Rhythms, Reward, and Blues: Consequences of Circadian Photoperiod on Affective & Reward Circuit Function

Abstract

Circadian disruptions, along with altered affective and reward states, are commonly associated with psychiatric disorders. In addition to genetics, the enduring influence of environmental factors in programming neural networks is of increased interest in assessing the underpinnings of mental health. The duration of daylight or photoperiod is known to impact both the serotonin and dopamine systems, which are implicated in mood and reward-based disorders. This review first examines the effects of circadian disruption and photoperiod in the serotonin system in both human and preclinical studies. We next highlight how brain regions crucial for the serotoninergic system (i.e., dorsal raphe nucleus; DRN), and dopaminergic (i.e., nucleus accumbens; NAc and ventral tegmental area; VTA) system are intertwined in overlapping circuitry, and play influential roles in the pathology of mood and reward-based disorders. We then focus on human and animal studies that demonstrate the impact of circadian factors on the dopaminergic system. Lastly, we discuss how environmental factors such as circadian photoperiod can impact the neural circuits that are responsible for regulating affective and reward states, offering novel insights into the biological mechanisms underlying the pathophysiology, systems, and therapeutic treatments necessary for mood and reward-based disorders.

Introduction

One key question in neuroscience research is: how do environmental factors induce enduring changes to the function of neural circuits resulting in increased risk for neuropsychiatric disorders? A pervasive environmental factor that appears to play a critical role in this line of questioning is the duration of daylight, or circadian photoperiod. Circadian influences have long been associated with psychiatric disorders (McClung, 2013); however, the neural substrates are still not fully understood. The circadian system is known to impact both mood and reward-based circuitry (Logan et al., 2014; Ketchesin et al., 2018), which are critical components of mental health. Especially relevant, photoperiod exposure results in plasticity and developmental programming of the circadian system itself (Ciarleglio et al., 2011b), along with downstream enduring effects in both the serotonin and dopamine systems (Green et al., 2015; Young et al., 2018), key nodes in the mood and reward systems. We argue that changes to circuits underlying mood and reward, driven through circadian encoding of photoperiod, can play an important role in the risk for psychiatric disorders. We propose that circadian photoperiod influences aspects of both of these overlapping circuits, and modulates risk for mood and reward-based disorders, which may be of great relevance for future diagnosis and treatment. In this review, we will discuss how circadian influences and environmental factors, such as photoperiod, play an integral role in modulating the underlying mechanisms and circuits necessary for mood and reward-based pathophysiology.

Influences on mood by the circadian and the serotonergic systems

Mood disorders have a significant health impact. Globally it is estimated that over 300 million individuals suffer from depression and 16 million Americans have reported experiencing at least one depressive episode ((NIMH), 2019; (WHO), 2019). This results in about 7% of the US population being affected by depression with an estimated economic cost of over $200 billion ((CDC), 2016; (NIMH), 2019). Anxiety disorders are estimated to impact 19% of individuals in the US with approximately 31% of adults experiencing an anxiety disorder during their lifetime ((NIMH), 2017).

Seasonal affective disorder or “winter depression” results in individuals presenting atypical depressive symptoms (Magnusson and Boivin, 2003) arising most frequently during the fall and winter seasons and typically regressing during the spring and summer seasons, providing a clear link between circadian alterations and mood (Zauderer and Ganzer, 2015). In addition, seasonal affective disorder is estimated to impact approximately 5% of the US population (Kurlansik and Ibay, 2012) and is commonly comorbid with other mood disorders such as major depression and anxiety (Winthorst et al., 2017). Lastly, treatments for seasonal affective disorder and major depression normally target the serotonin system with the use of antidepressants, however, light therapy has also been shown to be an effective treatment option for both of these disorders as well (Tuunainen et al., 2004; Terman and Terman, 2005; Even et al., 2008). Thus, affective disorders impact the global population, present a great economic cost, and due to high comorbidity suggest common underlying mechanisms and circuits, which may be specifically sensitive to circadian influences.

Clinical observations have consistently linked circadian disruption with several psychiatric disorders including major depression, anxiety, bipolar disorder, addiction, stress, and seasonal affective disorder (Benedetti et al., 2003; Magnusson and Boivin, 2003; Nievergelt et al., 2006; McClung, 2007a; Glickman, 2010; Sipilä et al., 2010; Landgraf et al., 2014a; Parekh et al., 2015; Geoffray et al., 2016; Ketchesin et al., 2020). In addition, these neuropsychiatric disorders are significantly associated with the serotonin system (Cook and Leventhal, 1996; Lesch et al., 1996; Mahmood and Silverstone, 2001; Magnusson and Boivin, 2003; Eley et al., 2004; Willeit et al., 2008; Daut and Fonken, 2019). The circadian and serotonin systems share underlying neural connections, with the central pace-maker in the brain, the suprachiasmatic nucleus (SCN), being upstream of and projecting to the dorsal raphe nucleus (DRN) (Ciarleglio et al., 2011a), the main hub for serotonin (5-HT) signaling in the brain (Gaspar et al., 2003) (Figure 1). Affective disorders, in particular, are classically associated with atypical signaling and circuitry in the serotonin system, are highly prevalent worldwide, and are thus a main focus of this review.

Figure 1.

Connecting the circadian (SCN), mood (DRN), and reward (VTA, NAc) systems. Note the solid lines represent direct connections, and the dashed lines represent indirect connections. SCN = suprachiasmatic nucleus, DRN = dorsal raphe nucleus, VTA = ventral tegmental area, and NAc = nucleus accumbens. Listed are the following direct connections: DRN to SCN (Kiyasova et al., 2011; Paulus and Mintz, 2012), DRN to VTA (Vertes and Linley, 2007), DRN to NAc (Vertes and Linley, 2007), VTA to SCN (Grippo et al., 2017), VTA to DRN (Kirouac et al., 2004), VTA to NAc (Russo and Nestler, 2013), and NAc to VTA (Russo and Nestler, 2013). Listed are the following indirect connections: SCN to DRN (Deurveilher and Semba, 2008) and SCN to VTA (Luo and Aston-Jones, 2009).

Studies in humans have identified genetic associations within both the circadian and serotonin systems with affective disorders (McClung, 2007a). For example, genes that form the core of the molecular circadian clockwork (i.e., clock genes), such as Clock Locomotor Output Cycles Kaput (CLOCK), Brain and Muscle ARNT-Like 1 (BMAL1), Period (Per), and Cryptochrome (Cry) have been associated with depression, anxiety, and bipolar disorder (Benedetti et al., 2003; Nievergelt et al., 2006; Benedetti et al., 2008; Sipilä et al., 2010; Partonen, 2012; Bunney et al., 2015; Buoli et al., 2018). Polymorphisms in key serotonin signaling genes such as the rate-limiting enzyme for serotonin synthesis, tryptophan hydroxylase (TPH), the serotonin 2A receptor (5-HT 2A), and specifically the serotonin transporter (SERT), are implicated in affective disorders as well (Lesch et al., 1996; Caspi et al., 2003; Eley et al., 2004; Zill et al., 2004; Canli and Lesch, 2007; Uher and McGuffin, 2008). Importantly, studies have also found associations between clock genes such as Per2, ARNTL, NPAS2, and risk for seasonal affective disorder or winter depression (Johansson et al., 2003; Partonen et al., 2007; Westrin and Lam, 2007). In addition, patients with seasonal affective disorder have shown alterations in the 5-HT 2A receptor gene along with SERT function and binding (Rosenthal et al., 1998; Enoch et al., 1999; Arias et al., 2001; Praschak-Rieder et al., 2008; Willeit et al., 2008; Kalbitzer et al., 2010; Tyrer et al., 2016). Serotonin depletion in these individuals can result in significant depressive symptoms during remission (Lam et al., 1996; Neumeister et al., 1997; Neumeister et al., 1998; Lam et al., 2000), and both selective serotonin reuptake inhibitors and light therapy are known to be beneficial treatment options (Terman et al., 1989; Swedo et al., 1997; Terman and Terman, 2005; Lam et al., 2006; Kurlansik and Ibay, 2012). Lastly, altered diurnal patterns of both mood and serotonin levels (Pietraszek et al., 1992; Peeters et al., 2006), shifted and decreased circadian gene expression patterns in brain regions implicated in depression (Li et al., 2013), and disrupted circadian activity patterns which correlates with decreased affect, have all been observed in individuals with major depression (Lyall et al., 2018). This underscores the fact that disruptions to and the interactions between the circadian and serotonin systems, in humans, may be integral in the pathophysiology of affective disorders (Daut and Fonken, 2019).

Preclinical animal model research has demonstrated reciprocal interactions and closely intertwined circuitry between the circadian and serotonin systems (Cagampang and Inouye, 1994; Recio et al., 1996; Cuesta et al., 2008; Ciarleglio et al., 2011a; Paulus and Mintz, 2012; Landgraf et al., 2016a). Studies have shown that activating the serotonin system by electrically stimulating the DRN or with pharmacology targeting 5-HT1A/7 receptors can result in release of 5-HT in the SCN, altered clock gene expression levels, and phase resetting of circadian behavior (Glass et al., 2000; Cuesta et al., 2008). Also, genetically knocking out Pet-1, a transcriptional factor needed for proper 5-HT neuron development, results in period lengthening of the circadian rhythms in the SCN and increased circadian behavior activity levels (Paulus and Mintz, 2013; Ciarleglio et al., 2014). Conversely, manipulating the circadian day length can increase DRN 5-HT content, shift the diurnal peak of serotonin content, and knocking out clock genes in animal models results in significant increases in mania and depressive-like behaviors (Cagampang et al., 1993; Roybal et al., 2007; Barnard and Nolan, 2008; Landgraf et al., 2014b). In addition to genetic disruptions, studies have investigated the effects of environmental factors such as jet lag or shift work on mental health (Vogel et al., 2012; Foster et al., 2013).

Rapid travel across multiple time zones (i.e., jet lag) or exposure to light during atypical work hours relative to the light/dark cycle (i.e., shift work) can result in significant disruptions to the SCN and circadian system (Navara and Nelson, 2007; Choy and Salbu, 2011). In addition, jet lag and shift work has been shown to impact the serotonin system and is associated with decreased mood and a higher incidence of affective disorders (Montange et al., 1981; Versteeg et al., 2015; Lee et al., 2017; Daut and Fonken, 2019). For example, altered circadian light cycles and light presented at night, either chronically or even acutely, can produce deleterious effects such as the desynchronization of circadian clock neurons, altered circadian rhythms and clock gene expression levels, learning and memory deficits, and elevated anxiety and depressive-like behaviors (Ohta et al., 2005; Ohta et al., 2006; Fonken et al., 2009; Bedrosian et al., 2011; Fonken and Nelson, 2011; LeGates et al., 2012; Bedrosian and Nelson, 2013; Fonken and Nelson, 2013; Bedrosian and Nelson, 2017; Walker et al., 2020a; Walker et al., 2020b). Interestingly, it was found that chronic selective serotonin reuptake inhibitor (SSRI) treatment could reverse some of these affective behavioral deficits due to circadian disruptions (Bedrosian et al., 2012). These studies demonstrate that manipulation of the circadian system via genetic alterations or environmental factors directly impacts affective behavior, and regulates key components of the serotonin system (Ciarleglio et al., 2011a).

Photoperiodic programming of the serotonin system

An important environmental factor, the duration of daylight or photoperiod has been implicated as a risk factor for numerous psychiatric disorders (Modai et al., 1994; Torrey et al., 1997; Castrogiovanni et al., 1998; Foster and Roenneberg, 2008; Lee et al., 2008; Disanto et al., 2012; Tonetti et al., 2012). Monoamine turnover of serotonin and dopamine is lower during the winter and fall seasons (Chotai and Adolfsson, 2002; Lambert et al., 2002), SERT binding can fluctuate across the seasons, (Praschak-Rieder et al., 2008; Kalbitzer et al., 2010; Tyrer et al., 2016), and an interaction between candidate genes for affective disorders and births in winter seasons has been identified, demonstrating a gene × environment risk for these disorders (Chotai et al., 2003). These findings are important because they highlight how photoperiod can modulate aspects of the serotonin and dopamine systems, underscoring the impact that photoperiod may have on the development and underlying mechanisms associated with affective disorders. Exposure to early life adversity and environmental factors such as stress and trauma during key neurodevelopmental time points including gestation and postnatal development have been associated with increased risk for psychiatric disorders later in life (Nestler et al., 2002; Andersen, 2015). Recently, human epidemiological work has demonstrated that high magnitude photoperiodic changes during the second trimester of gestation can result in decreased risk for depression in the offspring later in life (Devore et al., 2018). Thus, the day length or photoperiod is a critical environmental factor associated with psychiatric disorders, and importantly may play an underappreciated role in the development and pathophysiology of affective disorders (Figure 2).

Figure 2.

Hypothetical model connecting photoperiodic input with circadian (SCN), mood (DRN), and reward (VTA, NAc) systems and the associated psychiatric disorders. A) We hypothesize that that the SCN acts by mediating circadian entrainment under Long summer-like photoperiods (Ciarleglio et al., 2011b) and impacting the DRN, VTA, and NAc resulting in a potential decreased risk for anxiety, depression, or substance use disorders. B) We hypothesize that that the SCN acts by mediating circadian entrainment under Short winter-like photoperiods (Ciarleglio et al., 2011b) and impacting the DRN, VTA, and NAc resulting in a potential increased risk for anxiety, depression, or substance use disorders.

The serotonergic system is impacted by the duration of daylight and has been consistently implicated in affective disorders (Uher and McGuffin, 2008; Ciarleglio et al., 2011a; Spindelegger et al., 2012). Animal studies have shown that aberrant serotonin signaling during prenatal and perinatal development can result in dramatic and lasting molecular, systems level and behavioral changes in adulthood (Bennett et al., 2002; Gaspar et al., 2003; Bonnin et al., 2011; Bonnin and Levitt, 2011). Knockout rodent models targeting various aspects of the serotonergic system exhibit atypical serotonin (5-HT) neuronal development and viability, (Azmitia, 2001; Hendricks et al., 2003; Persico et al., 2003) altered circuit formation and monoamine content (Bennett-Clarke et al., 1993; Cases et al., 1995; Cases et al., 1996; Upton et al., 1999; Richardson-Jones et al., 2011; Migliarini et al., 2013) along with increased anxiety and depressive-like behaviors (Gross et al., 2002; Holmes et al., 2002; Holmes et al., 2003; Kalueff et al., 2007).

Preclinical rodent studies have shown that, depending on the daylight duration, photoperiodic exposure can significantly increase DRN serotonin neuronal firing rate (Green et al., 2015; Giannoni-Guzman, 2020), elevate monoamine signaling (Otsuka et al., 2014; Goda et al., 2015; Siemann, 2020), and reduce anxiety and depressive-like behaviors (Prendergast and Nelson, 2005; Einat et al., 2006; Pyter and Nelson, 2006; Krivisky et al., 2011; Xu et al., 2016; Takai et al., 2018). For example, prior work has shown that animals developed under Long (summer-like) photoperiods demonstrate enduringly increased firing rates in DRN serotonin neurons, elevated 5-HT and norepinephrine (NE) concentrations in the midbrain, along with decreased time spent immobile in the forced swim task and reduced time spent in the closed arms in the elevated zero maze task compared to mice developed under either Equinox or Short (winter-like) photoperiods (Green et al., 2015). Green et al also demonstrated that these photoperiodic effects were melatonin 1 receptor dependent, and the increased firing rate in DRN serotonin neurons is driven by developmental (i.e., prior to adolescence) rather than proximal photoperiodic effects and lasted up to 5 months after initial photoperiod exposure (Green et al., 2015). These were the first findings to show the enduring impact of photoperiod on 5-HT neurophysiology, monoamine signaling, and the associated affective behaviors. In addition, these results suggested that developmental periods may exist, which could be sensitive to photoperiodic programming, and further indicated a potential role in the pathology of affective disorders.

As a result of these novel findings, we have recently investigated the effects of photoperiodic programming during sensitive periods of prenatal and postnatal development. We found that prenatal Long photoperiodic exposure, in mice, results in enduring changes increasing DRN 5-HT neuronal firing rate, when evaluated in adolescence and adulthood (Siemann et al., 2019). In addition, there are critical temporal windows within postnatal development that are programmed by Long photoperiod resulting in lasting changes to gene expression, monoamine signaling, and affective behaviors (Siemann et al., 2019; Siemann, 2020). Specifically, Long (summer-like) photoperiod exposure during gestation consistently impacts the serotonin system by elevating DRN 5-HT neuronal firing rate, increases midbrain monoamine concentrations, and decreases affective behavior later in life (Siemann et al., 2019). We hypothesize that photoperiodic programming may occur in sequential steps. First, with lasting changes to DRN neuronal firing rate occurring prenatally, which may then “set the serotonin system”, followed by programming of monoamine signaling and behavior during sensitive periods of postnatal development. Importantly, there is emerging clinical and preclinical evidence implicating photoperiod in the development of mood disorders (Devore et al., 2018; Siemann et al., 2019; Siemann, 2020). While these effects are intriguing and warrant further investigation; independent of development, circadian photoperiod clearly impacts the cellular, signaling, and behavioral components of the serotonin system. This begs the question – are the outputs and targets from this system, specifically from the DRN, similarly impacted by circadian influences and photoperiod, resulting in lasting changes to underlying circuits critical for affective disorders?

Connecting mood & reward

There is overwhelming evidence demonstrating the role of the serotonin and dopamine systems in mood and reward (Nestler et al., 2002; Anguelova et al., 2003; Nestler and Carlezon Jr, 2006; Ishikawa et al., 2013; Russo and Nestler, 2013). Studies have shown genetic associations specifically between SERT and depression (Lesch et al., 1996; Jönsson et al., 1998; Canli and Lesch, 2007), the role of environmental factors such as stress on serotonergic and dopaminergic genes, demonstrating gene × environment associations (Pani et al., 2000; Caspi et al., 2003; Eley et al., 2004; Martens and van Loo, 2007; Uher and McGuffin, 2008; Heim and Binder, 2012), and most commonly prescribed antidepressants target the serotonin and dopamine systems (Pirraglia et al., 2003; Dailly et al., 2004). Serotonin is consistently implicated in affective disorders (Anguelova et al., 2003); however, studies have more recently investigated the role of this system in disorders of reward, including addiction (Kirby et al., 2011; Müller and Homberg, 2015; Heifets et al., 2019).

Classic hallmarks of reward-based disorders include dopamine gene variants, atypical dopaminergic signaling, and overall alterations to the dopamine system (Nestler and Carlezon Jr, 2006; Volkow et al., 2011). In addition, clinical observations have shown associations between affective disorders and the dopaminergic system, specifically in depression (Nestler et al., 2002; Dunlop and Nemeroff, 2007; Krishnan and Nestler, 2008; Fox and Lobo, 2019). Interestingly, comorbidities between substance abuse and mood disorders have been established (Volkow, 2004; Quello et al., 2005; Destoop et al., 2019), and alterations in the serotonin and dopamine systems have also been found in individuals with seasonal affective disorder (Neumeister et al., 1998; Rosenthal et al., 1998; Enoch et al., 1999; Lam et al., 2000; Praschak-Rieder and Willeit, 2011). These findings highlight that disruptions to the serotonin and dopamine systems are commonly found in affective and reward-based disorders, strongly suggesting shared and overlapping circuitry, and demonstrate that both of these systems are impacted by environmental factors, including circadian photoperiod.

Underlying circuitry connecting the serotonergic & dopaminergic systems

Major targets of the DRN include the ventral tegmental area (VTA) and nucleus accumbens (NAc), which are critical to the dopamine system and reward-based circuitry involved in mood disorders (Nestler et al., 2002; Nestler and Carlezon Jr, 2006; Russo and Nestler, 2013). The mesolimbic dopamine (DA) system has been extensively implicated in depression in both preclinical and clinical studies (Treadway and Zald, 2011; Friedman et al., 2014; Addy et al., 2015; Small et al., 2016). Both the NAc and VTA integrate information in the form of glutamatergic inputs from limbic, cortical, and thalamic regions and neuromodulatory inputs from key structures including serotonergic inputs from the dorsal raphe (Soghomonian et al., 1989; Brown and Molliver, 2000; Turner et al., 2017) (Figure 1). For example, electrical stimulation of the DRN can result in increased dopamine release in the NAc (De Deurwaerdère and Spampinato, 1999), along with modulation of DA neuronal firing rate and response properties in the VTA (Gervais and Rouillard, 2000).

The DRN, VTA, and NAc are key components of serotonergic and dopaminergic circuitry that have been associated with affective and reward-based disorders (Nestler and Carlezon Jr, 2006; Cohen et al., 2012; Cohen et al., 2015; Wong-Lin et al., 2017). To this point, two recent human imaging studies have demonstrated that patients with major depression who are either medication free or being treated with antidepressants show altered fMRI resting state activity and functional connectivity in the DRN and VTA (Wagner et al., 2017; Wohlschläger et al., 2018). In addition, preclinical model work has shown that ethanol consumption or self-administration results in altered 5-HT levels in the DRN and DA content in the VTA and NAc (McBride et al., 1993), and chronic cocaine administration increases both 5-HT and DA levels in the DRN and NAc (Parsons and Justice Jr, 1993). Studies have also begun to identify the neuronal populations and connections between these brain regions that are necessary for producing these behavioral effects. For example, inhibition of VTA DA neurons can result in depressive-like behavior and phasic activation of these neurons can reverse these behavioral deficits (Tye et al., 2013). While investigating the underlying mechanisms of depressive states, studies have observed that differences in VTA neuronal activity can result in susceptible or resilient behavioral populations (Friedman et al., 2014), the excitation or inhibition of the connections between the VTA and NAc can also produce susceptible or resilient behavioral responses to social-defeat stress (Chaudhury et al., 2013), and lesioning VTA neurons results in elevated depressive-like behavior, which can be reversed with the administration of an SSRI (Winter et al., 2007).

Recent studies have attempted to connect mood and reward-based pathways utilizing optogenetics to investigate the underlying circuitry between the serotonin and dopamine systems (Nakamura, 2013; Liu et al., 2014; Fonseca et al., 2015; Hayashi et al., 2015; Li et al., 2016; Browne et al., 2019; Li et al., 2019; Wang et al., 2019; Nagai et al., 2020). Studies have shown that nonserotonergic DRN neurons provide glutamatergic excitation on VTA DA neurons and are responsible for reinforcing behavior (McDevitt et al., 2014), DRN 5-HT neurons fire tonically for reward and phasically for reward acquisition, and DRN GABAergic neurons are inhibited for reward stimuli presentations (Li et al., 2016). In addition, DRN 5-HT neuronal activity over short compared to long time scales produce rewarding vs. aversive responses (Nakamura, 2013; Hayashi et al., 2015), and direct stimulation of DRN 5-HT terminals projecting to the VTA can impact reward-like behavior (Browne et al., 2019).

Prior work utilizing Pet-1 Cre-driven expression of channelrhodopsin to target optogenetic stimulation of DRN 5-HT neurons, found this results in the release of both 5-HT and glutamate leading to increased self-administration and reward-based behaviors (Liu et al., 2014). In addition, DRN 5-HT-vGlut3 terminals synapse onto DA VTA neurons, which activates the 5-HT terminals in the VTA causing the excitation of VTA DA cells (Wang et al., 2019). Excitation of these 5-HT terminals in the VTA contribute to increasing reward-like behavior via cocaine conditioned place preference (CPP) and downstream release of DA in the NAc (Wang et al., 2019). Also, using a Tph2 promoter, either stimulation or inhibition of DRN 5-HT neurons projecting to the VTA can result in significant increases and decreases in reward-based behavior, respectively (Nagai et al., 2020). Lastly, DRN 5-HT neuronal projections to the NAc have been associated with reward-based disorders (Chang et al., 2011; You et al., 2016). 5-HT neuronal projections and specifically 5-HT1A receptors are critical for DRN-NAc circuits involved in reducing reward behavior, via CPP, and producing anti-depressive-like behaviors (You et al., 2016). In addition, it has been found that 5-HT1B receptors in the NAc are necessary for significantly decreasing social reward behavior (Dölen et al., 2013). These elegant circuit-based studies demonstrate the complexity between the serotonin and dopamine systems, provide novel insights into the underlying mechanisms necessary for mood and reward, and present novel opportunities to investigate circadian-driven DRN 5-HT effects in the VTA and NAc.

There are clear connections and direct interactions between the key components of the serotonin (i.e., DRN) and dopamine (i.e., VTA and NAc) systems with studies identifying the underlying neural populations and circuits needed to produce affective and reward-based behaviors. Circadian influences and specifically photoperiodic exposure can result in dramatic and lasting changes to various components of the serotonin system. Alterations in the dopaminergic system are classic hallmarks of mood and in particular reward-based disorders. As the serotonergic and dopaminergic systems clearly share overlapping circuitry, it is then logical to ask 1) if circadian changes and specifically photoperiodic exposures produce enduring effects to dopamine rich nuclei, and 2) does this result in potential circadian-dependent changes in circuitry critical for mood and reward?

Bidirectional regulation of reward by the circadian and dopaminergic systems

In the US it is estimated that 20 million adults suffer from addiction or a substance abuse disorder ((NIMH), 2016), resulting in an estimated $740 billion economic cost annually ((NIDA), 2020), and approximately 8 million individuals present comorbidities with additional psychiatric disorders ((NIMH), 2016). In addition, there are known comorbidities between substance abuse and affective disorders (Quello et al., 2005) further highlighting the shared underlying mechanisms between mood and reward. Psychiatric disorders such as addiction, stress, major depression, bipolar, and substance abuse disorder are associated with the dopamine system (Pani et al., 2000; Nestler and Carlezon Jr, 2006; Moriam and Sobhani, 2013; Russo and Nestler, 2013; Belujon and Grace, 2015), with alterations in the circadian system being consistently identified as well (Benedetti et al., 2003; McClung, 2007a; 2013; Landgraf et al., 2014a; Logan et al., 2014). Substance abuse disorders are highly prevalent, present a significant economic cost, and based on the direct regulation by circadian influences, warrant further investigation into these underlying relationships.

Human genetic associations have been found between the circadian system and substance abuse disorders (McClung, 2007b; Falcón and McClung, 2009; Parekh et al., 2015; Partonen, 2015). For example, polymorphisms in Clock and Per genes have been implicated in alcohol, cocaine use, and addiction (Spanagel et al., 2005; Kovanen et al., 2010), and have been found in individuals displaying comorbid mood disorders as well (Sjöholm et al., 2010). Animal model studies have demonstrated that circadian disruptions via clock gene knockouts and knockdowns can disrupt dopamine rich nuclei, such as the VTA and NAc, and produce a variety of reward and addiction-like behaviors (Andretic et al., 1999; Falcón and McClung, 2009; Logan et al., 2014; Parekh et al., 2015). For example, knockout animal models or mutations to Per genes can result in altered alcohol consumption and cocaine reward-based behavior, however differential findings are observed when evaluating either Per1 or Per2 genes (Abarca et al., 2002; Spanagel et al., 2005; Zghoul et al., 2007; Perreau-Lenz et al., 2009). In addition, disruptions to Clock can result in elevated dopaminergic neuronal activity and tyrosine hydroxylase expression levels in the VTA along with increased reward and self-administrative behaviors (McClung et al., 2005; Coque et al., 2011). These Clock mutant mouse models also present manic-like behavior, such as hyperactivity, decreased sleep, and reduced anxiety and depressive-like behaviors, which can be reversed with lithium treatment along with viral administration of a functional Clock into the VTA (Roybal et al., 2007). Intriguingly, genetic knockdown of Clock specifically in the VTA results in altered affective and reward-based behaviors, and disrupts circadian rhythms in the SCN along with circadian behavior (Mukherjee et al., 2010). Lastly, Clock mutant animals and knockdown of VTA Clock expression levels produces altered glutamatergic tone, disruptions in clock gene expression levels in the VTA, and elevated ethanol consumption behavior (Ozburn et al., 2013).

In addition, environmental factors such as jet lag, shift work, and the time of day can modulate aspects of the dopaminergic system as well (Webb, 2017). Jet lag and shift work have been associated with increased risk for substance abuse disorders (Trinkoff and Storr, 1998; Bildt and Michélsen, 2002; Rogers and Reilly, 2002; Dorrian and Skinner, 2012), and preclinical work in rodents has shown that shifts to the circadian light/dark cycle can modulate alcohol consumption (Gauvin et al., 1997; Clark et al., 2007; Rosenwasser et al., 2010). In addition, the circadian and dopamine systems can regulate each other, demonstrating reciprocal interactions. For example, dopamine D1 receptors in the SCN are necessary for synchronizing circadian rhythms (Grippo et al., 2017), VTA dopaminergic neurons project to the SCN (Grippo and Güler, 2019), and dopamine and dopamine receptors demonstrate circadian diurnal expression levels in the VTA and NAc (Schade et al., 1995; Shieh et al., 1997; Weber et al., 2004). Lastly, the time of day can directly impact cocaine self-administration behavior in rodents (Baird and Gauvin, 2000; Sleipness et al., 2005), VTA dopamine neuronal firing rate along with tyrosine hydroxylase activity demonstrate diurnal rhythms, and this dopaminergic rhythmic activity throughout the day is responsible for manic-like behaviors (Sidor et al., 2015). Thus, there is evidence of not only genetic, but environmental circadian influences that can modulate various aspects of dopaminergic neurobiology and reward behaviors.

One clear link connecting serotoninergic and dopaminergic circuits is through the circadian system (Figure 1). Genetic manipulations of key clock genes in the VTA and NAc results in altered neuronal firing rate and anxiety and depressive-like behaviors (Mukherjee et al., 2010; Coque et al., 2011; Spencer et al., 2013; Landgraf et al., 2016b). Intriguing findings indicate that the VTA and NAc contain molecular circadian clockworks similar to the SCN central clock (Logan et al., 2014; Landgraf et al., 2016b; Porcu et al., 2020). Targeted knockdowns of Clock in the VTA results in dopaminergic neurophysiological and reward-based behavioral deficits similar to those observed utilizing global Clock mutations, and reintroducing a functional Clock into the VTA can reverse these effects (McClung et al., 2005; Roybal et al., 2007; Mukherjee et al., 2010; Coque et al., 2011; Ozburn et al., 2013). These findings critically demonstrate a direct relationship between circadian influences and dopaminergic nuclei such as the VTA, and these interactions appear to be critical for addiction and reward processing (Parekh et al., 2015).

Interestingly, disruption of NAc circadian molecular rhythms is associated with vulnerable rather than resilient populations in a learned helplessness model, indicating a link between NAc clock function and depressive-like behaviors (Landgraf et al., 2016b). Furthermore, specific genetic elements of the NAc clockworks regulate dopamine signaling in dopamine 1 receptor medium spiny neurons (D1-MSNs), and genetic disruption of the NAc circadian clock is associated with reduced helpless behavior (Porcu et al., 2020). Prior work has also demonstrated that developmental photoperiod drives enduring changes in the molecular waveform of the central SCN circadian clock, along with long-term changes in how the SCN responds to photoperiods later in life (Ciarleglio et al., 2011b). Clearly circadian genetic and environmental influences impact the dopaminergic system directly resulting in changes to reward and mood. This begs the question – can circadian photoperiod drive similar enduring changes in the molecular clockworks of the VTA and NAc, which impact affective and reward behaviors?

Mechanisms of photoperiodic programming on dopaminergic circuitry

Both human and animal model studies have shown that photoperiodic exposure can result in lasting changes to numerous aspects of the dopamine system, including neurophysiology (Meng et al., 2018), monoamine levels (Dulcis et al., 2013; Aumann et al., 2016; Itzhacki et al., 2018), and reward/addiction-like behaviors (Sorg et al., 2011; Young et al., 2018). Circadian variation in the dopamine transporter (DAT) function has been shown to underlie daily rhythms in dopaminergic tone in the NAc core (Ferris et al., 2014), seasonal changes can impact SERT function, and both SERT and DAT have been associated with seasonal affective disorder (Rosenthal et al., 1998; Neumeister et al., 2001; Praschak-Rieder et al., 2008; Kalbitzer et al., 2010; Praschak-Rieder and Willeit, 2011; Tyrer et al., 2016) suggesting that these transporters may be potential key points of regulation by circadian photoperiod. To this point, prior work has shown that circadian photoperiod can impact reward-based behavior differentially in DAT KO animals, highlighting a direct role of the effects of photoperiod on DAT for reward-like behaviors (Young et al., 2018). These findings demonstrate that circadian photoperiod can impact dopamine rich nuclei, however, future studies will need to investigate the neural components under differing photoperiodic conditions to more fully investigate these brain-behavior based relationships.

Long term functional changes in NAc neurons along with VTA dopaminergic and GABAergic neuron synapses are key features of mood disorders (Russo and Nestler, 2013; Fox and Lobo, 2019). The NAc and VTA receive convergent glutamatergic inputs from multiple regions, each thought to encode specific information, which is integrated and relayed through the reward system (Lüscher and Malenka, 2011; Grueter et al., 2012; Turner et al., 2017). Activity-dependent changes in synaptic strength at glutamatergic synapses are considered to be the fundamental mechanism in information processing and storage in the brain, and contribute to the development of neural circuit and experience-dependent behavioral plasticity. Consistently, the molecular and cellular bases of depressive-like behavior in animal models include changes in glutamatergic and GABAergic synapses, and membrane excitability in VTA and NAc neurons (Russo and Nestler, 2013). The NAc is comprised of parallel output loops (D1 and D2 dopamine receptor expressing medium spiny neurons; MSNs) whose function is differentially linked to behavioral outcomes (Lobo et al., 2010; Bock et al., 2013; Pascoli et al., 2015). The functional output of the NAc is gated by the strength of glutamatergic synapses onto D1 and D2 dopamine (DA) receptor-expressing GABAergic MSNs (Turner et al., 2017; Baimel et al., 2019). Adding to the complexity of the circuitry, the expression of a depressive-like behavioral state correlates with differential reorganization of excitatory synapses onto D1, but not D2 MSNs (Lim et al., 2012; Francis et al., 2017; Francis et al., 2019), and synaptic and membrane property adaptations at both DAergic and GABAergic neurons in the VTA have been reported as well (O’Dell and Parsons, 2004; Wang et al., 2019). Often overlooked, but functionally impactful are interneurons within the NAc (Schall et al., 2020). These microcircuits are necessary for behaviors associated with psychiatric disorders (Wang et al., 2018; Pisansky et al., 2019), and receive similar excitatory inputs that also express synaptic plasticity (Manz et al., 2020). However, there is a paucity of information regarding consequences of photoperiod on reward circuit synapses leading to the question: does photoperiod drive similar enduring synaptic plasticity of the NAc that then impact affective and reward-based behaviors?

Importantly, studies have evaluated the effects of clock gene disruption on dopaminergic synaptic plasticity in the NAc and the direct impact this has on the resultant reward-like behaviors (Parekh et al., 2018; Parekh et al., 2019). Clock mutant animals demonstrate reduced glutamate receptor (GLUA1) expression levels in the NAc, decreased synaptic excitatory strength and transmission along with lower intrinsic excitability in dopaminergic MSNs, and overexpressing GLUA1 directly in the NAc can rescue manic-like and reward-based behavioral deficits (Parekh et al., 2018). In addition, recent work has shown that knockdown of a key circadian gene, NPAS2, in the NAc, results in altered synaptic strength in specifically D1-MSNs, and NPAS2 expression in D1-MSNs, not D2-MSNs, is necessary for cocaine-induced behavior (Parekh et al., 2019). Based on these prior findings, we hypothesize that circadian photoperiodic exposure may differentially scale excitatory gain in the VTA and NAc, which may be responsible for the observed enduring changes in affective and reward-like behaviors. These studies highlight the cell-type specificity needed for circadian influences on reward-dependent circuitry and behavior, and allow for novel opportunities to investigate the effects of photoperiod on synaptic and neural drive in the VTA and NAc.

Enduring effects of circadian photoperiod on mood and reward

Studies have shown that photoperiod can program aspects of the circadian system including the neural firing rate in the master pacemaker of the brain, the SCN, along with circadian-based behavior (Ciarleglio et al., 2011a; Ciarleglio et al., 2011b; Tackenberg and McMahon, 2018). In addition, developmental photoperiod can impact brain regions downstream of the SCN, by enduringly programming neurons (Green et al., 2015; Siemann et al., 2019; Giannoni-Guzman, 2020), monoamine content (Otsuka et al., 2014; Goda et al., 2015), gene expression levels (Siemann, 2020), and affective behaviors associated with the DRN and serotonin system (Prendergast and Nelson, 2005; Einat et al., 2006; Pyter and Nelson, 2006; Krivisky et al., 2011). In addition, we have highlighted the sensitive periods of prenatal and postnatal development that are necessary for these effects (Takai et al., 2018; Siemann et al., 2019), and the first evidence, in humans, of prenatal photoperiodic exposure being associated with decreased risk for mood disorders in the offspring later in life (Devore et al., 2018). In preclinical and clinical studies, even relatively brief exposures to photoperiod result in lasting changes to dopaminergic firing rate (Domínguez-López et al., 2014; Meng et al., 2018), monoamine content (Dulcis et al., 2013; Aumann et al., 2016; Itzhacki et al., 2018), and reward-like behaviors (Sorg et al., 2011; Young et al., 2018). These findings not only provide evidence that photoperiodic exposure can enduringly impact brain regions across multiple neurotransmitter systems, but also highlights that circadian photoperiod exposure may produce significant circuit-level changes (Figure 2). Further investigations are needed to critically evaluate and determine the neural mechanisms underlying the interactions responsible for producing the system-level changes observed between photoperiod, mood, and reward.

Future studies should first evaluate the direct effects of circadian photoperiod on dopaminergic physiology, signaling, and reward-based behaviors. These investigations would then allow for several lines of questioning into how a pervasive environmental factor such as the duration of daylight or photoperiod can potentially have an enduring impact on circuits (i.e., serotonin and dopamine) that are critically intertwined with disorders of mood and reward. Monoaminergic dysregulation has been implicated to be central in the development of depression-like behavioral states, and the mesolimbic dopamine system, comprising the VTA and NAc, is a major target of the DRN (Soghomonian et al., 1989; Brown and Molliver, 2000). Elegant circuit-based studies have demonstrated how these underlying connections and specific neural populations are responsible for aspects of mood and reward-based behaviors (Cohen et al., 2012; Friedman et al., 2014; Liu et al., 2014; Ogawa et al., 2014; Cohen et al., 2015; Li et al., 2016; You et al., 2016; Wong-Lin et al., 2017; Wang et al., 2019; Nagai et al., 2020). However, there is limited evidence of studies evaluating the circadian-drive on serotonergic and dopaminergic circuitry. Specifically, the effects of photoperiodic programming of DRN 5-HT neurons in the VTA and NAc is unexplored. These findings demonstrate a clear opportunity to pursue circuit-based studies investigating the effects of circadian photoperiodic modulation of the serotonergic and dopaminergic systems. By utilizing tools such as electrophysiology, chemogenetics and optogenetics it would be possible to identify the molecular underpinnings and neuronal populations that are necessary and sufficient to produce these changes, and to further understand the underlying circuitry that is responsible for photoperiodic programming, with the ultimate goal that this would lead to more novel and specific therapeutic targets for affective and reward-based disorders.

Conclusions

Circadian influences directly impact neural circuits associated with both mood and reward-based disorders. Studies have investigated both the genetic components along with environmental factors such as circadian photoperiod to provide further understanding into the underlying mechanisms needed to produce these changes. Emerging evidence demonstrates that photoperiodic exposure results in lasting effects to the underlying neurophysiology, brain regions, and associated behaviors in the serotonin and dopamine systems, when measured in isolation. While evaluating each system individually has been informative in determining the influence of photoperiod on serotonin and dopamine signaling, it is most relevant to then utilize this knowledge to determine how photoperiod impacts the underlying mechanisms and connections between these two systems. More focused circuit-based studies investigating the role of photoperiod in multiple neurotransmitter systems would lead to further understanding into these effects, and may allow for the identification of specific neuronal populations responsible for the resultant behaviors underlying mood and reward. Therefore, evaluating the effects of circadian photoperiod in both the serotonin and dopamine systems, along with the known overlapping circuitry, could provide key insights into the development, pathophysiology, and eventually therapeutic treatments for mood and reward-related disorders.

Rhythms, Reward, and Blues: Consequences of Circadian Photoperiod on Affective & Reward Circuit Function – Highlights.

• The brain’s central biological clock is connected to key nuclei involved in mood and reward

• Circadian factors influence the serotonin and dopamine systems and affective and reward disorders

• Circadian photoperiod may play an underappreciated role in affective and addictive disorders