.gif "IP Gear")
- Joined
- Jan 12, 2012
- Messages
- 122
I found this in an Endocrinology Journal and thought you guys might be interested in it.
HORMONES AND SPORT
Insulin, growth hormone and sport
P H Sonksen
Guy’s, King’s and St Thomas’ School of Medicine, St Thomas’ Hospital, London SE1 7EH, UK; Email: [email protected]
Abstract
This review examines some interesting ‘new’ histories of
insulin and reviews our current understanding of its
physiological actions and synergy with GH in the regulation
of metabolism and body composition. It reviews the
history of GH abuse that antedates by many years the
awareness of endocrinologists to its potent anabolic actions.
Promising methods for detection of GH abuse have been
developed but have yet to be sufficiently well validated to
be ready for introduction into competitive sport. So far,
there are two promising avenues for detecting GH abuse.
The first uses immunoassays that can distinguish the
isomers of pituitary-derived GH from the monomer of
recombinant human GH. The second works through
demonstrating circulating concentrations of one or more
GH-sensitive substances that exceed the extremes of
normal physiological variability. Both methods require
blood rather than urine samples. The first method has a
window of opportunity lasting about 24 h after an injection
and is most suitable for ‘out of competition’ testing.
The second method has reasonable sensitivity for as long as
2 weeks after the last injection of GH and is uninfluenced
by extreme exercise and suitable for post-competition
samples. This method has a greater sensitivity in men than
in women. The specificity of both methods seems acceptably
high but lawyers need to decide what level of
scientific probability is needed to obtain a conviction. Both
methods need further validation before implementation.
Research work carried out as part of the fight against
doping in sport has opened up a new and exciting area of
endocrinology.
Journal of Endocrinology 170, 13–25
Introduction
Doping in sport has a very long history going as far back as
the original Olympic Games and is mainly driven by the
desire to win at all cost. In order to maintain an environment
where cheats do not win, it is essential to develop
methods of combating abuse of performance-enhancing
drugs. To this end the International Olympic Committee
(IOC) Medical Commission and Sub-Commission
‘Doping and Biochemistry in Sport’ publish annually a list
of ‘banned substances’ and have developed a sophisticated
system for detecting drug abuse. Recent evidence
indicates that the protein hormones insulin and growth
hormone (GH) have now become a significant threat to
the level playing field essential in sport. The IOC was
swift to ban GH and insulin but no tests are available to
detect their abuse. As endogenous substances secreted in
bursts they pose particular problems in developing
satisfactory methods of detection. Our understanding of
how insulin and GH work in the regulation of metabolism
and body composition has evolved to demonstrate a
remarkable degree of synergy, something the athletes may
have discovered before the endocrinologists.
GH has been used as a drug of abuse in sport since the
early 1980s – 10 years before endocrinologists recognised
and understood its potency as an anabolic agent and as a
hormone regulating body composition in adults. Insulin
appears to have a shorter history as a ‘doping agent’ – it was
at the Winter Olympic Games in Nagano in 1998 when
a Russian medical officer enquired as to whether the use
of insulin was restricted to insulin-dependent diabetes.
This drew attention to its role as a potential performanceenhancing
drug and the IOC were swift to act and
immediately placed it on its list of banned substances.
Subsequent evidence from a needle-exchange programme
in the UK (R T Dawson, personal communication) has
confirmed its widespread use in body building and other
sports although it remains unclear exactly how it is used.
This review presents a brief but important (and not very
well-known) history of the physiology of insulin that is
essential for understanding the way in which it is used as
a performance-enhancing agent. It points out how, somewhat
paradoxically, the advances of ‘modern science’,
through inappropriate extrapolation from in vitro to in vivo,
confused the thinking and teaching to hide the truth
behind insulin action right up to the present day!
Journal of Endocrinology 170, 13–25
0022–0795/01/0170–013 Society for Endocrinology Printed in Great Britain
Online version via Society for Endocrinology | Home
It attempts to ‘set the record straight’ about our current
understanding of how insulin and GH interact as anabolic
agents, about why athletes use them and outlines a way
forward in detecting their abuse. It says nothing about
anabolic steroids that are still major drugs of abuse and may
well be synergistic with the effects of insulin and GH.
Insulin physiology
Sir Edward Schafer was Professor of Physiology in
Edinburgh when he published in 1916 a wonderful book
called The Endocrine Organs. The book is based on a
series of lectures he delivered at Stanford University in
California in 1913 (Schafer 1916). As well as containing a
wealth of interesting insights into the early days of
endocrinology, this book is most notable for the fact that it
was the first time that the then hypothetical hormone
insulin was named (8 years before it was discovered). What
is even more remarkable, he predicted the formation of
insulin from ‘pro-insulin’ 54 years before it was actually
discovered!
Schafer was a contemporary of Baylis and Starling – two
eminent academic rivals from University College in
London. Shortly before Schafer delivered his lectures to his
American audiences, Baylis and Starling had isolated,
characterised and published about Secretin, the first
‘hormone’ (a term coined by them to describe a substance
produced in one part of the body, carried by the blood
stream and acting elsewhere in the body) to be isolated.
Schafer questioned the use of the word ‘hormone’ and
proposed two alternative names:
Autacoids – excitatory substances
Chalones – inhibitory substances
He went on to describe how ‘his’ new hypothetical
hormone ‘insuline’ exhibited properties that resembled
both autacoids and chalones and that the chalonic or ‘inhibitory’
actions were physiologically the most important. It
was, he proposed, lack of this chalonic (inhibitory) action of
insulin that led to a failure to store glucose in the liver with
the net result that the liver overproduced glucose and
glucose accumulated in the circulation, and this led to the
hyperglycaemia that is characteristic of diabetes. This was
indeed advanced thinking.
The ‘black ages’ of endocrinology followed early in vitro
experiments in the 1950s that showed insulin to be
capable of stimulating glucose uptake into bits of rat
muscle and fat. Before long the biochemists had extrapolated
from these experiments to conclude (wrongly) that
the hyperglycaemia of diabetes was due to a ‘damming
back’ of glucose in the blood stream as a result of a failure
of glucose to enter cells as a direct consequence of insulin
deficiency. The concept of glucose uptake into muscle
being ‘insulin dependent’ was born, and has been prevalent
in textbooks and teaching up to the present day, even
though it was shown to be rubbish 25 years ago.
The truth is that insulin acts exactly as Schafer had
predicted – it acts as both an autacoid and a chalone.
Through stimulating the translocation of ‘Glut 4’ glucose
transporters from the cytoplasm of muscle and adipose
tissue to the cell membrane it increases the rate of
glucose uptake to values greater than the uptake that takes
place in the basal state without insulin. This is most easily
shown in isolated adipocytes from young rats and is
illustrated in Fig. 1 (Thomas et al. 1979). What the 1950s
biochemists failed to take notice of was the considerable
uptake of glucose that takes place in all tissues even when
insulin is absent. We now know that there is a sufficient
population of glucose transporters in all cell membranes at
all times to ensure enough glucose uptake to satisfy the
cell’s respiration, even in the absence of insulin. Insulin
can and does increase the number of these transporters in
some cells but glucose uptake is never truly insulin dependent
– in fact, even in uncontrolled diabetic hyperglycaemia,
whole body glucose uptake is inevitably increased (unless
there is severe ketosis). Even under conditions of extreme
ketoacidosis there is no significant membrane barrier to
glucose uptake – the block occurs ‘lower down’ in the
metabolic pathway where the excess of ketones competitively
blocks the metabolites of glucose entering the Krebs
cycle. Under these conditions, glucose is freely transported
into the cell but the pathway of metabolism is effectively
blocked at the entry point to the Krebs cycle by the excess
of metabolites arising from fat and protein breakdown. As
a result of this competitive block at the entry point to the
Krebs cycle, intracellular glucose metabolites increase
‘damming back’ throughout the glycolytic pathway, leading
to accumulation of free intracellular glucose and
inhibiting initial glucose phosphorylation. As a result,
Figure 1 Insulin exhibits both inhibitory (chalonic) and excitatory
(autacoid) actions via the same receptor. In these experiments
carried out on rat adipose tissue, in vitro insulin simultaneously
inhibits lipolysis (the release of glycerol from stored triglyceride)
and stimulates lipogenesis (formation of stored triglyceride from
glucose). Thus its anabolic action is due to two mechanisms
working synergistically.
14 P H SONKSEN · Insulin, GH and sport
Journal of Endocrinology (2001) 170, 13–25 Society for Endocrinology | Home
Much of the ‘free’ intracellular glucose transported into the
cell is transported back out of the cell into the extracellular
fluid. Thus under conditions of ketoacidosis, glucose
metabolism (but not glucose uptake) is impaired as a direct
consequence of the metabolism of fat – the ‘glucose–fatty
acid’ cycle (Randle et al. 1965).
In Fig. 1 it can be seen that simultaneously with
insulin’s autacoid effect in stimulating lipogenesis it also
exhibits a chalonic effect in inhibiting glycerol release. It is
this inhibitory effect on lipolysis (and also glycolysis,
gluconeogenesis, ketogenesis and proteolysis) that accounts
for most of insulin’s physiological effects in vivo in man. It
is also this inhibitory effect that is mainly responsible for
insulin’s net anabolic actions.
It was the introduction of dynamic tracer studies that
enabled us to resolve the misunderstanding about how
insulin acts in vivo in man. Through infusing glucose (and
other substrates) labelled with either radioactive or stable
isotopes it is easily possible to measure accurately the rates
of glucose production (‘rate of appearance’ – Ra) and rates of
glucose utilisation (‘rate of disappearance’ – Rd) of glucose
in the circulating blood. When these techniques were
applied in people with uncontrolled diabetes it was found
that the fasting hyperglycaemia was associated with rates of
glucose appearance that were increased several fold above
normal. Somewhat unexpectedly, it was also found that
fasting glucose uptake was also increased. This is inevitable,
since the fasting hyperglycaemia in diabetes is
sustained and there is a ‘dynamic steady state’ where
Ra=Rd. Thus both Ra and Rd are elevated. This finding
clashed with the dogma obtained from extrapolating from
in vitro experiments that had become embedded in the
literature – namely that glucose uptake was ‘insulin
dependent’ and was reduced in states of insulin deficiency.
Despite a wealth of evidence confirming these early
studies, this ‘new concept’ of insulin action remains
unrecognised by the majority of teachers of physiology and
biochemistry.
A remarkable example of how difficult it
sometimes is to reverse dogma even when it has been
proven to be wrong (Sonksen & Sonksen 2000).
The facts are that in diabetes the fasting blood glucose is
a very good measure of the severity of insulin deficiency.
There is a linear correlation between the fasting blood
glucose and the rate of hepatic glucose production (Ra) and
thus with the rate of glucose disappearance (Rd). Since, in
diabetes, the fasting blood glucose exceeds the renal
threshold, not all glucose leaving the circulation is actually
being metabolised. By collecting the urine and quantifying
the urinary glucose losses it is easy to measure the true rate of
glucose utilisation and the rate of urinary glucose loss. Glycosuria
can account for as much as 30% of glucose turnover but even
under these conditions, after correcting whole body glucose
turnover for urinary glucose losses, tissue glucose utilisation
is increased compared with normal. Thus insulin is NOT
needed for glucose uptake and utilisation in man – glucose
uptake is NOT insulin dependent.
When insulin is administered to people with diabetes
who are fasting, blood glucose concentration falls. It is
generally assumed that this is because insulin increases
glucose uptake into tissues, particularly muscle. In fact this
is NOT the case and is another error arising from
extrapolating from in vitro rat data. It has been shown quite
unequivocally that insulin at concentrations that are within
the normal physiological range lowers blood glucose
through inhibiting hepatic glucose production (Ra) without
stimulating peripheral glucose uptake (Brown et al.
1978). As hepatic glucose output is ‘switched off’ by the
chalonic action of insulin, glucose concentration falls and
glucose uptake actually decreases. Contrary to most textbooks
and previous teaching, glucose uptake is therefore
actually increased in uncontrolled diabetes and decreased by
insulin administration! The explanation for this is that
because, even in the face of insulin deficiency, there are
plenty of glucose transporters in the cell membranes. The
factor determining glucose uptake under these conditions
is the concentration gradient across the cell membrane;
this is highest in uncontrolled diabetes and falls as insulin
lowers blood glucose concentration primarily (at physiological
insulin concentrations) through reducing hepatic
glucose production.
When insulin is given to patients with uncontrolled
diabetes it switches off a number of metabolic processes
(lipolysis, proteolysis, ketogenesis and gluconeogenesis) by
a similar chalonic action. The result is that free fatty acid
(FFA) concentrations fall effectively to zero within
minutes and ketogenesis inevitably stops through lack of
substrate. It takes a while for the ketones to clear from the
circulation, as the ‘body load’ is massive as they are water
and fat soluble and distribute within body water and body
fat. Since both ketones and FFA compete with glucose as
energy substrate at the point of entry of substrates into the
Krebs cycle, glucose metabolism increases inevitably as
FFA and ketone levels fall (despite the concomitant fall in
plasma glucose concentration). Thus insulin increases
glucose metabolism more through reducing FFA and
ketone levels than it does through recruiting more glucose
transporters into the muscle cell membrane.
In fact, insulin does have a direct action to recruit more
glucose transporters into muscle cell membranes and the
effect of this is to facilitate glucose uptake – this is reflected
as an increase in the metabolic clearance rate (MCR) of
glucose. The MCR measured with tracer technology is a
very important physiological measurement. It is defined as
‘the amount of blood irreversibly cleared of glucose in unit
time’. Expressed normally as ml/kg per min it is a nonlinear
function of blood glucose concentration (increasing
as glucose concentration falls) and is highly sensitive to
insulin (increasing with increasing insulin levels). It can
readily be measured relatively non-invasively in vivo using
non-radioactive tracers (or radioactive tracers but their use
has been superseded by stable isotopes) and is of great
importance in understanding substrate utilisation and its
Insulin, GH and sport · P H SONKSEN 15
Society for Endocrinology | Home Journal of Endocrinology (2001) 170, 13–25
control in vivo.
Although glucose is used here for demonstration
purposes, the process appears to be generic for all
polar (water-soluble) substrates, as ‘transporters’ are the
mechanism by which they are transported across the highly
non-polar (lipid) cell membranes.
Figure 2 illustrates our model of the role of substrate
transporters in facilitating water-soluble substrate entry
into cells and it shows the ways in which we believe
insulin exerts its normal control on glucose metabolism
(Boroujerdi et al. 1995). There is a direct chalonic action
inhibiting glucose production by the liver while simultaneously
insulin exerts an autacoid action stimulating
transporter translocation into the cell membrane and,
through that mechanism, increasing glucose uptake at any
given blood glucose concentration.
Experiments in normal subjects using hyperglycaemic
and hyperinsulinaemic ‘clamps’ have shown quantitatively
the importance of both glucose and insulin concentrations
in determining glucose uptake. Some studies illustrating
these points are shown in Fig. 3. In these experiments,
subjects were studied in the overnight-fasted state with
fasting insulin averaging 18 mU/l and on two other
occasions when they were infused with insulin at rates that
resulted in mean plasma insulin concentrations of 80 and
150 mU/l. They were also studied at the same insulin
concentrations but with plasma glucose increased and
maintained at a steady level by an exogenous glucose
infusion. Four glucose concentrations ranging from 5 to
10 mmol/l were studied with insulin levels maintained at
normal fasting values. During the insulin infusions, subjects
were studied at three glucose concentrations spanning
the same range. Using tracer methodology the authors
were able to calculate Ra, Rd and MCR at each glucose
and insulin concentration (Gottesman et al. 1982). The
data from Gottesman et al. (1982) have been fitted to
the model of glucose metabolism shown in Fig. 2. The
important points of note are as follows.
1. Total glucose uptake (Rd) is a non-linear function of
blood glucose concentration. Initially, uptake increases
as blood glucose concentration rises but plateaus at
higher glucose concentration. Although detectable
within the range of glucose concentrations studied, it
is made more obvious through extrapolation to higher
Figure 2 The entry of a water-soluble substrate such as glucose across an impermeable lipid bi-layer into a
cell requires a specific transport mechanism. These protein carriers are known as ‘transporters’. In the case
of glucose there are at least six types and they tend to be tissue-specific. In the case of muscle the
transporter is called ‘Glut 4’. It is normally present in excess in the cell membrane even in the absence of
insulin and is not rate limiting for glucose entry into the cell. Glucose transport into the cell is mainly
determined by the concentration gradient between the extracellular fluid and the intracellular ‘free’ glucose.
Normally, ‘free’ glucose is very low inside the cell as it is immediately phosphorylated. In uncontrolled
diabetes, particularly where there is a high concentration of FFA and ketones, glycolysis is inhibited,
phosphorylation of ‘free’ glucose stops and intracellular ‘free’ glucose rises. Insulin recruits more transporters
into the cell membrane from an intracellular pool. This increases the rate of glucose entry for a given glucose
concentration and this is reflected in vivo by an increase in the MCR of glucose. Thus MCR is an in vivo
measure of substrate transporter activity.
16 P H SONKSEN · Insulin, GH and sport
Journal of Endocrinology (2001) 170, 13–25 Society for Endocrinology | Home
These high glucose values are unobtainable in normal
subjects with existing technology. The shape of the
curve suggests simple ‘saturation’ kinetics obeying
Michaelis–Menten laws.
2. Glucose MCR falls with increasing plasma glucose
concentration irrespective of the ambient plasma
insulin concentration. This is in keeping with saturation
of the glucose transporter system as plasma
glucose rises.
3. MCR increases with increasing plasma insulin
concentration, irrespective of the ambient plasma
glucose concentration. This is in keeping with translocation
of more glucose transporters into the cell
membrane under the influence of increasing insulin
concentrations.
4. The parallel nature of the plots shown in Fig. 3C
(which is, in fact, a Scatchard plot of the data)
indicates that increasing insulin concentrations are
associated with increasing number of ‘receptors’ – in
this case, glucose transporters. There is no sign of a
change in ‘affinity’ of the transporters under the
influence of insulin, just the number present to
facilitate glucose entry into cells.
Is insulin a performance-enhancing drug?
Thus from our understanding of insulin physiology we
can see different ways in which insulin might be a
performance-enhancing agent.
1. Through facilitating glucose entry into cells in
amounts greater than needed for cellular respiration
it will stimulate glycogen formation. Thus hyperinsulinaemic
clamps will both increase muscle glycogen
concentrations prior to events and in the recovery
phase after events. Since performance in many events
is known to be a function of muscle glycogen stores,
Figure 3 The data used in this illustration were obtained from healthy normal subjects using a series of euglycaemic and
hyperglycaemic clamps at basal or increased insulin concentrations. Rd1: insulin independent glucose uptake. The data have been
fitted to the generic model shown in Fig. 2 (see text for details).
Insulin, GH and sport · P H SONKSEN 17
Society for Endocrinology | Home Journal of Endocrinology (2001) 170, 13–25
‘bulking up’ these stores will most probably enhance
performance. There is no documental proof that this
technique is being used but informed ‘street talk’
indicates that it is not uncommon.
2. Through use of similar hyperinsulinaemic clamps
post-event and during training, it is likely that
recovery and stamina will be improved.
3. ‘Street talk’ indicates that insulin is also being used in
a more haphazard way, particularly to increase muscle
bulk in body builders, weight lifters and power lifters.
This use is allegedly by regular injections of shortacting
insulin together with high carbohydrate diets.
Through this therapeutic regime it is almost certainly
possible to increase muscle bulk and performance not
only through increasing muscle glycogen stores on a
‘chronic’ basis but also by increasing muscle bulk
through inhibition of muscle protein breakdown. Just
as insulin has a chalonic action in inhibiting glucose
breakdown in muscle glycogen, it also has an equally
important chalonic action in inhibiting protein breakdown.
Indeed, the evidence now indicates that insulin
does NOT stimulate protein synthesis directly (this
process is under the control of GH and insulin-like
growth factor-I (IGF-I)). It has long been known that
insulin-treated patients with diabetes have an increase
in lean body mass when compared with matched
controls (Sinha et al. 1996).
Taken together, all these points support the concern
shown by the Russian medical officer in Nagano and the
immediate response of the IOC to ban the use of insulin in
those without diabetes.
GH and IGF-I
GH was written up as a potent performance-enhancing
anabolic agent in The Underground Steroid Handbook first
published in California in the early 1980s. The first
scientific studies demonstrating a clear regulatory role for
GH in adults was, however, only published in the peerreviewed
medical literature in 1989 (Jorgensen et al. 1989,
Salomon et al. 1989).
It had been alleged that many elite
athletes had been abusing GH for many years and indeed
several had confessed to having done so. The most
eminent of these being Ben Johnson who, after losing his
Gold Medal after testing positive for anabolic steroids at
the Seoul Olympic games, admitted during subsequent
investigation to using GH over many years (in combination
with anabolic steroids).
Although there are still no proper scientific studies proving GH to be performance enhancing in normal subjects (such studies would most
likely be considered unethical and would be unlikely to
receive research funding), few doubt this ability. GH has
now been shown to have a very important role in
regulating body composition in adult humans and also in
other species.
In cattle, GH is known as a ‘partitioning
agent’ – it specifically diverts calories in food towards
protein synthesis and away from fat synthesis. Animals
made transgenic for GH have greatly increased lean tissues
and reduced fat. Similar changes in body composition are
seen in humans with acromegaly. On the other hand,
GH-deficient (GHD) adults have reduced lean body mass
and increased fat mass, particularly central abdominal
fat mass. Recombinant GH given in physiological
‘replacement’ doses to adults with GHD results in
remarkable changes in body composition with, on average,
a 5 kg increase in lean body mass within the first month
(Salomon et al. 1989) and a comparable loss of 5 kg of
fat. The fat loss is particularly from the intra-abdominal
region where fat accumulates in the GHD state. In parallel
with these changes in body composition, the subnormal
exercise performance and strength of adults with GHD are
returned to normal (Cuneo et al. 1991a,b).
Evidence about the abuse of GH in sport is largely
anecdotal and circumstantial since, although banned by
the IOC, there is as yet no recognised test for detecting its
abuse. Perhaps the strongest circumstantial evidence for
GH abuse comes from the losses admitted by the pharmaceutical
industry of recombinant human GH (rhGH) from
the production line, the distribution networks and the
wholesale and retail outlets. There has also been a documented
case of the re-sale to an athlete of a medicinal
supply of rhGH by the mother of a GHD child.
Supplies of rhGH have been found in a team car during
the Tour de France, in the personal possession of a Chinese
swimmer in the World Championship in Perth, Australia
and in the personal baggage of a national team trainer
entering Australia for the Sydney 2000 Olympic Games.
Six months before these recent Sydney Olympic Games
there was a carefully targeted burglary on a wholesale
pharmacy in Sydney. A massive supply of rhGH was stolen
– nothing else was touched. It seems that none of this was
ever recovered as described in the following report.
Olympic Jitters at Power Drug Theft
By Deborah Cameron (16 February 2000)
‘‘The theft of a huge quantity of an undetectable
bodybuilding drug from a Sydney importer has raised
serious concerns among sports officials and doctors about
whether Australia’s elite athletes are ‘clean’. With just six
months to go until the Olympics, and as some sports
prepare for selection trials, the timing of the theft of 1575
(multidose) vials of human GH (hGH) is seen as highly
significant’’.
Perhaps of even greater concern is the knowledge that
GH is readily available from gymnasia and other sporting
establishments and that it is being used by school-children.
Supplies of pituitary-derived GH are still in circulation
indistinguishable from rhGH. There are several wellrecognised
sources of pituitary-derived GH appearing in
the world market and it is highly likely that some of these
18 P H SONKSEN · Insulin, GH and sport
Journal of Endocrinology (2001) 170, 13–25 Society for Endocrinology | Home
batches will be contaminated with Creutzfeld–Jacob prion.
There were press reports recently of the arrest of a Russian
who was found with a large jar with >1000 pickled human
pituitary glands in his apartment in Moscow.
GH physiology
GH is a polypeptide hormone secreted by the pituitary
gland. The predominant form has a molecular weight of
22 kDa and has a half-life in the plasma of between 15 and
20 min after secretion or intravenous injection. After
subcutaneous or intramuscular injection, blood concentrations
of GH reach a peak between 1 and 3 h after
injection and fall to undetectable levels after 24 h. As it is
a protein hormone it has to be administered by injection as
it is completely digested to its constituent amino acids
when administered by mouth. The circulating GH is
cleared from the blood stream through receptor-mediated
degradation, predominantly in the liver and kidney. The
liver and kidney internalise the GH–receptor complex
and completely degrade it to its basic amino acids. Only
minute quantities of GH appear in the urine and the
pattern of urinary excretion has been shown to be too low
and variable to be considered of any value in developing a
test of GH abuse.
GH receptors are present on all cells in the body. One
GH molecule binds to two receptors and leads to
dimerisation of the receptors. This dimerisation process is
essential for initiation of intracellular signalling. Analogues
of GH that prevent dimerisation are inhibitors of GH. The
pituitary gland secretes GH in bursts and the major stimuli
to GH secretion in man are sleep, exercise and stress. The
sleep-related burst of GH secretion occurs most consistently
during the phase of deep slow-wave sleep most
commonly occurring during the early hours of sleep. In
conditions where sleep pattern is disturbed, GH secretion
is impaired; the introduction of effective therapy for the
sleep disturbance (such as continuous positive airways
pressure for obstructive sleep apnoea) restores GH
secretion to normal. Hypnotics that reduce the period of
slow-wave sleep impair GH secretion while drugs that
enhance slow-wave sleep enhance GH secretion (Van
Cauter & Copinschi 2000).
GH secretion reaches its maximum around late teenage
life and falls progressively thereafter. The total amount of
GH secreted over 24 h in normal adults over the age of 65
is, in the majority of cases, overlapping with people with
organic GHD secondary to pituitary pathology or its
treatment (Toogood et al. 1996). There is thus evidence of
the development of functional GHD with increasing age –
the so-called ‘Somatopause’. The majority of middleaged
and elderly normal subjects may be considered
incompletely GHD.
GH stimulates many metabolic processes in all cells but
one of its best-known actions is the generation of IGF-I
(and its binding proteins). GH stimulates IGF-I gene
expression in all tissues. In most tissues, this IGF-I has local
‘autocrine’ and ‘paracrine’ actions but the liver actively
secretes IGF-I (and its binding proteins) into the circulation.
Until recently it was thought that this ‘hormonally’
secreted IGF-I produced by the liver was responsible for
many of GH’s in vivo actions. Recent data from hepatospecific
IGF-I knock-out mice have shed serious doubt
on this, since their growth and metabolism appear to be
quite normal despite very low circulating IGF-I levels.
Circulating IGF-I should now be considered more as a ‘marker’
of GH action on the liver than as the mechanism by which GH
exerts its effects. Hepatic IGF-I production is regulated by
factors other than GH, most notably nutritional and
thyroid status. Undernutrition, such as is seen in anorexia
nervosa and poorly controlled Type 1 diabetes are both
associated with low plasma IGF-I and high GH secretion.
It appears that in these circumstances it is the portal insulin
status that is one of the key factors regulating hepatic GH
receptor expression and hence hepatic sensitivity to GH.
IGF-I has many actions that resemble GH and at one
stage it was thought that most, if not all, of GH’s actions
were mediated through IGF-I. Recent experience with
the hepato-specific IGF-I knock-out mouse has shown
this may be an oversimplification of a complex system and
the exact roles of GH and IGF-I still have to be defined.
There are GH receptors on all cells in the body and it
appears that GH exerts effects on most, if not all, of these
cells. There are literally hundreds, if not thousands, of
GH-dependent ‘markers’ produced under the influence of
GH. IGF-I just happens to be the best known of these but
since the majority of circulating IGF-I comes from the
liver, we should now think of IGF-I more as a marker of
GH action on the liver rather than the ‘second messenger’
of GH action.
GH administration leads to the production of a whole
series of markers of its action that appear in the circulation
and, as we shall see later, these can be used as a way of
detecting GH abuse.
What are the metabolic effects of GH that make it
attractive as a drug of abuse?
GH’s major action is to stimulate protein synthesis. It is at
least as powerful as testosterone in this effect and, as they
both operate through distinct pathways, their individual
effects are additive or possibly even synergistic. In addition
to stimulating protein synthesis, GH simultaneously
mobilises fat by a direct lipolytic action. Together, these
two effects are responsible for the ‘partitioning’ action of
GH whereby it diverts nutritional calories to protein
synthesis, possibly through using the energy derived from
its lipolytic action. It most likely stimulates protein synthesis
through mobilisation of amino acid transporters in
a manner analogous to insulin and glucose transporters
Insulin, GH and sport · P H SONKSEN 19
Society for Endocrinology | Home Journal of Endocrinology (2001) 170, 13–25
(Fig. 2). This is reflected in vivo by an increase in amino
acid MCR and the process can be explained quantitatively
by a model of structure similar to that shown in Fig. 2.
IGF-I also acts directly to stimulate protein synthesis but it
has a weaker lipolytic action. GH, IGF-I and insulin thus
act in concert to stimulate protein synthesis. Using
Schafer’s conceptual terminology, GH and IGF-I act in
an autacoid manner to stimulate protein synthesis while
insulin acts in its characteristic chalonic manner to inhibit
protein breakdown. Thus they are synergistic in their
powerful anabolic action (Fig. 4). Insulin is essential for the
anabolic action of GH. GH administration in the absence
of adequate insulin reserves (as during fasting or in Type 1
diabetes) is in fact catabolic and its lipolytic and ketogenic
properties can induce diabetic ketoacidosis. Thus GH and
insulin are closely linked in normal physiology and it is of
great interest to see that athletes have discovered ways
in which this normal physiological dependence can be
exploited to enhance performance.
Why do athletes ‘dope’ with GH?
There are several reasons for this. It is well known from
surveys amongst elite athletes that they are prepared to
take risks in order to win medals. A survey carried out
in 1995 by a prominent sporting magazine in the USA
polled a series of elite athletes on a number of questions
(Bamberger & Yaeger 1997).
One of these was:
You are offered a banned performance-enhancing
substance with two guarantees:
you will not be caught
you will win
Would you take the substance?’
Answer: Yes, 195; No, 3.
Another was:
‘You are offered a performance-enhancing substance
that comes with two guarantees:
you will not be caught
you will win every competition you enter for the next
5 years and then you will die from the side-effects of the
substance
Would you take it?
Answer: Yes >50%.
Even though this may not be a truly representative
survey it is clear that the driving force to win is very strong
and is behind much of the drug abuse in sport but there are
other motives.
Injuries are common in most sports and athletes believe
that the prevention or mitigation of these is possible
Figure 4 This diagram illustrates our current understanding about the synergistic action between insulin, IGF-I and GH in regulating protein (P) synthesis. Without insulin, GH loses much (if not all) of its anabolic action. GH and IGF-I stimulate protein synthesis directly, while insulin is anabolic through inhibiting protein breakdown. The anabolic action of both GH and IGF-I appears to be mediated through induction of amino acid (Aa) transporters in the cell membrane. This is reflected in vivo by an increase in amino acid MCR. It is not yet clear how much of IGF-I’s action is through locally generated IGF-I (‘autocrine’ and ‘paracrine’) or through circulating IGF-I that is largely derived from the liver.
20 P H SONKSEN · Insulin, GH and sport
Journal of Endocrinology (2001) 170, 13–25 Society for Endocrinology | Home
Through judicious use of nutritional supplements and more
potent anabolic agents such as steroids and GH. Indeed the
original description of GH’s action in the 1983 edition of
the Underground Steroid Handbook stated that GH strengthened
tendons such that damage to them by weight and
power lifters was much reduced. There is also a view that
GH may prevent stress fractures and speeds the healing
process – experimental evidence from animal studies
indicates that they may well be right!
Perhaps the main reason why GH is such a threat to fair
play is, of course, the fact that it is a very potent anabolic
agent, readily available in unlimited quantities, pretty safe
and completely undetectable. The risks associated with
taking it are, in the short-term, minimal. The benefits are
potentially very considerable. Although the numbers
actually abusing it are likely to be quite small, it does pose
a very real threat to the majority of elite athletes who do
not abuse it or other performance-enhancing drugs. It is
therefore of great importance that not only should it be
banned but that there should be a test to catch and
disqualify those who use it to cheat.
How can we detect abuse with GH?
Unlike the majority of synthetic anabolic steroids, GH is
an ‘endogenous substance’ and indistinguishable from the
naturally occurring hormone. This makes it particularly
difficult to detect when used as a drug of abuse. The
pituitary gland secretes predominantly a 22 kDa isomer of
GH but there are other minor products of gene transcription
and post-transcription modification. The most prominent
of these is a 20 kDa isomer that is normally present
in the circulation at concentrations averaging around 10%
of the 22 kDa isomer.
It is possible to distinguish between endogenous GH
secretion and exogenous rhGH (but not pituitary-derived
GH) administration. If, for example, a blood sample is
obtained during a peak of endogenous GH secretion, then
10% of the immunoassayable GH should be 20 kDa GH.
Using assays based on a combination of relatively nonspecific
polyclonal antibodies and specific assays using two
monoclonal antibodies, each recognising specific epitopes
on the 22 kDa and 20 kDa molecules, Strasburger and
colleagues have developed a promising method of detecting
rhGH abuse. They have shown that the absence of the
20 kDa isomer when 22 kDa GH is present in substantial
amounts can be used as an indicator of exogenous rhGH
administration and can thus form the basis of a test of GH
abuse (Wu et al. 1999). This is one way of tackling the
detection of rhGH abuse but it will not, of course, detect
administration of exogenous pituitary-derived GH. Its
‘window of opportunity’ for detecting rhGH abuse will be
for less than 24 h after a dose of rhGH since, after then,
none will be detectable in the circulation. It could thus
best be the basis of an ‘out of competition’ test when an
unannounced blood sample may well be taken within 24 h
of the last rhGH injection. More work is needed to
validate this approach however.
Another approach has used the many substances in the
body regulated by GH as ‘markers’ of GH action as the
basis of a test for GH abuse. The principle of this approach
is that, when given in supra-physiological doses, GH
over-stimulates a number of processes normally regulated
by GH. Thus it should be possible to detect GH abuse if
these GH-dependent markers are present in amounts that
greatly exceed the normal physiological range. Clearly this
approach will not be able to distinguish the athlete with a
GH-secreting pituitary tumour but it should be able to
detect normal subjects administered supra-physiological
amounts of GH (people with acromegaly have been elite
athletes).
This was the main thrust of the international collaborative
research project ‘GH-2000’ jointly funded by the
European Union and the IOC, contract number
BMH4CT950678.
Since exercise is a powerful stimulus to GH secretion
and most samples for ‘doping control’ are taken immediately
post-competition, a pilot ‘washout’ study investigated
the effects of 1 week’s rhGH administration and acute
exercise on a wide range of substances influenced by GH.
The aim of this study was to determine a subgroup of
GH-dependent ‘markers’ that were sensitive to rhGH
administration but relatively insensitive to the acute effects
of exercise (Wallace et al. 1999, 2000) .
The results indicated that four markers produced by the
liver (IGF-I, IGF-binding protein-2 (IGFBP-2), IGFBP-3
and ALS) and four produced from collagen and bone
(osteocalcin, procollagen type III (P-III-P), type I collagen
telopeptide and C-terminal propeptide of type I collagen)
were suitable to take forward into other studies. These
markers and their points of origin are illustrated in
Fig. 5.
A ‘double-blind placebo-controlled’ study of 1 month’s
rhGH administration at two doses to more than 100
healthy volunteers was carried out in four countries. The
eight markers were measured on all the blood samples
before, during and for 3 months after the rhGH or placebo
administration. Analysis of a subset of the data indicated
that the best discrimination between active treatment and
placebo was obtained using two of the markers – IGF-I
and P-III-P. Using these two markers it was possible to
obtain complete separation between the active treatment
and placebo, whereas with either one of the markers there
was a degree of overlap. Further statistical developments
indicated that increased sensitivity and specificity
could be obtained by combining more of the markers
in the analysis. Analysis of the urine samples showed
much weaker discriminating power in separating active
treatment from placebo.
There was a need to construct a reference range of these
markers in elite athletes from different sporting disciplines.
Insulin, GH and sport · P H SONKSEN 21
Society for Endocrinology | Home Journal of Endocrinology (2001) 170, 13–25
Figure 5 A summary of the potential markers thought to be most useful in developing a test of GH abuse (see text for details). Dpd and Pyd are urinary metabolites of collagen markers.
22 P H SONKSEN · Insulin, GH and sport
Journal of Endocrinology (2001) 170, 13–25 Society for Endocrinology | Home
In order to do this, teams of research workers from the
project obtained permission to recruit volunteer elite
athletes from a variety of major national and international
sporting competitions. During the course of the project,
over 800 elite athletes gave explicit written informed
consent to a blood sample and for demographic data to
be collected immediately post-competition. On as many
occasions as possible a sample of urine was also obtained.
All eight markers were measured on each blood sample.
Analysis showed considerable apparent differences between
sports but when the results were adjusted for the effects
of age most of these apparent differences disappeared.
The results for all the GH-dependent markers showed a
very clear age-related fall. Although this is well recognised
for GH and IGF-I it was seen in all the markers. There
were gender and ethnic differences in some of the markers
but these were relatively minor compared with the agerelated
effects. The gender effect is well illustrated in
Fig. 6 which shows the effects of age on plasma IGF-I in
male and female elite athletes. Although there is a signifi-
cant difference between men and women it is obvious that
this is not a major one.
Of importance in developing a test for detecting GH
abuse is the stability of the markers over time and in
response to exercise, training, injury etc. A subset of those
who volunteered for the cross-sectional study used to
construct the reference range volunteered to provide
samples on other occasions during the year. Many of them
also agreed to undertake a simulated extreme exercise
event under ‘laboratory’ conditions. The results from this
‘longitudinal study’, taken with the information obtained
from the placebo-treated group during the double-blind
study, showed remarkable stability in the blood concentration
of the markers over time. It appears that the blood
concentration of the markers is most likely genetically
determined and relatively uninfluenced by day-to-day
environmental factors.
The effects of injury were examined by looking at the
levels of the markers in a group of volunteers from a
number of sports injury clinics in the UK. Although the
number of subjects studied was too small to form any
definitive conclusions, changes were seen after fractures
and injuries but, interestingly, although the concentration
of the collagen/bone markers rose in most cases, those of
Figure 6 Effects of age on serum IGF-I levels (mcg/l) in 800 elite athletes. Samples were taken within 2 h of the end of
competition. There is an exponential fall in IGF-I levels in both male and female elite athletes as in the normal population.
Although there are minor differences statistically between men and women these are of marginal importance. Red crosses:
female; blue crosses: male.
Insulin, GH and sport · P H SONKSEN 23
Society for Endocrinology | Home Journal of Endocrinology (2001) 170, 13–25
The hepatic IGF-I-related markers tended to fall. The
discriminant function based on the combination of IGF-I
and P-III-P remained remarkably stable.
Sensitivity and specificity of a test for GH abuse
Using simple discriminant function techniques and adjusting
for age, it is possible to obtain reasonable sensitivity in
detecting abuse of GH during its administration and for as
long as 2 weeks after it has stopped. The exact level of
sensitivity depends on the level of specificity that a court of
law would demand to uphold a prosecution. Since courts
of law do not normally deal in terms of ‘scientific
probabilities’ there is, at present, an important gap
between science and the law. In order to have a valid test
for GH abuse, this gap must be closed. Lawyers are
beginning to realise that the scientific issues of probability
must be incorporated into their process but they have
quite a long way to go before this issue is resolved
satisfactorily. Suffice it to say, a test based on the results of
the GH-2000 project has a sensitivity better than 90% of
picking up a man taking GH with a probability of less than
1:10 000 of being wrong. This is high degree of scientific
certainty but whether or not it is sufficient for a court of
law (or the Court of Arbitration in Sport (CAS)) will not
be clear until the system is actually tested with a case. It is
significant that one member of the CAS present at the
workshop, after seeing the data presented by GH-2000,
felt sufficiently convinced to remark that he would be
prepared to take an athlete caught cheating by these
criteria to court ‘. . . so long as he was a white European
male . . .’. This remark highlighted two of the many points
that still require further research: firstly that the large
majority of volunteers studies in the GH-2000 project
were white Europeans and secondly, the gender difference
in sensitivity to rhGH is quite marked. Although differences
in the criteria for detecting GH abuse between men
and women are minimal, the actual sensitivity of the test in
detecting GH abuse in women is much lower. On the
other hand, the specificity of the test is not lower in
women. So why he was less certain about taking a woman
to court whose values exceeded the reference range? We
need to do further research on the mechanisms behind this
gender difference.
When will we have a test for GH abuse?
The project was targeted at providing a test for the Sydney
Olympic Games. The final report was handed in to the
European Union and the IOC on 21 January 1999,
20 months before the Sydney Olympic Games. The results
were presented to the IOC medical commission at the
time of the World Doping Conference in Lausanne in
February 1999. The IOC convened a special workshop in
Rome in March 1999 to review the results. To this they
invited a number of outside experts to review the data
critically. It was at this workshop that a distinguished CAS
lawyer made the remarks already quoted and the conclusions
of the workshop were very much that the project had
done a remarkably good job. GH-2000 had shown that the
development of a test was feasible and exactly how it
should be done. There was, however, the need for some
further research to complete the rigorous scientific data
that are needed for disqualification of an athlete caught
cheating and upholding the decision in the face of an
appeal to CAS. There is also a need to publish all the
results in peer-reviewed scientific journals in order to
provide quality assurance on the research. This has taken a
great deal of time as the amount of data collected was
enormous and although many papers have been published
or are ‘in press’ it will be another year or two before that
is achieved.
GH-2000 was immensely successful – it brought
together a team of European endocrinologists and scientists
to work together in what turns out to be an exciting new
area – sport – and was able to show that what at the outset
had appeared to be an impossible task was, in fact,
achievable. It has shown the importance of the GH axis in
fitness and sport and has revealed that GH has important
regulatory control over an even greater range of processes
than we had previously understood. It has shown the
effects of ageing are not only on the GH–IGF-I axis but on
all areas where GH acts and it has shown that even elite
athletes who exercise at rates unachievable by most of us
still show a decline in GH production as they age. Is this
the reason why GH abuse is thought to be so prevalent in
sport? Does GH prolong active life? The answer to this is
not known but the data indicate that it is a valid question
and support the need for much further research work in
the area of GH and ageing and the possible therapeutic
role of GH in preventing loss of lean tissues with ageing.
Acknowledgements
This paper is based on an invited lecture given to the
British Endocrine Society meeting at Imperial College
London in November 2000. The results are those of the
GH-2000 project, not all these results have yet been
published. I would like to thank all those who participated
in the project including our volunteer athletes and all the
sports organisations who helped us recruit and study
our volunteers. Finally, I would like to acknowledge the
wonderful statistical support we had from Dr Eryl Bassett
and Professor Philip Brown from the University of Kent.
References
Bamberger M & Yaeger D 1997 Over the edge. Sports Illustrated 14
62–70.
24 P H SONKSEN · Insulin, GH and sport
Journal of Endocrinology (2001) 170, 13–25 Society for Endocrinology | Home
Boroujerdi MA, Umpleby AM, Jones RH & Sonksen PH 1995 A
simulation model for glucose kinetics and estimates of glucose
utilization rate in type 1 diabetic patients. American Journal of
Physiology 268 E766–E774.
Brown P, Tompkins C, Juul S & Sonksen PH 1978 Mechanism of
action of insulin in diabetic patients: a dose-related effect on
glucose production and utilisation. British Medical Journal 1
1239–1242.
Cuneo R, Salomon F, Wiles C, Hesp R & Sonksen PH 1991a
Growth hormone treatment in growth hormone-deficient adults.
I. Effects on muscle mass and strength. Journal of Applied Physiology
70 688–694.
Cuneo R, Salomon F, Wiles C, Hesp R & Sonksen PH 1991b
Growth hormone treatment in growth hormone-deficient adults.
II. Effects on exercise performance. Journal of Applied Physiology 70
695–700.
Gottesman I, Mandarino L, Verdonk C, RizzaR&Gerich J 1982
Insulin increases the maximum velocity of glucose uptake without
altering the Michaelis constant in man. Journal of Clinical
Investigation 70 1310–1314.
Jorgensen JO, Pedersen SA, Thuesen L, Jorgensen J, IngemannHansen
T, Skakkebaek NE & Christiansen JS 1989 Beneficial
effects of growth hormone in GH-deficient adults. Lancet i (8649)
1221–1225.
Randle PJ, Garland PB, Newsholme EA & Hales CN 1965 The
glucose fatty acid cycle in obesity and maturity onset diabetes
mellitus. Annals of the New York Academy of Sciences 131 324–333.
Salomon F, Cuneo R, Hesp R & Sonksen PH 1989 The effects of
treatment with recombinant human growth hormone on body
composition and metabolism in adults with growth hormone
deficiency. New England Journal of Medicine 321 1797–1803.
Schafer EA 1916 The Endocrine Organs – An Introduction to the Study of
Internal Secretions, pp 4–7 and 128–130. London: Longmans, Green
and Co.
Sinha A, Formica C, Tsalamandris C, Panagiotopoulos S, Hendrich E,
DeLuise M, Seeman E & Jerums G 1996 Effects of insulin on
body composition in patients with insulin-dependent and
non-insulin-dependent diabetes. Diabetic Medicine 13 40–46.
Sonksen PH & Sonksen J 2000 Insulin: understanding its action in
health and disease. British Journal of Anaesthesia 85 69–79.
Thomas SH, Wisher M, Brandenburg D & Sonksen PH 1979 Insulin
action on adipocytes, evidence that the anti-lipolytic and lipogenic
effects of insulin are mediated by the same receptor. Biochemical
Journal 184 355–360.
Toogood AA, O’Neill PA & Shalet SM 1996 Beyond the
somatopause: growth hormone deficiency in adults over the age
of 60 years. Journal of Clinical Endocrinology and Metabolism 81
460–465.
Van Cauter E & Copinschi G 2000 Interrelationships between
growth hormones and sleep. Growth Hormone and IGF Research
10 (Suppl B) S57–S62.
Wallace JD, Cuneo RC, Baxter R, Orskov H, Keay N, Pentecost C,
Dall R, Rosen T, Jorgensen JO, Cittadini A, Longobardi S, Sacca
L, Christiansen JS, Bengtsson B-A & Sonksen PH 1999 Responses
of the growth hormone (GH) and insulin-like growth factor axis to
exercise, GH administration, and GH withdrawal in trained adult
males: a potential test for GH abuse in sport. Journal of Clinical
Endocrinology and Metabolism 84 3591–3601.
Wallace JD, Cuneo RC, Lundberg PA, Rosen T, Jorgensen JOL,
Longobardi S, Keay N, Sacca L, Christiansen JS, Bengtsson B-A &
Sonksen PH 2000 Responses of markers of bone and collagen
turnover to exercise, growth hormone (GH) administration and GH
withdrawal in trained adult males. Journal of Clinical Endocrinology
and Metabolism 85 124–133.
Wu Z, Bidlingmaier M, Dall R & Strasburger CJ 1999 Detection of
doping with human growth hormone. Lancet 353 895.
Received 18 January 2001
Accepted 10 April 2001
Insulin, GH and sport · P H SONKSEN 25
Society for Endocrinology | Home Journal of Endocrinology (2001) 170, 13–25
HORMONES AND SPORT
Insulin, growth hormone and sport
P H Sonksen
Guy’s, King’s and St Thomas’ School of Medicine, St Thomas’ Hospital, London SE1 7EH, UK; Email: [email protected]
Abstract
This review examines some interesting ‘new’ histories of
insulin and reviews our current understanding of its
physiological actions and synergy with GH in the regulation
of metabolism and body composition. It reviews the
history of GH abuse that antedates by many years the
awareness of endocrinologists to its potent anabolic actions.
Promising methods for detection of GH abuse have been
developed but have yet to be sufficiently well validated to
be ready for introduction into competitive sport. So far,
there are two promising avenues for detecting GH abuse.
The first uses immunoassays that can distinguish the
isomers of pituitary-derived GH from the monomer of
recombinant human GH. The second works through
demonstrating circulating concentrations of one or more
GH-sensitive substances that exceed the extremes of
normal physiological variability. Both methods require
blood rather than urine samples. The first method has a
window of opportunity lasting about 24 h after an injection
and is most suitable for ‘out of competition’ testing.
The second method has reasonable sensitivity for as long as
2 weeks after the last injection of GH and is uninfluenced
by extreme exercise and suitable for post-competition
samples. This method has a greater sensitivity in men than
in women. The specificity of both methods seems acceptably
high but lawyers need to decide what level of
scientific probability is needed to obtain a conviction. Both
methods need further validation before implementation.
Research work carried out as part of the fight against
doping in sport has opened up a new and exciting area of
endocrinology.
Journal of Endocrinology 170, 13–25
Introduction
Doping in sport has a very long history going as far back as
the original Olympic Games and is mainly driven by the
desire to win at all cost. In order to maintain an environment
where cheats do not win, it is essential to develop
methods of combating abuse of performance-enhancing
drugs. To this end the International Olympic Committee
(IOC) Medical Commission and Sub-Commission
‘Doping and Biochemistry in Sport’ publish annually a list
of ‘banned substances’ and have developed a sophisticated
system for detecting drug abuse. Recent evidence
indicates that the protein hormones insulin and growth
hormone (GH) have now become a significant threat to
the level playing field essential in sport. The IOC was
swift to ban GH and insulin but no tests are available to
detect their abuse. As endogenous substances secreted in
bursts they pose particular problems in developing
satisfactory methods of detection. Our understanding of
how insulin and GH work in the regulation of metabolism
and body composition has evolved to demonstrate a
remarkable degree of synergy, something the athletes may
have discovered before the endocrinologists.
GH has been used as a drug of abuse in sport since the
early 1980s – 10 years before endocrinologists recognised
and understood its potency as an anabolic agent and as a
hormone regulating body composition in adults. Insulin
appears to have a shorter history as a ‘doping agent’ – it was
at the Winter Olympic Games in Nagano in 1998 when
a Russian medical officer enquired as to whether the use
of insulin was restricted to insulin-dependent diabetes.
This drew attention to its role as a potential performanceenhancing
drug and the IOC were swift to act and
immediately placed it on its list of banned substances.
Subsequent evidence from a needle-exchange programme
in the UK (R T Dawson, personal communication) has
confirmed its widespread use in body building and other
sports although it remains unclear exactly how it is used.
This review presents a brief but important (and not very
well-known) history of the physiology of insulin that is
essential for understanding the way in which it is used as
a performance-enhancing agent. It points out how, somewhat
paradoxically, the advances of ‘modern science’,
through inappropriate extrapolation from in vitro to in vivo,
confused the thinking and teaching to hide the truth
behind insulin action right up to the present day!
Journal of Endocrinology 170, 13–25
0022–0795/01/0170–013 Society for Endocrinology Printed in Great Britain
Online version via Society for Endocrinology | Home
It attempts to ‘set the record straight’ about our current
understanding of how insulin and GH interact as anabolic
agents, about why athletes use them and outlines a way
forward in detecting their abuse. It says nothing about
anabolic steroids that are still major drugs of abuse and may
well be synergistic with the effects of insulin and GH.
Insulin physiology
Sir Edward Schafer was Professor of Physiology in
Edinburgh when he published in 1916 a wonderful book
called The Endocrine Organs. The book is based on a
series of lectures he delivered at Stanford University in
California in 1913 (Schafer 1916). As well as containing a
wealth of interesting insights into the early days of
endocrinology, this book is most notable for the fact that it
was the first time that the then hypothetical hormone
insulin was named (8 years before it was discovered). What
is even more remarkable, he predicted the formation of
insulin from ‘pro-insulin’ 54 years before it was actually
discovered!
Schafer was a contemporary of Baylis and Starling – two
eminent academic rivals from University College in
London. Shortly before Schafer delivered his lectures to his
American audiences, Baylis and Starling had isolated,
characterised and published about Secretin, the first
‘hormone’ (a term coined by them to describe a substance
produced in one part of the body, carried by the blood
stream and acting elsewhere in the body) to be isolated.
Schafer questioned the use of the word ‘hormone’ and
proposed two alternative names:
Autacoids – excitatory substances
Chalones – inhibitory substances
He went on to describe how ‘his’ new hypothetical
hormone ‘insuline’ exhibited properties that resembled
both autacoids and chalones and that the chalonic or ‘inhibitory’
actions were physiologically the most important. It
was, he proposed, lack of this chalonic (inhibitory) action of
insulin that led to a failure to store glucose in the liver with
the net result that the liver overproduced glucose and
glucose accumulated in the circulation, and this led to the
hyperglycaemia that is characteristic of diabetes. This was
indeed advanced thinking.
The ‘black ages’ of endocrinology followed early in vitro
experiments in the 1950s that showed insulin to be
capable of stimulating glucose uptake into bits of rat
muscle and fat. Before long the biochemists had extrapolated
from these experiments to conclude (wrongly) that
the hyperglycaemia of diabetes was due to a ‘damming
back’ of glucose in the blood stream as a result of a failure
of glucose to enter cells as a direct consequence of insulin
deficiency. The concept of glucose uptake into muscle
being ‘insulin dependent’ was born, and has been prevalent
in textbooks and teaching up to the present day, even
though it was shown to be rubbish 25 years ago.
The truth is that insulin acts exactly as Schafer had
predicted – it acts as both an autacoid and a chalone.
Through stimulating the translocation of ‘Glut 4’ glucose
transporters from the cytoplasm of muscle and adipose
tissue to the cell membrane it increases the rate of
glucose uptake to values greater than the uptake that takes
place in the basal state without insulin. This is most easily
shown in isolated adipocytes from young rats and is
illustrated in Fig. 1 (Thomas et al. 1979). What the 1950s
biochemists failed to take notice of was the considerable
uptake of glucose that takes place in all tissues even when
insulin is absent. We now know that there is a sufficient
population of glucose transporters in all cell membranes at
all times to ensure enough glucose uptake to satisfy the
cell’s respiration, even in the absence of insulin. Insulin
can and does increase the number of these transporters in
some cells but glucose uptake is never truly insulin dependent
– in fact, even in uncontrolled diabetic hyperglycaemia,
whole body glucose uptake is inevitably increased (unless
there is severe ketosis). Even under conditions of extreme
ketoacidosis there is no significant membrane barrier to
glucose uptake – the block occurs ‘lower down’ in the
metabolic pathway where the excess of ketones competitively
blocks the metabolites of glucose entering the Krebs
cycle. Under these conditions, glucose is freely transported
into the cell but the pathway of metabolism is effectively
blocked at the entry point to the Krebs cycle by the excess
of metabolites arising from fat and protein breakdown. As
a result of this competitive block at the entry point to the
Krebs cycle, intracellular glucose metabolites increase
‘damming back’ throughout the glycolytic pathway, leading
to accumulation of free intracellular glucose and
inhibiting initial glucose phosphorylation. As a result,
Figure 1 Insulin exhibits both inhibitory (chalonic) and excitatory
(autacoid) actions via the same receptor. In these experiments
carried out on rat adipose tissue, in vitro insulin simultaneously
inhibits lipolysis (the release of glycerol from stored triglyceride)
and stimulates lipogenesis (formation of stored triglyceride from
glucose). Thus its anabolic action is due to two mechanisms
working synergistically.
14 P H SONKSEN · Insulin, GH and sport
Journal of Endocrinology (2001) 170, 13–25 Society for Endocrinology | Home
Much of the ‘free’ intracellular glucose transported into the
cell is transported back out of the cell into the extracellular
fluid. Thus under conditions of ketoacidosis, glucose
metabolism (but not glucose uptake) is impaired as a direct
consequence of the metabolism of fat – the ‘glucose–fatty
acid’ cycle (Randle et al. 1965).
In Fig. 1 it can be seen that simultaneously with
insulin’s autacoid effect in stimulating lipogenesis it also
exhibits a chalonic effect in inhibiting glycerol release. It is
this inhibitory effect on lipolysis (and also glycolysis,
gluconeogenesis, ketogenesis and proteolysis) that accounts
for most of insulin’s physiological effects in vivo in man. It
is also this inhibitory effect that is mainly responsible for
insulin’s net anabolic actions.
It was the introduction of dynamic tracer studies that
enabled us to resolve the misunderstanding about how
insulin acts in vivo in man. Through infusing glucose (and
other substrates) labelled with either radioactive or stable
isotopes it is easily possible to measure accurately the rates
of glucose production (‘rate of appearance’ – Ra) and rates of
glucose utilisation (‘rate of disappearance’ – Rd) of glucose
in the circulating blood. When these techniques were
applied in people with uncontrolled diabetes it was found
that the fasting hyperglycaemia was associated with rates of
glucose appearance that were increased several fold above
normal. Somewhat unexpectedly, it was also found that
fasting glucose uptake was also increased. This is inevitable,
since the fasting hyperglycaemia in diabetes is
sustained and there is a ‘dynamic steady state’ where
Ra=Rd. Thus both Ra and Rd are elevated. This finding
clashed with the dogma obtained from extrapolating from
in vitro experiments that had become embedded in the
literature – namely that glucose uptake was ‘insulin
dependent’ and was reduced in states of insulin deficiency.
Despite a wealth of evidence confirming these early
studies, this ‘new concept’ of insulin action remains
unrecognised by the majority of teachers of physiology and
biochemistry.
A remarkable example of how difficult it
sometimes is to reverse dogma even when it has been
proven to be wrong (Sonksen & Sonksen 2000).
The facts are that in diabetes the fasting blood glucose is
a very good measure of the severity of insulin deficiency.
There is a linear correlation between the fasting blood
glucose and the rate of hepatic glucose production (Ra) and
thus with the rate of glucose disappearance (Rd). Since, in
diabetes, the fasting blood glucose exceeds the renal
threshold, not all glucose leaving the circulation is actually
being metabolised. By collecting the urine and quantifying
the urinary glucose losses it is easy to measure the true rate of
glucose utilisation and the rate of urinary glucose loss. Glycosuria
can account for as much as 30% of glucose turnover but even
under these conditions, after correcting whole body glucose
turnover for urinary glucose losses, tissue glucose utilisation
is increased compared with normal. Thus insulin is NOT
needed for glucose uptake and utilisation in man – glucose
uptake is NOT insulin dependent.
When insulin is administered to people with diabetes
who are fasting, blood glucose concentration falls. It is
generally assumed that this is because insulin increases
glucose uptake into tissues, particularly muscle. In fact this
is NOT the case and is another error arising from
extrapolating from in vitro rat data. It has been shown quite
unequivocally that insulin at concentrations that are within
the normal physiological range lowers blood glucose
through inhibiting hepatic glucose production (Ra) without
stimulating peripheral glucose uptake (Brown et al.
1978). As hepatic glucose output is ‘switched off’ by the
chalonic action of insulin, glucose concentration falls and
glucose uptake actually decreases. Contrary to most textbooks
and previous teaching, glucose uptake is therefore
actually increased in uncontrolled diabetes and decreased by
insulin administration! The explanation for this is that
because, even in the face of insulin deficiency, there are
plenty of glucose transporters in the cell membranes. The
factor determining glucose uptake under these conditions
is the concentration gradient across the cell membrane;
this is highest in uncontrolled diabetes and falls as insulin
lowers blood glucose concentration primarily (at physiological
insulin concentrations) through reducing hepatic
glucose production.
When insulin is given to patients with uncontrolled
diabetes it switches off a number of metabolic processes
(lipolysis, proteolysis, ketogenesis and gluconeogenesis) by
a similar chalonic action. The result is that free fatty acid
(FFA) concentrations fall effectively to zero within
minutes and ketogenesis inevitably stops through lack of
substrate. It takes a while for the ketones to clear from the
circulation, as the ‘body load’ is massive as they are water
and fat soluble and distribute within body water and body
fat. Since both ketones and FFA compete with glucose as
energy substrate at the point of entry of substrates into the
Krebs cycle, glucose metabolism increases inevitably as
FFA and ketone levels fall (despite the concomitant fall in
plasma glucose concentration). Thus insulin increases
glucose metabolism more through reducing FFA and
ketone levels than it does through recruiting more glucose
transporters into the muscle cell membrane.
In fact, insulin does have a direct action to recruit more
glucose transporters into muscle cell membranes and the
effect of this is to facilitate glucose uptake – this is reflected
as an increase in the metabolic clearance rate (MCR) of
glucose. The MCR measured with tracer technology is a
very important physiological measurement. It is defined as
‘the amount of blood irreversibly cleared of glucose in unit
time’. Expressed normally as ml/kg per min it is a nonlinear
function of blood glucose concentration (increasing
as glucose concentration falls) and is highly sensitive to
insulin (increasing with increasing insulin levels). It can
readily be measured relatively non-invasively in vivo using
non-radioactive tracers (or radioactive tracers but their use
has been superseded by stable isotopes) and is of great
importance in understanding substrate utilisation and its
Insulin, GH and sport · P H SONKSEN 15
Society for Endocrinology | Home Journal of Endocrinology (2001) 170, 13–25
control in vivo.
Although glucose is used here for demonstration
purposes, the process appears to be generic for all
polar (water-soluble) substrates, as ‘transporters’ are the
mechanism by which they are transported across the highly
non-polar (lipid) cell membranes.
Figure 2 illustrates our model of the role of substrate
transporters in facilitating water-soluble substrate entry
into cells and it shows the ways in which we believe
insulin exerts its normal control on glucose metabolism
(Boroujerdi et al. 1995). There is a direct chalonic action
inhibiting glucose production by the liver while simultaneously
insulin exerts an autacoid action stimulating
transporter translocation into the cell membrane and,
through that mechanism, increasing glucose uptake at any
given blood glucose concentration.
Experiments in normal subjects using hyperglycaemic
and hyperinsulinaemic ‘clamps’ have shown quantitatively
the importance of both glucose and insulin concentrations
in determining glucose uptake. Some studies illustrating
these points are shown in Fig. 3. In these experiments,
subjects were studied in the overnight-fasted state with
fasting insulin averaging 18 mU/l and on two other
occasions when they were infused with insulin at rates that
resulted in mean plasma insulin concentrations of 80 and
150 mU/l. They were also studied at the same insulin
concentrations but with plasma glucose increased and
maintained at a steady level by an exogenous glucose
infusion. Four glucose concentrations ranging from 5 to
10 mmol/l were studied with insulin levels maintained at
normal fasting values. During the insulin infusions, subjects
were studied at three glucose concentrations spanning
the same range. Using tracer methodology the authors
were able to calculate Ra, Rd and MCR at each glucose
and insulin concentration (Gottesman et al. 1982). The
data from Gottesman et al. (1982) have been fitted to
the model of glucose metabolism shown in Fig. 2. The
important points of note are as follows.
1. Total glucose uptake (Rd) is a non-linear function of
blood glucose concentration. Initially, uptake increases
as blood glucose concentration rises but plateaus at
higher glucose concentration. Although detectable
within the range of glucose concentrations studied, it
is made more obvious through extrapolation to higher
Figure 2 The entry of a water-soluble substrate such as glucose across an impermeable lipid bi-layer into a
cell requires a specific transport mechanism. These protein carriers are known as ‘transporters’. In the case
of glucose there are at least six types and they tend to be tissue-specific. In the case of muscle the
transporter is called ‘Glut 4’. It is normally present in excess in the cell membrane even in the absence of
insulin and is not rate limiting for glucose entry into the cell. Glucose transport into the cell is mainly
determined by the concentration gradient between the extracellular fluid and the intracellular ‘free’ glucose.
Normally, ‘free’ glucose is very low inside the cell as it is immediately phosphorylated. In uncontrolled
diabetes, particularly where there is a high concentration of FFA and ketones, glycolysis is inhibited,
phosphorylation of ‘free’ glucose stops and intracellular ‘free’ glucose rises. Insulin recruits more transporters
into the cell membrane from an intracellular pool. This increases the rate of glucose entry for a given glucose
concentration and this is reflected in vivo by an increase in the MCR of glucose. Thus MCR is an in vivo
measure of substrate transporter activity.
16 P H SONKSEN · Insulin, GH and sport
Journal of Endocrinology (2001) 170, 13–25 Society for Endocrinology | Home
These high glucose values are unobtainable in normal
subjects with existing technology. The shape of the
curve suggests simple ‘saturation’ kinetics obeying
Michaelis–Menten laws.
2. Glucose MCR falls with increasing plasma glucose
concentration irrespective of the ambient plasma
insulin concentration. This is in keeping with saturation
of the glucose transporter system as plasma
glucose rises.
3. MCR increases with increasing plasma insulin
concentration, irrespective of the ambient plasma
glucose concentration. This is in keeping with translocation
of more glucose transporters into the cell
membrane under the influence of increasing insulin
concentrations.
4. The parallel nature of the plots shown in Fig. 3C
(which is, in fact, a Scatchard plot of the data)
indicates that increasing insulin concentrations are
associated with increasing number of ‘receptors’ – in
this case, glucose transporters. There is no sign of a
change in ‘affinity’ of the transporters under the
influence of insulin, just the number present to
facilitate glucose entry into cells.
Is insulin a performance-enhancing drug?
Thus from our understanding of insulin physiology we
can see different ways in which insulin might be a
performance-enhancing agent.
1. Through facilitating glucose entry into cells in
amounts greater than needed for cellular respiration
it will stimulate glycogen formation. Thus hyperinsulinaemic
clamps will both increase muscle glycogen
concentrations prior to events and in the recovery
phase after events. Since performance in many events
is known to be a function of muscle glycogen stores,
Figure 3 The data used in this illustration were obtained from healthy normal subjects using a series of euglycaemic and
hyperglycaemic clamps at basal or increased insulin concentrations. Rd1: insulin independent glucose uptake. The data have been
fitted to the generic model shown in Fig. 2 (see text for details).
Insulin, GH and sport · P H SONKSEN 17
Society for Endocrinology | Home Journal of Endocrinology (2001) 170, 13–25
‘bulking up’ these stores will most probably enhance
performance. There is no documental proof that this
technique is being used but informed ‘street talk’
indicates that it is not uncommon.
2. Through use of similar hyperinsulinaemic clamps
post-event and during training, it is likely that
recovery and stamina will be improved.
3. ‘Street talk’ indicates that insulin is also being used in
a more haphazard way, particularly to increase muscle
bulk in body builders, weight lifters and power lifters.
This use is allegedly by regular injections of shortacting
insulin together with high carbohydrate diets.
Through this therapeutic regime it is almost certainly
possible to increase muscle bulk and performance not
only through increasing muscle glycogen stores on a
‘chronic’ basis but also by increasing muscle bulk
through inhibition of muscle protein breakdown. Just
as insulin has a chalonic action in inhibiting glucose
breakdown in muscle glycogen, it also has an equally
important chalonic action in inhibiting protein breakdown.
Indeed, the evidence now indicates that insulin
does NOT stimulate protein synthesis directly (this
process is under the control of GH and insulin-like
growth factor-I (IGF-I)). It has long been known that
insulin-treated patients with diabetes have an increase
in lean body mass when compared with matched
controls (Sinha et al. 1996).
Taken together, all these points support the concern
shown by the Russian medical officer in Nagano and the
immediate response of the IOC to ban the use of insulin in
those without diabetes.
GH and IGF-I
GH was written up as a potent performance-enhancing
anabolic agent in The Underground Steroid Handbook first
published in California in the early 1980s. The first
scientific studies demonstrating a clear regulatory role for
GH in adults was, however, only published in the peerreviewed
medical literature in 1989 (Jorgensen et al. 1989,
Salomon et al. 1989).
It had been alleged that many elite
athletes had been abusing GH for many years and indeed
several had confessed to having done so. The most
eminent of these being Ben Johnson who, after losing his
Gold Medal after testing positive for anabolic steroids at
the Seoul Olympic games, admitted during subsequent
investigation to using GH over many years (in combination
with anabolic steroids).
Although there are still no proper scientific studies proving GH to be performance enhancing in normal subjects (such studies would most
likely be considered unethical and would be unlikely to
receive research funding), few doubt this ability. GH has
now been shown to have a very important role in
regulating body composition in adult humans and also in
other species.
In cattle, GH is known as a ‘partitioning
agent’ – it specifically diverts calories in food towards
protein synthesis and away from fat synthesis. Animals
made transgenic for GH have greatly increased lean tissues
and reduced fat. Similar changes in body composition are
seen in humans with acromegaly. On the other hand,
GH-deficient (GHD) adults have reduced lean body mass
and increased fat mass, particularly central abdominal
fat mass. Recombinant GH given in physiological
‘replacement’ doses to adults with GHD results in
remarkable changes in body composition with, on average,
a 5 kg increase in lean body mass within the first month
(Salomon et al. 1989) and a comparable loss of 5 kg of
fat. The fat loss is particularly from the intra-abdominal
region where fat accumulates in the GHD state. In parallel
with these changes in body composition, the subnormal
exercise performance and strength of adults with GHD are
returned to normal (Cuneo et al. 1991a,b).
Evidence about the abuse of GH in sport is largely
anecdotal and circumstantial since, although banned by
the IOC, there is as yet no recognised test for detecting its
abuse. Perhaps the strongest circumstantial evidence for
GH abuse comes from the losses admitted by the pharmaceutical
industry of recombinant human GH (rhGH) from
the production line, the distribution networks and the
wholesale and retail outlets. There has also been a documented
case of the re-sale to an athlete of a medicinal
supply of rhGH by the mother of a GHD child.
Supplies of rhGH have been found in a team car during
the Tour de France, in the personal possession of a Chinese
swimmer in the World Championship in Perth, Australia
and in the personal baggage of a national team trainer
entering Australia for the Sydney 2000 Olympic Games.
Six months before these recent Sydney Olympic Games
there was a carefully targeted burglary on a wholesale
pharmacy in Sydney. A massive supply of rhGH was stolen
– nothing else was touched. It seems that none of this was
ever recovered as described in the following report.
Olympic Jitters at Power Drug Theft
By Deborah Cameron (16 February 2000)
‘‘The theft of a huge quantity of an undetectable
bodybuilding drug from a Sydney importer has raised
serious concerns among sports officials and doctors about
whether Australia’s elite athletes are ‘clean’. With just six
months to go until the Olympics, and as some sports
prepare for selection trials, the timing of the theft of 1575
(multidose) vials of human GH (hGH) is seen as highly
significant’’.
Perhaps of even greater concern is the knowledge that
GH is readily available from gymnasia and other sporting
establishments and that it is being used by school-children.
Supplies of pituitary-derived GH are still in circulation
indistinguishable from rhGH. There are several wellrecognised
sources of pituitary-derived GH appearing in
the world market and it is highly likely that some of these
18 P H SONKSEN · Insulin, GH and sport
Journal of Endocrinology (2001) 170, 13–25 Society for Endocrinology | Home
batches will be contaminated with Creutzfeld–Jacob prion.
There were press reports recently of the arrest of a Russian
who was found with a large jar with >1000 pickled human
pituitary glands in his apartment in Moscow.
GH physiology
GH is a polypeptide hormone secreted by the pituitary
gland. The predominant form has a molecular weight of
22 kDa and has a half-life in the plasma of between 15 and
20 min after secretion or intravenous injection. After
subcutaneous or intramuscular injection, blood concentrations
of GH reach a peak between 1 and 3 h after
injection and fall to undetectable levels after 24 h. As it is
a protein hormone it has to be administered by injection as
it is completely digested to its constituent amino acids
when administered by mouth. The circulating GH is
cleared from the blood stream through receptor-mediated
degradation, predominantly in the liver and kidney. The
liver and kidney internalise the GH–receptor complex
and completely degrade it to its basic amino acids. Only
minute quantities of GH appear in the urine and the
pattern of urinary excretion has been shown to be too low
and variable to be considered of any value in developing a
test of GH abuse.
GH receptors are present on all cells in the body. One
GH molecule binds to two receptors and leads to
dimerisation of the receptors. This dimerisation process is
essential for initiation of intracellular signalling. Analogues
of GH that prevent dimerisation are inhibitors of GH. The
pituitary gland secretes GH in bursts and the major stimuli
to GH secretion in man are sleep, exercise and stress. The
sleep-related burst of GH secretion occurs most consistently
during the phase of deep slow-wave sleep most
commonly occurring during the early hours of sleep. In
conditions where sleep pattern is disturbed, GH secretion
is impaired; the introduction of effective therapy for the
sleep disturbance (such as continuous positive airways
pressure for obstructive sleep apnoea) restores GH
secretion to normal. Hypnotics that reduce the period of
slow-wave sleep impair GH secretion while drugs that
enhance slow-wave sleep enhance GH secretion (Van
Cauter & Copinschi 2000).
GH secretion reaches its maximum around late teenage
life and falls progressively thereafter. The total amount of
GH secreted over 24 h in normal adults over the age of 65
is, in the majority of cases, overlapping with people with
organic GHD secondary to pituitary pathology or its
treatment (Toogood et al. 1996). There is thus evidence of
the development of functional GHD with increasing age –
the so-called ‘Somatopause’. The majority of middleaged
and elderly normal subjects may be considered
incompletely GHD.
GH stimulates many metabolic processes in all cells but
one of its best-known actions is the generation of IGF-I
(and its binding proteins). GH stimulates IGF-I gene
expression in all tissues. In most tissues, this IGF-I has local
‘autocrine’ and ‘paracrine’ actions but the liver actively
secretes IGF-I (and its binding proteins) into the circulation.
Until recently it was thought that this ‘hormonally’
secreted IGF-I produced by the liver was responsible for
many of GH’s in vivo actions. Recent data from hepatospecific
IGF-I knock-out mice have shed serious doubt
on this, since their growth and metabolism appear to be
quite normal despite very low circulating IGF-I levels.
Circulating IGF-I should now be considered more as a ‘marker’
of GH action on the liver than as the mechanism by which GH
exerts its effects. Hepatic IGF-I production is regulated by
factors other than GH, most notably nutritional and
thyroid status. Undernutrition, such as is seen in anorexia
nervosa and poorly controlled Type 1 diabetes are both
associated with low plasma IGF-I and high GH secretion.
It appears that in these circumstances it is the portal insulin
status that is one of the key factors regulating hepatic GH
receptor expression and hence hepatic sensitivity to GH.
IGF-I has many actions that resemble GH and at one
stage it was thought that most, if not all, of GH’s actions
were mediated through IGF-I. Recent experience with
the hepato-specific IGF-I knock-out mouse has shown
this may be an oversimplification of a complex system and
the exact roles of GH and IGF-I still have to be defined.
There are GH receptors on all cells in the body and it
appears that GH exerts effects on most, if not all, of these
cells. There are literally hundreds, if not thousands, of
GH-dependent ‘markers’ produced under the influence of
GH. IGF-I just happens to be the best known of these but
since the majority of circulating IGF-I comes from the
liver, we should now think of IGF-I more as a marker of
GH action on the liver rather than the ‘second messenger’
of GH action.
GH administration leads to the production of a whole
series of markers of its action that appear in the circulation
and, as we shall see later, these can be used as a way of
detecting GH abuse.
What are the metabolic effects of GH that make it
attractive as a drug of abuse?
GH’s major action is to stimulate protein synthesis. It is at
least as powerful as testosterone in this effect and, as they
both operate through distinct pathways, their individual
effects are additive or possibly even synergistic. In addition
to stimulating protein synthesis, GH simultaneously
mobilises fat by a direct lipolytic action. Together, these
two effects are responsible for the ‘partitioning’ action of
GH whereby it diverts nutritional calories to protein
synthesis, possibly through using the energy derived from
its lipolytic action. It most likely stimulates protein synthesis
through mobilisation of amino acid transporters in
a manner analogous to insulin and glucose transporters
Insulin, GH and sport · P H SONKSEN 19
Society for Endocrinology | Home Journal of Endocrinology (2001) 170, 13–25
(Fig. 2). This is reflected in vivo by an increase in amino
acid MCR and the process can be explained quantitatively
by a model of structure similar to that shown in Fig. 2.
IGF-I also acts directly to stimulate protein synthesis but it
has a weaker lipolytic action. GH, IGF-I and insulin thus
act in concert to stimulate protein synthesis. Using
Schafer’s conceptual terminology, GH and IGF-I act in
an autacoid manner to stimulate protein synthesis while
insulin acts in its characteristic chalonic manner to inhibit
protein breakdown. Thus they are synergistic in their
powerful anabolic action (Fig. 4). Insulin is essential for the
anabolic action of GH. GH administration in the absence
of adequate insulin reserves (as during fasting or in Type 1
diabetes) is in fact catabolic and its lipolytic and ketogenic
properties can induce diabetic ketoacidosis. Thus GH and
insulin are closely linked in normal physiology and it is of
great interest to see that athletes have discovered ways
in which this normal physiological dependence can be
exploited to enhance performance.
Why do athletes ‘dope’ with GH?
There are several reasons for this. It is well known from
surveys amongst elite athletes that they are prepared to
take risks in order to win medals. A survey carried out
in 1995 by a prominent sporting magazine in the USA
polled a series of elite athletes on a number of questions
(Bamberger & Yaeger 1997).
One of these was:
You are offered a banned performance-enhancing
substance with two guarantees:
you will not be caught
you will win
Would you take the substance?’
Answer: Yes, 195; No, 3.
Another was:
‘You are offered a performance-enhancing substance
that comes with two guarantees:
you will not be caught
you will win every competition you enter for the next
5 years and then you will die from the side-effects of the
substance
Would you take it?
Answer: Yes >50%.
Even though this may not be a truly representative
survey it is clear that the driving force to win is very strong
and is behind much of the drug abuse in sport but there are
other motives.
Injuries are common in most sports and athletes believe
that the prevention or mitigation of these is possible
Figure 4 This diagram illustrates our current understanding about the synergistic action between insulin, IGF-I and GH in regulating protein (P) synthesis. Without insulin, GH loses much (if not all) of its anabolic action. GH and IGF-I stimulate protein synthesis directly, while insulin is anabolic through inhibiting protein breakdown. The anabolic action of both GH and IGF-I appears to be mediated through induction of amino acid (Aa) transporters in the cell membrane. This is reflected in vivo by an increase in amino acid MCR. It is not yet clear how much of IGF-I’s action is through locally generated IGF-I (‘autocrine’ and ‘paracrine’) or through circulating IGF-I that is largely derived from the liver.
20 P H SONKSEN · Insulin, GH and sport
Journal of Endocrinology (2001) 170, 13–25 Society for Endocrinology | Home
Through judicious use of nutritional supplements and more
potent anabolic agents such as steroids and GH. Indeed the
original description of GH’s action in the 1983 edition of
the Underground Steroid Handbook stated that GH strengthened
tendons such that damage to them by weight and
power lifters was much reduced. There is also a view that
GH may prevent stress fractures and speeds the healing
process – experimental evidence from animal studies
indicates that they may well be right!
Perhaps the main reason why GH is such a threat to fair
play is, of course, the fact that it is a very potent anabolic
agent, readily available in unlimited quantities, pretty safe
and completely undetectable. The risks associated with
taking it are, in the short-term, minimal. The benefits are
potentially very considerable. Although the numbers
actually abusing it are likely to be quite small, it does pose
a very real threat to the majority of elite athletes who do
not abuse it or other performance-enhancing drugs. It is
therefore of great importance that not only should it be
banned but that there should be a test to catch and
disqualify those who use it to cheat.
How can we detect abuse with GH?
Unlike the majority of synthetic anabolic steroids, GH is
an ‘endogenous substance’ and indistinguishable from the
naturally occurring hormone. This makes it particularly
difficult to detect when used as a drug of abuse. The
pituitary gland secretes predominantly a 22 kDa isomer of
GH but there are other minor products of gene transcription
and post-transcription modification. The most prominent
of these is a 20 kDa isomer that is normally present
in the circulation at concentrations averaging around 10%
of the 22 kDa isomer.
It is possible to distinguish between endogenous GH
secretion and exogenous rhGH (but not pituitary-derived
GH) administration. If, for example, a blood sample is
obtained during a peak of endogenous GH secretion, then
10% of the immunoassayable GH should be 20 kDa GH.
Using assays based on a combination of relatively nonspecific
polyclonal antibodies and specific assays using two
monoclonal antibodies, each recognising specific epitopes
on the 22 kDa and 20 kDa molecules, Strasburger and
colleagues have developed a promising method of detecting
rhGH abuse. They have shown that the absence of the
20 kDa isomer when 22 kDa GH is present in substantial
amounts can be used as an indicator of exogenous rhGH
administration and can thus form the basis of a test of GH
abuse (Wu et al. 1999). This is one way of tackling the
detection of rhGH abuse but it will not, of course, detect
administration of exogenous pituitary-derived GH. Its
‘window of opportunity’ for detecting rhGH abuse will be
for less than 24 h after a dose of rhGH since, after then,
none will be detectable in the circulation. It could thus
best be the basis of an ‘out of competition’ test when an
unannounced blood sample may well be taken within 24 h
of the last rhGH injection. More work is needed to
validate this approach however.
Another approach has used the many substances in the
body regulated by GH as ‘markers’ of GH action as the
basis of a test for GH abuse. The principle of this approach
is that, when given in supra-physiological doses, GH
over-stimulates a number of processes normally regulated
by GH. Thus it should be possible to detect GH abuse if
these GH-dependent markers are present in amounts that
greatly exceed the normal physiological range. Clearly this
approach will not be able to distinguish the athlete with a
GH-secreting pituitary tumour but it should be able to
detect normal subjects administered supra-physiological
amounts of GH (people with acromegaly have been elite
athletes).
This was the main thrust of the international collaborative
research project ‘GH-2000’ jointly funded by the
European Union and the IOC, contract number
BMH4CT950678.
Since exercise is a powerful stimulus to GH secretion
and most samples for ‘doping control’ are taken immediately
post-competition, a pilot ‘washout’ study investigated
the effects of 1 week’s rhGH administration and acute
exercise on a wide range of substances influenced by GH.
The aim of this study was to determine a subgroup of
GH-dependent ‘markers’ that were sensitive to rhGH
administration but relatively insensitive to the acute effects
of exercise (Wallace et al. 1999, 2000) .
The results indicated that four markers produced by the
liver (IGF-I, IGF-binding protein-2 (IGFBP-2), IGFBP-3
and ALS) and four produced from collagen and bone
(osteocalcin, procollagen type III (P-III-P), type I collagen
telopeptide and C-terminal propeptide of type I collagen)
were suitable to take forward into other studies. These
markers and their points of origin are illustrated in
Fig. 5.
A ‘double-blind placebo-controlled’ study of 1 month’s
rhGH administration at two doses to more than 100
healthy volunteers was carried out in four countries. The
eight markers were measured on all the blood samples
before, during and for 3 months after the rhGH or placebo
administration. Analysis of a subset of the data indicated
that the best discrimination between active treatment and
placebo was obtained using two of the markers – IGF-I
and P-III-P. Using these two markers it was possible to
obtain complete separation between the active treatment
and placebo, whereas with either one of the markers there
was a degree of overlap. Further statistical developments
indicated that increased sensitivity and specificity
could be obtained by combining more of the markers
in the analysis. Analysis of the urine samples showed
much weaker discriminating power in separating active
treatment from placebo.
There was a need to construct a reference range of these
markers in elite athletes from different sporting disciplines.
Insulin, GH and sport · P H SONKSEN 21
Society for Endocrinology | Home Journal of Endocrinology (2001) 170, 13–25
Figure 5 A summary of the potential markers thought to be most useful in developing a test of GH abuse (see text for details). Dpd and Pyd are urinary metabolites of collagen markers.
22 P H SONKSEN · Insulin, GH and sport
Journal of Endocrinology (2001) 170, 13–25 Society for Endocrinology | Home
In order to do this, teams of research workers from the
project obtained permission to recruit volunteer elite
athletes from a variety of major national and international
sporting competitions. During the course of the project,
over 800 elite athletes gave explicit written informed
consent to a blood sample and for demographic data to
be collected immediately post-competition. On as many
occasions as possible a sample of urine was also obtained.
All eight markers were measured on each blood sample.
Analysis showed considerable apparent differences between
sports but when the results were adjusted for the effects
of age most of these apparent differences disappeared.
The results for all the GH-dependent markers showed a
very clear age-related fall. Although this is well recognised
for GH and IGF-I it was seen in all the markers. There
were gender and ethnic differences in some of the markers
but these were relatively minor compared with the agerelated
effects. The gender effect is well illustrated in
Fig. 6 which shows the effects of age on plasma IGF-I in
male and female elite athletes. Although there is a signifi-
cant difference between men and women it is obvious that
this is not a major one.
Of importance in developing a test for detecting GH
abuse is the stability of the markers over time and in
response to exercise, training, injury etc. A subset of those
who volunteered for the cross-sectional study used to
construct the reference range volunteered to provide
samples on other occasions during the year. Many of them
also agreed to undertake a simulated extreme exercise
event under ‘laboratory’ conditions. The results from this
‘longitudinal study’, taken with the information obtained
from the placebo-treated group during the double-blind
study, showed remarkable stability in the blood concentration
of the markers over time. It appears that the blood
concentration of the markers is most likely genetically
determined and relatively uninfluenced by day-to-day
environmental factors.
The effects of injury were examined by looking at the
levels of the markers in a group of volunteers from a
number of sports injury clinics in the UK. Although the
number of subjects studied was too small to form any
definitive conclusions, changes were seen after fractures
and injuries but, interestingly, although the concentration
of the collagen/bone markers rose in most cases, those of
Figure 6 Effects of age on serum IGF-I levels (mcg/l) in 800 elite athletes. Samples were taken within 2 h of the end of
competition. There is an exponential fall in IGF-I levels in both male and female elite athletes as in the normal population.
Although there are minor differences statistically between men and women these are of marginal importance. Red crosses:
female; blue crosses: male.
Insulin, GH and sport · P H SONKSEN 23
Society for Endocrinology | Home Journal of Endocrinology (2001) 170, 13–25
The hepatic IGF-I-related markers tended to fall. The
discriminant function based on the combination of IGF-I
and P-III-P remained remarkably stable.
Sensitivity and specificity of a test for GH abuse
Using simple discriminant function techniques and adjusting
for age, it is possible to obtain reasonable sensitivity in
detecting abuse of GH during its administration and for as
long as 2 weeks after it has stopped. The exact level of
sensitivity depends on the level of specificity that a court of
law would demand to uphold a prosecution. Since courts
of law do not normally deal in terms of ‘scientific
probabilities’ there is, at present, an important gap
between science and the law. In order to have a valid test
for GH abuse, this gap must be closed. Lawyers are
beginning to realise that the scientific issues of probability
must be incorporated into their process but they have
quite a long way to go before this issue is resolved
satisfactorily. Suffice it to say, a test based on the results of
the GH-2000 project has a sensitivity better than 90% of
picking up a man taking GH with a probability of less than
1:10 000 of being wrong. This is high degree of scientific
certainty but whether or not it is sufficient for a court of
law (or the Court of Arbitration in Sport (CAS)) will not
be clear until the system is actually tested with a case. It is
significant that one member of the CAS present at the
workshop, after seeing the data presented by GH-2000,
felt sufficiently convinced to remark that he would be
prepared to take an athlete caught cheating by these
criteria to court ‘. . . so long as he was a white European
male . . .’. This remark highlighted two of the many points
that still require further research: firstly that the large
majority of volunteers studies in the GH-2000 project
were white Europeans and secondly, the gender difference
in sensitivity to rhGH is quite marked. Although differences
in the criteria for detecting GH abuse between men
and women are minimal, the actual sensitivity of the test in
detecting GH abuse in women is much lower. On the
other hand, the specificity of the test is not lower in
women. So why he was less certain about taking a woman
to court whose values exceeded the reference range? We
need to do further research on the mechanisms behind this
gender difference.
When will we have a test for GH abuse?
The project was targeted at providing a test for the Sydney
Olympic Games. The final report was handed in to the
European Union and the IOC on 21 January 1999,
20 months before the Sydney Olympic Games. The results
were presented to the IOC medical commission at the
time of the World Doping Conference in Lausanne in
February 1999. The IOC convened a special workshop in
Rome in March 1999 to review the results. To this they
invited a number of outside experts to review the data
critically. It was at this workshop that a distinguished CAS
lawyer made the remarks already quoted and the conclusions
of the workshop were very much that the project had
done a remarkably good job. GH-2000 had shown that the
development of a test was feasible and exactly how it
should be done. There was, however, the need for some
further research to complete the rigorous scientific data
that are needed for disqualification of an athlete caught
cheating and upholding the decision in the face of an
appeal to CAS. There is also a need to publish all the
results in peer-reviewed scientific journals in order to
provide quality assurance on the research. This has taken a
great deal of time as the amount of data collected was
enormous and although many papers have been published
or are ‘in press’ it will be another year or two before that
is achieved.
GH-2000 was immensely successful – it brought
together a team of European endocrinologists and scientists
to work together in what turns out to be an exciting new
area – sport – and was able to show that what at the outset
had appeared to be an impossible task was, in fact,
achievable. It has shown the importance of the GH axis in
fitness and sport and has revealed that GH has important
regulatory control over an even greater range of processes
than we had previously understood. It has shown the
effects of ageing are not only on the GH–IGF-I axis but on
all areas where GH acts and it has shown that even elite
athletes who exercise at rates unachievable by most of us
still show a decline in GH production as they age. Is this
the reason why GH abuse is thought to be so prevalent in
sport? Does GH prolong active life? The answer to this is
not known but the data indicate that it is a valid question
and support the need for much further research work in
the area of GH and ageing and the possible therapeutic
role of GH in preventing loss of lean tissues with ageing.
Acknowledgements
This paper is based on an invited lecture given to the
British Endocrine Society meeting at Imperial College
London in November 2000. The results are those of the
GH-2000 project, not all these results have yet been
published. I would like to thank all those who participated
in the project including our volunteer athletes and all the
sports organisations who helped us recruit and study
our volunteers. Finally, I would like to acknowledge the
wonderful statistical support we had from Dr Eryl Bassett
and Professor Philip Brown from the University of Kent.
References
Bamberger M & Yaeger D 1997 Over the edge. Sports Illustrated 14
62–70.
24 P H SONKSEN · Insulin, GH and sport
Journal of Endocrinology (2001) 170, 13–25 Society for Endocrinology | Home
Boroujerdi MA, Umpleby AM, Jones RH & Sonksen PH 1995 A
simulation model for glucose kinetics and estimates of glucose
utilization rate in type 1 diabetic patients. American Journal of
Physiology 268 E766–E774.
Brown P, Tompkins C, Juul S & Sonksen PH 1978 Mechanism of
action of insulin in diabetic patients: a dose-related effect on
glucose production and utilisation. British Medical Journal 1
1239–1242.
Cuneo R, Salomon F, Wiles C, Hesp R & Sonksen PH 1991a
Growth hormone treatment in growth hormone-deficient adults.
I. Effects on muscle mass and strength. Journal of Applied Physiology
70 688–694.
Cuneo R, Salomon F, Wiles C, Hesp R & Sonksen PH 1991b
Growth hormone treatment in growth hormone-deficient adults.
II. Effects on exercise performance. Journal of Applied Physiology 70
695–700.
Gottesman I, Mandarino L, Verdonk C, RizzaR&Gerich J 1982
Insulin increases the maximum velocity of glucose uptake without
altering the Michaelis constant in man. Journal of Clinical
Investigation 70 1310–1314.
Jorgensen JO, Pedersen SA, Thuesen L, Jorgensen J, IngemannHansen
T, Skakkebaek NE & Christiansen JS 1989 Beneficial
effects of growth hormone in GH-deficient adults. Lancet i (8649)
1221–1225.
Randle PJ, Garland PB, Newsholme EA & Hales CN 1965 The
glucose fatty acid cycle in obesity and maturity onset diabetes
mellitus. Annals of the New York Academy of Sciences 131 324–333.
Salomon F, Cuneo R, Hesp R & Sonksen PH 1989 The effects of
treatment with recombinant human growth hormone on body
composition and metabolism in adults with growth hormone
deficiency. New England Journal of Medicine 321 1797–1803.
Schafer EA 1916 The Endocrine Organs – An Introduction to the Study of
Internal Secretions, pp 4–7 and 128–130. London: Longmans, Green
and Co.
Sinha A, Formica C, Tsalamandris C, Panagiotopoulos S, Hendrich E,
DeLuise M, Seeman E & Jerums G 1996 Effects of insulin on
body composition in patients with insulin-dependent and
non-insulin-dependent diabetes. Diabetic Medicine 13 40–46.
Sonksen PH & Sonksen J 2000 Insulin: understanding its action in
health and disease. British Journal of Anaesthesia 85 69–79.
Thomas SH, Wisher M, Brandenburg D & Sonksen PH 1979 Insulin
action on adipocytes, evidence that the anti-lipolytic and lipogenic
effects of insulin are mediated by the same receptor. Biochemical
Journal 184 355–360.
Toogood AA, O’Neill PA & Shalet SM 1996 Beyond the
somatopause: growth hormone deficiency in adults over the age
of 60 years. Journal of Clinical Endocrinology and Metabolism 81
460–465.
Van Cauter E & Copinschi G 2000 Interrelationships between
growth hormones and sleep. Growth Hormone and IGF Research
10 (Suppl B) S57–S62.
Wallace JD, Cuneo RC, Baxter R, Orskov H, Keay N, Pentecost C,
Dall R, Rosen T, Jorgensen JO, Cittadini A, Longobardi S, Sacca
L, Christiansen JS, Bengtsson B-A & Sonksen PH 1999 Responses
of the growth hormone (GH) and insulin-like growth factor axis to
exercise, GH administration, and GH withdrawal in trained adult
males: a potential test for GH abuse in sport. Journal of Clinical
Endocrinology and Metabolism 84 3591–3601.
Wallace JD, Cuneo RC, Lundberg PA, Rosen T, Jorgensen JOL,
Longobardi S, Keay N, Sacca L, Christiansen JS, Bengtsson B-A &
Sonksen PH 2000 Responses of markers of bone and collagen
turnover to exercise, growth hormone (GH) administration and GH
withdrawal in trained adult males. Journal of Clinical Endocrinology
and Metabolism 85 124–133.
Wu Z, Bidlingmaier M, Dall R & Strasburger CJ 1999 Detection of
doping with human growth hormone. Lancet 353 895.
Received 18 January 2001
Accepted 10 April 2001
Insulin, GH and sport · P H SONKSEN 25
Society for Endocrinology | Home Journal of Endocrinology (2001) 170, 13–25
