Canadian Medical Association Journal 1995; 153: 585-588
[résumé]
When Gloria Shaffer Tannenbaum received a career award from the Quebec government in 1994 she was being recognized for work that began with her doctoral research almost 20 years ago. She had discovered that the pituitary gland of the male rat releases growth hormone in a strikingly rhythmic pattern, and she focused her career on unlocking the complex mechanisms that regulate that rhythm.
Tannenbaum's research at the McGill University- Montreal Children's Hospital Research Institute has led to the development of a simple test to differentiate children who are deficient in growth hormone from those who are short but growing normally. Growth hormone is essential for normal growth and bone maturation.
This work is part of a new and growing body of research into "dynamical diseases." Medical scientists now realize that cardiovascular and respiratory problems, tremors and other movement disorders, endocrine and other abnomalities and menopausal hot flashes can be understood and treated in terms of coupled interactions. In her own area of study, Tannenbaum has discovered that "from the release of growth hormone into the blood to the level of gene expression there is an exquisite rhythmicity."
Tannenbaum has focused her research on four aspects of growth hormone secretion: the interactions of neurohormones in the brain and pituitary gland that regulate the release of growth hormone, feedback mechanisms in this system, sexual differences in secretory patterns, and clinical applications.
A professor in the departments of Pediatrics and of Neurology and Neurosurgery at McGill, Tannenbaum receives funding for her research from the Medical Research Council and the Fonds de la recherche en santé du Québec (FRSQ). She received a Chercheur-boursier de mérite exceptionnel award from the FRSQ from 1989 to 1994 and, in 1994, the new Chercheur de carrière award, which is held by one senior research scientist at each of Quebec's four medical schools.
In the past, Tannenbaum says, researchers found huge variations in the baseline levels of circulating growth hormone. "If you took 10 basal blood samples from normal animals, the values [would be] all over the map. You couldn't study growth hormone regulation properly."
Another problem was that hormone secretion is sensitive to stress. While working on her doctorate at McGill with Dr. Joseph Martin, Tannenbaum developed a method that mimicked as closely as possible the normal physiologic state of the test animals and therefore subjected them to a minimum of stress. She still uses this model: her test animals, free-moving rats, have a catheter inserted in the jugular vein to permit blood samples to be taken frequently over a long period without distressing the animals.
What she found was a striking ultradian rhythm in the male rat: every 3.3 hours there were huge bursts in plasma growth hormone levels, i.e., seven surges in 24 hours. Peak levels often exceeded 200 ng/mL. After a brief period, these levels fell off until they were undetectable (less than 1 ng/mL). In the controlled environment of the lab, this pattern of secretion was so regular that all of the animals showed peaks and troughs of growth hormone release at the same time of day. "This was a major breakthrough in the area of growth hormone regulation and provided an explanation for the wide variation in basal growth hormone levels noted earlier,"
Tannenbaum says, adding that pulsatile patterns of growth hormone release have subsequently been documented in every mammalian species studied, including humans. Tannenbaum then turned her attention to the mechanisms in the brain that regulate this rhythm, focusing on two hypothalamic neuropeptides: somatotropin release-inhibiting factor (SRIF), also called somatostatin, and growth hormone-releasing hormone (GHRH). In 1984 she hypothesized that these peptides are themselves released from the brain rhythmically.
SRIF is produced in the preoptic-anterior region of the hypothalamus; GHRH is produced in the ventromedial-arcuate region. These peptides are released from axon terminals in the median eminence into the hypophysial portal circulation, and from there are transported to the pituitary gland. A series of experiments demonstrated clearly that GHRH stimulates the release of growth hormone from the pituitary gland, whereas SRIF inhibits secretion.
To shed light on the roles of GHRH and SRIF in generating the ultradian rhythm of growth hormone secretion, Tannenbaum tried to block their effects by administering specific antibodies to each peptide. Her results lent support to the view that surges in growth hormone secretion are caused by the periodic release of hypothalamic GHRH, whereas troughs are regulated by periodic SRIF release.
Tannenbaum also investigated the ability of GHRH to stimulate the release of growth hormone at different times in the peak-trough cycle. She gave the same dose of GHRH when there was usually a surge in growth hormone secretion and when there was usually a trough. Again, the results were striking. When GHRH was given when secretion was normally high, there was a more than two-fold increase in the release of growth hormone. But when the same dose was given during a trough period, minimal growth hormone was released.
There is evidence that these neuropeptides also interact within the central nervous system. Here, SRIF seems to influence the secretion of GHRH, but the exact mechanisms and morphologic basis for this action are not entirely clear. One study suggests that SRIF directly affects the GHRH-containing cells of the arcuate nucleus, given that a subpopulation of these cells have SRIF binding sites. Another study suggests that there may be an ultradian oscillation in SRIF binding to arcuate neurons in relation to the peaks and troughs of the growth hormone secretory rhythm. Finally, there is evidence that the levels of messenger RNA coding for GHRH and SRIF in the hypothalamus change in a manner consistent with the existence of an ultradian rhythm in gene expression. Thus, there appear to be multiple levels of interaction between SRIF and GHRH in the genesis of rhythmic growth hormone secretion.
It is one thing to understand these patterns in caged rats for which environmental influences are rigidly controlled and quite another to know whether a child who arrives at a clinic for growth hormone level testing is in a peak or a trough period of secretion. Furthermore, the tests currently used to detect growth hormone deficiency are, in some cases, risky in themselves. False results, which do occur, can cause some children to be given growth hormone therapy when they don't need it -- with unknown long-term consequences, and at a cost of approximately $15 000 a year. "This is why it is important to have a reliable diagnostic test to distinguish normal short children from those who are truly growth hormone deficient," says Tannenbaum.
In 1993 Tannenbaum and her associates Drs. Harvey Guyda and Zvi Dickerman developed a test that is cheaper and more reliable than those currently in use. "This test is noninvasive and has no side effects," Tannenbaum explains. It is based on her earlier discovery that, in addition to inhibiting growth hormone secretion, SRIF also sensitizes pituitary somatotrophs -- the cells of the adenohypophysis that produce growth hormone -- to GHRH. Tannenbaum did a further series of experiments on rats using a long-acting SRIF analogue. Pretreatment with this analogue enhanced the responsiveness of growth hormone to GHRH by allowing stores of growth hormone to build up in the pituitary gland.
The researchers first conducted a study to determine whether this approach could maximize pituitary growth hormone response in humans. Their study involved 37 normal short children. One group of children was given the SRIF analogue; a second group acted as controls. Five hours later both groups were given GHRH. The children who had received the SRIF analogue secreted two to three times the amount of growth hormone secreted by those in the control group.11 "The somatostatin pretreatment puts a brake on the pituitary gland and allows the stores of growth hormone to build up," Tannenbaum explains. "The subsequent GHRH challenge exerts an accentuated effect because of a larger, readily releasable pool of growth hormone."
The next, recently completed, study was designed to see whether this approach could be used to discriminate between normal short children and children who were truly deficient in growth hormone. It involved 24 children who came to clinics at Montreal Children's Hospital and Hôpital Sainte-Justine to be tested for growth hormone deficiency. "In addition to conventional diagnostic tests, we used our test to see whether we could better discriminate between them. The children were given long-acting SRIF and then GHRH 5 hours later. The kids who were deficient hardly released any growth hormone, while children who were normal released a huge amount. This was an excellent discriminator of the two populations: the combined SRIF/GHRH test clearly discriminated normal short children and, more importantly, those with borderline responses to conventional tests, from those with growth hormone deficiency, as there was no overlap of the growth hormone responses between the groups."
"I'm pleased to have been involved in the development of a clinical application based on our animal research so rapidly," says Tannenbaum, adding that she hopes this test will eventually become standard. It is equally applicable to boys and girls, although Tannenbaum's research shows dramatic sexual dimorphism in growth hormone release patterns in rats. Male rats grow at three times the rate of females. Although the overall amount of growth hormone secreted is the same in males and females, the patterns of growth hormone secretion in females are more erratic: growth hormone peaks are lower and more frequent in females than in males, and the trough levels never approach zero. In females, hypothalamic SRIF appears to be secreted continuously into the hypophysial-portal circulation, rather than cyclically, as in males. "The pattern of growth hormone secretion, including the period of quiescence regulated by somatostatin, is ultimately the crucial determinant for the animal's growth," Tannenbaum concludes.
Although GHRH and SIRf control the pattern of growth hormone secretion, it has become evident that sex steroids also play a role. When Tannenbaum gave estradiol to male rats they showed a female secretory pattern. Similarly, female rats who were given testosterone showed a male pattern of secretion. Tannenbaum thinks that sex steroids probably alter the release patterns of SRIF and GHRH in the brain. Tannenbaum has also obtained evidence that growth hormone can regulate its own secretion by means of a negative feedback system, possibly by stimulating hypothalamic SRIF secretion. She is now investigating the feedback effects of insulin-like growth factors (IGFs), the hormones that promote longitudinal bone growth. There may be a synergistic interaction between IGF-1 and IGF-2 in the brain that modulates the pulses of growth hormone secretion.
The clinical implications of Tannenbaum's work are already being tested. For example, treatment of growth hormone deficiency may be optimized by mimicking the natural pattern of secretion; researchers such as Michael Thorner at the University of Virginia in Charlottesville, Va., have used a programmable pump to give GHRH -- rather than growth hormone -- to children deficient in growth hormone. By giving the hormone in seven doses over a 24-hour period they were able to maximise the child's growth potential.
Growth hormone is receiving increased attention these days. Physicians once believed its role ended once longitudinal growth was completed. Now researchers in Europe and North America are studying the effects of growth hormone replacement therapy in adults deficient in growth hormone. "This has marked effects in terms of psychologic well-being, body composition and metabolism," notes Tannenbaum. Other researchers, investigating the rejuvenating effects of growth hormone to see if it can halt or even reverse some of the degenerative changes associated with aging, have seen improved muscle tone and a greater sense of well-being in their patients.
Recently, Tannenbaum turned her attention to the role of growth hormone in aged rats. "We are trying to understand why growth hormone secretion declines in aging. Once we understand what the basic mechanisms are, we will know whether and how to replace it properly."