Joyce Lim:
Thank you for expressing your difficulty in your attempt to find a working conceptual model of GH - IGF-1 actions. I too had similar difficulties understanding regulation of GH by hypoglycemia. I have attempted to summarize some of the facts below and hope you find this helpful.
Some of the typical diagrams of GH-IGF-1 relationships tend to give us the impression that IGF-1 secretion by the liver is entirely under the control of growth hormone but this is not true. Under conditions of caloric restriction, Humans synthesize less IGF-I and IGF-I synthesis in the liver is refractory to GH stimulation. Secondly, IGF-1 levels are low in poorly controlled diabetes mellitus, and administration of insulin via the portal vein restores IGF-1 synthetic response of the liver. These data taken together indicate that insulin, nutritional state also regulate the synthesis of IGF-1 besides GH.
Another point is IGF-1 has both endocrine and paracrine effects. The primary source of circulating IGF-1 is the liver. The growth promoting effects of GH on the epiphyseal plate are likely mediated by the direct actions of growth hormone as well as actions of IGF-1 locally generated in the epiphyseal plate. The evidence for this comes from observations that combined deficiency of growth hormone and IGF-1 cause a greater restriction in linear growth compared to deficiency of either GH or IGF-1 alone.
For the purposes of remembering or developing a working model, one can hold GH - IGF-1 axis as one positive regulator of postnatal growth. Deficiency of IGF-1 in intrauterine life results in considerable growth retardation, and importantly, IGF-1 is not primarily under the control of GH in intrauterine life. Whereas in postnatal life, IGF-1 is an important negative feedback regulator of GH release from the anterior pituitary.
But GH has other acute metabolic effects expressed under conditions of nutrient deprivation. These are not apparent from its name. I too used to find the fact that hypoglycemia is a potent stimulus for release of growth hormone somewhat "paradoxical".
Metabolic actions of GH in the fasted state: It is not paradoxical when one considers that GH drives lipolysis and consequently mobilization of free fatty acids. It restrains the effects of insulin on glucose uptake by muscle - a mechanism that allows glucose to be prioritized for tissues such as brain and red blood cells that rely upon glucose for ATP generation. Effectively, what GH does under conditions of hypoglycemia is allow for a switch in the major substrate oxidized in muscle from glucose to fatty acids, and consequently, protein to be conserved. Because sustained glucose homeostasis in the face of starvation requires gluconeogenesis, the starvation response would consist of greater protein catabolism (and loss of lean body mass) if it were not for lipolytic effects of GH and the switch alluded to above.
To summarize, insulin is a major anabolic hormone during conditions of energy surplus (hormone of plenty) whereas growth hormone functions as a structure preserving hormone under conditions of famine and energy deficiency.
In the well fed state however, GH together with IGF-1 and insulin allow for anabolism (protein, fat and glycogen) and tissue growth. Now, finally to the effect of IGF-1 on glucose metabolism: Nicholas describes this very well (see his post below). With regard to the magnitude of the 'hypoglycemic' effect of IGF-1, I wish to add that IGF-1 is much less (12 times less potent) potent in relation to insulin as a blood glucose lowering hormone. Furthermore, a larger fraction of IGF-1 is bound to proteins in plasma, when compared to insulin. I am not denying that IGF-1 promotes the uptake of glucose by muscle and other tissues, and it does suppress hepatic glucose output. But, unless IGF-1 or IGF-2 levels are unchecked as may happen with a tumor secreting IGF-II, 'hypoglycemia' due to IGF is very unlikely. There are several reports of IGF-II mediated hypoglycemia in cancers, but this certainly is not a physiologic effect of IGF-I or even IGF-II.
Also, as Nicholas describes, the effects of IGF-1 may become important under conditions of insulin deficiency. First, normal nutrient levels and insulin levels are required for IGF-1 expression. Since IGF-1 is the major negative feedback regulator of GH release and GH fosters insulin resistance, IGF-1 deficiency may promote insulin resistance through a relative and "chronic" excess of GH. A single dose of IGF-1 has been shown to reduce "relative" GH excess in type I diabetes mellitus. Continuous infusions of IGF-1 and IGF binding protein 3 has also been shown to reduce insulin requirements in individuals with type I diabetes [3]. Normally, raised insulin resistance due to GH wanes within a few hours, however, that does not happen if GH secretion by IGF-1 is unchecked. That is we have a positive feedback loop here with insulin deficiency leading to insulin resistance.
Bottom line: Circulating IGF-1 is predominantly stimulated by GH but this requires a positive energy balance, and elevated insulin levels, such as in the postabsorptive state. This hormonal milieu is compatible with anabolism and growth. Whereas, under conditions of nutrient deprivation, the lipolytic and hyperglycemic effects of GH predominate.
References:
[1] Moller et al. http://edrv.endojournals.org/cgi/content/short/30/2/152 [2] Ohlsson et al. http://edrv.endojournals.org/cgi/content/full/30/5/494
[3]
Clemmons et al. http://jcem.endojournals.org/cgi/content/abstract/85/4/1518 All the best
Prakash
*****************************
On Sun, May 1, 2011 at 11:41 AM, Nicholas Cheong
<nicholas...@gmail.com> wrote:
I have a doubt.
Growth hormone (GH)
antagonizes actions of insulin, thus increasing glucose levels in blood. Another fact is
that GH effects/actions are mostly mediated by insulin-like
growth factor (IGF). But why does IGF cause hypoglycemia?
"Shouldn't" it cause hyperglycemia,
since it is supposed to mediate the actions of GH? Going by its name, I understand why IGF causes hypoglycemia. But since it is also known as somatomedin, "shouldn't' it cause hyperglycemia like GH? Please enlighten me.
GH is lipolytic while IGF is antilipolytic - another confusion for me.
Attached is the diagram showing the correlation of GH and IGF obtained from Ganong.
Thank you for your
help.
Joyce Lim
MBBS Batch 15 (Year 2)
Dear Joyce,
Insulin-like growth factors are thus named because it has
functional and structural similarities with Insulin. IGF’s and insulin cross react with each others’
receptors. Some tissues express hybrid receptors that combine insulin and IGF-I
receptor subunits.
Since insulin facilitates glucose entry into cells, one
would expect IGF to do the same.
IGF was initially named somatomedin…mediator of somatotropin
(GH) because of its action on the bones and cartilage, and not so much because
of its actions on the intermediary metabolism.
We can say that IGF-1 is a growth promoting hormone and its metabolic actions appear similar to that of insulin. I'd like to think of IGF's as insulin-like factors which promote growth.
IGF’s or somatomedins have an
inhibitory effect on GH secretion by stimulating somatostatin release from the neuroendocrine
neurons of the periventricular nucleus of the hypothalamus. If IGF is reduced
for whatever reasons, the negative feedback on GH secretion is lost and GH
secretion increases. GH increases glucose output and decreases insulin
sensitivity. There is a compensatory
increase of insulin secretion by the endocrine pancreas, giving rise to a
hyperinsulinemic state. What we see here is a case of insulin resistance (elevated
basal as well as glucose-induced insulin levels but normal glucose levels).
What is worse is that the pancreas might be overworked due to excess GH
secretion and pancreatic beta cells might be damaged. The significance of this
is that, decreased IGF production or defective IGF signaling may predispose an individual
to diabetes. And IGF-1 administration could be an option for treatment of DM
type-2.
Experimental rats with deleted IGF genes show hyperinsulinemic state.
http://diabetes.diabetesjournals.org/content/50/7/1539.full
The antilipolytic effect was demonstrated by the deletion of IGF genes
in lab rats. These rats appeared leaner. This implies that IGF is lipogenic.
In experimental rats with deleted
IGF genes, leptin levels were found to be increased. Leptin is known to
decrease adiposity ( one way it does so is by acting on higher center to
decease appetite).
In these rats, IGF’s produced by
liver were absent. Hence Insulin may surge secondary to GH oversecretion.
Insulin has been shown to increase plasma leptin.(please follow the link I have included above for further reading)
Dr Prakash will be able to provide a better explanation. In the mean time, I hope this helps a little.
All the best,
Nicholas Cheong
MBBS Batch 14 (Year 3)
--
Visit us on the Web at https://sites.google.com/site/medicalphysiologyataufom/home
To unsubscribe, send an e-mail to dresp...@gmail.com from the address at which you are subscribed with the Subject Line "Unsubscribe from Medical Physiology at SOM, AIMST"