| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Turku PET Centre (P.I., V.O., J.B., T.G., P.N., J.K.), Department of Nuclear Medicine, and Department of Medicine (T.T., P.N.), University of Turku, FI-20521 Turku, Finland; PET Centre (P.I., E.F.), National Research Council, Institute of Clinical Physiology, 56100 Pisa, Italy; Department of Medicine (A.K.T.), University Hospital of Kuopio, FIN-70210 Kuopio, Finland; and Department of Internal Medicine (E.F.), University of Pisa School of Medicine, 56100 Pisa, Italy
Address all correspondence and requests for reprints to: Patricia Iozzo, M.D., Institute of Clinical Physiology, National Research Council, Via Moruzzi 1, 56100 Pisa, Italy. E-mail: patricia.iozzo{at}ifc.cnr.it.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We conclude that, in IGT patients, the ability of the liver to extract FFA from the circulation appears to be impaired. The reciprocal relationship between hepatic FFA extraction and glucose/lactate flux may derive from intrahepatic substrate competition.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The liver is an important site of FFA removal from the blood. Animal studies conducted during the 1960–1970s have shown that approximately one fourth to one third of isotopically labeled FFAs are extracted by the liver from perfusing blood and are subsequently retained within the organ (14, 15). Their intracellular fate is critically dependent on nutritional status because ß-oxidation prevails in the postabsorptive state, whereas incorporation into triglycerides predominates postprandially (16).
Very little is known about the regulation of hepatic uptake of circulating FFAs in vivo, especially in humans. Splanchnic catheterization studies measuring the net balance between FFA uptake and release across multiple organs can provide some information. However, although there is general agreement that an acute rise in plasma glucose and insulin levels impairs splanchnic ß-oxidation (17, 18, 19, 20, 21, 22), controversial findings have been reported as to whether FFA uptake and extraction also are affected; they were either unchanged during modest hyperglycemia (17) or reduced after a glucose load (18, 19, 20) and were increased under conditions of enhanced hepatic FFA oxidation, such as exercise (21, 22). In hepatic balance studies in rats, FFA fractional extraction was stimulated by prolonged starvation, a condition associated with increased ß-oxidation (23). Because of the high proportion of circulating FFA normally disposed of by the liver and the potential contribution of this organ to the excess plasma levels observed in insulin resistance states, there is a need for a better understanding of this function in man.
The present study was designed to investigate whether an impairment of liver FFA fractional extraction and/or uptake is present in patients with impaired glucose tolerance (IGT), i.e. a model of spontaneous, chronic hyperglycemia and insulin resistance. Tissue measurements were performed noninvasively by using the FFA analog 14(R,S)- [18F]fluoro-6-thia-heptadecanoic acid ([18F]FTHA) and positron emission tomography.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Ten male subjects with IGT by World Health Organization criteria (24) participated in the study. Eight healthy men with no family history of diabetes and normal physical examination, blood pressure, and routine blood tests served as controls. None of the patients or control subjects was taking any medication. Written informed consent was obtained from all subjects after the study was approved by the Ethics Committee of Turku University Central Hospital and Kuopio University Hospital. FFA uptake data in skeletal muscle and myocardium of some of these patients have been previously published (25).
Positron emission tomography scanning
Studies were conducted after a 12–15-h overnight fast using an ECAT 931–08/12 scanner (CTI Inc, Knoxville, TN). [18F]FTHA was synthesized as previously described, leading to a radiochemical purity greater than 98% (26). Two venous catheters were inserted, one into an antecubital vein for tracer injection and one into a heated vein (
70 C) for sampling of arterialized venous blood. After optimization of the subject’s position, transmission scans were obtained with a 68Ge removable ring source to correct all subsequent emission data for tissue attenuation of 
photons. Then, [18F]FTHA (171 ± 5 MBq) was injected, and a 32-min dynamic scan was carried out to image the liver and the cardiac chamber. Arterialized blood samples were collected throughout the study to measure glucose, insulin, FFA, and plasma [18F]FTHA metabolites as previously described (25, 26). Average values of plasma glucose, insulin, FFA, and lactate during the scan were used in the analyses.
Image processing
All sinograms were corrected for tissue attenuation, dead time, and decay and reconstructed through standard reconstruction algorithms. Final in-plane resolution was 9.5 mm full width at half maximum. Large circular regions of interest for hepatic [18F]FTHA time-activity measurements were placed on two to four consecutive image planes in the right lobe of the liver. Such measurements were averaged to generate one tissue time-activity curve in each subject. Smaller regions of interest were drawn on two adjacent image planes in the left ventricular chamber of the heart for the measurement of radioactivity in arterial blood; special attention was paid to avoid contamination from surrounding myocardial tissue. Input functions were corrected for time delay by comparison with corresponding tissue time-activity curves, based on previously proposed methods (27). In our application, a two-tissue compartment model was used (27), and dispersion was excluded from the model.
Data analysis
The nonmetabolized fraction of [18F]FTHA was determined by high-performance liquid chromatography from nine blood samples, and these data were used to correct the input function. Arterial and tissue time activity curves were analyzed graphically to derive the influx rate constant in liver (LKi; min–1), which is given by the slope of the linear fit of the data.
Graphical analysis was developed to quantify irreversible regional uptake of a test substance by using its sequential tissue and blood concentrations (28); it was subsequently modified to account for situations of incomplete tracer trapping (29). In this model, a graph is generated by plotting the following equation:
 |
LKi is a clearance parameter (ml·min–1·ml–1 of tissue) and can be used here interchangeably with liver FFA extraction fraction (LFEF) given that the latter is related to the former through plasma flow, which approximates unity (i.e. 1 liter/min) in the liver (31). The liver FFA uptake index (LFU) was computed as the product of LKi and plasma FFA levels. No lumped constant term was used.
Maximal aerobic power was determined using oxygen detectors and a cycle ergometer (Model 800S; Ergoline, Mijnhardt, The Netherlands), with a continuous incremental protocol. Body fat content was estimated from four skinfolds (subscapular, triceps brachii, biceps brachii, and crista iliaca), as measured with a caliper.
Statistical analysis
All data are presented as mean ± SEM. One-way ANOVA was used for group comparisons. Regression analyses were carried out according to standard techniques. Significance was set at P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
). In addition, IGT subjects had higher circulating levels of glucose, insulin, and triglycerides, and lower high-density lipoprotein cholesterol levels, whereas serum FFA levels were similar (Table 2). The homeostasis model assessment index (32) suggested significant insulin resistance in the IGT group.
|
|
Liver data
Rapid accumulation of tracer in the liver was observed with time, leading to progressively higher tissue to blood radioactivity ratios. Linear fit of the data were similar in the two groups, with mean r values exceeding 0.99 in both groups (0.998 ± 0.001 vs. 0.993 ± 0.004, P = not significant), as exemplified in Fig. 1.
|
). Separation of the more hyperglycemic subjects into a separate group (metabolic data of subgroups are shown in Table 3) did not change this result despite the small sample size. Accordingly, individual data showed that LFEF was inversely related to plasma glucose and glycosylated hemoglobin levels (Fig. 3); this correlation with plasma glucose levels was even stronger after excluding the one outlier with higher fasting glycemia seen in Fig. 3 (r = 0.58; P = 0.01).
|
|
|
), which, however, fell just short of statistical significance. As shown by the individual values (Fig. 4), this result was entirely accounted for by one outlier, in whom LFU exceeded the mean value of healthy subjects due to higher serum FFA levels (0.9 mmol/liter).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The current results demonstrate that chronic, mild hyperglycemia, fasting hyperinsulinemia, and insulin resistance are associated with a reduced uptake and fractional extraction of circulating FFA by the liver. Interestingly, the defect was even more pronounced in those few patients who had higher fasting plasma glucose levels on the day of the study, thus supporting a quantitative relationship between hyperglycemia and LFEF (Fig. 2). At normal serum FFA levels, LFU was reduced in the majority of IGT subjects compared with the mean value of healthy subjects, while being increased in one in whom the higher serum FFA levels were likely overriding the LFEF impairment.
Given a liver plasma flow of 0.9 liter/min, a plasma FFA level of 0.5 mmol/liter, and a hepatic FFA extraction ratio of 0.35, liver FFA uptake averages 0.16 mmol/min (or 0.2 µmol/min/ml of liver volume, Fig. 1), which corresponds to approximately 20% of the fasting FFA turnover rate of healthy adults (35). Consequently, it is expected that a reduction in LFEF, such as observed in our IGT group, should increase fasting FFA concentrations only marginally; this is in keeping with our findings.
In the fasting state, ß-oxidation is the primary route for intracellular FFA utilization (16), and FFA uptake is dependent on the efficiency of this metabolic pathway. In the splanchnic bed as a whole, a defect in fatty acid oxidation has been induced in healthy subjects by acutely raising plasma glucose and insulin levels (17). Our data support that this mechanism may be operative in the liver of patients with chronic, spontaneous hyperglycemia and insulin resistance. This interpretation is in line with most previous reports, which show hepatic FFA fractional extraction and oxidation to be concordantly decreased after a glucose load (18, 19, 20) and LFEF to be increased under conditions of enhanced hepatic FFA oxidation, such as exercise (splanchnic) (21, 22) and starvation (hepatic) (23). The discrepancies arising from balance studies of the splanchnic area (17, 18, 19, 20) might be due to differences in lipid metabolism between hepatic and extrahepatic tissues; this would also explain the lack of a correlation between ß-oxidation and FFA uptake reported in some of these studies (17). Additionally, differences in study design, including the acute nature of the experiment and the characteristics of the study population (17, 18, 19), or the use of different animal (20, 22, 23) and disease models (21, 22) in previous studies make it difficult to compare their results.
Skeletal muscle FFA extraction (25, 36) and/or oxidation (36, 37, 38) also have been reported to be impaired in insulin resistance states. Together with this evidence, our data do not seem to support that competition between FFA and glucose for oxidation is a major mechanism for the peripheral insulin resistance seen in IGT as long as circulating FFA are not increased. Conversely, increased lactate flux from peripheral tissues to the liver can stimulate gluconeogenesis and increase the intracellular availability of precursors for both FFA re-esterification and, to a lesser extent, de novo fatty acid synthesis. Also, glucose entry and oxidation are strongly self-regulated processes in hepatocytes, as opposed to FFA access into mitochondria, which is transfer limited (39); the latter process seems to be down-regulated by glucose through the suppression of carnitine-palmitoyl transferase-1 activity (40). According to this view, mild hyperglycemia diverts FFA away from oxidation (17), resulting in lipid accumulation within the liver (10) and enhanced export of triglyceride-rich lipoproteins into the plasma compartment (Table 2). Such reverse substrate competition (41) appears to prevail in prediabetic subjects with normal plasma FFA.
Hyperinsulinemia was marked in our IGT patients. Insulin has been shown to promote the formation of acetyl-coenzyme A (CoA), which may in turn inhibit ß-oxidation (42); it has also been suggested that the hormone activates acetyl-CoA carboxylase, leading to malonyl-CoA synthesis (43). The latter depresses the activity of carnitine-palmitoyl transferase-1 (44). In accord with the earlier hypothesis of reverse substrate competition, insulin enhances glucose uptake in the liver (45, 46), reinforcing the mass action effect of hyperglycemia on the glucose oxidative pathway. In turn, glucose appears to be necessary for insulin-mediated malonyl-CoA synthesis. Thus, in our IGT patients, hyperinsulinemia and hyperglycemia represented a synergistic combination to inhibit FFA oxidation. The recent evidence that intrahepatocellular lipids are higher in insulin-resistant type 2 diabetic subjects and are further increased after long-term insulin infusion in proportion to baseline insulin sensitivity supports the concept that impaired oxidation diverts FFA into triglyceride synthesis (47). It is interesting that ß-oxidation and FFA extraction appear to be either normal or increased in splanchnic organs of type 1 diabetic patients (21) and in the liver of depancreatized dogs (22), suggesting that the defect might be a prerogative of insulin-resistant, hyperinsulinemic (not just hyperglycemic) states.
The hypothesis of reverse substrate competition may sound in contrast with previous findings by us and by others, which show that insulin-mediated hepatic glucose uptake is impaired in patients with type 2 diabetes (48, 49). Although in normal glucose-tolerant, insulin-resistant subjects we could not detect any reduction in liver glucose disposal (48), a defect was described by others in IGT subjects with normal insulin sensitivity and with impaired first-phase insulin secretion (50). These previous observations are not easily extrapolated to the present study because of the differences in disease stage (IGT vs. overt diabetes vs. normal glucose tolerance), in plasma insulin levels (fasting vs. postprandial vs. mild hyperinsulinemia) and glucose levels (euglycemia vs. hyperglycemia), in the route of glucose administration (oral vs. peripheral vs. none), and in the ethnicity of the study populations (American vs. Northern European vs. Southern European). More importantly, the acute nature of these experiments, in which the levels of insulin and/or glucose were set to be equal between study groups, differs from the chronic condition of our subjects, in whom serum substrate/hormone levels were left in their spontaneous equilibrium. Although fasting hepatic glucose uptake data are required to draw definitive conclusions concerning reverse substrate competition in the liver of IGT patients, it is reasonable to anticipate that the putative regional glucose uptake defect might be overridden by the concurrent hyperinsulinemia and hyperglycemia.
Although several components of the metabolic syndrome can modify the intracellular fate of FFA, our data do not rule out the primacy of a defect in FFA metabolism, inducing a cascade of compensatory events, clustering as metabolic syndrome. As long as insulin hypersecretion is able to restrain lipolysis and confine FFA within the adipose depots, IGT is accompanied by normal serum FFA concentrations.
Our findings appear to extend to the liver the generalized impairment of fasting FFA oxidation observed in the whole organism and in skeletal muscle in association with insulin resistance (25, 36, 37, 38, 51, 52, 53). Interestingly, the capacity to augment lipid oxidation during fasting conditions, which was impaired in obesity and enhanced by exercise, was a strong predictor of the severity of insulin resistance (in the case of obesity) (51) or of the improvement in insulin sensitivity (with exercise) (52). Metabolic inflexibility has been described in obese individuals, who fail to suppress lipid utilization postprandially (38). It has been suggested that lower rates of FFA oxidation during fasting and postprandial inflexibility may be mechanisms leading to excess lipid accumulation within skeletal muscle and that, in turn, the adverse metabolic consequences of this defect may depend upon the capacity for efficient lipid utilization during fasting. Because exercise increases the disposal of a glucose load by the liver (54), it is tempting to speculate that, similar to skeletal muscle, the reduced fasting FFA uptake observed in this study, which reflects FFA oxidation, may be predictive of the hepatic insulin resistance that was described in IGT subjects during glucose ingestion (50).
Methodological considerations
Based on the knowledge that, in the fasting state, FFAs entering the liver are mostly oxidized while small amounts are stored into triglycerides for later release (15, 17) and on the fact that FTHA, after its transport into mitochondria and formation of two acetyl-CoA moieties, cannot be further metabolized being trapped in proportion to its uptake, we opted to use graphical analysis to quantify liver FFA uptake, a choice that was supported by a strong concordance between fitted and measured data, as exemplified in Fig. 1. As long as steady-state conditions are maintained, the iv injection of radiolabeled tracers satisfies most of the requirements of graphical analysis, as underlined by Patlak et al. (28). Our patients were under steady fasting conditions during the relatively short duration of the study. The occurrence of tracer trapping was verified in our study subjects by the goodness of the linear fit to the measured data, a test that is considered as proof of the assumption (28, 29, 30). Untrapped plasma metabolites were accounted for in the input function. Given the short duration of the present experiments, tracer trapping would include FFA entering the oxidative pathway and, to a lesser extent, their storage in triglycerides. Graphical analysis has the additional advantage of taking into account the integral of the input function over earlier time frames and its ratio to single time-point values in later intervals, thus minimizing the influence of tracer dispersion in blood passing through the extrahepatic splanchnic bed to the portal vein, which mostly affects the peak of the input curve (55).
Because it is not known whether a lumped constant term, accounting for differences in carrier affinities between FFA and FTHA, is required, no value was used; the lack of such a constant may slightly influence absolute LFU values, but it should not affect group comparisons. Notably, circulating FFA comprise a spectrum of different molecules; FTHA is a 17-carbon FFA analog, of which the uptake was shown to be in good agreement with that of 11C-labeled palmitate in the myocardium (56), a demonstration that supports the use of this tracer as representative of long-chain fatty acids. The use of more sophisticated modeling approaches, which provide information on the intracellular fate of our tracer that could be extrapolated to an FFA pool, awaits further characterization. With the current knowledge, the influence of single intracellular enzyme affinities in the two study groups could not be accounted for, and the study was conducted under fasting conditions to minimize the effect of enzymes implicated in triglyceride synthesis.
Although our assessment of fractional FFA extraction was based on a direct measurement, we used an arterial input function to model our data, whereas the liver is perfused by both portal venous and arterial blood. Thus, a difference in the amount of visceral fat between study groups may account for a diverse proportion of tracer being delivered to the liver. Although a lower uptake or higher release of tracer by visceral fat would reinforce the difference observed here between study groups, the reverse possibility cannot be presently excluded. By the same line of reasoning, the calculation of liver uptake represented a minimal estimate because peripheral FFA concentrations were used, and a higher FFA flux to the liver through the portal vein could not be excluded. We cannot dismiss the possibility that, in IGT, the LFEF impairment might have been partly overcome or even determined by a higher FFA delivery from visceral organs. In this case, down-regulation of hepatic FFA extraction would protect the organ against an exaggerated FFA flux in the face of a reduced oxidative capacity. However, abdominal visceral fat correlates with systemic lipolysis (57), and the peripheral delivery of FFA through the hepatic vein is strongly related to visceral adipose tissue FFA release (58). We could find neither a correlation between fat mass and LFU nor between-group differences in the contribution of visceral fat to peripheral FFA levels. Thus, although the simultaneous evaluation of visceral fat FFA metabolism is necessary to support this interpretation, our data indicate that higher plasma insulin levels in IGT subjects were likely sufficient to restrain lipolysis in a similar manner as in controls.
In conclusion, our data suggest that, in patients with IGT, the ability of the liver to extract FFA from the circulation is impaired; however, information on visceral fat FFA metabolism will be necessary for the full understanding of this finding. Such a defect in hepatic FFA metabolism may be causally related to the increased export of FFA as triglyceride-rich lipoproteins. The reciprocal relationship between hepatic FFA extraction and glucose/lactate levels is compatible with intrahepatic substrate competition.
|
Footnotes |
|---|
Received July 3, 2003.
Accepted March 28, 2004.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  G. H. Tomkin Targets for Intervention in Dyslipidemia in Diabetes Diabetes Care, February 1, 2008; 31(Supplement_2): S241 - S248. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Guiducci, T. Gronroos, M. J. Jarvisalo, J. Kiss, A. Viljanen, A. G. Naum, T. Viljanen, T. Savunen, J. Knuuti, E. Ferrannini, et al. Biodistribution of the Fatty Acid Analogue 18F-FTHA: Plasma and Tissue Partitioning Between Lipid Pools During Fasting and Hyperinsulinemia J. Nucl. Med., March 1, 2007; 48(3): 455 - 462. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. S. Burgert, S. E. Taksali, J. Dziura, T. R. Goodman, C. W. Yeckel, X. Papademetris, R. T. Constable, R. Weiss, W. V. Tamborlane, M. Savoye, et al. Alanine Aminotransferase Levels and Fatty Liver in Childhood Obesity: Associations with Insulin Resistance, Adiponectin, and Visceral Fat J. Clin. Endocrinol. Metab., November 1, 2006; 91(11): 4287 - 4294. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. R. Barrows, M. T. Timlin, and E. J. Parks Spillover of Dietary Fatty Acids and Use of Serum Nonesterified Fatty Acids for the Synthesis of VLDL-Triacylglycerol Under Two Different Feeding Regimens Diabetes, September 1, 2005; 54(9): 2668 - 2673. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, P = 0.087 vs. control group; *, P = 0.0009 vs. control, and P = 0.018 vs. IGT.
, healthy subjects; 
, IGT patients).