss and partial degeneration of the circadian clock. We have investigated the molecular evolution of the nonvisual photoreceptor melanopsin opn4m2, whose mutation contributes to the inability of peripheral clocks to respond to light. Our intra- and inter-species analyses suggest that the 'blind' clock in P. andruzzii evolved because of the loss of selective constraints on a trait that was no longer adaptive. Based on this change in selective regime, we estimate that the functional constraint on cavefish opn4m2 was relaxed at ∼5.3 Myr. This implies a long subterranean history, about half in complete isolation from the surface. The visual photoreceptor rhodopsin, expressed in the brain and implicated in photophobic behavior, shows similar evolutionary patterns, suggesting that extreme isolation in darkness led to a general weakening of evolutionary constraints on light-responsive mechanisms. Conversely, the same genes are still conserved in Garra barreimiae, a cavefish from Oman, that independently and more recently colonized subterranean waters and evolved troglomorphic traits. Our results contribute substantially to the open debate on the genetic bases of regressive evolution.

Transcriptional regulation of arylalkylamine-N-acetyltransferase-2 gene in the pineal gland of the gilthead seabream.

Pineal serotonin-N-acetyltransferase (arylalkylamine-N-acetyltransferase; AANAT) is considered the key enzyme in the generation of circulating melatonin rhythms; the rate of melatonin production is determined by AANAT activity. In all the examined species, AANAT activity is regulated at the post-translational level and, to a variable degree, also at the transcriptional level. Here, the transcriptional regulation of pineal aanat (aanat2) of the gilthead seabream (Sparus aurata) was investigated. Real-time polymerase chain reaction quantification of aanat2 mRNA levels in the pineal gland collected throughout the 24-h cycle revealed a rhythmic expression pattern. In cultured pineal glands, the amplitude was reduced, but the daily rhythmic expression pattern was maintained under constant illumination, indicating a circadian clock-controlled regulation of seabream aanat2. DNA constructs were prepared in which green fluorescent protein was driven by the aanat2 promoters of seabream and Northern pike. In vivo transient expression analyses in zebrafish embryos indicated that these promoters contain the necessary elements to drive enhanced expression in the pineal gland. In the light-entrainable clock-containing PAC-2 zebrafish cell line, a stably transfected seabream aanat2 promoter-luciferase DNA construct exhibited a clock-controlled circadian rhythm of luciferase activity, characteristic for an E-box-driven expression. In NIH-3T3 cells, the seabream aanat2 promoter was activated by a synergistic action of BMAL/CLOCK and orthodenticle homeobox 5 (OTX5). Promoter sequence analyses revealed the presence of the photoreceptor conserved element and an extended E-box (i.e. the binding sites for BMAL/CLOCK and OTX5 that have been previously associated with pineal-specific and rhythmic gene expression). These results suggest that seabream aanat2 is a clock-controlled gene that is regulated by conserved mechanisms.

Zebrafish arylalkylamine-N-acetyltransferase genes - targets for regulation of the circadian clock.

Daily rhythms of melatonin production are controlled by changes in the activity of arylalkylamine-N-acetyltransferase (AANAT). Zebrafish possess two aanats, aanat1 and aanat2; the former is expressed only in the retina and the latter is expressed in both the retina and the pineal gland. Here, their differential expression and regulation were studied using transcript quantification and transient and stable in vivo and in vitro transfection assays. In the pineal gland, the aanat2 promoter exhibited circadian clock-controlled activity, as indicated by circadian rhythms of Enhanced green fluorescent protein (EGFP) mRNA in AANAT2:EGFP transgenic fish. In vivo transient expression analyses of the aanat2 promoter indicated that E-box and photoreceptor conserved elements (PCE) are required for expression in the pineal gland. In the retina, the expression of both genes was characterized by a robust circadian rhythm of their transcript levels. In constant darkness, the rhythmic expression of retinal aanat2 persisted while the aanat1 rhythm disappeared; indicating that the former is controlled by a circadian clock and the latter is also light driven. In the light-entrainable clock-containing PAC-2 zebrafish cell line, both stably transfected aanat1 and aanat2 promoters exhibited a clock-controlled circadian rhythm, characteristic for an E-box-driven expression. Transient co-transfection experiments in NIH-3T3 cells revealed that the two, E-box- and PCE-containing, promoters are driven by the synergistic action of BMAL/CLOCK and orthehodenticle homeobox 5. This study has revealed a shared mechanism for the regulation of two related genes, yet describes their differential phases and photic responses which may be driven by other gene-specific regulatory mechanisms and tissue-specific transcription factor profiles.

Flies and fish: birds of a feather.

The identification of specific clock-containing structures has been a major endeavour of the circadian field for many years. This has lead to the identification of many key components of the circadian system, including the suprachiasmatic nucleus in mammals, and the eyes and pineal glands in lower vertebrates. However, the idea that these structures represent the only clocks in animals has been challenged by the discovery of peripheral pacemakers in most organs and tissues, and even a number of cell lines. In Drosophila, and vertebrates such as the zebrafish, these peripheral clocks appear to be highly autonomous, being set directly by the environmental light/dark cycle. However, a hierarchy of clocks may still exist in mammals. In this review, we examine some of the current views regarding peripheral clocks, their organization and how they are entrained.

A clockwork organ.

The vertebrate circadian clock was thought to be highly localized to specific anatomical structures: the mammalian suprachiasmatic nucleus (SCN), and the retina and pineal gland in lower vertebrates. However, recent findings in the zebrafish, rat and in cultured cells have suggested that the vertebrate circadian timing system may in fact be highly distributed, with most if not all cells containing a clock. Our understanding of the clock mechanism has progressed extensively through the use of mutant screening and forward genetic approaches. The first vertebrate clock gene was identified only a few years ago in the mouse by such an approach. More recently, using a syntenic comparative genetic approach, the molecular basis of the the tau mutation in the hamster was determined. The tau gene in the hamster appears to encode casein kinase 1 epsilon, a protein previously shown to be important for PER protein turnover in the Drosophila circadian system. A number of additional clock genes have now been described. These proteins appear to play central roles in the transcription-translation negative feedback loop responsible for clock function. Post-translational modification, protein dimerization and nuclear transport all appear to be essential features of how clocks are thought to tick.

Asynchronous oscillations of two zebrafish CLOCK partners reveal differential clock control and function.

Most clock genes encode transcription factors that interact to elicit cooperative control of clock function. Using a two-hybrid system approach, we have isolated two different partners of zebrafish (zf) CLOCK, which are similar to the mammalian BMAL1 (brain and muscle arylhydrocarbon receptor nuclear translocator-like protein 1). The two homologs, zfBMAL1 and zfBMAL2, contain conserved basic helix-loop-helix-PAS (Period-Arylhydrocarbon receptor-Singleminded) domains but diverge in the carboxyl termini, thus bearing different transcriptional activation potential. As for zfClock, the expression of both zfBmals oscillates in most tissues in the animal. However, in many tissues, the peak, levels, and kinetics of expression are different between the two genes and for the same gene from tissue to tissue. These results support the existence of independent peripheral oscillators and suggest that zfBMAL1 and zfBMAL2 may exert distinct circadian functions, interacting differentially with zfCLOCK at various times in different tissues. Our findings also indicate that multiple controls may be exerted by the central clock and/or that peripheral oscillators can differentially interpret central clock signals.

Rhythmic transcription: the molecular basis of oscillatory melatonin synthesis.

Pulsatile hormone synthesis and secretion are characteristic features of various oscillatory biological systems. Circadian rhythms are critical in the regulation of most physiological functions, and much interest has been centred on the understanding of the molecular mechanisms governing them. Adaptation to a changing environment is an essential feature of physiological regulation. The day-night rhythm is translated into hormonal oscillations governing the metabolism of all living organisms. In mammals the pineal gland is responsible for the circadian synthesis of the hormone melatonin in response to signals originating from the endogenous clock located in the hypothalamic suprachiasmatic nucleus (SCN). The molecular mechanisms involved in rhythmic synthesis of melatonin involve the CREM gene, which encodes transcription factors responsive to activation of the cAMP signalling pathway. The CREM product, ICER, is rhythmically expressed and participates in a transcriptional autoregulatory loop which also controls the amplitude of oscillations of serotonin N-acetyl transferase, the rate-limiting enzyme of melatonin synthesis. Thus, a transcription factor modulates the oscillatory levels of a hormone.

Light acts directly on organs and cells in culture to set the vertebrate circadian clock.

The expression of clock genes in vertebrates is widespread and not restricted to classical clock structures. The expression of the Clock gene in zebrafish shows a strong circadian oscillation in many tissues in vivo and in culture, showing that endogenous oscillators exist in peripheral organs. A defining feature of circadian clocks is that they can be set or entrained to local time, usually by the environmental light-dark cycle. An important question is whether peripheral oscillators are entrained to local time by signals from central pacemakers such as the eyes or are themselves directly light-responsive. Here we show that the peripheral organ clocks of zebrafish are set by light-dark cycles in culture. We also show that a zebrafish-derived cell line contains a circadian oscillator, which is also directly light entrained.

PASting together the mammalian clock.

Over the past year, the first components of the mammalian clock have been identified; Clock, bmal1 and three homologs of Drosophila period have been cloned, all of which encode PAS proteins. Expression of the mammalian period gene oscillates in many tissues in vivo and in immortalized cell cultures in vitro. Now, can we say that every cell has a circadian clock?

Stress-induced expression of transcriptional repressor ICER in the adrenal gland.

Second messenger cyclic AMP plays a central role in signalling within the hypothalamo-pituitary-adrenal (HPA) axis. Changes in gene expression are central to long-term adaptations made in response to stress in the adrenal gland. Here we demonstrate that expression of the cAMP inducible transcriptional repressor, ICER (Inducible cAMP Early Repressor), is rapidly and powerfully induced in response to surgical stress in the rat adrenal gland. Hypophysectomisation blocks stress-induced ICER expression. Finally we demonstrate that injection of the pituitary hormone ACTH (Adrenocorticotropin Hormone) induces robust ICER expression in the adrenal cortex. Thus, induction of the transcriptional repressor ICER is coupled to the HPA axis response to stress.

Zebrafish Clock rhythmic expression reveals independent peripheral circadian oscillators.

The only vertebrate clock gene identified by mutagenesis is mouse Clock, which encodes a bHLH-PAS transcription factor. We have cloned Clock in zebrafish and show that, in contrast to its mouse homologue, it is expressed with a pronounced circadian rhythm in the brain and in two defined pacemaker structures, the eye and the pineal gland. Clock oscillation was also found in other tissues, including kidney and heart. In these tissues, expression of Clock continues to oscillate in vitro. This demonstrates that self-sustaining circadian oscillators exist in several vertebrate organs, as was previously reported for invertebrates.

Rhythmic transcription: the molecular basis of circadian melatonin synthesis.

Adaptation to a changing environment is an essential feature of physiological regulation. The day-night rhythm is translated into hormonal oscillations governing the metabolism of all living organisms. In mammals the pineal gland is responsible for the synthesis of the hormone melatonin in response to signals originating from the endogenous clock located in the hypothalamic suprachiasmatic nucleus (SCN). The molecular mechanisms involved in rhythmic synthesis of melatonin involve the cAMP response element modulator (crem) gene, which encodes transcription factors responsive to activation of the cAMP signalling pathway. The CREM product, inducible cAMP early repressor (ICER), is rhythmically expressed and participates in a transcriptional autoregulatory loop that also controls the amplitude of oscillations of 5-HT N-acetyl transferase, the rate-limiting enzyme of melatonin synthesis. Thus, a transcription factor modulates the oscillatory levels of a hormone.

Ectopic ICER expression in pituitary corticotroph AtT20 cells: effects on morphology, cell cycle, and hormonal production.

The products of the cAMP response element modulator (CREM) gene play an important role in the transcriptional response to cAMP in endocrine cells. By virtue of an alternative, intronic promoter within the gene, the inducible cAMP early repressor (ICER) isoform is generated. ICER was shown to act as a dominant negative regulator and to be cAMP-inducible in various neuroendocrine cells and tissues. ICER negatively autoregulates its own expression and has been postulated to participate in the molecular events governing oscillatory hormonal regulations. To elucidate ICER function in pituitary physiology, we have generated AtT20 corticotroph cell lines expressing the sense or antisense ICER transcript under the control of the cadmium-inducible human methallothionein IIA promoter. Here we demonstrate that changes in the regulated levels of ICER have drastic consequences on the physiology of the corticotrophs. Ectopic ICER expression induces remarkable modifications in AtT20 morphology. Cells with persistent, nonregulated high levels of ICER are blocked in the G2/M phase of the cell cycle, while the opposite effect is obtained in cells expressing an antisense ICER transcript. We show that the effect of ICER on the AtT20 cell cycle is correlated to a direct down-regulation of the cyclin A gene promoter by ICER. Finally, we show that ACTH hormonal secretion from the corticotrophs is completely blocked by ICER ectopic expression. Interestingly, this effect is not due to a direct regulation of the POMC gene, but is mediated by a transcriptional control of the prohormone convertase 1 gene. These results point to a key regulatory function of CREM in pituitary physiology.

The transcriptional repressor ICER and cAMP-induced programmed cell death.

The cAMP pathway plays a central role in the response to hormonal signals for cell proliferation, differentiation and apoptosis. In IPC-81 leukaemia cells, activation of the cAMP pathway by prostaglandin E1 treatment, or other cAMP-elevating agents, induces apoptosis within 4-6 h. Inhibition of mRNA or protein synthesis during the first 2 h of cAMP induction protects cells from apoptosis, suggesting a requirement for early gene expression. cAMP-dependent protein kinase phosphorylates a class of nuclear factors and thereby regulates the transcription of a specific set of genes. Here we show that CREM (cAMP Responsive Element Modulator) expression is induced rapidly upon prostaglandin E1 treatment of IPC-81 cells. The induced transcripts correspond to the early product ICER (Inducible cAMP Early Repressor). ICER expression remains elevated until the burst of cell death. Protein synthesis inhibitors which prevent cAMP-induced apoptosis also block de novo ICER synthesis. Transfected IPC-81 cell lines, constitutively expressing high level of ICER are resistant to cAMP-induced cell death. In these transfected cells, cAMP fails to upregulate the ICER transcripts demonstrating that ICER exerts strongly its repressor function on CRE-containing genes. That an early expression of ICER blocks apoptosis, suggests that gene repression by endogenous ICER in IPC-81 is insufficient or occurs too late to protect cells against death. ICER transfected cells rescued from cAMP-induced apoptosis are growth arrested. It shows for the first time that CREM activation directly participates to the decision of the cell to die. ICER, by sequentially repressing distinct sets of CRE-containing genes could modulate cell fate.

Cyclic AMP signalling pathway and cellular proliferation: induction of CREM during liver regeneration.

The CREM gene encodes both activators and repressors of cAMP-induced gene expression. An isoform of CREM encodes the powerful transcriptional repressor ICER (Inducible cAMP Early Repressor), which has been shown to be inducible by virtue of an alternative, intronic promoter. The CREM gene belongs to the early response class and displays a characteristic neuroendocrine cell- and tissue-specific expression. To date ICER inducibility has been described in non-replicating, terminally differentiated tissues. In this paper we document a robust induction of CREM expression in the regenerating rat liver after partial hepatectomy. This represents the first link of inducible CREM expression to the phenomenon of cellular proliferation. Furthermore, it represents the first example of transcriptional activation of a cAMP-responsive factor in the regenerating liver. This has significant physiological relevance since the adenylate cyclase signalling pathway is strongly implicated in liver regeneration. Finally, we show that the repressor ICER is inducible in the hepatoma cell line H35 upon activation of the adenylate cyclase and phosphorylation of the activator CREB.

Rhythmic transcription: the molecular basis of circadian melatonin synthesis.

Adaptation to a changing environment is an essential feature of physiological regulation. The day/night rhythm is translated into hormonal oscillations governing the physiology of all living organisms. In mammals the pineal gland is responsible for the synthesis of the hormone melatonin in response to signals originating from the endogenous clock located in the hypothalamic suprachiasmatic nucleus (SCN). The molecular mechanisms involved in rhythmic synthesis of melatonin involve the CREM gene, which encodes transcription factors responsive to activation of the cAMP signalling pathway. The CREM product, ICER, is rhythmically expressed and participates in a transcriptional autoregulatory loop which also controls the amplitude of oscillations of serotonin N-acetyl transferase (AANAT), the rate-limiting enzyme of melatonin synthesis. In contrast, chick pinealocytes possess an endogenous circadian pacemaker which directs AANAT rhythmic expression. cAMP-responsive activator transcription factors CREB and ATF1 and the repressor ICER are highly conserved in the chick with the notable exception of ATF1 that possesses two glutamine-rich domains in contrast to the single domain encountered to date in mammalian systems. ICER is cAMP inducible and undergoes a characteristic day-night oscillation in expression reminiscent of AA-NAT, but with a peak towards the end of the night. Interestingly CREB appears to be phosphorylated constitutively with a transient fall occurring at the beginning of the night. Thus, a transcription factor modulates the oscillatory levels of a hormone.