what-when-how
In Depth Tutorials and Information
from the hypothalamus and circulating GH, usually
bound to GH-binding protein (GHBP, the extracellular
portion of the GH receptor
32
), induces the production
of IGF-1 by the liver and other peripheral tissues. Liver-
derived IGF-1 constitutes the majority of systemic IGF-1
and circulates as part of a ternary IGF complex that con-
tains IGF-1, IGF-binding protein-3 (IGFBP-3 or -5) and
the acid labile subunit (ALS), also produced by the liver
in response to GH.
33
Circulating IGF-1 provides nega-
tive feedback directly by inhibiting GH production from
somatotrophs and indirectly by stimulating hypotha-
lamic somatostatin.
31
Additional IGFBPs (IGFBP-1, -2, -4
and -6) modulate the activity of IGF-1 in various tissues,
and IGFBPs can also exert their own effects independent
of IGF-1, providing an additional layer of regulatory
complexity to the GH/IGF-1 axis.
34
As a central regulator of growth, the GH/IGF-1 axis
is also influenced by a number of other hormone path-
ways, integrating cues related to nutrition, physical activ-
ity, stress and sexual reproduction. For example, ghrelin
is a small peptide produced by the gastrointestinal tract
that induces GH production and regulates appetite.
35,36
Growth hormone levels are also regulated positively by
thyroid hormones
37
and negatively by glucocorticoids.
38
Sex steroids have a profound influence on the GH/IGF-1
axis that is particularly evident during puberty. Growth
hormone secretion patterns are similar in boys and girls
prior to puberty, with low-level pulses throughout the
day and maximal secretion during early sleep.
39
Parallel
with the increase of sex steroid levels during puberty,
mean 24-hour GH levels also increase, owing to increased
GH pulse amplitude.
40
Although elevated GH and IGF-1
levels are observed in girls earlier than boys, both GH
and IGF-1 decrease after Tanner stage 5 in both sexes.
41
The sexually dimorphic pattern of GH secretion persists,
however, with women having more frequent, uniform
pulses throughout the day and higher interpulse GH
levels than men, who by contrast, have large nocturnal
pulses and relatively low daytime GH secretion.
42
The peak levels of GH (and IGF-1) attained during
puberty correlate with peak longitudinal growth veloc-
ity during the pubertal growth spurt.
43,44
This longitudi-
nal growth occurs through the process of endochondral
ossification in long bones. In this process, growth plate
chondrocytes differentiate in a linear fashion and become
hypertrophic. These differentiated chondrocytes form a
cartilaginous template that becomes new trabecular bone
formed by osteoblasts, thereby elongating the metaphy-
sis. As the bone elongates, those trabeculae near the cen-
ter of long bones are resorbed to form the marrow cavity,
and trabeculae near the outer edges of the bone coalesce
and are remodeled by osteoclasts to form the metaphyseal
cortex. In the diaphysis, cross-sectional growth is medi-
ated by the combination of periosteal cortical apposition
and endosteal resorption.
45
Salmon and Daughaday first
proposed that GH stimulated this longitudinal epiphy-
seal growth through circulating, liver-derived IGF-1 in
the original somatomedin hypothesis.
21
However, subse-
quent evidence of direct effects of GH on chondrocytes
led to the formation of a second hypothesis by Green
in 1985: the dual effector theory.
46
This working model
proposed that GH would induce differentiation of chon-
drocytes in the germinal zone, as well as local IGF-1
synthesis, that would then drive the subsequent clonal
expansion of chondrocyte columns and hypertrophy
through an autocrine/paracrine mechanism.
Studies in genetic mouse models suggest that it is
likely a combination of both mechanisms by which GH
and IGF-1 contribute to longitudinal bone growth, with
site, timing and relative GH/IGF-1 levels determining
the apparent mechanism. For example, mice with con-
ditional deletion of
Igf1
from only the liver challenged
the somatomedin hypothesis, since significantly reduced
circulating IGF-1 levels resulted in no effect on postnatal
growth.
47,48
Suggesting that both of these models failed
to delete liver
Igf1
prior to the post-weaning growth
spurt, Stratikopoulos and colleagues generated a mouse
in which
Igf1
was conditionally expressed in the liver,
on an
Igf1
-null background (effectively producing only
liver-derived IGF-1).
49
In this model, liver-derived IGF-1
contributed to 30% of adult body size and sustained
postnatal development, supporting the somatomedin
hypothesis that GH stimulates longitudinal growth (at
least in part) through liver IGF-1 production. In addi-
tional support of the somatomedin hypothesis, admin-
istration of exogenous IGF-1 to
Igf1
-null mice increases
growth, while exogenous GH treatment is ineffective.
50
It is clear, however, that not all the growth promoting
effects of GH are dependent upon IGF-1 production,
since mice with disruption of both
Ghr
and
Igf1
display
a more severe growth retardation than either mutant
individually.
50
Further,
Igf1
-null mice (with increased
circulating GH due to loss of negative feedback) have
an enlarged germinal zone and reduced chondrocyte
hypertrophy. By contrast, mice lacking
Ghr
exhibit both
defective chondrocyte generation and hypertrophy, sup-
porting the dual effector theory and suggesting distinct
functions of GH and IGF-1 in the growth plate.
50
In addition to their impact on longitudinal bone
growth through chondrocytes, GH and IGF-1 also exert
significant effects on bone mass through osteoblasts
(and osteoclasts, although less well studied). GH and
IGF-1 both promote osteoblast proliferation and survival
in vitro
,
51-53
although the effects of GH on osteoblast pro-
liferation appear to depend upon GH-induced IGF-1
production.
53
IGF-1 is also critical for osteoblast differ-
entiation and mineralization, with a biphasic expression
during osteoblast differentiation
in vitro
that initially
promotes maturation
54
and later augments type I colla-
gen synthesis and inhibits collagen degradation.
55,56
