The Effects of Fatty Acids in the Disorder of Insulin
Resistance And Non-insulin Dependent Diabetes Mellitus
Somebody predicted that Diabetes will become a
plague in the future and I think that it's not impossible since almost 50% of the population in our country, the Philippines, has
it. This disease has been such an issue especially with obese people, and as of now, there's
still no cure for it.
Named because of its "sweet urine" characteristic, diabetes mellitus is one of the major
diseases seen in this country today. More frequently referred to as simply diabetes, this
disorder has been known since ancient times. There are several descriptions of the clinical
signs of diabetes in the Egyptian, Greek and Roman literature. During the Middle ages,
diabetes was diagnosed by tasting the urine to detect the large amount of glucose excreted.
Fortunately, we have more reliable (and less distasteful) methods for detecting glucose
loss today.
Before going further let us first discuss the types of diabetes. The primary metabolic
problem in diabetes mellitus is an ability of the body to utilize glucose. Absorption of of
this carbohydrate across the intestinal wall is unimpaired, and glucose enters the
blood stream unhindered. However, the process of glucose transfer into the cell for further
metabolism is impaired. At the heart of the issue is insulin, the hormone responsible for
cellular glucose uptake. In some situations, the synthesis and release of insulin by the
pancreas is greatly decreased. Therefore, less insulin is available to exert its metabolic
effect. In other categories of diabetes, however, the amount if insulin is adequate (and
often more than adequate), but its biochemical effect is significantly diminished owing to
the changes in the number and type of cell receptors responsible for glucose uptake.
Diagnosis of diabetes mellitus from clinical symptoms is often difficult because of the
various ways the disease manifest itself. In one type of the disorder, the clinical picture is
one of weight loss, thirst, and loss of glucose in the urine (glucosuria). Frequently we also
see the presence of other substances in the urine known as ketone bodies. In another
category is the patient who is overweight, physically inactive, and has glucosuria. A third
situation may be that of a woman who shows transient abnormalities of glucose metabolism only during pregnancy. The picture is complicated and sometimes very
confusing.
A brief comparison of the characteristics of the two major types of diabetes mellitus
(Table 1-1) is helpful ( the transient stage in pregnancy is omitted from this discussion).
Insulin-dependent diabetes mellitus (type 1) usually manifests itself at an early age (generally below 20 years old) and is seen in less than 10% of all diabetics. Over 90% of
those with diabetes have the noninsulin dependent variety (type II), which appears much
later in life.
In the insulin dependent form, there are decreased number of beta cells in the
pancreas and little production of insulin. Frequently, inflammation of the pancreas is also
seen. The noninsulin dependent variety is generally characterized by excessive insulin
production, with little or no pancreatic damage until much later in the course of the
disease. Whereas the insulin dependent diabetic often experiences significant weight loss
during the onset of the disease , the patients with non-insulin dependent diabetes is usually
obese, a major contributing factor for the disease. Even the therapies are implied by the
nomenclature. Insulin injection is the major form of treatment for the insulin-dependent
diabetic, but the non-insulin dependent patient is usually treated by diet control and
weight loss. These latter patients do not normally require insulin administration to regulate
blood glucose levels.
--------------------------------------------------------------------------------------------------------------
TABLE 1-1
CHARACTERISTICS OF TYPES OF
DIABETES MELLITUS
INSULIN-
NON-INSULIN-
DEPENDENT
DEPENDENT FEATURE
(TYPE I)
(TYPE II)
--------------------------------------------------------------------------------------------------------------
Age of onset Usually <20
Usually >40
Percentage of
Diabetics
<10%
>90%
Ketone bodies Usually
Rarely
Obesity at onse Rare
Common
Serum Insulin Very low
Normal or high
--------------------------------------------------------------------------------------------------------------
Evidence is reviewed that free fatty acids (FFAs) are one important link
between obesity and insulin resistance and non-insulin dependent diabetes mellitus. Fist is that plasma FFA levels are increased in most obese people. Second, is the fact that
physiological increase in plas FFA concentrations inhibit insulin stimulated peripheral glucose uptake
in a dose-dependent manner in normal controls and in patients with NIDDM. There were
two mechanisms that were discovered by Goenther Buden.
1) A fat-related inhibition of glucose transport or phosphorylation, which appears after 3-4 hours of fat infusion, and
2) a decrease in muscle glycogen synthase activity, which appears after 4-6 hours of fat
infusion.
Third, FFAs stimulate insulin secretion in nondiabetic individuals. Some of this
insulin is transmitted in the peripheral circulation and is able to compensate for
FFA-mediated peripheral insulin resistance. FFA- mediated portal hyperinsulinemia counteracts
the stimulation of FFAs on hepatic glucose production (HGP) and thus prevents hepatic
glucose overproduction. It is speculated that, in obese individuals who are genetically
predisposed to develop NIDDM, FFAs will eventually fail to promote insulin secretion.
The stimulatory effect of FFAs on HGP would then become unchecked, that leads to
leads to hyperglycemia. Hence, continuosly elevated levels of plasma FFAs may play a key
role in the pathogenesis of NIDDM in predisposed individuals by impairing peripheral
glucose utilization and by promoting hepatic glucose overproduction.
NIDDM affects between 5 and 20% of the population in Western industrialized
societies and also among Filipinos, and is responsible for a significant amount of
morbidity and mortality. Unfortunately, the pathogenesis of this disease remains
incompletely understood despite decades of investigative reports. It has recently been
suggested that NIDDM may have more to do with abnormalities in fat than carbohydrate
metabolism. Supporting this notion are the well-known facts that 85% of patients with
NIDDM are obese and that obesity is virtually always associated with insulin resistance,
which is arguably the earliest detectable and dominant metabolic defect in patients with
this disease. Moreover, there is evidence to suggest that the association between obesity
and insulin resistance may be a cause and effect relationship. For example, it has been
shown in humans and in animals that weight gain decreased, while weight loss increased,
insulin sensitivity and glucose tolerance. It remains uncertain, however, how obesity
produces insulin resistance. One possible mechanism would be the generation of one or
more metabolic messengers by the adipose tissue, which when released would inhibit
insulin action on muscle and/or the liver. Here, we review the evidence, gained from
studies in humans, that FFAs are likely candidates for such messengers. They are elevated
in most obese individuals primarily because of an increase in the rate of lipolysis from the
expanded cell mass. Elevated plasma FFA concentrations produce peripheral and hepatic
insulin resistance, which in normal subjects is compensated by FFA induced potentiation
of glucose stimulated insulin secretion. It was proposed that in the development if NIDDM
FFAs fail to stimulate insulin secretion, which leaves hepatic and peripheral insulin
resistance unchecked resulting in hepatic overproduction and peripheral underutilization of
glucose.
FREE FATTY ACIDS AND GLUCOSE UPTAKE IN MUSCLE
Historical review. The concept that
elevated blood levels of FFAs play a key role in the development of insulin resistance in obesity and NIDDM was first proposed by Randle et
al. more than 30 years ago. Based on their demonstration that the increased availability of
FFAs decreased carbohydrate oxidation in isolated perfused , rat hearts and
hemidiaphragm. Randle et al proposed a glucose-fatty acid cycle. The key points of this cycle
were: The increased availability of FFAs in blood produces an increase in intramuscular
acetyl CoA and citrate content: acetyl-CoA inhibits pyruvate dehydrogenase allosterically
and this in turn reduces glucose oxidation; citrate inhibits phosphofructokinase 1 and
thus glycolysis itself, eventually leading in the impairment of glucose uptake. The initial
enthusiasm for this intriguing concept was dampened when several groups were unable to
reproduce the inhibitory effects of fatty acids on glucose uptake in rat skeletal muscle that
Randle et al, had demonstrated in rat heart. More recently, several groups have re-examined glucose-fatty acids interactions in vivo, using indirect calorimetry ( to determine
rates of carbohydrate and fat oxidation) in combination with hyperinsulinemic clamping
(to determine insulin sensitivity). Practically all of the groups found that raising plasma
fatty acids concentrations increased fat oxidation and inhibited carbohydrate oxidation.
Some also found inhibitory effects of fat on glucose uptake, but most did not. Hence, while
the suppressive effect of fatty acids on carbohydrate oxidation was generally confirmed, it
remained controversial where fatty acids also inhibited insulin stimulated glucose uptake
(i.e. caused peripheral insulin resistance).
Recent Development. It has recently been showed in healthy volunteers that the fatty acid
mediated inhibition of insulin stimulated carbohydrate oxidation occurred early (i.e.
within 1-2 hours) whereas the inhibition of glucose uptake developed only after - 4 hours
of fat infusion. Thus, insufficient time of fat plus insulin infusion (2 hours in most studies)
was the most likely reason why the inhibitory effect of fatty acids on glucose uptake was
not found in most studies. It was also shown that FFAs inhibited glucose uptake in a dose-dependent fashion throughout the physiological range of plasma FFA concentrations
(from -50 to -80 umol/l ). Further, fatty acids have been demonstrated to inhibit insulin-stimulated glucose uptake in healthy subjects and patients with NIDDM. Three groups,
who had previously failed to find inhibition of glucose uptake by fatty acids, had infused
fat plus insulin together for only 2 hours, which was not long enough to develop inhibition
of glucose uptake.
Also of interest was the finding that, under conditions of comparable euglycemia and
low plasma FFAs (-100umol/l), insulin-stimulated glucose uptake was 2 times higher in
normal controls than in patients with NIDDM, indicating that FFAs could account for only
a part of the insulin resistance in diabetic patients and that a major part, perhaps as much as
50%, was unrelated to fatty acids.
Hence, there is currently strong evidence, in normal as well as in diabetic subjects, that
physiological elevations of plasma FFA levels lower peripheral insulin sensitivity dose-dependently. It is reasonable to assume, therefore that chronically elevated plasma levels of
FFA , perhaps together with FFAs released from intramuscular fat depots, are contributing
to the insulin resistance usually seen in obese patients. It should be pointed out that FFA
induced insulin resistance serves as an important physiological role , preserving glucose
for oxidation in the central nervous system when glucose is scarce, for instance, during
fasting, prolonged exercise, or late pregnancy. In obesity, These same mechanisms can
become counterproductive, inhibiting glucose utilization when there is no need to spare
glucose.
Most obese individuals have normal glucose tolerance because their insulin resistance
is matched by enhanced insulin secretion. The mechanism responsible for the increased
insulin secretion in obesity is not clear, particularly in cases where blood glucose
concentrations are not elevated. Hyperinsulinemia has been suggested to be at least in part,
a consequence of reduced insulin uptake by the liver, caused by exposure of the liver to be
elevated plasma FFA levels coming from enlarged intra-abdominal fat depots. Recent in
vitro and in vivo studies from some laboratory, however, have failed to support the concept
that hyperinsulinemia associated with central obesity is caused by FFA- mediated reduction
in hepatic insulin clearance. Alternatively, it has been proposed that, in obesity, FFAs
could produce both insulin resistance and a compensatory increase in insulin secretion.
FATTY ACIDS AND INSULIN SECRETION
The acute stimulation of insulin secretion by FFAs has been well established. It has
recently been reported, however, that prolonged exposure to elevated concentrations of
FFAs ( for 48 hours) produced a biphasic response; an initial stimulation ( after 3 and 6
hours) was followed later (after 24 and 48 hours) by severe inhibition of insulin secretion
from isolated islets or isolated perfused pancreas. These in vitro findings appeared to be
incompatible with the notion that FFAs were the link between insulin resistance and
compensatory insulin secretion . Instead, they suggested that increased FFA levels
produced a second major pathogenic defect (i.e. decreased insulin secretion). A recent
study, however, showed that the in vivo effects of FFAs obtained in normal subjects were
different from the in vitro effects obtained in animals. High plasma FFA concentrations
were not only not associated with decreased insulin secretion, but actually increased
insulin secretion rates for as long as 48 hours. Glycerol, which was infused together with
lipid, could not have been responsible for the increase in insulin secretion since glycerol is
not an insulin secretagogue. Compared with glucose, FFAs were weak insulin
secretgogue. Nevertheless, some of the insulin secreted into the portal circulation in response to FFAs
was transmitted into the peripheral circulation and normalized the previously suppressed
abnormal glucose utilization.
Obviously, obesity-related hyperlipidemia lasting for months or years may have
different effects than hyperlipidemia lasting for 2 days. Moreover, the demonstration that a
prolonged (48 hours) elevation of plasma FFAs stimulated insulin secretion in normal
subjects did not exclude the possibility that FFA- mediated insulin secretion may be
reduced in patients with NIDDM or subjects who genetically predisposed to develop
NIDDM, the FFA stimulation of insulin secretion may be reduced, compared with normal
subjects. This reduction may result in uncompensated peripheral insulin resistance and,
in addition, may have consequences with respect to hepatic glucose production (HGP).
FFAS AND HGP
There is good in vitro evidence to show that FFAs promote gluconeogenesis. The
proposed mechanisms include increased production of ATP and NADH and the activation
of pyruvate carboxylase by the Acetyl-CoA that is generated via fatty acid oxidation. The
available in vivo evidence is less strong. On one hand, Rebrin et al. have recently reported
a very strong relationship in conscious dogs between plasma FFAs and HGP a steady state and during dynamic insulin changes. On the other hand, the lowering of plasma FFAs with
nicotinic acid or acipimox, a nicotininc acid analog, has been reported to decrease, to
increase or not to change HGP. When plasma FFAs were raised during euglycemic- or
hyperglycemic-hyperinsulinemic clamping in normal controls of patients with
NIDDM, the insulin suppression of HGP was partially inhibited . Additional evidence in favor of a
stimulatory effect of FFAs on HGP was obtained in overnight fasted normal individuals
in whom plasma FFAs were raised (by infusing triglycerides and heparin), while insulin
was clamped at basal concentrations (by the infusion of somatostatins and basal insulin
replacements). Under these conditions, HGP and plasma glucose levels rose dramatically.
On balance, the available human data suggests that FFAs increase HGP but that the extent
of the increase is determined by the FFA-mediated stimulation of insulin secretion.
MECHANISMS
The mechanism responsible responsible for the inhibitory effect of fatty acids on
carbohydrate oxidation, first demonstrated by Randle et al. in rat hearts, has been
confirmed in humans where fat infusion produced large increments in acetyl-CoA (432%)
and in acetyl-coA/free CoA (489%) and the inhibition of pyruvate dehydrogenase in
skeletal muscle. It needs to be emphasized, however, that the inhibition of carbohydrate
oxidation by FFAs alone does not affect insulin-stimulated glucose uptake since glucose
uptake during lipid infusions remained unchanged for 3-4 hours while carbohydrate
oxidation was inhibited. An explanation for this phenomenon became apparent when it was
found that for the initial 3-4 hours FFAs had little or no effect on glycolytic flux and that
the glucose carbons that could not be oxidized were shunted into lactate and alanine
production.
Relevant to the understanding of the mechanisms responsible for the inhibitory effects
of fatty acids on glucose uptake was the finding that fatty acids did not affect basal (i.e.
insulin-dependent) glucose uptake or the glucose uptake stimulated by hyperglycemia.
This indicated that fatty acids selectively inhibited glucose uptake stimulated by insulin.
In an attempt to localize the fat-induced defects on glucose uptake, glucose fluxes
through all major pathways of intracellular glucose utilization were determined using non-invasive methods that have been recently validated. It was found that insulin stimulated
rates of glucose uptake, glycogen synthesis, and glycolysis are inhibited to about the same
extent. These results were most compatible with an FFA- induced defect in glucose
transport or phosphorylation since the primary inhibitions of glycogen synthesis or
glycolysis would be expected to result in disproportionally reduced rates of these
pathways. The presence of a transport or a phophorylation defect was also supported by
another independent finding, namely, that glycogen synthase activity was normal in muscle
biopsies taken 4 hours after the start of fat infusion ( i.e., at a time when glucose uptake
was significantly inhibited). On the other hand, 2 hours later (i.e., between 4 and 6 hours of
fat infusion) plasma FFA concentrations >500 umol/l caused a decrease in muscle
glycogen synthase activity and an increase in muscle glucose-6-phosphate concentrations. Thus,
there appeared to be at least two mechanisms by which FFAs inhibited insulin stimulated
glucose uptake:1) by the inhibition of glucose transport or phosphorylation ( after 3-4
hours of fat infusion) and 2) by decreasing muscle glycogen synthase activity ( after more
than 4 hours of fat infusion).
The cellular and molecular mechanisms responsible for these transport/ phosphorylation
and glycogen synthase defects are not known. Possibilities include the fatty acid- induced
inhibition of insulin stimulated glucose transport via the accumulation of glucosamine
pathway metabolites, including N-acetylglucosamine, N-acetylglucosamine-6-phosphate,
or uridine-diphosphate N-acetylglucosamine-6-phosphate. A similar mechanism has been
proposed by Marshall et al. to explain the decrease in insulin sensitivity that occurs after
hyperglycemia (glucose toxicity). Alternatively, FFAs could interfere with GLUT4 gene
expression in muscle and adipose tissue. This would explain the relatively long lag period
of -4 hours needed for the development of the transformation/phosphorylation defect. In
support of this possibility, Long and Pekala have recently shown that several long chain
fatty acids decreased GLUT4 mRNA levels in fully differentiated 3T3-L1 cells by decreasing GLUT4-gene transcription and by destabilizing the GLUT4 message.
Another possibility to be considered would be fatty acid induced changes in membrane
fluidity. Insulin receptors are embedded in the lipid bilayer of plasma membranes, and
there is evidence to suggest that altering the fatty acid content of membranes can alter
insulin receptor accessibility, insulin binding and action. Generally, it has been found that
increasing polyunsaturated fatty acid content can increase membrane fluidity, insulin
binding and action, whereas decreasing their content has the opposite effect. It is not
known, however, whether 4 hours of fat infusion (the time necessary to produce insulin
resistance) can produce changes in membrane fatty acid composition of sufficient degree
to inhibit insulin action.
TUMOR NECROSIS FACTOR (TNF)-& AND INSULIN RESISTANCE
It has recently been suggested that the cytokine TNF-& may play a pivotal role in the
pathogenesis of peripheral insulin resistance in obesity. This hypothesis is supported by the
observation that TNF-& is overexpressed in the adipose tissue of obese insulin-resistant
rodents and humans and that the neutralization of TNF-& in falfa Zucker rats decreased
insulin resistance and increased insulin-receptor tyrosine kinase activity in adipose tissue
and muscle. Several possible mechanisms have been suggested by which TNF-& could
produce insulin resistance in obese subjects.
1) TNF-& released from adipose tissue could produce insulin resistance in
muscle, although this endocrine mode of actin seems unlikely since TNF-& is undetectably low in the circulation. Supporting this notion, Hurel et al.
have recently reported that the neutralization of TNF-& with an antibody over a 4-week
period had no effect on insulin sensitivity in obese patients with NIDDM.
2) TNF-& could inhibit insulin action through local (paracrine) action on muscle.
And
3) TNF-& may act indirectly through another factor that is released into the circulation and inhibits insulin
action on muscle. Inasmuch as TNF-& has been shown to increase lipolysis , this factor
could well be FFAs; hence, the effect of TNF-& on glucose uptake may be mediated, at
least in part, by fatty acids. This notion is supported by the observations that the
neutralizations of TNF-& in Zucker rats was associated not only with an increase in insulin
sensitivity, but also with the decrease in plasma FFA levels and that infusion of TNF-& in
humans increased plasma FFAs. Clearly, the role of TNF-& as a link between human
obesity and insulin resistance and the interrelationship between TNF-& and FFAs need to
be explored further.
PHYSIOLOGICAL AND CLINICAL SIGNIFICANCE
The 4-hour lag period between the rise in plasma FFAs and the onset of inhibition of
insulin stimulated glucose uptake most likely prevents insulin resistance in normal weight
healthy individuals after eating a fat-rich meal as plasma fatty acid levels rarely remain
elevated for that long. The situation is likely to be different in obese healthy individuals in
whom fatty acid levels are elevated. There, elevated plasma FFAs can be expected to cause
peripheral insulin resistance. They also stimulate insulin secretion, which is critically
important. The increased insulin secretion will not only compensate for the increased
peripheral insulin resistance, but will also prevent FFAs from increasing HGP.
Obese individuals who are genetically predisposed to develop NIDDM may eventually
lose their ability to increase insulin secretion in response to elevated plasma FFAs levels.
The FFA induced stimulation of HGP would then become unchecked and peripheral underutilization, together with the hepatic overproduction of glucose, would result in postprandial (early) and fasting (late) hyperglycemia. This could initiate a vicious cycle with
the hyperglycemia producing progressively more B-cell desensitization and more peripheral
insulin resistance.
REFERENCES:
1. Harris MI: Impaired glucose tolerance in the US population. Diabetes Care 12:464-174, 1989
2. Randle PJ, Garland PB, Hales CN, Newsholmer EA: The glucose fatty acid
3. McGarry JD : What if Mankowski had been agenesic? An alternative angle on
diabetes. Science 258:766 771, 1992
4. Guenther Boden; Diabetes, Vol. 46 , January 1997
5. Donald F. Calbreath: Clinical Chemistry, A Fundamental Textbook; 1992
.