Physiological mechanisms underlying children’s circannual growth patterns and their contributions to the obesity epidemic in elementary school age children

Summary

Several studies since the 1990s have demonstrated that children increase their body mass index at a faster rate during summer months compared with the school year, leading some to conclude that the out-of-school summer environment is responsible. Other studies, however, have suggested that seasonality may play a role in children’s height and weight changes across the year. This article reviews evidence for seasonal differences in the rate of children’s height and weight gain and proposes potential physiological mechanisms that may explain these seasonal variations.

1 |. INTRODUCTION

Numerous studies from the late 1700s to as recently as 2018^1–8 have reported seasonal variation in the velocity of children’s height and weight growth and metabolic rate.⁹ Seasonality of these variables in children suggests that growth may be driven by endogenous factors (ie, circannual rhythms) and synchronized by environmental cues such as seasonal changes in day length, timing of sunrise, light intensity, and temperature¹⁰ (see Table 1 for definitions of key terms). Understanding how seasonality impacts children’s growth has important implications for our interpretation of epidemiological studies showing that summer is a time when children gain weight at an accelerated rate.⁴ In westernized agrarian societies, summer is often conceptualized as a time when children have greater opportunity to engage in obesogenic behaviors (eg, consumption of sugar and sedentary activities) while the structure of the school year promotes healthy weight related behaviors.^4,11,12 Although structure has been shown to be important for children’s development¹³ and likely contributes to the promotion of a healthy weight status,^14,15 previous research has focused almost exclusively on the effect of behavioral and social influences on weight status while overlooking the various endogenous influences on children’s growth patterns. A better understanding of these biological influences may lead to improved identification of problematic weight gain as opposed to normal weight gain and identification of novel treatment targets.¹⁰ This paper reviews the evidence for circannual rhythms in children’s growth and environmental factors involved in the synchronization of these rhythms. Drawing from the biological rhythms literature, we posit that biological mechanisms likely underlie seasonal rhythms in children’s growth and we offer avenues for future research. The goal of this line of inquiry is to begin to identify potential biological mechanisms affecting children’s growth to enhance our understanding of short-term growth trajectories and improve behavioral approaches to the prevention of obesity in children.

TABLE 1.

Key terms defined

Key Term	Definition
Biological rhythm	A biological, endogenous (occurring within the body) rhythm that persists in the absence of external cues, can be synchronized or entrained by environmental cues such as the earth’s light dark cycle and meal timing, and maintains a similar duration regardless of temperature.
Behavioral rhythm	An overt observable rhythm controlled by the circadian clock (eg, sleep/wake, eating, and activity).
Biological night	The period in a 24-hour cycle that corresponds to when physiological processes supporting sleep and an overnight fast are most prevalent.
Chronotype	Individual differences in the timing of activities, such as wake and sleep and peak functioning.
Circadian rhythm	A biological rhythm with a period or length of about 24 hours.
Circannual rhythm	A biological rhythm with a period or length of about 1 year.
Cortisol	A steroid hormone, in the glucocorticoid family. Cortisol release exhibits a circadian pattern with increasing levels coinciding with wake times and low levels occurring in the evening.
Entrainment	Adjustment or synchronization of the endogenous biological rhythms to equal the period or length of the environmental rhythm (eg, 24 hours or 1 year), and the temporal relationship between the endogenous and environmental rhythms are stable.
Free running period	The length of a rhythm under constant conditions such as constant light or constant darkness (in the absence of environmental cues or zeitgebers). In the case of circadian rhythms, free running period is slightly longer than 24 hours in the majority of humans, while a free running circannual period is slightly shorter than 1 year (10–11 months).
Gene (DNA) Methylation	Process by which methyl groups are added to the DNA molecule to change the activity or function of a DNA segment without changing the sequence, thereby affecting gene expression.
Glucocorticoids	Glucocorticoids are corticosteroids that bind to the glucocorticoid receptor and are present in almost all vertebrates and support life sustaining physiological systems including metabolic and cardiovascular systems. Cortisol is an example of a glucocorticoid.
Glucose homeostasis	A process by which insulin and glucagon regulate levels of blood glucose which are maintained within a narrow or tight range.
Intrinsic	Originating from inside an organism or cell.
Leptin	A hormone that regulates energy balance by inhibiting hunger and increasing feelings of fullness. Leptin is released primarily by the adipocyte.
Light-dark cycle	Refers to the Earth’s 24-hour alternating periods of light (daytime) and dark (night) created by the Earth’s rotation on its axis.
Lipogenesis	The metabolic process through which acetyl-coa is converted to fatty acids, contributing to a small portion of the overall total fat balance in humans.
Lipolysis	Metabolic process through which triglycerides breakdown into glycerol and fatty acids within the cell.
Melanopsin	A photopigment in the retina belonging to the opsin protein family. It plays an important role in the entrainment of circadian rhythms via sensitivity to the light dark cycle as well as other functions.
Melatonin	A neurohormone that is a marker of the biological night in humans. It is often referred to as the hormone of darkness because its release typically occurs at night and can be inhibited or suppressed by light. In humans, melatonin levels are typically low during the day, increase closer to habitual bedtime, remain high throughout the night and decrease around habitual wake time.
Photoperiod	Length of time that an individual or organism is exposed to light (either natural or electrical) in a 24-hour period.
Pineal gland	A small gland organ in the brain of most vertebrates that produces and secretes melatonin.
Superchiasmatic nucleus (SCN)	Paired nucleus located in the hypothalamus directly above the optic chiasm, it is responsible for the regulation and timing of aspects of behavior and physiology that show circadian rhythmicity, including sleep, physical activity, alertness, hormone levels, body temperature, immune function, and digestive activity. When the SCN is lesioned, circadian rhythmicity is lost.
Zeitgeber	An external or environmental cue that synchronizes an organism’s biological rhythms.

2 |. CIRCANNUAL RHYTHMS AND THEIR ENTRAINMENT

Living organisms including humans have evolved to anticipate or respond to the earth’s 24-hour cycle and the seasonal (ie, winter, spring, summer, and fall) changes in the light-dark cycle caused by the earth’s annual orbit around the sun.¹⁶ Animals’ ability to anticipate these seasonal changes permits physiological processes such as reproduction and weight gain to occur at optimal times thus increasing the chances of survival.¹⁶ Circannual rhythms are synchronized or entrained by exposure to environmental cues such as seasonal changes in day length and the timing of sunrise and sunset.¹⁶ Circannual rhythms have a free running period of about 10 to 11 months meaning that, when animals are exposed to a constant photoperiod (12-hour light: 12-hour dark year round) for several years, they continue to exhibit the circannual rhythms (ie, molt, reproductive changes) with a period of 10 to 11 months, resulting in the occurrence of these cycles progressively earlier each year.¹⁷ Because circannual cycles are shorter than 12 months (ie, one full orbit around the sun), input from the seasonal light-dark cycle is needed to maintain a 12-month circannual rhythm that promotes alignment with the seasons.¹⁷ Similar to circadian rhythms in which people are characterized as early birds (ie, larks) or night owls, there is individual variation in the timing of annual rhythms in mammals, with some individuals having an early or late annual chronotype relative to other individuals.¹⁷ Humans and other mammals exhibit evidence of photoperiodic entrainment of circannual rhythms in sleeping metabolic rate,¹⁸ brown adipose tissue (BAT) activation,¹⁹ genetic regulation of glucose homeostasis,²⁰ and neurohormones such as melatonin,^21,22 cortisol,^23–25 and leptin,^26,27 all of which may have important implications for children’s growth patterns.^28–30

Similar to circadian rhythms, seasonal rhythms are regulated by the daily encoding of day length via inputs to melanopsin-containing visual receptors in the eye known as intrinsically photosensitive retinal ganglion cells. The visual receptors transmit day length information to the suprachiasmatic nucleus (SCN) through the retinohypothalamic tract, which is a tract of neurons connecting the eye to the hypothalamus where the SCN is located.^31,32 The SCN or the body’s “master clock” uses this information to synchronize physiological rhythms within the body and to encode seasonal changes in the light/dark cycle by creating an internal representation of day length.²¹ Information about day length is signaled to other areas of the brain such as the pineal gland (ie, area of the brain that releases melatonin which signals the dark phase).³³ The length of melatonin release, marking the biological night, varies seasonally in response to changes in the length of the earth’s dark period.^21,22 A bi-oscillator model of circadian regulation suggests that seasonal adaptation to the light-dark cycle is facilitated by two oscillators, one entrained (ie, synchronized with an environmental cue such as light) via dusk, controlling the onset of melatonin and the other entrained via dawn controlling the offset of melatonin.^34,35 Within the SCN, the phase of the ventral and dorsal SCN regions diverge from one another during the longer days of summer.³⁶ If SCN firing during daylight hours inhibits melatonin release, the separation between the ventral and dorsal regions during long summer days will shorten the melatonin release duration during the summer. Under natural lighting conditions in which the dark period is much longer in winter compared with summer, humans exhibit a longer melatonin release in winter compared with summer.²²

Analysis of gene expression levels in white blood cells and adipose tissue from adults and children demonstrated two distinct seasonal patterns, with one set upregulated or “turned on” in summer and another set upregulated in winter.³⁷ Opposing patterns in the seasonal up regulation of genes were found in northern and southern hemispheres while an equatorial sample demonstrated patterns of genetic expression corresponding to rainy versus dry seasons. This suggests that exposure to the seasonal light-dark cycle synchronizes genetic expression resulting in seasonal variability in human physiology such as immune function³⁷ and possibly growth.

2.1 |. Evidence for the seasonality of children’s height gain

There is compelling evidence for seasonal variation in the light-dark cycle predicting height growth in children through photoperiodic entrainment.^2,9,38–40 Examination of growth patterns among blind, partially sighted and sighted children living in Southern England indicated sighted children demonstrated maximum gains in height between January and June when day length is increasing, while periods of maximum growth in blind, and partially sighted children were evenly distributed throughout the year.³⁸ Another study exposed 45 Swedish boys to sun lamp treatment for 7 hours daily during winter. Compared with a control group of 292 boys not receiving treatment,^2,9 the boys who received light therapy during winter grew at an accelerated rate during treatment while weight gain slowed. During summer, the control group grew at a faster rate and demonstrated a concurrent slowing of weight gain. Over the entire year, there was no difference in total height gain between groups; however, the season during which the greatest growth occurred varied. In addition, observational studies have established periodicity or seasonality in height patterns suggesting that day length and intensity of light exposure synchronize the annual timing of changes in height velocity.^2,39,40

2.2 |. Potential mechanisms governing circannual patterns of height gain

Similar to normally growing children, growth hormone deficient children treated with recombinant human growth hormone demonstrate comparable seasonal variation in height gain. Children treated with growth hormone also exhibit faster height velocity during periods of greater daylight, supporting a relationship between height gain and light exposure.^8,41 Greater response to growth hormone therapy was achieved among children with high levels of daylight exposure compared with those with intermediate or low levels of daylight exposure.⁸ Causal network analysis to evaluate upstream regulation of transcriptional factors tested three theories to explain these findings (circadian rhythms, melatonin, or vitamin D). The melatonin pathway best explained the daylight exposure and height velocity correlation.⁸ Melatonin regulation, however, is closely related to the circadian regulation of cortisol, a glucocorticoid.⁴² The regulation of melatonin and the glucocorticoid pathway are both synchronized by the light-dark cycle,⁴³ suggesting that the circannual cortisol rhythm is also similarly but independently entrained by the seasonal photoperiod.⁴⁴ Glucocorticoids have demonstrated a strong relationship to growth in children.^45–47 Excess glucocorticoid during childhood due to Cushing syndrome, for example, retards growth.⁴⁷ In addition, ARNTL gene, also known as the clock gene BMAL1, demonstrates seasonal variation in its expression in the peripheral blood mononuclear cells and adipose tissue of children and adults.³⁷ It becomes increasingly more upregulated as day length increases, peaking following the summer solstice, then increasingly downregulated as day length shortens with a nadir following the winter solstice.³⁷ ARNTL has been strongly correlated with the glucocorticoid receptor NR3C1³⁷; hepatic glucocorticoid receptors have been shown to regulate growth in mice.⁴⁸ The seasonal regulation of ARTNL could provide a potential pathway through which epigenetic changes in gene expression may regulate seasonal variation in growth.

The timing and amount of the daily morning rise in cortisol varies seasonally in response to changes in the light-dark cycle with higher levels of cortisol occurring in the winter when photoperiod is shorter and lower levels in the summer when day length in longer. Higher levels of cortisol are typically found in winter when the slowest rates of height gain are typically found.^23,24,49 Timing of sunrise rather than season have been shown to better explain seasonal variation in cortisol than day length.²⁴ Further, less variation in cortisol levels have been observed closer to the equator compared with more northerly or southerly latitudes.²³ Among children, less seasonal variation in the light-dark cycle due to living closer to the equator has been associated with less height gain; possibly due to less seasonal variation in cortisol and melatonin.^{7,23,24,50,51} These observational studies suggest that cortisol rhythms are entrained by changes in the timing of sunrise resulting in variations across latitudes, which likely correspond to differences in children’s growth patterns.

The effect of classroom lighting on children’s circannual cortisol rhythms was examined among 90 Swedish children assigned to one of four classrooms, each differing in the amount of natural light and type of fluorescent lighting.⁴⁹ Two classrooms had natural lighting that differed in the type of fluorescent lighting (warm-white lamps versus daylight lamps) and two windowless classrooms that differed similarly in the type of fluorescent lighting. The type of lighting to which the children were exposed affected their circannual cortisol rhythm.⁴⁹ Children in the windowless room without daylight fluorescent tubes demonstrated delayed cortisol rhythms.⁴⁹ Height was only assessed at the beginning and end of the school year and no differences in growth were observed across classrooms; however, higher morning cortisol levels were associated with less growth after controlling for the effect of height on cortisol concentration. This association was stronger during winter when all but the windowless classroom without daytime bulbs demonstrated a decrease in morning cortisol levels. Lack of daytime natural lighting may shift children’s circannual rhythms, which may affect the timing of increases in height velocity and possibly the magnitude of these increases.

2.3 |. Evidence for the seasonality of children’s weight gain

To our knowledge, no experiments have been conducted to examine the effect of day length on the timing of maximum increases in weight among children. However, numerous studies have examined changes in children’s body mass index (BMI) during the school year and summer. They demonstrated that children increased their BMI at a faster rate during summer compared with the school year.^4,52 Most of these studies, however, included only two yearly measurements of BMI limiting the ability to examine the effect of season versus school year or school holiday environments on weight gain velocity. The limitations of biannual measurements were highlighted by a cross-sectional examination of monthly differences in close to 70,000 children’s standardized BMI (BMIz).⁵³ When researchers simulated what their results would have been if they had only used biannual measurements as was done in the studies based on the academic school year, they found similar patterns of improvements in BMIz during the school year and increases during the summer. However, examination of monthly variation in children’s BMIz revealed lower BMIz values in late spring and early summer (May-June) followed by stasis in July and then higher values in late summer and fall (August-September). These findings suggest an alternative phenomenon other than the effects of the in-school and out-of-school year environment.⁵³ Similarly, a literature review of prospective studies that obtained more frequent weight measurements throughout the year found seasonal weight gain among school age children with fall (September-December) being of time of substantial weight gain, followed in order of decreasing magnitude by summer (June-August,) with winter and spring being times of minimum gains.¹

A study of 246 children aged 5 to 17 years conducted in 1974 to 1975 in Guatemala City found that, for the overall sample, children gained weight faster during the dry season when daylight is more abundant compared with the rainy season when day light is less abundant, though analyses within subgroups (authors divided the sample into six subgroups based on age and gender [sample size ranged from 35 to 46]) only remained significant for females aged 11 to 13 and males aged 14 to 17.⁵⁴ While results across studies have varied, these findings suggest a possible seasonal pattern of weight gain regulated by changes in the light dark cycle, though not as convincing as the evidence for seasonality in height. The following reviews potential mechanisms that may explain seasonal weight gain.

2.4 |. Potential mechanisms governing circannual patterns of weight gain

Humans demonstrate a 24-hour pattern of energy balance. Energy expenditure is lowest during sleep,⁵⁵ resulting in the burning of fewer calories during the biological night. However, the biological night is dominated by lipid oxidation and physiological processes that maintain blood glucose levels during the overnight fast, resulting in a small but potentially important impact on overall energy expenditure and adiposity.^55,56 The biological day is dominated by carbohydrate oxidation (ie, processing of glucose), uptake of glucose into the muscle, and storage of excess glucose.⁵⁵ The balance between lipid oxidation during the biological night and carbohydrate oxidation and glucose storage during the day support energy homeostasis. Melatonin synchronizes the function of the adipocyte promoting lipolysis (ie, fat or lipid breakdown) during the biological night and lipogenesis (ie, lipid creation) during the biological day.^57,58 Seasonal or artificial changes in day length due to the use of electrical lights at night affect the ratio of the duration of the biological day to duration of the biological night,⁵⁹ in turn affecting the duration of melatonin secretion. Adults exposed to either a natural lighting environment in summer and winter or a modern lighting environment in which light schedules could be self-selected in summer and winter demonstrated evidence of seasonality in melatonin duration under natural lighting conditions but not under modern lighting conditions. However, this has not been examined in children. A study examining seasonal differences in elementary school age children’s sleep duration found children exhibited a shortened sleep duration during the summer vacation compared with the school year and that later timing of sleep predicted a shorten sleep duration as children failed to sleep in to make up for the later timing of sleep.⁶⁰ Taken together, these studies suggest that, during summer, children go to bed late, which, in addition to the shortened natural photoperiod, may contribute to shortened melatonin duration during summer.

Figure 1 illustrates the differences in the timing of children’s sleep during the school year and the summer and the proposed effects on melatonin duration and other factors related to energy balance. The small interior gray numbers represent clock time, while the larger interior black/white numbers represent hours since habitual wake time. It is proposed that during the winter, school requires children to go to bed earlier and to wake earlier. During summer, children may go to bed later and sleep in. Figure 1 shows the seasonal photoperiod due to Earth’s light dark cycle. Note, however, that children are likely exposed to electric light when they are awake (ie, awake time represents exposure to natural and or electrical light). Based on data from preadolescents, we estimate that the onset of melatonin will occur about 1 hour before habitual bedtime.⁶¹ Due to lack of data available on children, adult data²² were used to hypothesize that melatonin offset would likely occur about an hour after wake in winter and about 30 minutes after wake in summer due to sleeping in after sunrise. While, in the adult study, melatonin offset occurred before sleep offset in a weekend environment (more similar to summer), we hypothesize that, if children repeatedly sleep in, their melatonin offset will shift so that offset occurs after typical wake time. It is proposed that the seasonal or artificial alteration of the ratio of the length of the biological night to the length of the biological day will likely modify energy balance and thereby adiposity (Figure 1).

FIGURE 1.

Physiological mechanisms supporting energy balance. The opposing influences of the biological day and night promote energy balance. Seasonal changes in day length may alter this balance through changes in the length of the biological day/night, resulting in changes the rate of weight gain.

Two possible pathways explain how a shortened biological night during summer months may lead to weight gain: (a) shifts in the diurnal patterns of glucose homeostatis leading to a reduction in the ratio of time spent in lipolysis versus lipogenesis and (b) activation of thermo-genesis (ie, oxidation of fatty acids) in BAT.

Glucose homeostasis and lipolysis.

Healthy humans exhibit a normal 24-hour rhythm of glucose tolerance with higher glucose tolerance in the morning, followed by poorer glucose tolerance in the evening (Figure 1).^62,63 This facilitates energy storage during the day and increases availability of substrates during the biologic night to support energy needs during the overnight fast.^56,64 The optimal timing of glucose homeostasis to align with sleep and wake behaviors is controlled by the SCN, a complex process. Sleep/wake, feeding, and activity behaviors are both outputs of the circadian system yet also influence entrainment, for example 24-hour insulin patterning is influenced by feeding patterns, the timing of which are controlled by the SCN.^65,66 The SCN also regulates 24-hour fluctuations for other drivers of glucose tolerance, such as insulin-independent glucose uptake and pancreatic β-cell function.^66–68 Circadian fluctuations in glucose tolerance reflect transcription-translation feedback loops with the SCN receiving inputs from other clock controlled hormones (eg, adiponectin, leptin, cortisol, and insulin) and sleep controlled inputs (eg, growth hormone).

During the day, glucose is primarily derived from dietary glucose.⁶⁹ Glucose provides fuel to the brain, adipose tissue, muscle, and liver⁶⁹; however, insulin is a contributor to glucose uptake in adipose tissue and muscle.⁷⁰ During the biological day, insulin sensitivity is optimized promoting increased uptake of glucose into the muscle and adipose tissue in humans without obesity. In a fed state, insulin facilitates the entry of glucose into the adipocyte and is used to synthesize glycerol and fatty acids (eg, triglycerides) increasing energy storage in the adipocyte (ie, lipogenesis).^71,72 Leptin is released by the adipocyte in response to a fed state (ie, sufficient energy reserves), reducing the release of ghrelin, a hormone released by the stomach that increases appetite.⁷³ Leptin follows a circadian pattern with low levels in the morning, promoting food intake and peak levels during the biological night, promoting an overnight fast.⁷⁴ Cortisol levels typically peak in the early morning and are markers of the start of the biological day.⁷⁵ High levels of cortisol stimulate appetite, assist with the utilization of glucose in the morning, and serve to partially regulate leptin.⁷⁶

Adiponectin, produced by the adipocyte, improves the sensitivity of the adipocyte to insulin⁷⁷ by facilitating the uptake of glucose into the muscle, reducing the production of glucose in the liver, and increasing fatty acid oxidation in the liver and muscle.^78,79 Adiponectin expression by the adipocyte is clock controlled, but also partially regulated by sleep-dependent growth hormone and prolactin with levels peaking in the early afternoon and low levels during the biological night in humans (see Figure 1).^62,74,80 In addition, adiponectin increases appetite and decreases energy expenditure by acting on the hypothalamus.⁸¹ High levels of adiponectin during the biological day contribute to the storage of energy in the adipocyte, while low levels at night support maintenance of blood glucose levels during the extended fast by decreasing sensitivity to insulin.⁷⁴ The ADIPOR1 receptor for adiponectin has demonstrated seasonal variation and is more highly expressed in summer in humans.³⁷ Growth hormone also aids in the maintenance of blood glucose levels during the overnight fast by stimulating production of glucose by the liver and decreasing insulin release by the pancreas.⁵⁶ Melatonin administration increases adiponectin expression in rodents resulting in improved insulin sensitivity during the daytime.⁷⁴ In humans, melatonin administration was positively correlated with the amplitude of adiponectin⁸² and resulted in restoration of normal adiponectin levels in individuals with nonalcoholic steatohepatitis and characteristically low levels of adiponectin.⁸³ Taken together, when eating patterns follow the expected diurnal fluctuations, the biological daytime is when lipogenesis principally occurs, increasing adiposity.

During sleep (ie, the overnight fast) data from BMAL1 knockout mice suggest that the downregulation of BMAL1 in white adipose tissue, and skeletal muscle, as seen in humans at night⁸⁴ inhibits lipogenesis⁸⁵ and reduces both insulin-stimulated and AICAR-stimulated glucose uptake.⁸⁶ At night, melatonin inhibits insulin secretion.^87,88 Insulin regulates the synthesis of melatonin just after lights out and just before lights on, creating a negative feedback loop.^89,90

At the same time, brain utilization of glucose decreases such that enough glucose is produced by the body from energy reserves in the liver, a process facilitated by growth hormone, and fatty acids can serve as the principle source of fuel. Enabled by low levels of insulin, fatty acids are made available through the breakdown of fats and other lipids, including adipose cells, known as lipolysis. Further, muscle that is at rest, as is largely the case during sleep, utilizes fatty acids.⁹¹ Thus, in contrast, the biological day is a time of adiposity-promoting lipogenesis, while the biological night is characterized by adiposity-reducing lipolysis.

Glucose homeostasis and lipolysis during the biological night play a key role in energy balance. During summer, we posit that less time spent in lipolysis during the shortened biological night and greater lipogenesis due to longer days contribute to seasonal weight gain patterns observed in children. These seasonal differences in metabolic processes are likely regulated by seasonal variation in DNA methylation.³⁷

Thermogenic activity.

Melatonin receptors are found throughout the body (eg, central nervous and cardiovascular systems, liver, skin, pancreas, skeletal muscle, and adipocyte cells),^74,92 and, is one way, the SCN synchronizes the body’s circadian and seasonal rhythms. While known for its role in the timing of sleep, melatonin also promotes optimal energy balance and metabolism through synchronization of leptin, fat metabolism, cortisol, adiponectin and sleep/wake behavior, which triggers the release of prolactin and growth hormone.⁹² In regard to seasonal weight gain, melatonin synchronizes metabolic function of the adipocytes promoting high lipolysis during the biological night and high lipogenesis (ie, the creation of lipids that are stored in the adipocyte) during biological day.⁵⁸ This synchronization also occurs through sympathetic activation of white adipose tissue.⁵⁷ Among Siberian hamsters that do not gain weight in winter, short winter-like days led to longer nocturnal melatonin release. Longer melatonin duration led to greater stimulation of melatonin receptors in the forebrain, thereby engaging sympathetic activation of white adipose tissue, resulting in lipolysis and a decrease in seasonal adiposity.⁵⁷ Melatonin induces browning of white adipose tissue in rodents (ie, increasing recruitment of beige adipose tissue).⁹³ In rodents, BAT increases energy expenditure by converting fatty acids and glucose into heat (ie, increasing their thermogenic activity), resulting in weight loss.⁹⁴ In humans, BAT cells uptake glucose,⁹⁵ and increase energy expenditure via glycolysis (ie, thermogenesis), particularly in the postprandial state, but also during the overnight fast.⁹⁶ Whether the thermogenic effects of BAT are great enough to contribute to seasonable variations in adiposity is unclear; BAT has been found in up to 50% children; and, supporting data in adults, BAT activity is inversely correlated with BMI percentile in children.^97,98 Greater activation of BAT due to longer melatonin secretory patterns during longer winter nights may explain seasonal weight changes in response to seasonal changes in day length in rodents.⁹⁹

While melatonin is involved in the recruitment (ie, browning) and activation of BAT, highlighting its potential role in the prevention and treatment of obesity,⁹⁴ prolonged light exposure has been shown to increase adiposity through its effects on thermogenesis even in mice who have had their pineal glands removed.¹⁰⁰ In the absence of melatonin, these pinealectomized mice increased their white adipose tissue fat mass through reduced uptake of triglycerides and glucose into BAT and decreased activation of BAT resulting in decreased thermogenesis.¹⁰⁰ This study suggests that, while melatonin is a marker of the biological night and plays a role in the synchronization of circadian processes in the body, in its absence, light is a strong enough zeitgeber affecting other hormones in the body that are able to compensate for the absence of melatonin, at least in this rodent model.¹⁰⁰ The extent to which melatonin is critical for circadian and circannual regulation of lipolysis and thermogenesis is unknown.

3 |. SUMMARY

There is convincing evidence that children gain height and weight in a seasonal manner, suggesting that children’s exposure to the earth’s seasonal light/dark cycle likely entrains the timing of these rhythms. There is promising evidence that seasonal variation in the velocity of height gain may be related to seasonal changes in cortisol and that seasonal changes in the length of the biological light may affect time spent in lipolysis, lipogenesis, and thermogenesis, resulting in decreases in energy expenditure and increases in the velocity of weight gain during summer. It is hypothesized that these seasonal changes are entrained by the light-dark cycle that synchronizes genetic expression resulting in seasonal variability in human physiology and children’s growth patterns.^17,37 The timing of the summer break from school, which likely encourages more irregular schedules and potentially greater circadian misalignment in young children at a time when children are primed for accelerated weight gain, may promote obesity.¹⁰

3.1 |. Areas for future research

While the evidence supporting the presence of seasonality in children’s growth is compelling, further testing that this phenomenon is regulated by endogenous rhythms and the environmental cues that entrain these rhythms is needed. First, the extent to which children exhibit circannual rhythms or seasonal differences in melatonin duration is unclear. There is a lack of evidence regarding how children respond to seasonal changes in light exposure under natural lighting conditions (in the absence of electrical lighting) and under modern lighting conditions. Studies comparing the growth patterns and seasonal differences in melatonin and cortisol rhythms of children in areas of extreme seasonal changes in light/dark (eg, Norway, Sweden, and Finland) could offer insight in how the natural light/dark cycle affects circannual rhythms in children. Additionally, comparing growth and circannual changes in melatonin among children in pre-industrialized communities without electricity and communities which only recently have been introduced to electricity may help to clarify the role of exposure to artificial light at night on circannual rhythms in children. Further, behavioral changes such as the timing of sleep, eating, activity, and light exposure patterns resulting from transitions from the school environment to the home environment may disrupt sleep patterns and circannual rhythms. Finally, examination of children in traditional schools and year-round schools may help to clarify whether growth patterns continue to occur seasonally or whether the social demands of the school and summer environment influence changes in children’s behavioral rhythms (eg, sleep, eating, and activity patterns) resulting in changes in the velocity of height and weight gain. Studies should control for the effect of the school and summer environment on light exposure in order to rule out the possibility that the school and summer environment may result in changes in children’s light exposure which may influence growth. Understanding how the school and summer environment and natural and electrical lighting influence children’s growth will facilitate the development of guidelines for parents regarding children’s optimal sleep and light exposure patterns across seasons. Further excess weight gain can be triggered by behavior and not just circannual rhythms as evidenced by weight gain associated with holiday feasting,¹⁰¹ highlighting the need to disentangle the effects of a seasonal versus holiday weight gain effect.

Evidence from rodent models and humans suggest that weight status is associated with circadian rhythmicity.^102,103 Given the possible relationship between weight status and circadian rhythmicity, individuals with obesity may be more prone to circannual disruption and, in turn, explain the higher degree of accelerated weight gain observed in children with overweight and obesity. Further research is needed to understand whether an obese state negatively affects the body’s ability to regulate circadian and circannual rhythms and whether behavioral interventions can improve regulation. Finally, the extent to which normal circannual variation in height and weight contributes to an unhealthy weight status should be explored. It is not expected that normal seasonal variation in growth velocity would contribute to development of an unhealthy weight status in the absence of other diet and activity influences; however, this should be tested empirically.

4 |. IMPLICATIONS AND CONCLUSION

Summertime increases in BMI have been observed since the 1990s and have been interpreted as consequences of the obesogenic modern society.^4,11 However, evidence of seasonal increases in height and weight consistent with more recently observed changes in BMI have been observed since the late 1700s.^2,9 Children may exhibit seasonality in their growth due to biological influences such as circadian and circannual rhythms. Understanding how these biological mechanisms contribute to obesity is an important area of future inquiry. A better understanding of biological mechanisms affecting height and weight and environmental determinants of these patterns should lead to identification of more effective behavioral interventions for the prevention of overweight and obesity in children. The multiple links by which misalignment of circadian and circannual rhythms relates to obesity suggests that promising intervention targets should include reducing exposure to electric light at night, promoting consistent bedtimes (ie, supporting exposure to a consistent light/dark cycle), and limiting eating episodes during the biological night when efficiency of glucose utilization is poor.

Abbreviations:

SCN

suprachiasmatic nucleus

BMI

body mass index