Tesi sul tema "Circadian clock"

Cyanobakterien zŠhlen zu den Šltesten Lebewesen auf der Erde. Diese Bakterien, auch Blaualgen genannt, trugen wesentlich zur Sauerstoffanreicherung der Erde bei, da sie eine ausgeprŠgte FŠhigkeit zur Photosynthese besitzen. Der produzerte Sauerstoff der Photosynthese hemmt jedoch eine weitere AktivitŠt von Cyanobakterien, die Stickstofffixierung. Um die Hemmung zu vermeiden, werden diese AktivitŠten zeitlich getrennt und optimal dem tŠglichen Hell-Dunkel-Rhythmus angepasst. Ein evolutionŠrer Vorteil wird erzielt, wenn der Organismus diesen Rhythmus antizipiert und sich darauf vorbereitet. Aus diesem Grund haben Cyanobakterien eine innere Uhr entwickelt, deren Rhythmus zirkadian ist, ãzirka diemÒ bedeutet ãungefŠhr ein TagÒ. Cyanobakterien der Spezies Synechococcus elongatus PCC 7942 haben sich als Modellorganismus etabliert, weil in ihnen die ersten bakteriellen zirkadianen Oszillationen auf molekularer Ebene entdeckt worden sind. Ihre zirkadiane Uhr entspringt dreier, auf der DNS beieinanderliegenden, Gene (kaiA, kaiB, kaiC) und ihrer dazugehšrigen Proteine. Phosphorylierte KaiC-Proteine Ÿben eine RŸckkopplung auf die Transkription von kaiB und kaiC aus, wodurch die AktivitŠt des kaiBC-Promotors zirkadian oszilliert. Eines der wichtigsten Experimente der letzten Jahre hat gezeigt, dass dieser Transkriptions-Translations-Oszillator mit einem weiteren Oszillator gekoppelt ist, der nicht von Transkription und Translation abhŠngt. Das Experiment des Kondo Labors rekonstruiert zirkadiane Oszillationen mit nur drei Proteinen KaiA, KaiB, KaiC und ATP. Die Proteine bilden Komplexe verschiedener Stoichiometrie, die durchschnittliche Phosphorylierung des Proteins KaiC zeigt stabile Oszillationen mit einer zirkadianen Periode. Da ein Entfernen von einem der Proteine zum Verlust der Oszillationen fŸhrt, wird dieser Post-Translations-Oszillator auch als Kernoszillator bezeichnet. Der Phosphorylierungszyklus von KaiC wird bestimmt durch fortlaufende Phosphorylierung und Dephosphorylierung an zwei Positionen des Proteins, den AminosŠuren Serin 431 und Threonin 432. Die Phase des Kernoszillators kann an der Verteilung der vier PhosphorylierungszustŠnde (nicht-, serin-, threonin- und doppeltphosphoryliert) abgelesen werden. KaiC wechselwirkt mit KaiA und KaiB, damit verschieden phosphorylierte KaiC synchronisieren und die Uhr Ÿber mehrere Tage konstante Oszillationen zeigt. Die Details dieser Wechselwirkung sind jedoch unbekannt. In dieser Dissertation erstelle ich ein mathematisches Modell des Kernoszillators und simuliere die vorliegenden Experimente des O''Shea Labors. Die Simulation reproduziert den KaiC Phosphorylierungszyklus der Uhr quantitativ. Um die wichtigsten experimentellen Nebenbedingungen zu erfŸllen, muss das theoretische Modell zwei molekulare Eigenschaften von KaiC berŸcksichtigen, wodurch ich wichtige Vorhersagen treffe. Die erste Nebenbedingung ist durch die Robustheit des Systems gegeben. Die KaiC-Phosphorylierung Šndert sich nicht, wenn die Gesamtkonzentrationen der drei Proteine in gleicher Weise variiert werden. Um diese Bedingung zu erfŸllen, muss das Modell zwei verschiedenartige Komplexe von KaiA und KaiC berŸcksichtigen. ZusŠtzlich zu einem KaiAC Komplex, der die Autophosphorylierung von KaiC unterstŸtzt, muss KaiC den grš§ten Teil von KaiA unabhŠngig vom Phosphorylierungszustand sequestrieren. Diese zweite Bindestelle ist meine erste theoretische Vorhersage. Die zweite Nebenbedingung ist durch das Ÿbergangsverhalten nach Hinzugabe von KaiB gegeben. KaiB induziert eine Dephosphorylierung von KaiC, die abhŠngig vom Phosphorylierungsniveau ist. Ein Umschalten zwischen phosphoylierendem und dephosphorylierendem KaiC ist deshalb nur in bestimmten Zeitfenstern mšglich. Um die gemessenen Zeitfenster in der Simulation zu reproduzieren, postuliere ich im Modell, dass sechsfach Serin phosphorylierte KaiBC Komplexe KaiA inaktivieren. Diese hochgradig nichtlineare RŸckkopplung ist meine zweite theoretische Vorhersage. Die beiden Vorhersagen werden anschlie§end experimentell ŸberprŸft. HierfŸr werden aufgereinigte Kai-Proteine mit ATP gemischt. Proben an ausgewŠhlten Zeitpunkten werden mit der nativen Massenspektrometrie untersucht. Diese ist eine neuartige Methode, die es erlaubt, intakte Proteinkomplexe zu untersuchen. Die Spektren bestŠtigen sowohl die zweite KaiAC-Bindestelle als auch die nichtlineare RŸckkopplung. Das mathematische Modell erlaubt es au§erdem, die drei definierenden Prinzipien von zirkadianen Uhren fŸr den Kernoszillator zu erklŠren. Erstens sichern konstante Phosphorylierungs- und Dephosphorylierungsraten von KaiC und ein pŸnktliches Umschalten zwischen beiden Phasen den Freilauf des Oszillators. Dieser Freilauf bewirkt, dass die zirkadiane Uhr auch unter konstanten Bedingungen, vor allem gleichbleibenden LichtverhŠltnissen, weiterlaufen kann. Zweitens muss die Periodendauer des Oszillators zu unterschiedlichen Šu§eren Bedingungen erhalten bleiben (Temperaturkompensation). Diese Bedingung wird realisiert, indem temperaturabhŠngige Dissoziationskonstanten von KaiAC und KaiBC Komplexen Phasenverschiebungen erzeugen, die sich gegenseitig kompensieren. Drittens muss die Phase des Oszillators sich dem Tagesrhythmus anpassen kšnnen. Diese Anpassung folgt aus einem Šu§eren Warm-Kalt-Rhythmus, der die drei temperaturabhŠngigen Phasenverschiebungen nur zum Teil einschaltet und damit die Kompensation verhindert. Eine in silico Evolutionsanalyse zeigt, dass eine zweite phosphorylierbare AminosŠure einen evolutionŠren Vorteil bringt und die Verteilung der PhosphorylierungszustŠnde optimiert ist, um eindeutig die Zeit zu bestimmen. Das Ergebnis weist darauf hin, dass diese Verteilung die physiologisch wichtige Ausgangsgrš§e der Uhr ist und die vier PhosphroylierungszustŠnde die Funktionen der zirkadianen Uhr von Cyanobakterien sichern.
Biological activities in cyanobacteria are coordinated by an internal clock. The rhythm of the cyanobacterium Synechococcus elongatus PCC 7942 originates from the kai gene cluster and its corresponding proteins. In a test tube, the proteins KaiA, KaiB and KaiC form complexes of various stoichiometry and the average phosphorylation level of KaiC exhibits robust circadian oscillations in the presence of ATP. The characteristic cycle of individual KaiC proteins is determined by phosphorylation of serine 431 and threonine 432. Differently phosphorylated KaiC synchronize due to an interaction with KaiA and KaiB. However, the details of this interaction are unknown. Here, I quantitatively investigate the experimentally observed characteristic phosphorylation cycle of the KaiABC clockwork using mathematical modeling. I thereby predict the binding properties of KaiA to both KaiC and KaiBC complexes by analyzing the two most important experimental constraints for the model. In order to reproduce the KaiB-induced dephosphorylation of KaiC a highly non-linear feedback loop has been identified. This feedback originates from KaiBC complexes, which are exclusively phosphorylated at the serine residue. The observed robustness of the KaiC phosphorylation level to concerted changes of the total protein concentrations demands an inclusion of two KaiC binding sites to KaiA in the mathematical model. Besides the formation of KaiAC complexes enhancing the autophosphorylation activity of KaiC, the model accounts for a KaiC binding site, which constantly sequestrates a large fraction of free KaiA. These theoretical predictions have been confirmed by the novel method of native mass spectrometry, which was applied in collaboration with the Heck laboratory. The mathematical model elucidates the mechanism by which the circadian clock satisfies three defining principles. First, the highly non-linear feedback loop assures a rapid and punctual switch to dephosphorylation which is essential for a precise period of approximately 24 h (free-running rhythm). Second, the dissociation of the protein complexes increases with increasing temperatures. These perturbations induce opposing phase shifts, which exactly compensate during one period (temperature compensation). Third, a shifted external rhythm of low and high temperature affects only a part of the three compensating phase perturbations, which leads to phase shifts (phase entrainment). An in silico evolution analysis shows that the existing second phosphorylatable residue of KaiC is not necessary for the existence of sustained oscillations but provides an evolutionary benefit. The analysis demonstrates that the distribution of four phosphorylated states of KaiC is optimized in order for the organism to uniquely distinguish between dusk and dawn. Consequently, this thesis emphasizes the importance of the four phosphorylated states of KaiC, which assure the outstanding performance of the core oscillator.

Circadian clocks are intrinsic, time-tracking systems that enable organisms to maintain their physiological state and their synchrony with the 24-hour rotation of the Earth, by partitioning behavioural and metabolic processes according to time of day within each tissue. They are entrained to the external environment by light/dark cycles and by food timing, which act as clock synchronizers. Emerging evidence suggests that circadian regulation is intimately linked to metabolic homeostasis and that dysregulation of circadian rhythms can contribute to disease. Conversely, metabolic signals also feed back into the circadian system, modulating circadian gene expression and behaviour. Conventional experimental approach of circadian clock in vitro studies is based on single-pulse stimulation of only one metabolite or hormone, while in vivo peripheral tissues are exposed to periodic oscillating stimuli of a large number of metabolites and hormones, whose variations are in most cases interconnected, as for example glucose and insulin. Moreover, only one or few clock genes are generally considered, while it is known that a large number of genes, thus biological processes, are under circadian regulation. Therefore, this Ph.D. research work is aimed at the development of technologies and data analysis tools to investigate the entrainment of peripheral mammalian circadian clock to frequency-encoded metabolic stimuli, which well mimic physiological oscillations at which peripheral tissues are exposed in vivo. Technologies, and, more specifically, microtechnologies have been developed to investigate the effects of periodic metabolic entrainment, showing that in murine fibroblasts oscillatory periodic metabolic stimulations entrain the expression of Per2, one of the core genes of the circadian molecular mechanism. Moreover, it has been proven that only by metabolic oscillations it is possible to completely reset the phase of cell-autonomous clocks. In order to develop a physiological and pathological in vitro model, achieving a high spatio-temporal control of cell culture microenvironment, frequency-encoded perturbations have been automated in a newly designed microfluidic platform for circadian applications. Finally, to broaden the description of genes expressed with a circadian temporal pattern, a new data analysis method has been proposed and characterized, that allows to identify circadian genes in whole transcriptome data, to group genes based on the phase of their expression, to visualize transcriptome data at a glance and clearly identifying modifications at the transcriptome level from one biological condition to another one.
I ritmi circadiani sono meccanismi biologici di organizzazione temporale intrinseci e autosostenuti, che consentono agli organismi di anticipare i cambiamenti ambientali e permettono loro di adattare il loro comportamento e la loro fisiologia nell’arco della giornata. L’orologio circadiano è sincronizzato dai cicli luce/buio e dall’ora dei pasti. La funzione biologica essenziale del ritmo circadiano è mantenere lo stato fisiologico dell’organismo e la sua sincronia comportamentale e metabolica con l’ambiente esterno. Recentemente è stato dimostrato che l’orologio circadiano garantisce il mantenimento dell’omeostasi metabolica, e che una distruzione del ritmo circadiano è causa di numerose malattie. L’approccio sperimentale convenzionale per lo studio dell’orologio circadiano in vitro è basato su una singola stimolazione di un solo metabolita o ormone, mentre in vivo i tessuti sono esposti in continuo a stimoli oscillatori periodici di una grande vastità di metaboliti e ormoni, le cui variazioni sono spesso interconnesse, come nel caso di glucosio e insulina. Inoltre, nell’analisi sperimentale convenzionale, sono studiati solo uno o pochi geni noti per essere implicati nell’orologio circadiano, mentre è noto che un elevato numero di geni sono espressi in modo circadiano. Lo scopo di questo progetto di ricerca è quindi sviluppare tecnologie e metodi di analisi per studiare l’effetto di stimoli metabolici in frequenza sull’orologio circadiano di tessuti periferici. Questi stimoli riproducono infatti in vitro le oscillazioni metaboliche a cui i tessuti sono esposti in vivo. Tecnologie, e più nello specifico, microtecnologie sono state sviluppate per studiare gli effetti di stimoli metabolici oscillatori, ed è stato dimostrato che in fibroblasti murini l’espressione di Per2 (uno dei geni principali del meccanismo molecolare dell’orologio circadiano) è sincronizzata da stimoli metabolici oscillatori. Inoltre, è stato dimostrato che le oscillazioni metaboliche sono di per sé sufficienti per allineare l’orologio circadiano nei tessuti periferici. Per sviluppare un modello che riproducesse in vitro condizioni sia fisiologiche che patologiche, raggiungendo un controllo spazio-temporale preciso del microambiente cellulare, le stimolazioni in frequenza sono state automatizzate in un dispositivo microfluidico progettato in modo dedicato per studi del ritmo circadiano. Infine, per estendere lo studio ai geni espressi con un pattern temporale circadiano, un nuovo metodo di analisi è stato proposto e caratterizzato. Il metodo permette di identificare geni circadiani da dati di trascrittomica, di suddividere i geni basandosi sulla fase della loro espressione, di visualizzare dati di trascrittomica nel loro complesso e di individuare rapidamente e in modo semplice modifiche a livello trascrizionale da una condizione biologica ad un’altra.

Cette thèse a étudié comment une dérégulation de l'horloge circadienne, qui n'avait pas été clairement associée à une maladie rétinienne jusqu'à présent, pourrait contribuer à la dégénérescence et influencer le développement et la fonction de la rétine. L'inactivation spécifique du gène horloge Bmal1 (rod-Bmal1KO) dans la lignée de souris portant la mutation P23H de la rhodopsine aggrave les symptômes de dégénérescence rétinienne, tels que la réduction de la réponse ERG et la perte de bâtonnets, induits par la seule mutation P23H. Ces observations ont été corroborées par l'analyse RNA-Seq qui a révélé des changements majeurs dans l'expression des gènes, liés à la phototransduction et aux processus métaboliques, entre le double mutant (rod-Bmal1KO/P23H) et les rétines P23H. Nous avons montré qu'au cours du développement, l’invalidation des gènes horloge Per1 et Per2 chez la souris affecte de manière significative l'expression des gènes de la phototransduction et du cycle cellulaire. Nous avons observé que les souris adultes déficientes en Per1 et Per2 ne modulent pas quotidiennement leur sensibilité à la lumière, dans des conditions scotopiques et mésopiques. Nous avons également constaté une altération de la régulation journalière de la sensibilité à la lumière chez les souris déficientes en gène d'horloge Bmal1 dans les bâtonnets. De plus, nous avons investigué comment la dégénérescence des bâtonnets pourrait influencer la capacité rythmique globale de la rétine en mesurant les rythmes de bioluminescence PER2::LUC chez des souris P23H. Nos résultats montrent que l'horloge rétinienne chez les souris hétérozygotes P23H/+ présente des rythmes circadiens avec une robustesse et une amplitude significativement accrues. Ces effets impliquent probablement l’activation des cellules gliales
This thesis investigated how circadian clock misregulation, which has not been clearly associated with retinal genetic disease so far, could contribute to degeneration and influence development and function in the retina. The rod-specific knockout of Bmal1 clock gene (rod-Bmal1KO) from the mouse line carrying the P23H mutation of rhodopsin exacerbated the retinal degeneration phenotypes, such as reduction in ERG response and rods loss, induced by the P23H mutation alone. These observations were corroborated by RNA-Seq analysis, where we found major changes in expression of genes related to phototransduction and metabolic processes, between the (rod-Bmal1KO/P23H) double mutant and P23H retinas. We showed that during development, Per1 and Per2 clock genes deficiency in mice significantly affects gene expression of phototransduction and cell cycle components. We found that adult mice deficient for Per1 and Per2 genes lack a daily modulation of light sensitivity, under scotopic and mesopic conditions. We also found an impaired daily modulation of light sensitivity in mice deficient for Bmal1 clock gene in rods. Additionally, we investigated how rod degeneration could impact on the global rhythmic capacity of the retina by measuring PER2::LUC bioluminescence rhythms in P23H mice. We showed that the retinal clock in P23H/+ heterozygous mice displays circadian rhythms with significantly increased robustness and amplitude. These effects likely involve activation of glial cells

Circadian and photoperiodic phenomena serve to organize the temporal pattern of various biological processes. While the former generates endogenous daily rhythms, the latter is related to seasonality. In Drosophila melanogaster, the gene timeless (tim) encodes a cardinal component of the circadian clock and also contributes to photoperiodism, which is observed as an adult reproductive diapause. In this work, natural tim variants were examined for diapause across different temperatures and photoperiods. The newly derived allele, ls-tim, exhibited consistently higher diapause levels than the ancestral one, s-tim, implicating a putative adaptive advantage in the seasonal European environment and providing a perfect substrate for the recently proposed scenario of directional selection. To investigate further genetic links between circadian and photoperiodic mechanisms, classical clock mutations and transgenes were placed on a natural congenic background and assayed for locomotor activity behaviour and diapause response. Surprisingly, the results not only highlighted the importance of tim, and its natural alleles, but also revealed the participation of other clock components in diapause, suggesting that both daily and seasonal timers might have molecularly coevolved. The phenotypic effects promoted by ls-tim arise from the protein isoform LTIM, which expresses an additional N-terminal fragment. To study the adaptive significance of the N-terminal residues, including putative phosphorylation sites, a number of mutagenized TIM constructs were generated and functionally analysed. At the molecular level, it was demonstrated that both the N-terminus length and the order of its residues are important variables modulating the interaction dynamics between TIM and CRYPTOCHROME (CRY). At the behaviour level, the overall amino acid composition, rather than a particular order, appeared to be more critical for the phaseshift responses. Interestingly, despite the functional importance of the N-terminus, a deletion mapping analysis revealed that CRY directly binds to a protein sequence located at TIM C-terminus.

It is well established that elderly individuals have increased difficulty sleeping at night combined with falling asleep and waking up earlier. Although these age-related declines in circadian output are clearly observable in activity recordings of laboratory animals, the underlying changes in molecular and neuronal activity remain unknown. The fruit fly, Drosophila melanogaster, has long been used as a model for studying the circadian system and for ageing research. In this thesis Drosophila was used as a model to study the effect of ageing on circadian and sleep behaviour. Circadian behaviour was measured using the Drosophila Activity Monitoring system, recording activity of flies at various stages of the ageing process, demonstrating a linear decline in rhythm strength with age combined with an increase in period length. Weakened circadian output is combined with significant alterations of diurnal behaviour of Drosophila, namely a reduction in morning and evening anticipatory behaviour. Ageing also has a significant impact on sleep behaviour, significantly increasing sleep duration whilst reducing latency, with larger effects observed on day- time sleep. Age-related changes in neuronal activity were investigated using whole-cell patch clamp electrophysiology to record from large lateral ventral (l-LNV) clock neurons, finding that ageing was associated with a significant decrease in input resistance, but no significant changes in spontaneous electrical activity or membrane potential. Manipulating the electrical properties of the circadian system by knocking down expression of candidate ion channels in all clock neurons had significant effects on behaviour, linking electrical activity with clock outputs. The results presented in this thesis demonstrate the suitability of Drosophila as a model to interrogate how ageing effects the circadian clock, identifying Alterations in the electrical properties of the l-LNV neurons may underlie observed changes in diurnal activity and sleep, while decreased remodelling of the s-LNV neurons can explain weakened circadian behaviour.

Obesity, metabolic syndrome, and diabetes represent a major source of morbidity and mortality in the United States and worldwide. Chronic misalignment of an organism’s internal circadian clock with diurnal, cyclic changes in the external environment, prevalent in professions that require shift work, contributes significantly to Type 2 Diabetes development. Experimentally, only short-term models of circadian disruption have been explored. Therefore, the goal of this study was to establish an animal model of chronic circadian disruption, which would more closely mimic the harmful misalignment associated with metabolic syndrome in clinical studies. Moreover, since high fat diet consumption alters circadian behavior and rhythmic gene expression, contributing to the diet-induced phenotype, I hypothesized that chronic circadian disruption interacts with a high fat diet to worsen metabolic syndrome. To investigate circadian misalignment and diet-induced metabolic syndrome, I examined the contribution of the Aryl Hydrocarbon Receptor (AhR). AhR has similar PAS domain containing motifs as circadian clock proteins allowing for protein/protein interactions and crosstalk between AhR signaling and circadian rhythms. Furthermore, AhR activation is implicated in Type 2 Diabetes risk. To examine chronic circadian disruption, male wild-type (WT; C57Bl/6J) and AhR +/- mice were entrained to 12/12-hour light/dark cycles where lights were on from 10pm-10am and off from 10am-10pm. Misalignment was initiated by delaying the time of lights on by 8 hours on Monday. Mice were exposed to the misalignment schedule Monday-Friday then returned to the entrainment schedule Saturday and Sunday to mimic readjustment to society during the weekend. Circadian misaligned mice were exposed to the altered light schedule for 15 weeks and control animals remained on the12/12-hour light/dark cycle. Mice were fed a normal chow diet (10% fat) or a high fat diet (60% fat). Animals were sacrificed and samples were collected at 4-hour intervals on day 2 of the weekend. Exposure to chronic circadian misalignment by light disruption or high fat diet altered circadian rhythms of behavior, metabolic outputs, and expression of circadian clock, clock-controlled nuclear receptor, and lipid metabolism genes. A combination of light misalignment and high fat diet exacerbated the effects of either treatment alone further disrupting behavior, enhancing % body fat and fasting glucose, and dampening circadian clock gene expression. AhR +/- mice also were protected from the metabolic consequences of chronic misalignment and a high fat diet by resistance to altered behavioral and molecular circadian rhythms and disruption of metabolic outputs. With metabolic syndrome and Type 2 Diabetes occurrence on the rise, it is important to understand all contributing factors, including circadian disruption. Differences between chronic circadian misalignment and high fat diet-induced obesity in WT and AhR +/- mice furthers our understanding of the complex mechanisms that underlie Type 2 Diabetes development and advocates the discovery of potential therapeutic targets for the development of novel treatment options.

The circadian clock is an endogenous time-keeping mechanism that allows an organism to coordinate its biology with the 24 hour variations in its external environment. Epidemiological studies have linked clock disruption to an increase incidence of cancer, in particular breast malignancy. On a molecular level, clock components have been shown to regulate cell cycle gene expression and its progression in a number of models. This thesis set out to further dissect the link between these two important systems. Within the zebrafish cell-line, PAC2, mitosis was demonstrated to be under circadian control via clock regulation of the cell cycle mediator, Cyclin B1. Techniques used were then translated into a human cell-line model to study species specific clock function, with particular reference to breast epithelial tissue. Glucocorticoids, putative clock synchronisation agents in vivo, were observed to induce cellular clock synchronisation in HEK 293 cells and the benign breast epithelial cell-line MCF10A. Clock disruption inhibited cell growth. Study of the breast epithelial cell cycle mutant, MDA-MB-231, demonstrated a functional clock, revealing no reciprocal regulation. In contrast, decreased expression levels of clock gene and putative tumour suppressor, Per1, were observed within the malignant breast epithelial cell-line, MCF7, leading to greatly disrupted clock function and circadian independent cell growth. Unlike the zebrafish model, no intracellular clock regulation of cell cycle genes expression or function was observed, expression being preferentially modified by homeral circadian regulators such as glucocorticoids and melatonin. This also contradicts mammalian in vivo studies, leading to the hypothesis that the clock and cell cycle maybe uncoupled in immortalised cell cultures. In conclusion this study has demonstrated that clock regulation of the cell cycle in mammalian system is a multifactorial process and that disruption of this system leads to changes in the character of the cell cycle within the host tissue. Further work must explore this relationship in an in vivo setting.

Circadian clocks are endogenous oscillators that control many physiological processes and confer functional and adaptive advantages in various organisms. These molecular oscillators comprise several interlocked feedback loops at the gene expression level. In plants, the circadian clock was recently shown to be organ specific. The root clock seemed to involve only a morning loop whereas the shoot clock also includes an evening loop in a more complex structure. My work aimed at refining the differences and similarities between the shoot and root clocks, using a combination of experimental and theoretical approaches. I developed an imaging method to obtain more data from the shoot and root clocks over time in various conditions. Some previous results were confirmed: the free running periods (FRPs) are longer in roots compared to shoots under constant light (LL). In addition, the amplitude of clock gene expression rhythms is lower in roots compared to shoots. However, the expression of several evening genes is circadian in roots, contrary to previous conclusions. This was confirmed with qPCR, and was observed in both light- and dark-grown roots. Yet light affects clock gene expression in roots, so an automatic covering system was designed to keep the roots in darkness and obtain data in more physiological conditions. Clock genes behaved differently in shoots and light-grown roots that were in the same environmental conditions, and may be differentially affected by blue and red light. However shoot and root clocks were more similar under constant darkness (DD). My imaging and RT-qPCR data, together with new microarray results and preliminary studies on clock mutants suggest that shoot and root circadian systems may have a similar structure but different input pathways. Entrainment is a fundamental property of circadian systems, which can be reset by cues such as light/dark (LD) cycles. I demonstrated that light can directly entrain the root clock in decapitated plants. The root clock could be entrained by a broad range of T cycles using low light intensity. In addition, rhythms were preferably entrained by low light than by any putative signal from shoots in experiments using conflicting LD cycles of different strengths. My results indicate that direct entrainment by LD cycles could be the main mechanism that synchronise the shoot and root clocks at constant temperature. This is physiologically relevant because dark-grown roots can perceive light channelled by the exposed tissues, in a fibre optic way. I also showed for the first time that clock and output genes could be rapidly entrained by temperature cycles in roots. Several mathematical models of the shoot circadian clock were used to try and fit the root clock data by optimising some parameters. The best set of parameters gave a good qualitative fit to root data under LD, LL and DD. It reproduced the long FRP observed in roots under LL and captured the entrainment under LD with lower amplitude in roots. The parameters that were changed for these simulations were all related to light input, which supports the idea of similar clock structures in shoots and roots but with different input pathways. Together my results confirmed that the plant circadian clock is organ specific and suggest that it is organ autonomous.

I have investigated the co-evolutionary divergence of two central circadian clock genes; period (per) and timeless (tim). The molecular components of the clock were derived from a common ancestor in higher eukaryotes, so that closely related species possess homologous genes and mechanisms. By substituting a clock gene from one species into another closely related one, it is probable that the clock will still be able to function to some degree because it maintains its interactions with other clock proteins. Previous behavioural analysis of Drosophila melanogaster per01 transfromants carrying a single copy of the, Musca domestica per gene, revealed a successful rescue of rhymicity, plus Musca - specific circadian phenotypes were transferred to the transformants. I have generated a series of M. domestica / D. melanogaster chimeric per transgenes and transformed them into a per-null background, in an attempt to define the regions of the per gene which give species-specific patterns of behaviour.;To further understand how PER and TIM interact via the PAS domain of PER, I have used an Evolutionary Trace procedure to predict functionally important residues of the PASA region. I have generated a number of PAS mutations based on the results of this bioinformatics analysis and studied the effect of these mutations on PER-TIM interactions using both the yeast two-hybrid system and for nuclear transport and retention in Schneider 2 cells.;Finally I have focused on TIM and its interaction with Drosophila Axin (dAxin), a protein that was originally discovered in the wingless pathway. dAxin is known to facilitate the phosphorylation and subsequent degradation of ARM (protein product of the armadillo gene). dAxin does this by acting as a scaffold between ARM and the serine-threonine kinase SGG (protein product of the shaggy gene). SGG has been shown to phosphorylate TIM, promoting nuclear localization.

Les neurones à Kisspeptine (Kp) de l'AVPV sont essentiels pour la survenue du pic de LH. Celle-ci est conditionnée par les concentrations circulantes d'oestrogènes (E2) et le moment du jour. Nous avons étudié si les neurones à Kp de l'AVPV étaient le lieu d'intégration de deux messages chez des souris sauvages intactes : un message E2, et un message temporel. Nous voulions savoir si ces neurones hébergeaient une horloge secondaire impliquée dans la temporalité du pic de LH. Durant l'après-midi du proestrus, une baisse drastique de l'immunoréactivité (ir) de Kp apparaît 2h avant la survenue du pic de LH au moment où l'expression de l'ARNm Kiss1 est élevée. Au contraire durant le diestrus, Kpir,l'expression de l'ARNm Kiss1 et les concentrations circulantes de LH restent basses. Les neurones à Kp de l'AVPV expriment une protéine horloge PER1 avec un rythme journalier exhibant un retard de phase de 2.8 h en diestrus comparativement au proestrus. Des explants d'AVPV exprimant les Kp provenant de souris PER2::LUCIFERASE dévoilent des oscillations circadiennes soutenues avec une période de 23.2h, significativement plus courte que celle observée dans les NSC. L'incubation des explants d'AVPV en présence d'E2 (10nM) rallonge la période d'une heure. En conclusion, cette étude indique que les neurones à Kp de l'AVPV présentent un rythme journalier dépendant des E2, qui pourrait être piloté par la présence d'une horloge secondaire au sein de ces neurones
The kisspeptin (Kp) neurons in the anteroventral periventricular nucleus (AVPV) are essential for the preovulatory LH surge, which is gated by circulating estradiol (E2) and the time of day. We investigated whether AVPV Kp neurons in intact female mice may be the site in which both E2 and daily signals are integrated and whether these neurons may host a circadian oscillator involved in the timed LH surge. In the afternoon of proestrous day, Kp immunoreactivity displayed a marked and transient decrease 2 hours before the LH surge. In contrast, Kp content was stable throughout the day of diestrus, when LH levels are constantly low. AVPV Kp neurons expressed the clock protein period1 (PER1) with a daily rhythm that is phase delayed compared with the PER1 rhythm measured in the main clock of the suprachiasmatic nuclei (SCN). PER1 rhythm in the AVPV, but not in the SCN,exhibited a significant phase delay of 2.8 hours in diestrus as compared with proestrus. Isolated Kp expressing AVPV explants from PER2::LUCIFERASE mice displayed sustained circadian oscillations of bioluminescence with a circadian period (23.2 h) significantly shorter than that of SCN explants(24.5 h). Furthermore, in AVPV explants incubated with E2 (10 nM to 1 μM), the circadian period was lengthened by 1 hour, whereas the SCN clock remained unaltered. In conclusion, these findings indicate that AVPV Kp neurons display an E2-dependent daily rhythm, which may possibly be driven by an intrinsic circadian clock acting in combination with the SCN timing signal

El reloj circadiano es un mecanismo celular endógeno capaz de medir el paso del tiempo y traducir las señales medioambientales, principalmente luz y temperatura, en respuestas temporales que resultan en ritmos metabólicos y fisiológicos de aproximadamente 24 horas. Esta coordinación temporal les permite a los seres vivos predecir y anticipar cambios periódicos en el medioambiente. A pesar de su importancia para la adaptación y supervivencia de las plantas, la posible función regulatoria del reloj circadiano sobre la actividad y homeostasis mitocondrial ha sido difícil de elucidar. En esta Tesis Doctoral, hemos seguido un enfoque integral para demostrar el mecanismo molecular mediante el cual uno de los componentes clave del reloj circadiano, TOC1 (TIMING OF CAB EXPRESSION 1), controla la actividad mitocondrial. Con este fin, hemos estudiado la fluctuación in vivo de los niveles del ATP citosólico mediante la utilización de un biosensor de ATP basado en la tecnología FRET. También hemos realizado análisis transcriptómicos correlacionándolos con datos de los cambios en la acumulación de metabolitos observados en plantas sobre-expresantes y mutantes de TOC1. Hemos identificado el mecanismo molecular por el cual TOC1 regula la actividad mitocondrial a través de la unión directa al promotor del gen relacionado con el ciclo del ácido tricarboxílico, FUMARASE 2. Nuestros estudios de interacción genética también han validado este mecanismo. Las plantas que sobre-expresan TOC1 acumulan menos biomasa y tienden a presentar un fenotipo similar al de plantas sometidas a inanición. La sobre-expresión del gen FUMARASE 2 en estas plantas ayuda a la recuperación de la biomasa y alivia el fenotipo de inanición. En general, con este estudio se ha demostrado el papel que ejerce el reloj circadiano en la regulación de la demanda energética celular en sincronización con el medioambiente.
Circadian clocks are molecular timekeeping mechanisms that translate environmental cues, mostly light and temperature, into temporal information to generate ~24h rhythms in metabolism and physiology. The temporal coordination by the clock enables organisms to predict and anticipate periodic changes in the environment. Despite its importance for plant fitness and survival, the possible role of the circadian clock directly regulating plant mitochondrial activity and energy homeostasis has remained elusive. In this Doctoral Thesis, we have followed a comprehensive approach to demonstrate the molecular mechanism by which the key clock component TOC1 (TIMING OF CAB EXPRESSION 1) sets the time of mitochondrial activity. To that end, we have followed the in vivo dynamics of cytosolic ATP production using a FRET-based ATP biosensor. We have also performed transcriptomic analyses and examined their correlation with actual changes in metabolite content using plants miss-expressing TOC1. We have identified the molecular mechanism by which TOC1 regulates the mitochondrial activity through direct binding to the promoter of the tricarboxylic acid cycle related gene FUMARASE 2. Our genetic interaction studies have validated this mechanism, as over-expression of FUMARASE 2 in TOC1 over-expressing plants alleviates the reduced biomass and the starvation-like phenotypes observed in TOC1 overexpressing plants. Overall, ours studies uncover the role of the circadian clock controlling the cell energetic demands in synchronization with the environment.

The circadian system provides animals with a means to adapt internal physiology to the constantly changing environmental stimuli that exists on a rotating planet. Light information is translated into molecular timing mechanisms within individual pacemaker cells of the mammalian hypothalamic suprachiasmatic nucleus (SCN) via transcriptionaltranslational feedback loops. Humoral and neural outputs from this master clock result in circadian rhythms of physiology and behavior. The hierarchy of the circadian system involves SCN synchronization of cellular clocks within peripheral tissues so that differential transcriptional profiles in individual organs reflect their specific function. The first step to investigating equine circadian regulation was to identify and isolate the core components of the molecular clock in the horse. Successful isolation and sequencing of equine Bmal1, Per2, Cry1 and Clock cDNAs revealed high sequence homology with their human counterparts. Real Time RT-PCR assays were subsequently designed to quantitatively assess clock gene expression in equine peripheral tissues. Synchronization of equine fibroblasts revealed temporal profiles of clock gene expression identical to those of the SCN and peripheral tissues of other species. However, while clock gene expression varies over time in equine adipose tissue, there was no observable oscillation of clock gene transcripts in equine blood. Spurred by recent reports of immune-circadian interactions, this novel finding prompted an investigation of clock gene expression in equine blood during a systemic inflammatory response. The results demonstrated that acute inflammation upregulates Per2 and Bmal1 in equine blood. Subsequent experiments identified neutrophils as the source of this upregulation and highlighted exciting new immunecircadian interplay during an innate immune response. Finally, the effect of a 6-h phase advance of the light/dark cycle, mimicking an easterly transmeridian journey, on circadian melatonin and core body temperature rhythms was investigated. In contrast to the gradual adaptation observed in other species, these markers of equine circadian phase adapt immediately to a time zone transition. Combined, the results of these experiments highlight important interspecies differences in circadian regulation with practical implications regarding the potential impact of jet lag on equine athletes. Furthermore, the results underline the relevance of chronobiological investigation in a large mammalian species such as the horse.

El rellotge circadià està sincronitzat per senyals mediambientals externes, principalment la llum i la temperatura. Entendre com respon el rellotge circadià de la planta a les oscil·lacions de temperatura és crucial per comprendre la capacitat de resposta de la planta a l'entorn. En aquesta tesi doctoral, trobem una funció prevalent depenent de la temperatura de l'component de el rellotge d'Arabidopsis EARLY Flowering 4 (ELF4) en el rellotge circadià de l'arrel. En plantes en les quals l'àpex aeri s'ha eliminat, el rellotge pot funcionar en les arrels, tot i que exhibeix un període més curt i una fase avançada en comparació amb les arrels de plantes completes. Els assajos de microempelt mostren que ELF4 es mou des de l'àpex aeri per regular els ritmes en les arrels. El moviment de la proteïna ELF4 no transmet informació fotoperiòdica, sinó que és essencial per controlar el període de el rellotge circadià en l'arrel d'una manera depenent de la temperatura. Les baixes temperatures afavoreixen la mobilitat de ELF4, el que resulta en un rellotge de de ritme lent, mentre que les altes temperatures disminueixen el moviment, el que porta a un rellotge més ràpid. Per tant, el moviment de la proteïna ELF4 mòbil proporciona informació sobre la temperatura i ajuda a establir un diàleg entre l'àpex aeri i l'arrel de la planta per controlar el ritme circadià en l'arrel.
El reloj circadiano está sincronizado por señales medioambientales externas, principalmente la luz y la temperatura. Entender cómo responde el reloj circadiano de la planta a las oscilaciones de temperatura es crucial para comprender la capacidad de respuesta de la planta al medio ambiente. En esta Tesis Doctoral, encontramos una función prevalente dependiente de la temperatura del componente del reloj de Arabidopsis EARLY FLOWERING 4 (ELF4) en el reloj circadiano de la raíz. En plantas en las que el ápice aéreo se ha eliminado, el reloj puede funcionar correctamente en las raíces, aunque exhibe un período más corto y una fase avanzada en comparación con las raíces de plantas completas. Los ensayos de microinjerto muestran que ELF4 se mueve desde el ápice aéreo para regular los ritmos en las raíces. El movimiento de la proteína ELF4 no transmite información fotoperiódica, sino que es esencial para controlar el período del reloj circadiano en la raíz de una manera dependiente de la temperatura. Las bajas temperaturas favorecen la movilidad de ELF4, lo que resulta en un reloj de de ritmo lento, mientras que las altas temperaturas disminuyen el movimiento, lo que lleva a un reloj más rápido. Por lo tanto, el movimiento de la proteína ELF4 móvil proporciona información sobre la temperatura y ayuda a establecer un diálogo entre el ápice aéreo y la raíz de la planta para controlar el ritmo circadiano en la raíz.
The circadian clock is synchronized by external environment cues, mostly through light and temperature. Explaining how the plant circadian clock responds to temperature oscillations is crucial to understanding plant responsiveness to the environment. In this thesis, we found a prevalent temperature-dependent function of the Arabidopsis clock component EARLY FLOWERING 4 (ELF4) in the root clock. The clocks in roots are able to run properly in the absence of shoots although shoot excision leads to a shorter period and advanced phase in excised roots compared to entire roots. Micrografting assays show that ELF4 moves from shoots to regulate rhythms in roots. ELF4 movement does not convey photoperiodic information, but trafficking is essential for controlling the period of the root clock in a temperature-dependent manner. Low temperatures favour ELF4 mobility, resulting in a slow paced root clock, whereas high temperatures decrease movement, leading to a faster clock. Hence, the mobile ELF4 delivers temperature information and establishes a shoot-to-root dialogue that sets the pace of the clock in roots.
Universitat Autònoma de Barcelona. Programa de Doctorat en Biologia i Biotecnologia Vegetal

The first experiment used the male Syrian hamster as the subject, as it has good wheel-running rhythms with consistent and robust activity onsets and also a well defined phase delay and advance region of the phase response curve (PRC). In this study, a double light pulse paradigm was used to examine if the clock could reset within 2 h. The second light pulse administered 2 h after an initial pulse was used to map the effects of the first light pulse on the resetting behaviour to determine whether the clock had reset to primary light pulse before receiving the second one. The results obtained in both the phase delay and phase advance portions of the PRC were consistent with the hypothesis that the clock of the Syrian hamster could reset within 2 h of a light pulse. The next experiment investigated whether the immediate early genes (IEGs) are a reliable marker of clock resetting and if they have a role in clock resetting. The same double light pulse paradigm employed in the first experiment was used to ascertain if c-fos and egr-l could be induced to two delaying light pulses. The study revealed that both IEGs were induced to each light pulse demonstrating that the clock could differentiate between the two pulses of light and that the immediate early genes may play a role in phase resetting to light. Having established that the clock of the Syrian hamster was reset in 2 h the outbred mouse was used to ascertain if this rapid resetting is common to other mammalian species. The results obtained using multiple light pulses revealed that the mouse clock is also reset within 2 h, supporting the findings in the Syrian hamster and demonstrating cross-species homology. In addition, investigations on CREB phosphorylation, c-Fos induction and AP-l binding activity revealed that all three transcriptional regulators were induced in response to a light pulse during subjective night and that there was a strong correlation with photic phase resetting of the mammalian clock.

The majority of organisms on Earth, from cyanobacteria upwards, contain an internal timing mechanism that allows them to make optimum use of the rhythmic abundance of many environmental resources (e.g. sunlight). In mammals, the master circadian (i.e. daily) clock is contained within the suprachiasmatic nuclei (SCN) of the anterior hypothalamus. The clock is thought to consist, at the simplest level, of a number of autoregulatory negative feedback loops involving the products of the cryptochrome (mCry1 and mCry2) and period (mPer1, mPer2 and mPer3) genes. These loops are driven by the positive factors encoded by the clock and Bmal 1 genes. Chapter 3 of this dissertation presents immunocytochemical data showing that the expression of mPER1 and mPER2 protein is rhythmic in the SCN, both in mice entrained to a light-dark cycle or in continuous darkness. In contrast, the expression of mTIM is constitutive under all lighting conditions. These are thus functional differences in gene expression between mice and the fruitfly Drosophila, where tim is a central, rhythmic clock component. The clock can be reset (i.e. phase advanced or delayed) by light incident on the retina during the night. Experimental data, both in vivo and in vitro, suggest that mCry genes are central oscillator components insensitive to light, whereas mPer1 and mPer2 are up-regulated by resetting light pulses. Resetting by light pulses which delay the clock occurs rapidly (within 1-2 cycles), whereas the full expression of advances of the clock can take several cycles to be completed.

Plants, like most organisms, have developed elaborate mechanisms for anticipating periodic environmental changes. The circadian clock allows an organism to adapt its metabolic, developmental and physiological processes to coincide with favourable environmental conditions. At the centre of the Arabidopsis thaliana clock, linking environmental inputs and driving the overt biological rhythm is a central oscillator that consists of multiple interlocked transcriptional/translational negative feedback loops. What is known about the structure of the central oscillator comes primarily from genetic analysis. Less clear, is how putative oscillator proteins perform their perceived functions in circadian rhythm maintenance. Described are the cloning, expression and purification of clock-associated proteins; TOCI-PRR, ELF4, LUX and LIPl. Purified ELF4 was subjected to unsuccessful crystallisation trials, probably due to its intrinsically unstructured nature. A truncated form of LIP 1 was shown to be an active OTPase, representing the first example of an active OTPase in the plant clock. In addition, a protocol for the production of ssDNA aptamers has been developed (against SRRl), which can be used to replace antibody-based experimentation. The work presented discusses the difficulty in obtaining the novel, plant-specific proteins in quantities required for crystallisation, and suggests alternative methods for structural and biochemical analysis of these proteins. Moreover, this thesis combines experimental data with a range of Bioinformatic tools to aid design for subsequent biochemical expression, purification and crystallisation trials.

The endogenous circadian clock adjusts the physiology and behaviour of an organism to advantageous periods of the day, and represents an adaptation to daily environmental cycles, such as light and temperature. Locomotor activity in Drosophila melanogaster represents a robust behavioural rhythm used to study the clock. This clock is located in the lateral and dorsal neurons of the fly and in the suprachiasmatic nuclei (SCN) of the hypothalamus in the mammal. The molecular bases of underlying circadian timing mechanisms in insects and mammals are conserved. Although we have a basic knowledge of the Drosophila molecular clock circuits functioning, many questions regarding the nature of the protein complexes that subserve circadian pacemakers, the connections between the oscillator and the overt rhythms and the entrainment signals to the clock remain unanswered. To identify new D. melanogaster circadian components I used three different approaches. The first is based on immunoprecipitation of protein complexes using tagged CYC, a dedicated clock protein, to pull down its partners. The second employs a comparative approach with the mammalian circadian SCN proteome and the third uses a tap-tagging design which is used to screen the proteome. Expression studies of candidate proteins, and behavioural analyses using mutants and transgenes to disrupt and silence some of these factors, have revealed a number of candidate genes that may affect aspects of clock function. Two novel genes involved in glutamate metabolism are particularly compelling, and appear to contribute to the circadian mechanism by mediating the neurons that are important for light input. A further synaptic gene may be involved in setting the clock pacemaker.

The mammalian suprachiasmatic nuclei (SCN) are located in the ventral part of the hypothalamus and orchestrate circadian rhythms in physiology and behaviour. The ~20.000 neurones of the murine SCN express key molecular clock components including the Cryptochrome (Cry1/2) and Period (Per1/2/3) genes and their protein products CRY1/2 and PER1/2/3. Using different mouse models, this work demonstrates that with disrupted expression of CRY in the after-hours (Afh/Afh) mouse, cells of the ventral part of the SCN (vSCN) have a propensity to desynchronise. They receive increased GABAergic inputs and are less excitable during the projected night but not during the day compared to congenic wildtype (+/+). The linkage between CRY protein expression and the reduced excitability at night is supported by recordings from SCN cells of Cry2 deficient mice (Cry2-/-), which exhibit similar electrophysiological behaviour. Luminometrical recordings of single cell Per2 expression confirms the involvement of GABAergic signalling in both, maintaining a coherent rhythm in synchronised SCN cells from +/+ controls and the propensity of Afh/Afh SCN cells to desynchronise. A mechanism by which neuronal excitability is regulated in mammals, is the modulation of activity of small-conductance Ca2+-activated K+ (SK) channels. Western blot analysis demonstrates the expression of SK2 and SK3 channel protein in SCN neurones. Functionally, we show with whole cell electrophysiology, calcium imaging and luminometry how SK channels regulate the levels of intracellular calcium ([Ca2+]i) from day to night. In the more hyperpolarised SCN network of the Afh/Afh genotype at night, SK channel activity is altered and contributes to the lower single cell excitability. Vasoactive intestinal polypeptide (VIP) and its cognate receptor, VPAC2, are synthesised by SCN neurones and this intercellular signal facilitates coordination of suprachiasmatic neuronal activity. How the loss of VPAC2 receptor signalling affects the electrophysiology of SCN neurones and their response to excitatory inputs is unclear. Here we made patch clamp recordings of SCN neurones in brain slices prepared from animals that do not express VPAC2 receptors (Vipr2-/- mice) as well as non-transgenic animals (Vipr2+/+ mice). While Vipr2+/+ SCN neurones exhibit coordinated day-night variation in their electrical state, Vipr2-/- neurones do not and instead manifest a range of states during both day and night. We find that Vipr2+/+ neurones vary the membrane threshold potential at which they start to fire actions potentials from day to night, while Vipr2-/- neurones lack this variation. This is due to Vipr2-/- neurones lacking a voltage-gated sodium current. Subsequently we determine that this aberrant temporal control of neuronal state and excitability alters appropriate neuronal responses to a neurochemical mimic of the light-input pathway to the SCN. Conclusively, these results highlight the critical role intercellular signalling plays in the activity of individual neuronal state and their response to neural input as well as ensemble activity and function of the suprachiasmatic neural network.

As a strategy to survive to the upcoming winter, many insects enter diapause (a typical overwintering response that results on their developmental arrest). Drosophila melanogaster undergoes an adult or reproductive diapause that can be easily spotted by looking at the stage of development of the females’ ovaries. The possibility of the circadian clock influencing this phenotype was proposed to explain photoperiodic differences in induction levels. Nevertheless, to the date the debate is still on. In this thesis, I looked at several canonical clock mutants and assessed their impact on diapause, finding that 1) depending on the temperature in which they were reared the effects on the adult flies varied enormously 2) most of the clock mutants gave a strong effect in one or other growing conditions. In particular, Pdf0 and ClkJrk mutants behave in completely opposite ways. A second part of the project consisted on looking at the effects of period temperature-sensitive splicing in diapause. Using splicing locked transgenic flies provided by Isaac Edery, I found that expression of the summer isoform impaired the ability of the flies to undergo diapause. Hence, I cloned the different splicing variants into a pUAST vector and generated UAS lines to perform a neuroanatomical dissection of the phenotype. Also, related with the previous project, I decided to look if any miRNA could be regulating diapause by affecting any of the splicing variants. I found several possible miRNAs that could target the summer (intron-containing) non-splicing isoform. I found that one particular, miRNA-276b, was having a huge effect on diapause. Using a sponge particularly against this miRNA (which would result in its downregulation) diapause levels halved compared to all the controls that were performed in parallel.

Circadian rhythms are daily internal predictive biological cycles. Numerous studies have associated human circadian gene polymorphisms with phenotypic changes and multiple disease risks. Subsequent replications have often been negative in other populations, possibly due to poor linkage disequilibrium (LD) between the reported single nucleotide polymorphisms (SNPs) and underlying phenomena, or differing LD across populations. Alternatively, non-human circadian studies have reported population- specific effects, often expressed as latitudinal dines. Genome-wide genotyping results can produce distinctions. Haplotypes including separate association SNPs can unify multiple results. Strong LD across populations improves the likelihood of population-specific effects. Unusual patterns can possibly predict undiscovered effects. The importance of in silico techniques increases as ongoing projects seek to sequence all common SNPs. The present study has examined the variation patterns of 35 upstream, downstream, and core clock genes. 20 genes displayed unusual results using multiple screening measures. A possible 7-gene African-population haplogroup network was identified. While selection across the class of circadian genes was ruled out, CLOCK and FBXW11 are candidates for balancing and positive selection, supported by multiple genome- wide frequency spectrum distributions. CLOCK was examined in a UK-based sample in the most dense diurnal preference candidate gene study yet performed. An eveningness haplotype independent of the previously reported 3111C allele was identified. This haplotype and 3111C combined were important for intermediate preference. Strong LD, including 3 haplogroups and a 122-SNP 3111C haplotype shared by all 3 HapMap populations, united distinct association study results into population-specific phenotypes unlikely to be due to differing or poor LD across populations. The eveningness association was age-dependent. A non-linear relationship between ageing and diurnal preference was also uncovered, particularly for morning functioning and evening sleepiness. A possible circadian-climate interaction was examined. 14 genes had strongly impacted LD patterns. Near-exact relationships between climate (but not latitude) SNPs and diurnal preference were found in PERI and PER2, while an extrapolated relationship affected PER3. Strong relationships demonstrated negative or even opposite results in other populations. Combined, these results suggest that human circadian phenotypes, and possibly related disease risks, are likely altered by the environment. This raises the importance of future multiple-population studies, but predicts increased complexity in their interpretation.

Includes bibliographical references (leaves 61-84).
The circadian system has a regulatory role in almost all aspects of a plant's life. In Arabidopsis thaliana, almost 36% of the genome has been shown to be circadianly regulated and many genes that are circadianly regulated have been shown to be light responsive or involved in light responses. Rhythmic histone acetylation has been demonstrated in the promoter of TIMING OF CAB EXPRESSION1 (TOC1). Here, I used semi-quantitative Reverse Transcriptase Polymerase Chain Reaction (semi-quantitative RT -PCR) to investigate which enzymes are involved in the rhythmic expression of TOC1. I also determined whether loss-of-function histone acetylation and methylation mutants could affect the overall functioning of the circadian oscillator by measuring their circadian leaf movement and delayed fluorescence (DF) rhythms. GCN5/ HAG1 mutant plants (gcn5) exhibited erratic TOC1 expression in both constant dark (DD) and constant light (LL) conditions. Although TOC1 expression appeared to be rhythmic in both DD and LL conditions, the waveform of the rhythm was altered in TATA-binding protein associated factor 1 (taf1) mutants. This suggested that TAF1 and GCN5 might play different roles in the rhythmic histone acetylation affecting TOC1 expression. DF data and leaf movement data indicated that both TAF1 and GCN5 might play a role in the overall functioning of the A. thaliana circadian clock. Arrhythmic TOC1 expression and DF was observed in histone deacetylase 1 (hd1) mutants, suggesting that HD1 is not only involved in the rhythmic histone deacetylation affecting TOC1 expression but in the overall functioning of the circadian clock. Semi-quantitative RTPCR, DF and leaf movement studies demonstrated that CURLY LEAF (CLF), a histone methylase is involved in both the histone methylation affecting TOC1 expression and in the overall functioning of the A. thaliana circadian clock.

The genetic identification of molecular mechanisms responsible for circadian rhythm generation has advanced tremendously over the past 25 years. However the molecular identities of the avian clock remain largely unexplored. The present studies seek to determine candidate clock components in the avian species Gallus domesticus. Construction and examination of the transcriptional profiles of the pineal gland and retina using DNA microarray analysis provided a clear view into the avian clock mechanism. Investigation of the pineal and retina transcriptomes determined the mRNA profiles of several thousand genes over the course of one day in LD (daily) and one day in DD (circadian) conditions. Several avian orthologs of mammalian clock genes were identified and many exhibited oscillating patterns of mRNA abundance including several of the putative avian clock genes. Comparison of the pineal transcriptional profile to that of the retina revealed several intriguing candidate genes that may function as core clock components. Including the putative avian clock genes and several others implicated in phototransduction, metabolism, and immune response. A more detailed examination of several candidate photoisomerase/photopigment genes identified from our transcriptional profiling was conducted. These include peropsin (rrh), RGR-opsin (rgr), melanopsin (opn4) and cryptochrome 2 (cry2) genes. This analysis revealed several interesting patterns of mRNA distribution and regulation for these genes in the chick. First, the mRNA of all 4 genes is located within the Inner Nuclear Layer (INL) and Retinal Ganglion cell Layers (RGL) of the ocular retina, where circadian photoreception is present. Second, opn4 and cry2 mRNA is expressed in the photoreceptor layer of the chick retina where melatonin biosynthesis occurs. Lastly, the mRNA for all 4 candidate photopigment genes is regulated on a circadian basis in the pineal gland. As a whole these data yield significant insight into the mechanisms of the avian circadian system and present several candidate genes that may function to integrate photic information, and/or regulate circadian rhythm generation in birds.

In mammals, the circadian clock coordinates multiple behavioural and physical processes, including energy homeostasis. At the centre of these rhythms lies the circadian clock machinery, a precisely coordinated transcription-translation feedback system required to maintain the correct time. Metabolic homeostasis requires accurate and coordinated circadian timing within individual cells and tissues of the body. Moreover, recent evidence has shown that the coupling of circadian and metabolic circuits involves reciprocal regulatory feedback. In line with this, mounting evidence suggests that disruption of the clock contributes to the development of obesity and its comorbidities. This is particularly concerning given that modern lifestyles often undermine our bodies’ clock. However, the casual mechanisms which link circadian disruption to metabolic disease are not well defined. This work aims to gain a further understanding of clock control of metabolic homeostasis and especially regulation of lipid metabolism. This work uses dietary challenge to determine which peripheral clocks and downstream metabolic pathways are particularly susceptible to diet induced obesity (DIO). We demonstrate that although behavioural rhythmicity was maintained in DIO, gene expression profiling revealed tissue-specific alteration to the phase and amplitude of the molecular clockwork. Clock function was most significantly attenuated in visceral white adipose tissue (WAT) of DIO mice, and was coincident with elevated tissue inflammation, and dysregulation of clock-coupled metabolic regulators PPARα/γ.The rhythmic expression of Rev-erbα, a nuclear receptor involved in the circadian clock, was particularly affected in DIO mice. This study uses the Rev-erbα-/- mouse to explore clock-metabolic coupling, specifically lipid metabolism. In line with published work, Rev-erbα-/- mice exhibit an obese phenotype with associated upregulation in gWAT of lipogenic (Dgat2, Fasn) and fatty acid liberation (Lpl) genes. Differences in fat mobilization are observed as Rev-erbα-/- mice show a heightened insulin stimulated lipogenic drive and an attenuation of the lipolytic drive in the fasted state, suggesting an increased propensity for fat accumulation. The role of the clock was further investigated in adipose tissue by deletion of Bmal1 (clock ablation) or Rev-erbα (clock manipulation) specifically in adipocytes using Cre-Lox methodology. AdipoCREBmal1flox/flox mice showed attenuated feeding rhythms, indicating a direct effect of the adipocyte circadian clock on hypothalamic feeding centres and severe dysregulation of metabolic genes. However, AdipoCRERev-erbαflox/flox displayed very little phenotypic difference compared to control littermates, suggesting that global loss of Rev-erbα may have reinforcing metabolic consequences. This work suggests a key role of the clock in lipid handling and the pathogenesis of obesity. Insights into this link may lead to novel targets for treating both obesity and metabolic complications.

The circadian central oscillator of Drosophila melanogaster consists of at least two interlocked negative transcriptional feedback loops. This has been taken to be a general model for higher eukaryotes with the core components conserved but their regulation altered. The work presented here indicates that in Musca domestica, a dipteran closely related to Drosophila, one of these regulatory loops, involving PERIOD (PER) and TIMELESS (TIM), functions in a completely different manner. This study shows that in contrast to Drosophila, Musca PER remains constant in western studies in any lighting condition, whereas like Drosophila TIM cycles in both LD and DD and is constantly degraded in LL. In addition within the central brain immunostaining revealed that even in the small set of cells thought to contain the central pacemaker PER staining was restricted exclusively to the cytoplasm. However following the Drosophila model PER was observed to cycle in the cytoplasm of these cells. Although TIM co-localises with PER in these cells, unlike PER, TIM does become nuclear. This indicates that the negative feedback model illustrated by analysis of the Drosophila is inadequate to explain clock function in Musca. A putative Musca PER nuclear export sequence which functions in other species was tested in GFP constructs but not shown to be involved in altered localisation. In contrast in peripheral tissue such as photoreceptor cells both PER and TIM cycle and both proteins become nuclear late at night as in Drosophila. Stability of Musca PER in LL and an altered relationship between transgenic Musca PER and Drosophila DOUBLETIME indicates an altered relationship between PER and the DBT kinase that may be responsible for PER stability. Thus although it can be seen that a different model is required for other insect species how these proteins act remains to be elucidated.

In the model plant Arabidopsis thaliana, the circadian clock is believed to be composed of a number of coupled transcriptional negative feedback loops. The LATE ELONGATED HYPOCOTYL (LHY) gene is thought to form part of at least two of these transcriptional feedback loops, as well as playing a role in the perception of light signals by the clock. To better understand how multiple transcriptional feedback loops might be integrated in the transcriptional regulation of LHY, we have performed an analysis of the cis-regulation of this gene. Through deletion analysis of reporter gene constructs, we have identified a 957 basepair region of the LHY promoter which contains sufficient sequence to direct the characteristic expression profile of LHY. Furthermore, we provide evidence that at least two circadian signals converge on this region. Electrophoretic mobility shift assays identified four classes of candidate cis-elements within the LHY promoter including a poly-CTT tract, an AAAAA motif, a candidate MYB-binding site and a G-box motif. Through mutational analysis of these elements, we have been able to determine aspects of their in vivo regulatory function. We report that a G-box motif and the previously uncharacterized AAAAA element are implicated in the regulation of LHY transcription by light signals. In etiolated seedlings, the region of the LHY promoter containing the MYB-binding site motif and multiple copies of the poly-CTT motif mediates regulation of LHY by both light-responsive and circadian signals.

Drosophila melanogaster locomotor activity responds to seasonal conditions by modulating the “evening” activity component. During simulated winters of cold temperature and short days an advanced evening locomotor peak occurs with more daytime locomotor activity; on the other hand long photoperiods and warm temperatures give a delay in the evening peak, thereby avoiding a possible desiccation during the hottest times of the day. This pattern of activity is related to a thermosensitive splicing event that occurs in a 3’ intron in the period gene, with a higher level of splicing and earlier accumulation of PERIOD in short days and low temperatures. A mutation in norpA which encodes a phospholipase C, generates a high level of spliced per at warmer temperature, so mutants behave as if it is colder than it actually is. The relation between norpA, per splicing and the circadian neurons has been analysed. Initially, norpA expression has been investigated via in situ hybridisation and immunocytochemistry. norpA transcript has been localised among the clock pacemakers but not NORPA. Subsequently, norpA expression has been knocked-down by RNAi in specific subset of neurons. The resulting locomotor behaviour shows seasonally related effects implicating the photoreceptors, lateral and dorsal clock neurons as structures involved in timing the locomotor behaviour. In parallel, the thermal role of a second PLCβ, plc21C, has been investigated via RNAi among circadian pacemakers. It has been possible to show that plc21C expression in the photoreceptors, lateral and dorsal neurons is required to set different locomotor behaviours at different temperatures, but not via per and tim splicing. Finally, in contrast to reports that the double photoreceptor mutants involving glass and cryptochrome are “circadian blind”, these flies have been observed to entrain to light-dark cycles at moderate temperatures. Candidate orphan G protein coupled receptors have been screened in order to identify a further set of putative circadian-relevant photoreceptors contributing to this residual entrainment in glass60jcryb mutants. In constant light conditions, the RNAi of CG7497 and CG16958 generates rhythmic or arrhythmic flies depending on the genetic background tested.

Circadian clocks involve feedback loops that generate rhythmic expression of key genes. Molecular genetic studies in the higher plant Arabidopsis theliene have revealed a complex clock network. We begin by modelling the first part of the Arabidopsis clock network to be identified, a transcriptional feedback loop comprising TIMING OF CAB EXPRESSION 1 (TOCl), LATE ELONGATED HYPOCOTYL (LHY) and CIRCADIAN CLOCK ASSOCIATED 1 (CCA1). As for many biological systems, there are no experimental values for the parameters in our model, and the data available for parameter fitting is noisy and varied. To tackle this we construct a cost function, which quantifies the agreement between our model and various key experimental features. We then undertake a global search of parameter space, to test whether the proposed circuit can fit the experimental data. Our optimized solution for the Arabidopsis clock model is unable to account for significant experimental data. Thanks to our search of parameter space, we are able to interpret this as a failure of the network architecture. We develop an extended clock model that is based upon a wider range of data and accurately predicts additional experimental results. The model comprises two interlocking feedback loops comparable to those identified experimentally in other circadian systems. We propose that each loop receives input signals from light, and that each loop includes a hypothetical component that had not been explicitly identified. Analysis of the model predicts the properties of these components, including an acute light induction at dawn that is rapidly repressed by LHY and CCAL We find this unexpected regulation in RNA levels of the evening-expressed gene GIGANTEA (GI), supporting our proposed network and making GI a strong candidate for this component. We go on to develop reduced models of the Arabidopsis clock to aid conceptual understanding, and add a further proposed feedback loop to develop a 3-loop model of the circadian clock. This 3-loop model is able to reproduce further key experimental data.

The circadian clock of Drosophila melanogaster consists of at least two interlocked feedback loops. In the first, the period and timeless gene products negatively regulate their own transcription. CRYPTOCHROME (CRY) is the dedicated circadian photoreceptor, and flies carrying a strong hypomorphic mutation in the cry gene have severely blunted circadian photoresponses. CRY physically interacts with the core components of the clock, PERIOD (PER) and TIMELESS (TIM) in a light-dependent manner. Previous work carried out in the laboratory showed that removing 20 amino acids at the C-terminus of CRY to create CRYDelta results in the loss of light-dependency of CRY interactions in yeast two-hybrid assays. Based on this work, the aim of my project was to study the role of the CRY C-terminus in vivo by clock neurons targeted overexpression of CRYDelta with the hypothesis that it should behave as a constitutively active form of the protein. CRYDelta flies have long period of locomotor activity in constant darkness, show abnormal responses to light and exhibit altered oscillation of the PER and TIM proteins in central and peripheral clocks. These phenotypes are reminiscent of responses observed when wild-type flies are kept under continuous low-light intensity. Therefore, this study provides strong behavioural, molecular and immunohistochemical evidence confirming that CRYDelta is constitutively active, and elicits continuous light responses. Moreover, previous work demonstrated that CRY role in the Drosophila clock exclusively involves light signalling to the core components of the clock. This study identified a potential new light-dependent function for CRY in vivo.

Circadian rhythms are daily molecular oscillations within cells ranging from prokaryotes to humans. This rhythm is self sustaining, and receives external cues in order to synchronize an organism's behavior and physiology with the environment. Many metabolites utilized in metabolic processes seem to follow a pattern of circadian oscillation. Iron, an essential component in cellular processes such as respiration and DNA synthesis, is obtained almost exclusively through diet, yet little is known about how the clock governs iron metabolism. The regulation of iron within the cell is very tightly controlled, as iron is highly reactive in the generation of oxidative stress and the excretion of excess iron is very limited. There are limited findings indicating that there are molecular ties between the circadian clock and the regulation of iron metabolism. The first half of my dissertation focuses on the role of the circadian clock in modulating expression of iron metabolic components. We found that key components of iron import, in TFRC, and export, in SLC40A1, show altered expression in response to changes in the expression of clock transcription components. Furthermore, in circadian synchronized HepG2 hepatocytes TFRC and SLC40A1 showed rhythms in their mRNA expression, although expression of these genes was highly altered in conditions of high iron availability. We also examined IREB2, which expresses a master regulator of iron concentration in IRP2. IRP2 showed rhythms in phase with circadian component PER2, and IRP2's rhythmicity was lost under iron overload conditions. We observed that the ability of these three critical iron metabolic components to respond to sudden increases in available iron was mitigated in cells with clock impairment. Whole cistrome and transcriptome analysis was used to determine that rhythmicity in TFRC and SLC40A1 are not equal in their recruitment of circadian protein binding or in the stage of transcription in which circadian rhythms are generated. The cumulative effect of all of this regulation is that rhythmic variation in intracellular hepatic ferrous iron is clock controlled. The second half of my dissertation focuses on understanding how iron uptake influences clock resetting. Initially, iron was added to the cells in the form of ferrous sulfate, or chelated out of the cells using 2-2'-dipyridyl and clock gene expression was monitored. Altered rhythmicity of these components was seen at both the mRNA and protein level in cells with disrupted iron homeostasis. Then, we measured changes in period, phase, and amplitude of these rhythms, ultimately using a luciferase reporter cell line to demonstrate that even slight changes in cellular iron produce an effect on rhythmic period. We find that the circadian clock and iron metabolism pathway are intimately related, and that the intracellular iron concentration plays a role in circadian clock behavior. Overall, our research illustrates the importance of the circadian clock in liver metabolism and physiology. Improper iron metabolism due to genetic or dietary shortcomings is common in humans, and our work builds on the importance of chronotherapy in treatment of these conditions. Conversely, our research into the effect intracellular iron has on the clock contributes to the growing body of research into how circadian clocks, especially the peripheral clock of the liver, receive input from a range of metabolites in conjunction with signals from the master oscillator of the suprachiasmatic nucleus.
Ph. D.

La rotation de la Terre oblige les organismes vivants à s’adapter aux modifications cycliques de l’environnement, et tout particulièrement aux changements de lumière et de température. Des unicellulaires à l’Homme, la plupart des espèces ont développé des horloges circadiennes, qui leur permettent d’anticiper les transitions jour-nuit. La lumière constitue le signal majeur pour la synchronisation de l’horloge. En cycles jour-nuit, les drosophiles présentent un profil d’activité locomotrice bimodal, avec un premier pic autour de l’aube et le deuxième au crépuscule. Chez cet insecte, la perception de la lumière est assurée à la fois par un système complexe, constitué des yeux composés, des ocelles et de l’eyelet d’Hofbauer-Buchner. Ces organes contiennent des photorécepteurs (PRs) exprimant six protéines photosensibles différentes, les rhodopsines (Rh1 à Rh6). Une septième rhodopsine (Rh7) a été décrite dans quelques neurones de l’horloge cérébrale. La lumière est également perçue directement dans la plupart des neurones d’horloge grâce à une protéine photosensible, le cryptochrome (Cry). Les différentes études du rôle de la lumière sur l’entraînement de l’horloge ont essentiellement porté sur la voie cry-dépendante, en utilisant de courts flashs lumineux pour recaler l’horloge cérébrale. Notre étude s’est intéressée à l’entraînement de l’horloge via les rhodopsines. Quels types de photorécepteur sont impliqués ? Après l’activation de la cascade de phototransduction et la libération de l’histamine par les photorécepteurs, quels neurones, exprimant les récepteurs à l’histamine Ort et Hiscl1, participent à l’entraînement de l’horloge circadienne ? Une première partie présente l’étude de l’implication des 6 rhodopsines dans l’entraînement circadien. Tout d’abord, nous avons mis en évidence la fonction de photorécepteurs spécifiques (exprimant Rh1 ou Rh6) dans la voie NorpA-dépendante (Saint-Charles et al. J Comp Neurol 2016). Nous avons ensuite généré des lignées de drosophiles n’exprimant aucune ou qu’une seule rhodopsine. Sans rhodopsine ni Cry les mouches sont incapables de se synchroniser sur les cycles jour-nuit, quelle que soit l’intensité lumineuse. En lumière faible, l’input pour l’entraînement vient principalement des photorécepteurs exprimant Rh1 et Rh6. En forte lumière, chacune des 6 rhodopsines des différents photorécepteurs est capable d’entrainer l’horloge, Rh1, Rh5 et Rh6 étant les plus efficaces ( Alejevski et al., in prep). Une deuxième partie présente la caractérisation des voies neuronales connectant directement ou indirectement les PRs à l’horloge cérébrale. L’horloge circadienne de mouches mutantes, à la fois pour le cryptochrome et les 2 récepteurs à l’histamine, est « aveugle » alors que les mutantes pour Cry mais possédant l’un ou l’autre récepteur à l’histamine sont capables de se synchroniser sur les cycles de lumière. La ré-expression chez les mutants de Ort ou Hiscl1 dans les neurones d’horloge ne restaure pas l’entraînement, suggérant ainsi l’absence de connexions directes entre les PRs histaminergiques et les neurones d’horloge. Nos expériences de sauvetage comportemental mettent en évidence des connexions fonctionnelles entre certains interneurones Ort des lobes optiques et les neurones d’horloge. En revanche et de façon inattendue, nous n’observons d’entraînement circadien que lorsque nous ré-exprimons Hiscl1 dans les seuls PRs Rh6. Nos résultats révèlent que les photorécepteurs interviennent dans l’entraînement à la fois comme photorécepteurs et comme interneurones, cibles d’input histaminergique, rappelant ainsi le double rôle des cellules ganglionnaires de la rétine exprimant la mélanopsine chez les mammifères (Alejevski et al. Nat Commun, in revision)
The rotation of the earth forces living organisms to adapt to its cyclic environment, in particular light and temperature changes. From unicellular organisms to humans, almost all species have evolved circadian clocks, which allow them to anticipate day-night transitions and use light as the most powerful synchronizing cue. In light-dark cycles, D. melanogaster flies display a bimodal locomotor activity with peaks around dawn and dusk. To perceive light, Drosophila has evolved a complex visual system, composed of compound eyes, ocelli and Hofbauer-Buchner eyelet. These organs contain photoreceptors (PRs) expressing six different light receptors named rhodopsins (Rh1 to Rh6). In addition, one rhodopsin (Rh7) is found in some of the clock neurons in the brain. Most of the clock cells also express another type of light receptor, Cryptochrome (Cry). Most studies about clock entrainment by light have focused on the Cry-dependent light input, which allows short light pulses to reset the brain clock. The present thesis focuses on the entrainment of the brain clock through rhodopsins. In photoreceptors, rhodopsins capture photons and activate a transduction cascade, where a key player is the phospholipase C (PLC) encoded by norpA. Mutants deficient for Cry and NorpA do not synchronize at low light intensity but still entrain with high light, indicating that an unknown NorpA-independent pathway is also used by the clock. Light induces a depolarization of the PRs, which release histamine as a neurotransmitter, but their role in circadian entrainment is unknown. Which type of rhodopsine-expressing photoreceptors are implicated? After the phototransduction cascade activation and the release of histamine from the photoreceptors, which downstream neurons expressing the histamine-gated chloride channels Ort and Hiscl1 (whose function has been studied in the visual behavior) are involved in the circadian entrainment? The first part of the thesis was to study the function of the 6 PR rhodopsins in circadian entrainment. I first contributed to studying the function of the specific photoreceptors in the NorpA-dependent pathway (Saint-Charles et al. J Comp Neurol 2016). Then, we generated genotypes having either none or only one of the six PR rhodopsins. Mutants with no Cry and none of the 6 PR rhodopsins could not synchronize with light-dark (LD) cycles (low light or high light). In low light, Rh1 and Rh6 were the main light input for entrainment. In high-light, each one of the 6 PR rhodopsins can provide entrainment, with Rh1, Rh5 and Rh6 being the most efficient (Alejevski et al., in prep).The second part of the work was to identify the neuronal pathways that connect the PRs to the brain circadian clock. Flies deficient for Cry and the two histamine receptors are circadianly blind, whereas Cry mutants having either Ort or Hiscl1 are able to entrain. Thus, each one of the two receptors supports circadian entrainment. Rescuing Ort or Hiscl1 in the clock cells could not restore entrainment, indicating that there is no direct histaminergic connection between PRs and clock neurons. Our rescue experiments revealed several pathways in otic lobes that rely on Ort-expressing interneurons to entrain the clock. In contrast and unexpectedly, we observed that the expression of Hiscl1 in PRs but not in interneurons was involved in circadian entrainment. In fact, only Hiscl1 expression in Rh6 PRs mediates entrainment. Our work thus reveals Rh6-expressing PRs as both photoreceptors and histamine-receiving interneurons in the rhodopsin-dependent entrainment pathway, which recalls the role of melanopsin-expressing retinal ganglion cells in the mammalian retina (Alejevski et al. Nat Commun, in revision)