DISCOVERY MEDICINE ®

www.discoverymedicine.com

191

ISSN: 1539-6509; eISSN: 1944-7930

ContinuationofGradualWeightGainNecessary fortheOnsetofPubertyMayBeResponsible forObesityLaterinLife SteveN Lehrer Abstract: A continuation of the gradual weight gain necessary for the onset of puberty may be responsible for obesity later in life. Hypothetically, a group of brain nuclei form components of a single pubertal clock mechanism that drives pre-pubertal weight gain and governs the onset of puberty and fertility. No mechanism evolved to shut off pre-pubertal and pubertal weight and body fat gain after puberty. The weight gain continues unabated throughout life. A better understanding of the mechanism of puberty and pre-pubertal weight gain could provide new insights into obesity and diseases associated with obesity such as type 2 diabetes, dyslipidemia, hypertension, heart disease, depression, etc. [Discovery Medicine 20(110):191-196, October 2015]

What causes us to gain weight as we age? Why don’t we lose weight as we get older? Mozaffarian et al. (2011) wrote that average long-term weight gain in non-obese populations is gradual but accumulates over time, and can be caused by life style changes, diet, and exercise. Another cause may be the steady weight and body fat gain required for the onset of puberty. Hypothetically, a group of brain nuclei form components of a single pubertal clock mechanism that drives pre-pubertal weight gain and governs the onset of puberty and fertility. However, no mechanism evolved to shut off the constant weight and body fat gain needed for puberty, after puberty has occurred. The pre-pubertal and puberSteven Lehrer, M.D. Department of Radiation Oncology, Icahn School of Medicine at Mount Sinai, Box 1236, 1 Gustave L. Levy Place, New York, NY 10029, USA.

tal weight gain continues throughout life. Fertility and the Light:Dark Cycle The reproductive cycle of mammals is closely related to the 24-hour light:dark cycle and circadian rhythms (Tamarkin et al., 1985). For example, during long nights and short days or after blinding, gonadal atrophy will occur in the Syrian golden hamster, Mesocricetus auratus. This phenomenon is mediated by stimulation of the pineal gland (Hoffman and Reiter, 1965; Reiter, 1980; Reiter et al., 2011a; 2011b). Moreover, the fourday estrus cycle of hamsters is closely coupled to the length of the light:dark cycle. A normal hamster living in a 24-hour light:dark cycle has an estrus cycle of 96 hours (4 x 24); whereas a hamster maintained under constant dim illumination with a free running circadian rhythm of 24.5 hours has an estrus cycle of 98 hours, that is, 4 x 24.5 (Zucker, 1980). It appears that cycle durations are quite similar per one cycle. But over time (weeks or months for animals and years for humans), the small difference per cycle will add up, leading to earlier or later occurrence of puberty or menarche. Oscillators (Pulse Generators) in the Brain The suprachiasmatic nucleus (SCN) of the mammalian hypothalamus has been referred to as the master circadian pacemaker that drives daily rhythms in behavior and physiology (Abe et al., 2002). There is also evidence for extra-SCN circadian oscillators, in particular the arcuate nucleus, which is responsible for the hourly gonadotropin pulses necessary for puberty and fertility (Pohl and Knobil, 1982). Indeed, the brain contains multiple damped circadian oscillators outside the SCN. The phasing of these oscillators to one another may play a critical role in coordinating brain activity and its adjustment to changes in the light:dark cycle (Abe et al., 2002).

Corresponding Email: [email protected] © DiscoveryMedicine. All rights reserved.

Discovery Medicine, volume 20, Number 110, Pages 191-196, October 2015

192

Weight Gain for Puberty May Be responsible for Obesity Later in Life

Mechanism of Puberty A pivotal event in the onset of puberty in mammals is the resumption of pulsatile release of gonadotropinreleasing hormone (GnRH) from neurons of the hypothalamus. Known influences on the timing of this event in mammals include the light:dark cycle, leptin levels, and the increased expression of neurokinin B, kisspeptin, and their receptors, NK3R and KISS1R (Hughes, 2013). One familial version of precocious puberty is the result of mutations in a paternally expressed imprinted gene, MKRN3, which encodes makorin RING-finger protein 3 (Abreu et al., 2013). Imprinted genes have a sex bias. They are expressed only from the maternal or the paternal chromosome. Some genes are paternally imprinted, whereas others are maternally imprinted. MKRN3 is maternally imprinted, and expression from the maternally inherited copy of the gene is silenced. MKRN3 protein is synthesized from RNA transcribed only from the paternally inherited copy of the gene. Makorin proteins are prolifically expressed in the developing brain, especially within the arcuate nucleus, the function of which is necessary for puberty and fertility. Delayed puberty and estrogen resistance have been described in a woman with an estrogen receptor α variant. This woman had no breast development and distinctly elevated serum levels of estrogens, as well as bilateral multi-cystic ovaries (Quaynor et al., 2013).

Yet despite many physiologic studies, there is no agreement as to what mechanism or mechanisms may actually trigger the onset of puberty. It is not known why puberty starts at about the age at the junction of the first and second decades of human life. Lehrer (1983) has proposed that clock genes in the brain (Hastings, 1998) and the resonance of brain oscillators and circadian rhythms cause puberty (Sizonenko, 2000). This resonance mechanism can explain three apparently unrelated characteristics of puberty: 1. The gonadotropins are elevated at birth. Luteinizing hormone (LH), for example, is high not only at puberty but also at birth in both rats and humans (Figure 1). In humans, LH falls between the ages of two and four, then rises again at puberty (Grumbach, 1980). This fluctuation, which corresponds with the period of infant sexuality and subsequent latent period described by Freud (Freud, 1905), is especially prominent in agonadal children. The high LH levels at birth and puberty, in both normal and agonadal girls, can be accounted for by continual reduction in arcuate nucleus intrinsic frequency after birth, with resonance at birth and puberty, the resonance at birth occurring at twice the intrinsic frequency of the resonance at puberty. 2. Blind girls have their menarche earlier than sighted girls (Zacharias and Wurtman, 1964), and rats reared in constant darkness have vaginal opening earlier than rats reared in an eight hour light:sixteen hour dark cycle (Relkin, 1967). The onset of puberty is probably earlier in blind girls and rats reared in darkness because they have circadian rhythms which are more rapid than usual; the entrainment of these rhythms and the SCN to the daily light:dark cycle does not occur and they free-run. Therefore, the intrinsic frequency of the arcuate nucleus does not have to diminish as much to resonate with the same harmonic (frequency multiple) of the suprachiasmatic nucleus rhythm, and puberty can occur at an earlier age (Figure 2).

Figure 1. The high LH levels at birth and puberty, in both normal and agonadal girls, can be accounted for by a continual reduction of arcuate nucleus intrinsic frequency after birth. (Modified from Grumbach, 1980). Discovery Medicine, volume 20, Number 110, October 2015

3. Hourly LH pulses occur only during sleep in early puberty (Boyar et al., 1972), but occur throughout the day and night in adulthood. At early puberty, the intrinsic frequency of the arcuate nucleus is very close, but not equal to, the frequency of the brain oscillator entrained to the SCN, with which it resonates at sexual maturity. When two

Weight Gain for Puberty May Be responsible for Obesity Later in Life

oscillators of very nearly the same frequency interact, the “beat” phenomenon occurs (Figure 3). The rhythms combine to give a rhythm whose amplitude varies periodically with time. Such a mechanism is familiar in mammals, having already been invoked to explain the hormone level variations in the rat estrus cycle (Yochim and Shirer, 1981). At puberty, the beats correspond with the LH pulses during sleep. Note that the resonance mechanism for puberty is basically the description of a clock. A mechanical clock has an oscillator (pendulum) and counts pendulum swings (oscillations). When the clock mechanism has counted a specified number of oscillations, it chimes. When the pubertal clock counts a specific number of oscillations, i.e., light:dark cycles, puberty occurs.

193

Pre-puberty (childhood) obesity is recognized as a major risk factor for adult obesity, and has been rising dramatically in the West (Poskitt and Edmunds, 2008). Analyses from two nationally representative samples of U.S. girls suggest a drop of about 2½ months in the average age of menarche during the time period between 1963-1970 and 1988-1994. This drop was paralleled by a concurrent shift in the population distribution of body mass index toward higher relative weights (Anderson et al., 2003). Although boys do not have a marker of puberty as distinctive as menarche, their body mass index increases constantly throughout childhood and they too seem to require the same critical fatness for puberty to occur (Mantzoros et al., 1997; Vizmanos and Marti-

Testable Hypothesis #1 Characteristics 1 and 3 combined provide a testable hypothesis of the resonance mechanism of puberty. As the LH pulsations in early childhood decline in amplitude (Figure 1), at about the age of three years the pulses may occur only during sleep as they do during the increase in amplitude during early puberty. This would be consistent with a common mechanism underlying LH changes in both periods. The gonadotropin pulses are present in early childhood, though at significantly diminished amplitude, insufficient to stimulate gonadal activity (Hayes and Crowley, Jr, 1998). But no 24 hour studies of their levels, comparable to the studies of early puberty (Boyar et al., 1972), have been done.

Figure 2. Blind girls (and rats reared in constant darkness) probably have higher than normal suprachiasmatic nucleus (SCN) frequencies and corresponding harmonics. As a result, the intrinsic frequency of the arcuate nucleus does not have to diminish as much for resonance to occur, and puberty can take place at an earlier age.

Body Fat, Weight Gain, Puberty, and the Light: Dark Cycle Frisch and colleagues reported that reduction in fat/lean ratio in young ballerinas and college athletes delays onset of menarche (Frisch et al., 1980). Moreover, estrus is simultaneous with vaginal opening in 81% of rats on a high-fat diet, in comparison to 48% on low-fat diets, suggesting that a critical body composition of fatness is essential for fertility of the rat and the human female (Frisch et al., 1975).

Figure 3. The beat phenomenon. Two rhythms of slightly different frequencies, shown in (a), combine in (b) to produce a rhythm whose amplitude (broken line) varies periodically with time. This variation may account for the LH pulses which occur only during sleep in early puberty. Discovery Medicine, volume 20, Number 110, October 2015

194

Weight Gain for Puberty May Be responsible for Obesity Later in Life

Henneberg, 2000).

Testable Hypothesis #2

Pubertal weight gain is related to the length of the daily light:dark cycle. Rats on a 22.5 hour light:dark cycle have early vaginal opening (Lehrer, 1986) and more weight gain than rats on a 25 hour cycle (Vilaplana et al., 1995). The pubertal clock mechanism needs to count a fixed number of light:dark cycles to trigger puberty, and in rats on the 22.5 hour cycle the necessary number occurs sooner (Sizonenko and Aubert, 1986; Sizonenko, 2000).

Women whose menarche was delayed will have lower BMI in later life than women with normal menarche, controlling for BMI at the age of normal menarche. Since abnormally low BMI at the age of normal menarche has been proposed as a cause of delayed menarche, its effect -- beyond the effect of early BMI on later BMI -- will be examined.

Hypothetically, gain of body mass and fatness are manifestations of the pubertal clock mechanism within the brain. The pubertal clock, possibly mediated by leptin (Mantzoros et al., 1997), drives the weight gain, as well as the other physiologic and hormonal changes needed for puberty. However, no mechanism evolved to shut off the constant weight and body fat gain needed for puberty, after puberty has occurred, and the pre-pubertal weight gain continues throughout life (Figure 4). The drop in BMI up to age 5 in Figure 4 parallels the fall in plasma LH levels shown in Figure 1. The narrow left segment and widened right segment of the sinusoidal BMI curve in Figure 4 may correspond to the fact that gonadotropin (arcuate nucleus) pulse frequency decreases with aging in post-menopausal women (Hall et al., 2000).

In the course of testing hypothesis 2, the effect of abnormally low BMI at the time of normal menarche on adult BMI will be tested. If there is such an effect in women, the same effect may be present in men. Published studies have examined effects of high, but not low, BMI. For example, earlier age at menarche is associated with a significantly higher adult BMI. This inverse association of age at menarche with BMI and obesity in middle age is not explained by confounding by early childhood BMI (Pierce and Leon, 2005). Earlier age at menarche is also associated with a significantly higher risk of diabetes, and specifically type 2 diabetes, in later life. But in one study the effect of age at menarche was reduced by adjustment for adult BMI and was no longer significant (Pierce et al., 2012).

Figure 4. BMI vs. age in females. Data from girls in 50th percentile of BMI ages 2 years to 19 years are from Centers For Disease Control, Clinical Growth Charts (http://www.cdc.gov/growthcharts/clinical_charts.htm#Set1); data for women above age 20 (7 data points) are from reference Kuczmarski et al., 1994, women in the 50th BMI percentile. The drop in BMI up to age 5 parallels the fall in plasma LH levels shown in Figure 1. The narrow left segment and widened right segment of the sinusoidal BMI curve above may correspond to the fact that gonadotropin (arcuate nucleus) pulse frequency decreases with aging in post-menopausal women (Hall et al., 2000). Note the five points above BMI 25. These women are overweight. 55% of American women fall into this category (Must et al., 1999). Discovery Medicine, volume 20, Number 110, October 2015

Weight Gain for Puberty May Be responsible for Obesity Later in Life

Conclusion In centuries past, obesity and continual weight gain after puberty were not of particular concern. Primitive man had to survive long enough only to reproduce, and no natural shutoff mechanism for pre-pubertal weight gain evolved. Until mid-twentieth century, infectious diseases were the major public health problem. Cigarette smoking and resultant cardiovascular disease took their place. Today, obesity is the major threat to public health. More than 60% of adults in the United States are obese, and the problem has spread to children and adolescents, with dismal consequences. Hypertension, type 2 diabetes, and dyslipidemia have already reduced life expectancy by more than five years among individuals who have these health issues. Depression, another consequence of obesity, is also increasing. The cost of medical care to treat obesity-related illness is spiraling upward, along with diminished productivity and lost income (Wyatt et al., 2006). The obesity epidemic is a fairly new phenomenon. Before this epidemic, 50 years (or 100 years) ago for example, the same puberty development in human history should also have applied to weight gain during puberty; however, a far smaller proportion of the population had obesity even though no mechanism evolved to shut off the constant weight and body fat gain needed for puberty, after puberty has occurred. From another angle, both before and after the obesity epidemic, a significant number of individuals (a large majority of the population before the epidemic), who had gone through puberty, remained in the normal weight range. These two observations can be explained by the continual extension of the human lifespan. There has been a striking increase in average life expectancy during the 20th century, which counts as one of society’s greatest achievements. Although most babies born in 1900 did not live past age 50, life expectancy at birth now exceeds 83 years in Japan and is at least 81 years in several other countries. Extension of life expectancy from 50 to 83 has led in many people to obesity resulting from the pubertal weight gain mechanism that is never shut off. As life expectancy continues to increase, so does the prevalence of obesity. A majority of people who have gone through puberty do not have obesity just after puberty because pre-pubertal and pubertal weight gain are gradual and obesity develops over a period of many years. Therefore, more people are obese later in life.

195

Some people (the majority in many populations) even with today’s long lifespan (for example, the majority of Japanese and Chinese people) do not have obesity later in life. Did weight gain during puberty play a role in these non-obese people? The pubertal weight gain occurs irregardless, but obesity is a complex disease with multiple risk factors, in adults (Chou et al., 2004) and children (Reilly et al., 2005). Some of these factors may lessen or suppress the impact of pubertal weight gain on the development of obesity in some individuals. A better understanding of the mechanism of puberty and pre-pubertal weight gain could provide new insights into obesity and may help to stem an obesity epidemic. Acknowledgment Supported by NIH grant R01 CA49506. Disclosure The author reports no conflicts of interest. References Abe M, Herzog ED, Yamazaki S, Straume M, Tei H, Sakaki Y, Menaker M, Block GD. Circadian rhythms in isolated brain regions. J Neurosci 22(1):350, 2002. Abreu AP, Dauber A, Macedo DB, Noel SD, Brito VN, Gill JC, Cukier P, Thompson IR, Navarro VM, Gagliardi PC, Rodrigues T, Kochi C, Longui CA, Beckers D, de ZF, Montenegro LR, Mendonca BB, Carroll RS, Hirschhorn JN, Latronico AC, et al. Central precocious puberty caused by mutations in the imprinted gene MKRN3. N Engl J Med 368(26):2467-2475, 2013. Anderson SE, Dallal GE, Must A. Relative weight and race influence average age at menarche: results from two nationally representative surveys of US girls studied 25 years apart. Pediatrics 111(4):844, 2003. Boyar R, Finkelstein J, Roffwarg H, Kapen S, Weitzman E, Hellman L. Synchronization of augmented luteinizing hormone secretion with sleep during puberty. N Engl J Med 287(12):582586, 1972. Chou SY, Grossman M, Saffer H. An economic analysis of adult obesity: results from the Behavioral Risk Factor Surveillance System. J Health Econ 23(3):565-587, 2004. Freud S. Three essays on the theory of sexuality. Perseus Books, New York, NY, USA, 1905. Frisch RE, Hegsted DM, Yoshinaga K. Body weight and food intake at early estrus of rats on a high-fat diet. Proc Natl Acad Sci U S A 72(10):4172-4176, 1975. Frisch RE, Wyshak G, Vincent L. Delayed menarche and amenorrhea in ballet dancers. N Engl J Med 303(1):17, 1980. Grumbach MM. The neuroendocrinology of puberty. In: Neuroendocrinology. (Eds. Krieger DT, Hughes JC). pp249-258.

Discovery Medicine, volume 20, Number 110, October 2015

196

Weight Gain for Puberty May Be responsible for Obesity Later in Life

Sinauer Associates, Sunderland, MA, USA, 1980.

ance in a woman with estrogen receptor alpha variant. N Engl J Med 369(2):164-171, 2013.

Hall JE, Lavoie HB, Marsh EE, Martin KA. Decrease in gonadotropin-releasing hormone (GnRH) pulse frequency with aging in postmenopausal women. J Clin Endocrinol Metab 85(5):1794, 2000.

Reilly JJ, Armstrong J, Dorosty AR, Emmett PM, Ness A, Rogers I, Steer C, Sherriff A. Early life risk factors for obesity in childhood: cohort study. BMJ 330(7504):1357, 2005.

Hastings M. The brain, circadian rhythms, and clock genes. BMJ 317(7174):1704, 1998.

Reiter RJ. The pineal and its hormones in the control of reproduction in mammals. Endocr Rev 1(2):109, 1980.

Hayes FJ, Crowley WF, Jr. Gonadotropin pulsations across development. Horm Res 49(3-4):163-168, 1998.

Reiter RJ, Rosales-Corral S, Coto-Montes A, Boga JA, Tan DX, Davis JM, Konturek PC, Konturek SJ, Brzozowski T. The photoperiod, circadian regulation and chronodisruption: the requisite interplay between the suprachiasmatic nuclei and the pineal and gut melatonin. J Physiol Pharmacol 62(3):269-274, 2011a.

Hoffman RA, Reiter RJ. Pineal gland: influence on gonads of male hamsters. Science 148(3677):1609, 1965. Hughes IA. Releasing the brake on puberty. N Engl J Med 368(26):2513-2515, 2013. Kuczmarski RJ, Flegal KM, Campbell SM, Johnson CL. Increasing prevalence of overweight among US adults. JAMA 272(3):205-211, 1994. Lehrer S. Puberty and resonance: a hypothesis. Mt Sinai J Med 50(1):39-43, 1983. Lehrer S. Rats on 22.5-hr light:dark cycles have vaginal opening earlier than rats on 26-hr light:dark cycles. J Pineal Res 3(4):375378, 1986. Mantzoros CS, Flier JS, Rogol AD. A longitudinal assessment of hormonal and physical alterations during normal puberty in boys. V. Rising leptin levels may signal the onset of puberty. J Clin Endocrinol Metab 82(4):1066-1070, 1997. Mozaffarian D, Hao T, Rimm EB, Willett WC, Hu FB. Changes in diet and lifestyle and long-term weight gain in women and men. N Engl J Med 364(25):2392-2404, 2011. Must A, Spadano J, Coakley EH, Field AE, Colditz G, Dietz WH. The disease burden associated with overweight and obesity. JAMA 282(16):1523-1529, 1999. Pierce MB, Kuh D, Hardy R. The role of BMI across the life course in the relationship between age at menarche and diabetes, in a British Birth Cohort. Diabet Med 29(5):600-603, 2012. Pierce MB, Leon DA. Age at menarche and adult BMI in the Aberdeen children of the 1950s cohort study. Am J Clin Nutr 82(4):733-739, 2005. Pohl CR, Knobil E. The role of the central nervous system in the control of ovarian function in higher primates. Annu Rev Physiol 44(1):583-593, 1982. Poskitt E, Edmunds L. Management of Childhood Obesity. Cambridge University Press, Cambridge, UK, 2008. Quaynor SD, Stradtman EW, Jr, Kim HG, Shen Y, Chorich LP, Schreihofer DA, Layman LC. Delayed puberty and estrogen resist-

Discovery Medicine, volume 20, Number 110, October 2015

Reiter RJ, Tan DX, Korkmaz A, Ma S. Obesity and metabolic syndrome: Association with chronodisruption, sleep deprivation, and melatonin suppression. Ann Med, epub ahead of print, Jun. 13, 2011b. Relkin R. Pineal function relation to absolute darkness and sexual maturation. American J Physiol 213(4):999, 1967. Sizonenko PC. Role of sex steroids during development-integration. In: The Onset of Puberty in Perspective: Proceedings of the 5th International Conference on the Control of the Onset of Puberty, Liège, Belgium, 26-28 September 1999. (Eds. Bourguignon JP, Plant TM). pp299-306. Elsevier, Amsterdam, The Netherlands & New York, NY, USA, 2000. Sizonenko PC, Aubert ML. Neuroendocrine changes characteristic of sexual maturation. J Neural Transm Suppl 21:159-181, 1986. Tamarkin L, Baird CJ, Almeida OF. Melatonin: a coordinating signal for mammalian reproduction? Science 227(4688):714, 1985. Vilaplana J, Madrid JA, Sanchez-Vazquez J, Campuzano A, Cambras T, ez-Noguera A. Influence of period length of light/dark cycles on the body weight and food intake of young rats. Physiol Behav 58(1):9-13, 1995. Vizmanos B, Marti-Henneberg C. Puberty begins with a characteristic subcutaneous body fat mass in each sex. Eur J Clin Nutr 54(3):203, 2000. Wyatt SB, Winters KP, Dubbert PM. Overweight and obesity: prevalence, consequences, and causes of a growing public health problem. Am J Med Sci 331(4):166-174, 2006. Yochim JM, Shirer HW. Evidence for a photoperiod-sensitive pacemaker for estrous cycle of the rat. Am J Physiol Endocrinol Metab 241(3):E261, 1981. Zacharias L, Wurtman RJ. Blindness: its relation to age of menarche. Science 144(3622):1154, 1964. Zucker I. Light, behavior, and biologic rhythms. In: Neuroendocrinology. (Eds. Krieger DT, Hughes JC). pp93-101. Sinauer Associates, Sunderland, MA, USA, 1980.

THE MOUNT SINAI JOURNAL OF MEDICINE

Vol . 50, No. 1, January-February, 1983 Printed in U.S.A .

Puberty and Resonance : A Hypothesis STEVEN LEHRER, M .D .

Abstract A new mechanism for puberty is proposed . Puberty appears to result from the interaction of two physiologic oscillators within the hypothalamus : the arcuate nucleus, which produces the gonadotropin-releasing hormone, and the suprachiasmatic nucleus, which is a master oscillator that regulates many circadian rhythms . Puberty results when the frequency of the arcuate nucleus has slowed sufficiently to resonate with a harmonic of the suprachiasmatic nucleus rhythm . The onset of puberty is earlier in blind girls and rats reared in darkness because they have circadian rhythms which are more rapid than usual . Therefore, the frequency of the arcuate nucleus does not have to slow as much to resonate with the same harmonic of the suprachiasmatic nucleus rhythm, and puberty can occur at an earlier age . The proposed mechanism also accounts for the occurrence of luteinizing hormone pulses only during sleep in early puberty, and for the elevation of gonadotropins at birth.

Puberty, the transitional period between childhood and adulthood, is accompanied by the appearance of secondary sexual characteristics and the achievement of fertility . In mammals, the central nervous system exercises the only restraint to puberty onset . This neuroendocrine control is mediated by the gonadotropin releasing hormone (GnRH) secreting neurons in the arcuate nucleus of the hypothalamus (1) . For many years, a "gonadostat" or feedback mechanism was believed to explain the relatively sudden onset of puberty . Recent evidence, however, has cast doubt on the gonadostat hypothesis ; few investigators in the field today still believe that it is correct . This article proposes a new explanation for the onset of puberty, the physical phenomenon of resonance .

autumn of 1909 (2, 3) . In 1936 Hohlweg found that estradiol treatment was more effective suppressing castration hypertrophy in prepubertal rats than in adult rats (4) . Hohlweg's observation provided the rationale for the view, later put forward by Ramirez and McCann (5), that puberty is caused by a sudden decrease in sensitivity of the hypothalamic-pituitary unit to sex hormone feedback . The immature gonads were known to produce small amounts of sex steroids, even in very young animals . These hormones were believed to inhibit the hypothalamic-pituitary production of the gonadotropins, follicle stimulating hormone (FSH), and luteinizing hormone (LH) . It was hypothesized that around the time of puberty, the hypothalamic-pituitary axis lost its sensitivity to such small amounts of sex steroids, a sudden outpouring of gonadotropins occurred, and puberty was the result. In other words, the "gonadostat" had been reset and would only respond to the feedback effect of much higher levels of sex hormones . In the past few years, assay techniques that make possible the continuous measurement of gonadotropin levels have cast doubt on the gonadostat hypothesis . These measurements reveal that LH, for example, is elevated not only at puberty but also at birth in both rats and humans (Fig . 1) . In humans, LH falls between the ages of two and four, then rises again at puberty . This fluctuation, which corresponds with the period of

The Gonadostat Hypothesis The basis of the gonadostat hypothesis is the pituitary hypertrophy that occurs after castration .' This phenomenon was familiar to physiologists at the beginning of this century and is clearly described in a comprehensive article on the pituitary written by Harvey Cushing in the From the Radiation Therapy Department, Veterans Administration Hospital, Bronx, N .Y ., 10468 . Requests for reprints should be sent to Dr. S . Lehrer, Radiation Therapy Department, VA Hospital, 130 W . Kingsbridge Rd ., Bronx, N .Y . 10468 . 39



40

THE MOUNT SINAI JOURNAL OF MEDICINE

Arcuate Nucleus Intrinsic Frequency (hr 1) 0 .67

1.33 1 .00 i

20 Age (years)

FIG. 1 . (Modified from Grumbach, reference no . 1) : The high

LH levels at birth and puberty, in both normal and agonadal girls, can be accounted for by a continually slowing arcuate nucleus intrinsic frequency after birth .

infant sexuality and subsequent latent period posited by Freud (6), is especially prominent in agonadal children ; it cannot be explained adequately on the basis of the gonadostat hypothesis . Further, the sensitivity of the pituitary-hypothalamic axis does not appear to change appreciably during development (7) . These observations imply that the phenomenon observed by Hohlweg is a result rather than a cause of puberty onset (8) . Fertility and Circadian Rhythms The reproductive cycle of mammals is closely related to the 24-hour light : dark cycle and circadian rhythms (9) . For example, during long nights and short days or after blinding, gonadal atrophy will occur in the Syrian golden hamster, Mesocricetus auratus . This phenomenon has been shown to be mediated by stimulation of the pineal gland (10) . Moreover, the four-day estrus cycle of hamsters is closely coupled to the length of the light : dark cycle . A normal hamster living in a 24-hour light : dark cycle has an estrus cycle of 96 hours (4 x 24) ; whereas a hamster maintained under constant dim illumination with a free running circadian rhythm of 24 .5 hours has an estrus cycle of 98 hours, that is, 4 x 24 .5 (9). Puberty, too, is affected by the light :dark cycle . Blind girls have their menarche earlier than sighted girls (11, 12), and rats reared in constant darkness have vaginal opening earlier than rats reared in an eight hour light : sixteen hour dark cycle (13) . Other investigators have attributed this acceleration of puberty to a pineal effect . However, such an effect would be paradoxical,

January-February 1983

since blindness or darkness stimulates the pineal, resulting in gonadal atrophy (10) . A more plausible explanation is that acceleration of the normal 24 hour circadian rhythms has accelerated puberty . These rhythms are generated in rats, hamsters, and probably higher mammals as well by a hypothalamic structure, the suprachiasmatic nucleus or SCN (14) . The SCN rhythm is exactly synchronized with the external light-dark cycle by impulses received from the eyes through a retinohypothalamic projection . In a blind animal or an animal kept in constant darkness or constant dim illumination, the SCN rhythm is free running, that is, somewhat greater or less in frequency than one cycle in 24 hours . The SCN is believed to be a master clock or zeitgeber, probably driving other physiologic rhythms, including plasma corticosterone levels and pineal N-acetyltransferase levels in rats . In blinded animals these rhythms are also free running, though synchronized with each other (15) . Blind adult human beings are known to have a free running frequency of about one cycle in 25 hours (16), but the free running frequency of children is unknown . One may, therefore, suggest that the free running frequency of a child is greater than one cycle in 24 hours-perhaps one cycle in 21 hours-and slows with age . Such slowing with age has recently been demonstrated in hamsters (17) . There are two possible ways of explaining how accelerated circadian rhythms could accelerate puberty . One way is to postulate that an animal needs a certain critical number of rhythmic stimuli to trigger puberty . In the rat, which has vaginal opening at about 38 days, 38 such stimuli would obviously be necessary . But in a rat on a 21-hour light : dark cycle, 38 stimuli could be given in about 33 days (that is, 21/24 X 38), thus accelerating puberty (Fig . 2), a fact which has been demonstrated experimentally (18) . Attractive though this explanation may be, it fails to account for two important facets of the childhood-puberty-adulthood sequence : (a) the elevation of gonadotropins at birth, and (b) the pulses of LH occurring every 1 1/2 hours only during sleep at the onset of puberty in humans ; these pulses later occur continuously every 1 1/2 hours throughout the day (19) . However, the physical conditions of oscillation and resonance can explain all of these factors . Oscillation and Resonance Work by Knobil and others suggests that the gonadotropin-releasing hormone secreting cells of



41

PUBERTY AND . RESONANCE : A HYPOTHESIS-LEHRER

21 hour 110 rats) (10.5 hrs IIghL 10.5 bs dark)

I'll,

j

iy 24 hour cycle (15 WO t (4h" II ht 20 h,s dark)

FIG. 2 . Rats reared on a 21-hour light : dark cycle (10.5 : 10.5) had vaginal opening significantly (p < 0 .01) earlier than rats reared on two variants of a 24-hour light : dark cycle (12 : 12 or 4 : 20) .

the arcuate nucleus are an independently functioning oscillator with their own intrinsic or free-running frequency (20, 21) . But in a normal, intact animal, the oscillation of the arcuate nucleus interacts with the zeitgeber, the SCN . As the animal approaches puberty, the frequency of the arcuate nucleus slows and resonance, which is well documented in animals (22), occurs. Hormonal surges increase in intensity at this time because in any oscillating system, the amplitude of the oscillations surges to a maximum at resonance . This may be likened to the motion of a swing being pushed periodically . When the pushes occur with a frequency other than the intrinsic one of the swing, the displacement of the swing is rather small . But as the frequency of the pushes approaches the intrinsic frequency of the swing, the displacement of the swing becomes larger and larger . Resonance is said to occur when an oscillating system, be it swing or arcuate nucleus, is acted on by a periodic series of impulses (from the SCN in the case of puberty) having a frequency equal or nearly equal to its intrinsic frequency . This explanation is most satisfactory if the SCN is postulated to be acting on the arcuate nucleus with an ultradian frequency of one cycle in 1I/2 hours (that is, 0 .67 cycle per hour) rather than one cycle in 24 hours . Such an ultradian frequency is possible according to the laws of physics because it is an integral multiple or harmonic of the fundamental frequency of the SCN (that is, 16 x 1/24) . Further, ultradian oscillations of this frequency and the existence of more than one in-

trinsic frequency are well documented in humans and other animals (22, 23) . An animal on a 21-hour light : dark cycle would have an ultradian rhythm more rapid than 0 .67 cycle per hour, since this rhythm is a multiple of and coupled to the SCN rhythm . As a result, the arcuate nucleus rhythm would not have to slow as much for resonance to occur, and puberty would take place at an earlier age (Fig . 3). The high LH level at birth can also be explained . If the frequency of the arcuate nucleus of an infant is very much higher than in an adult, specifically 1 .33 cycles per hour, resonance with the ultradian rhythm of the SCN will occur, since 1 .33 cycles per hour is an integral multiple (harmonic) of 0 .67 cycle per hour (that is, almost 2 X 0.67) . As the frequency of the arcuate nucleus slows between four and eight years of age, resonance is lost and LH levels drop . But when the frequency of the arcuate nucleus has slowed to 0 .67 cycle per hour, resonance again occurs, resulting in the gonadotropin surge necessary for secondary sexual development (Fig . 1) . The surges at puberty may be greater than the neonatal surges because of the growth and maturation of the brain that have taken place during the interval, or because resonance may not occur as readily at the higher fundamental frequency, 1 .33 cycles per hour . An interesting characteristic of the data in Fig . 1 is the apparent logarithmic slowing of the intrinsic frequency of the arcuate nucleus with time, especially evident in the agonadal girls . The peak LH level in infancy occurs at age 1 .6, the minimum at age 5, and the peak pubertal level at

Arcuate nucleus intrinsic frequency

r U ww D ww U. . U

I

I --------------------Harmonic of SCN frequency-normal girls II I Early puberty I Normal puberty I in normal girls in blind girls I

~1

If

-

AGE FIG . 3. Blind girls (and rats reared in constant darkness) probably have higher than normal suprachiasmatic nucleus (SCN) frequencies and corresponding harmonics . As a result, the intrinsic frequency of the arcuate nucleus does not have to slow as much for resonance to occur, and puberty can take place at an earlier age .



42

THE MOUNT SINAI JOURNAL OF MEDICINE

January-February 1983

groups of oscillating cells within the brain, the arcuate nucleus and the suprachiasmatic nucleus .

Arcuate nucleus rhythm ~Suprachiasmatic nucleus rhythm

TIME .

References

b. i

FIG . 4. The beat phenomenon . Two rhythms of slightly different frequencies, shown in (a), combine in (b) to produce a rhythm whose amplitude (broken line) varies periodically with time . This variation may account for the LH pulses which occur only during sleep in early puberty .

age 16 (1); the logarithms for each of these ages are as follows : age 1 .6 5 16 50 log age 0 .2 0.7 1.2 1.7 The logarithmic difference between each of these age-points is 0 .5. Thus, the next theoretically complete loss of resonance, comparable to that at age 5, would occur at age 50-around the time of menopause. The mechanism of resonance can account for the LH pulses that occur during sleep in early puberty . At early puberty, the frequency of the arcuate nucleus is very close to the ultradian frequency of the SCN. When two oscillators of very nearly the same frequency interact, the phenomenon known as "beats" will occur (Fig . 4) ; the rhythms combine to give a rhythm whose amplitude varies periodically with time . Such a mechanism is familiar in mammals, having already been invoked to explain the hormone level variations in the rat estrus cycle (24). In the phenomenon of puberty, the beats correspond with the LH pulses during sleep . In addition, the beat phenomenon appears to be related to the connection between time-of puberty onset in girls and body weight . Very lean girls undergo menarche later than girls with more adipose tissue (25) ; a specific fat-to-lean mass ratio must be present for the onset of menarche . The resonance mechanism of puberty suggests that extreme leanness increases the frequency of the arcuate nucleus . One would therefore expect that rapid loss of weight in a young postpubertal woman might restore LH "beats" during sleep ; this, in fact, has been observed to occur (26) . Thus, as this article attempts to demonstrate, many of the characteristics of puberty can be explained as the simple interaction between two

1 . Grumbach M . The neuroendocrinology of puberty . In : Krieger DT, Hughes JC, eds. Neuroendocrinology . Sunderland, Mass . : Sinauer Associates, 1980 :249 . 2. Cushing H . The functions of the pituitary body . Am J Med Sci 1910 ; 139 :473-484 . 3 . Christy NP. Harvey Cushing as clinical investigator and laboratory worker . Am J Med Sci 1981 ; 281 :79-96 . 4 . Hohlweg W. Der Mechanismus der Wirkung von gonadotropen Substanzen auf das Ovar der infantilen Ratte . Klin Wochenschr 1936 ; 15 :1832-1835 . 5 . Ramirez VD, McCann SM . Comparison of the regulation of luteinizing hormone (LH) secretion in immature and adult rats . Endocrinology 1963 ; 73 :452-464 . 6 . Hall CS . A Primer of Freudian Psychology . Mentor Books, New York, 1954, 110 . 7 . Johnson JH, Davis CL . Increased pituitary sensitivity to luteinizing hormone releasing hormone at puberty : an event of proestrus . Proc Soc Exp Biol Med 1981 ; 167 :434-437 . 8 . Ojeda SR, Advis JP, Andrews WW . Neuroendocrine control of the onset of puberty in the rat . Fed Proc 1980; 39 :2365-2371 . 9 . Zucker I . Light, behavior, and biologic rhythms . In : Krieger DT, Hughes JC, eds . Neuroendocrinology . Sunderland, Mass . : Sinauer Associates, 1980 :93 . 10 . Hoffman RA, Reiter RJ . Pineal gland : influence on the gonads of male hamsters . Science 1965 ; 148:1609 . 11 . Zacharias L, Wurtman R. Blindness: its relation to age of menarche . Science 1964; 144 :1154-1155 . 12 . Magee K, Basinska J, Quarrington B, Stancer HC . Blindness and menarche . Life Sciences 1970 ; 9 :7-12 . 13 . Relkin R. Pineal function relation to absolute darkness and sexual maturation . Am J Physiol 1967 ; 213 :999-1002 . 14 . Mosko S, Moore R . Neonatal ablation of the suprachiasmatic nucleus-effects on the development of the pituitary-gonadal axis in the female rat . Neuroendocrinology 1979; 29 :350-361 . 15 . Pohl CR, Gibbs FP . Circadian rhythms in blinded rats : correlation between pineal and activity cycles . Am J . Physiol 1978 ; 234 :R110-14. 16. Miles LE, Raynal DM, Wilson MA . Blind man living in normal society has circadian rhythms of 24 .9 hours . Science 1977 ; 198 :421-423 . 17 . Davis FC, Menaker M. Hamster through time's window: temporal structure of hamster locomotor rhythmicity . Am J Physiol 1980 ; 239 :R149-155 . 18 . Lehrer S. Twenty-one-hour light-dark cycle accelerates vaginal opening in the rat . Bull NY Acad Med 1981 ; 57 :705-708 . 19 . Boyar RM, Finkelstein JW, Roffwarg H, Kapen S, Weitzman ED, Hellman L . Synchronization of augmented luteinizing hormone secretion with sleep during puberty. N Engl J Med 1972 ; 287 :582-586 . 20 . Knobil E . Patterns of hypophisiotropic signals and gonadotropin secretion in the rhesus monkey . Biol Reprod 1981 ; 24 :44-49 . 21 . Krey LC, Hess DL, Butler WR, Espinosa Campos H, Lu KH, Piva F, Plant TM, Knobil E . Medial basal



Vol . 50, No . 1

PUBERTY AND RESONANCE : A HYPOTHESIS-LEHRER

43

hypothalamic disconnection and the onset of puberty in the female rhesus monkey . Endocrinology 1981 ; 108 :1944-1948 . 22 . Veerman A, Vaz Nunes M . Circadian rhythmicity participates in the photoperiodic determination of diapause in spider mites. Nature 1980 ; 287 :140-141 . 23 . Klein R, Armitage R . Rhythms in human performance . 1 1/z hour oscillations in cognitive style . Science 1979 ; 204 :1326-1328.

sensitive pacemaker for estrous cycle of the rat . Am J Physiol 1981 ; 241 :E261-267 . 25 . Frisch RE . Pubertal adipose tissue : is it necessary for normal sexual maturation? Evidence from the rat and the human female . Fed Proc 1980 ; 39 :2395-2400 . 26 . Kapen S, Sternthal E, Braverman L . A pubertal 24 hour luteinizing homrone (LH) secretory pattern following weight loss in the absence of anorexia nervosa . Psychosom Med 1981 ; 43 :177-182 .

24 . Yochim JY, Shirer HW . Evidence for a photoperiod-

Submitted for publication March 1982 .