Insulin-like growth factor-1 deficiency and metabolic syndrome

An assortment of epidemiological and clinical studies have stated glucose and lipid
metabolism alterations, insulin resistance, and central obesity as predominant factors
for the development of MetS 38], 212].

Similarities between insulin and IGF-1 suggest the possible role of IGF-1 in the pathological
process of this syndrome, therefore several studies have attempted to correlate IGF-1
plasma levels with MetS. Figure 2 represents the pathophysiology of an altered IGF-1/GH/insulin axis, and potential
beneficial actions of IGF-1 therapy.

Fig. 2. Metabolic effects of IGF-1 and GH under pathological conditions. The figure summarises
schematically some of the metabolic mechanisms altered in obesity and the role that
IGF-1 and GH exert on them

A general finding is that obese patients fulfilling criteria for MetS together with
low IGF-1 plasma levels tend to develop a worse cardiovascular disease outcome than
those with mid-normal to high-normal IGF-1 levels 213]. Nevertheless, many of them also present insulin resistance and inflammatory cytokine
secretion, so it is difficult to determine the exact role of each component in the
cardiovascular outcome.

Nonetheless, low IGF-1 circulating levels are also associated with reduced insulin
sensitivity 207], glucose intolerance, and T2D 207], 208], 214]. Moreover, some inflammatory cytokines are known to reduce IGF-1 levels in animal
models 215]. Additionally, the IGF-1/IGFBP-3 ratio, a common rough of free IGF-1 levels, is significantly
decreased in obesity 216], however no IGF-1 bioactivity was estimated. This parameter has been further studied,
showing that those men and women in the lowest quartile of the IGF-1/IGFBP3 ratio
are threefold more likely to meet the Adult Treatment Panel III (ATP-III) definition
for MetS, and twice as likely to be insulin resistant—that IGF-1/IGFBP-3 ratio decreases
notably as the number of MetS components increases 217]. Furthermore, visceral adipose tissue mass has been inversely correlated with circulating
IGF-1 levels 216]. Nevertheless, the mechanism of this possible inverse relationship between MetS and
free IGF-1 levels remains unclear.

A very interesting epidemiological study that supports this idea showed that in normal
subjects, IGF-1 contributes to glucose homeostasis. This study analysed a group of
Dutch Caucasians with a polymorphism in the promoter of the IGF-1 gene 129], 218]. Results within this group showed a reduced IGF-1 secretion—40 % lower than those
without the polymorphism. These subjects are 2.1 cm shorter and have 2.2-fold increase
in T2D prevalence after the age of 60 129], 219]. A different study 220] tested IGF-1 in response to energy intake in the Gujarati migrant community in Sandwell
(UK) and data was compared with people still resident in their village of origin in
India. Total energy and total fat intake were higher in UK migrants, as were IGFBP-3
and IGF-1, but IGFBP-1 was lower in UK migrants. At both sites, IGF-1 and IGFBP-3
correlated positively with total energy and fat. Conversely, in Indian Gujaratis,
IGFBP-1 fell with increasing total energy and fat intake but not in UK Gujaratis.

Several other studies (one of them being a large scale community-based Framingham
Heart Study 221]) have suggested a role for IGF-1 in the prevalence of insulin resistance and MetS
221]–223]. Biomarkers correlating the lower IGF-1 concentration to increased waist-to-hip ratio
or to impaired glucose tolerance are also being studied 224], 225].

Insulin-like growth factor-1, as discussed earlier, has implications on lipid and
glucose metabolism 87], 88], and its exogenous administration enhances insulin sensitivity in healthy adults
139], 140] as well as those with T2D 155].

A fascinating study including over 500 patients revealed that IGF-1 concentrations
were independently associated with insulin sensitivity accounting for 10.8 % of its
variation. The results were assessed by HOMA-S together with anthropometric measurements,
HDL, TG and blood pressure, and found correlations between these parameters and IGF-1
plasma levels. Additionally, they established that according to the WHO definition
for MetS, each unit increase in log-transformed IGF-1 concentrations, was associated
with a 90.5 % reduction in the risk of MetS 226].

Salmon et al. 227] showed that transgenic mice with reduced levels of IGF-1 can induce female insulin
resistance 227]. Moreover, global deletion of IGF-1 gene expression in mice does not result in glucose
intolerance. It has to be mentioned that KO mice for IGF-1 gene are not viable and
studies in such mice have to be done in the first days of life, and thus results are
not very conclusive as metabolism is not properly established at this stage 129], 228]. However, if a partial deletion is present, the mice will develop glucose intolerance
if starved. Additional studies found that elimination of hepatic IGF-1 gene expression
results in a compensatory threefold increase in GH secretion—recall that IGF-1 is
secreted by GH stimulation in hepatocytes. This combination of lowered serum IGF-1
and increased GH secretion leads to increased insulin resistance—as in the systemic
deletion but also developed glucose intolerance 129], 209]. Interestingly, glucose intolerance could be improved when IGF-1 was systemically
administered. This response was caused primarily by GH hypersecretion, as expression
of a GH antagonist resulted in improvement of glucose homeostasis 129], 229]. Additionally, administration of IGF-1 in the presence of this antagonist results
in a further improvement in insulin sensitivity; suggesting that, at high concentrations,
IGF-1 has effects not simply mediated by suppressing the effect of GH on hepatic insulin
sensitivity 129], 209]. In a similar study, the pivotal role for the IGF-1 in insulin sensitivity has received
further support from liver-specific IGF-1 KO mice which exhibited overt insulin resistance
and hyperinsulinaemia that was reversed by the administration of IGF-1 161].

Moreover, results from the aforementioned Framingham heart study also demonstrated
the correlation between low IGF-1 and the increasing metabolic syndrome markers 221]. A good example is the finding that low circulating levels of IGF-1 are independently
associated with hyperglycaemia and insulin resistance in adults 160], 230]–232]. To the contrary, high to normal levels of circulating IGF-1 correlate with a rise
in adiponectin levels and a reduced prevalence of MetS is found 233].

Since the liver is the major site of IGF-1 production, when steatosis develops lowering
insulin sensitivity, the severity of steatosis at different stages of insulin resistance
and metabolic syndrome seems to be correlated with worsened circulating IGF-1 levels
234]. In addition, low IGF-1 subjects in a study were found to possess up-regulated fatty
acid metabolism along with down-regulated GLUT-1 gene (in charge of glucose uptake
in erythrocytes, brain endothelial cells, eye, peripheral nerve and also responsible
for materno-placental glucose transfer) 210].

In summary, reconciling all discussed aspects relevant to insulin, IGF-1 improves
insulin sensitivity by suppressing insulin and GH secretion and by improving insulin
signalling indirectly reducing FFA flux.

When we look to IGF-1 levels in sera from T2D patients, the results found are very
wide 235]. It must be considered that multiple factors interact to control IGF-1 levels, many
of which are disturbed in T2D, namely: increased inflammatory cytokines, decreased
hepatic insulin action due to resistance, concomitant changes in IGFBPs, and the effects
of obesity. In addition, T2D is the result of a complex interaction of environmental
and genetic factors, being difficult to establish the role of each one in the pathogenesis
of diabetes and in the levels of IGF-1. In an experimental model, transgenic mice
expressing a kinase-deficient IGF-1R ?-subunit (thus displaying reduced signal transduction
in both IGF-1R and hybrid receptors) developed diabetes early on life 162], 236]. Also, mice carrying a genetic mutation that lack one of the igf1r alleles (igf1r^+/?
) show a 10 % reduction in post-natal growth, insulin resistance and glucose intolerance
237]. Additionally, infants born small for gestational age who exhibit low IGF-1 levels
presented a higher risk for the onset and development of T2D in adult life than those
born with normal weight 238], 239]. Nevertheless, it is noteworthy that the inverse correlation between IGF-1 and diabetes
only prevails in younger individuals (65 years) 231], establishing that this deficiency can lead to MetS—as aging can be considered an
IGF-1 deficiency condition 107].

Abnormal IGF-1 and GH levels have been proposed to play a key role in obesity 130], 240], 241]. Obese human and animal models are generally accompanied with abnormal circulating
IGF-1 levels 130], as well as IGF-1/IGFBP-3 ratio; and an inverse relationship between IGF-1 and visceral
fat mass distribution has been described 242]–246] as previously mentioned.

Moreover, some studies revealed that high fat diet promoted reactive oxygen species
(ROS) and cytokine production, apoptosis, protein and mitochondrial damage, and reduced
ATP content. These defects were accompanied by disrupted phosphorylation of the IRS,
as well as down-regulated expression of mitochondrial proteins PPAR? co-activator
1? (PGC1?) and uncoupling protein-2 (UCP-2) 200]. All of these factors can be alleviated by the cytoprotective and anti-inflammatory
actions of IGF-1 105], 123], 247]–249].

The culmination of all of these studies come to light with the finding of several
(some reported and many from our group still publication pending) Laron’s Syndrome
(congenital IGF-1 deficiency or GH insensitivity) never treated patients who underwent
progression to MetS and ultimately to T2D together with diabetic retinopathy when
they reached their late 30 s, 250], 251].

The role of insulin-like growth factor binding proteins in metabolic syndrome

Recall that IGFBPs have a very important role, not only in modulating free IGF-1,
half-life and localisation, but also by possessing IGF-independent activities mediated
by their own receptors. Changes in IGFBPs have been correlated with certain parameters
of MetS. For instance, low IGFBP-1 levels with high C-reactive protein values are
strong predictors of MetS 252], 253]. Low IGFBP-2 also acts as a good marker for MetS along with high fasting glucose
254]. Even though the data is extensive it is not enough to explain any relationship between
these molecules and the independent MetS factors, or any mechanism by which these
molecules could be involved in the pathophysiology.

As a result of IGF-1 changes, IGFBPs are also altered in T2D 255]. Studies in prediabetes suggest that IGFBP-1 shows normal levels before developing
T2D 256]. This may be caused by hyperinsulinaemia during the onset of the disease. Elevated
insulin causes an increase of free serum IGF-1, nevertheless, as resistance to insulin
progresses, the liver becomes insensitive to insulin-mediated suppression of IGFBP-1
254], 256]. On the other hand, IGFBP-3 undergoes augmented proteolysis on diabetic patients
which results in a sudden rise of free IGF-1 concentrations 257], 258].

Overexpressed IGFBP-1 in transgenic mice shows that its increase produces hyperinsulinaemia
accompanied by glucose intolerance, and that this process is dependent upon IGFBP-1
phosphorylation state—which is increased in diabetes 259]. In another study it was found that after weight loss there is a rise in IGFBP-1
and IGFBP-2, both in children and adults 260], 261]. Considering that IGF-1 regulates IGFBP-1 and IGFBP-2, it is reasonable to conclude
that early changes in insulin resistance alter their concentrations.

A recent study tested whether IGFBP-1 concentrations were able to predict the development
of T2D in women. The outcomes showed that women with the lowest fasting IGFBP-1 at
baseline had a higher risk for developing diabetes within 8 years, also showing an
impaired IGFBP-1 suppression after oral glucose loading 256]. Conversely, in another study, 615 patients with IGF-1 values in the lower half of
the normal range were closely observed during 4.5 years. It was found that there was
an increased predisposition to develop glucose intolerance or T2D, and that this change
was independent from IGFBP-1 208].

Moreover, IGFBP-2 is decreased in obese patients and thus increasing free IGF-1 262]. This fact is relevant as transgenic mice overexpressing IGFBP-2 were resistant to
developing obesity when fed with high fat diet 174], suggesting that this molecule could have a direct effect on preadipocyte differentiation.

IGF-1 treatment: future and limitations

The FDA approved the use of rhIGF-1 (Mecasermin, Increlexâ„¢) for treatment of severe
primary IGF-1 deficiency in 2003. By the same time, the FDA approved the use of a
equimolar combination of IGF-1 and IGFBP-3 (Mecasermin Rinfabate, iPlexâ„¢), suggesting
that it would be a better choice, seeming as it will require lower doses because of
the augmented half-life of the molecule and “buffering” effect on concentration. Additionally
this complex can bypass the IGF-2 displacement effect. Because IGFBP-3 also carries
IGF-1, when IGF-1 is administered on its own, the carrier IGFBP-3 saturates with IGF-1
displacing IGF-2 and thus augments its free circulating concentration. However, subsequent
studies revealed that no significant difference (among with some patient issues) could
be observed, so now IGF-1 alone seems to be an efficient treatment. Recombinant human
IGF-1 is usually synthesised in E. coli and subsequently purified. Its purification is a very insidious process that elevates
the cost, and this fact in combination to the limited applicability that has been
linked to date, elevates the price of the treatment.

The use of analogues has shown multiple and substantial clinical benefits along with
metabolic improvement. A promising therapeutic approach being studied uses IGF-1 analogues
such as PEG-IGF-1 263], which offers promising protection against acute contraction-induced muscle injury.

To date several clinical trials have been conducted to test IGF-1 under several conditions
and an infinity of animal models have been used to investigate its deficiency, treatment
and to exploit its actions. Phase I studies from the late 1980s using IGF-1 to treat
Laron’s Syndrome (GH insensitivity) revealed effective increase in linear growth reaching
adult heights using doses ranging from 80 to 240 µg/kg/day 264]–268]. Different groups of patients and candidates for IGF-1 treatment have been those
with an IGF-1 gene deletion and others with idiopathic short stature. Furthermore,
phase II studies were undertaken to prove the efficacy of mecasermin rinfabate (IGF-1/IGFBP-3)
in paediatric severe burns and had promising outcomes when doses of 1–4 mg/kg/day
269]–273] were used. Moreover IGF-1 was tested in patients with osteopenia/osteoporosis linked
to anorexia nervosa and severe bone fractures with positive results utilising doses
ranging from 30 µg to 1 mg/kg/day for 2–9 months 274], 275].

When treating metabolic disorders with IGF-1, phase II trials until now conducted
by Clemmons et al. including adult patients with either T1D or T2D treating with mecasermin
rinfabate (1–2 mg/kg/day for 14 days) showed lowered exogenous insulin requirements
while improving glycaemic control. These results suggest improvement in insulin sensitivity,
and moreover, no additional side effects to those found with placebo were observed
in T1D, meanwhile in T2D patients oedema, jaw pain and arthralgias were 4 % lower
than previous reports 276], 277]. Evidence revealed that endogenous and exogenous IGF-1 administration protects against
the onset and progression of diabetic cardiomyopathy 278], 279]. Consistent with this, patients with T2D responded to IGF-1 treatment with improved
glucose tolerance, hyperinsulinaemia, and hyperlipidaemia as previously stated 280]. Up to date, limited information is available regarding obesity treatment by regulation
of the GH-IGF-1 system. Apart from all of these conditions tested under IGF-1 treatment,
there are a number of other ones, among others: cystic fibrosis, AIDS, Chron’s, multiple
sclerosis and ALS—which are carefully revised by Rosenbloom 281].

Despite all the above-mentioned clinical benefits, the safety of long-term administration
of this hormone remains controversial. There is supporting evidence reporting adverse
effects from long-term rhIGF-1 treatment including neoplastic formation, cataract
and renal hypertrophy, all of which seem to be the most severe effects observed and
are usually rare 123], 282]–284]. Additionally these complications were transient, easily treated and tolerated without
treatment discontinuation. More common side effects within the mentioned studies revealed
mild to moderate effects like pain at site of injection or headache that was transient
and disappeared after 1 month 268]. Other side effects reported ranged from lipohypertrophy at injection site, papilloedema
(related to cranial hypertension), and facial nerve paralysis 276], 285], however symptoms did not prevail after treatment pause and restarting with lower
doses 268]. The usual concern with regards to treatment with IGF-1 has historically been hypoglycaemia,
however in the trials done so far it has not always occurred and had been lessened
when administered with meals—it was also usually connected to an appearance in a loss
of appetite 264], 286]. Another reported effect has been growth of lymphoid tissue (specially acromegaly
and tonsillar hypertrophy), renal enlargement (with normal kidney function) and in
rare cases facial coarsening of features and incremented hair growth 264].

Of further importance is the well-known effect of exercise in IGF-1 plasma level rise.
It has been determined for years that after only single bouts of moderate to high-intensity
total IGF-1 plasma levels rise up to ?10–30 % and peak only after 5–10 min after the
onset of the training 287]. This rapid increase has been argued to be due either to a release of IGF-1 stores
in tissues or due to a proteolytic cleavage of IGFBP-3 either in the plasma or in
tissue. Also it must be noted that during exercise pH drops dramatically and this
affects negatively to IGFBPs affinity for IGF-1/2. It has been somehow elucidated
that muscle tissue is the one movilising most of the IGF-1 from cells to plasma 288]. This muscle autocrine/paracrine/endocrine effect has various effects. Apart from
the obvious differentiation of satellite cells, which may account for many of the
beneficial actions of exercise training, without excluding the classical fat burning,
cardiovascular tonification, neuroendocrine system activation and chronic inflammation
grade lowering. These actions have been reviewed herein, but of special mention during
exercise is the recruitment of hippocampal neuroprogenitor cells and improved neuroglia
function 289], 290]. Moreover, it has been recently proven that IGF-1 central administration also induced
improvement in insulin sensitivity 291], meaning that this sudden rise after exercise or treatment could be of benefit not
only in the periphery, but acting centrally. It must be reminded that in our experience,
IGF-1 treatment does not only raise IGF-1 plasma levels, but also stimulates tissues
to synthesise their own IGF-1 by a mechanism still not uncovered, aiming these mechanisms
of central stimulation of peripheral improvement. These facts are in accordance with
the perspective of this review which suggests the central and nuclear role of IGF-1
in metabolism. Because exercise is the current best option when it comes to restore
metabolism and obesity inflammation problems (i.e. for MetS and T2D), and because
exercise is one of the most potent IGF-1 synthesis/freeing mechanisms, it seems logical
to correlate them and establish IGF-1 as a target for future options in the multifactorial
treatment for metabolic syndrome.

Inasmuch, these complications have caused fear and controversy among studies and opinions
towards the safety of the treatment even when there are findings supporting the need
for further investigation in the field of metabolic disorders. In our experience,
and the problem we often see in these trials, comes to the dosage administered which
are brutally over the target of restoring the physiological normal levels of the hormone.
All studies have been conducted using 80 µg–4 mg/kg/day and ranging in a wide spectrum
of lengths (weeks to months) and routes of administration. In all our previous murine
studies we have used short cycles (10–14 days) of very low subcutaneal doses (20 µg/kg/day)
after carefully assessing its circulating concentrations and using the minimum amount
that was sufficient to adequately restore physiological values of the molecule. Using
these concentration we have not yet perceived any of the reported adverse affects,
including hypoglycaemias, retinopathies or any others. We are conscious of the limitations
of the animal models 123], however in a clinical trial undertaken by this group to asses IGF-1 in liver function
under cirrhosis with these same dosage (20 µg/kg/day) no side effects were reported
and liver function greatly increased 108]. So, as this group has been experimenting over the last years, the problem with IGF-1
is a matter of dosage. Even more intriguing is the fact that none of the inclusion
criteria for the above-mentioned studies encompassed tumour markers to potentially
discard any ongoing or potential tumour process that, obviously, IGF-1 could accelerate.
IGF-1, as happens with other hormone therapies (thyroid hormone or insulin), should
never been used lightly without firstly assessing its deficiency (local, central or
systemic) and secondly by ascertaining that no tumours are near to be generated. Careful
monitoring during treatment should be carried out to guarantee a safe outcome.