Defective regulation of triglyceride metabolism by insulin in the liver in NIDDM

Abstract

Features of diabetic dyslipidaemia in non-insulin-dependent diabetes mellitus (NIDDM), such as a low HDL cholesterol concentration, the preponderance of small dense LDL and postprandial fat intolerance are considered to be metabolic consequences of elevated plasma triglycerides [1]. The mechanism underlying hypertriglyceridaemia in NIDDM is still unclear. A majority of studies indicate that the elevation of plasma triglycerides in NIDDM result from an overproduction of VLDL triglyceride [2,3] and apo B [4,5] but it is unclear and a matter of controversy whether the increased production of VLDL particles is driven by a direct effect of hyperinsulinaemia or is a consequence of defective suppression of VLDL production by insulin [1]. In healthy subjects acute insulin administration suppresses VLDL apo B production [6][7][8]. Data on the suppression of VLDL production by insulin in insulin resistant states are controversial. Lewis et al. [6] showed an impaired suppression of VLDL apo B production in obese subjects while Cummings et al. [9] did not find any defect in the ability of insulin to suppress VLDL apo B production in patients with NIDDM.Two major subclasses of VLDL particles are recognised, large triglyceride-rich VLDL1 particles Diabetologia (1997) 40: 454-462 Defective regulation of triglyceride metabolism by insulin in the liver in NIDDM Summary Insulin administration to healthy subjects inhibits the production of very low density lipoprotein (VLDL)1 (Svedbergs flotation (Sf) rate 60-400) without affecting that of VLDL2 (Sf 20-60) subclass. This study was designed to test whether this hormonal action is impaired in non-insulin-dependent diabetes mellitus (NIDDM). We studied six men with NIDDM (age 53 ± 3 years, body mass index 27.0 ± 1.0 kg/m 2 , plasma triglycerides 1.89 ± 0.22 mmol/l) during an 8.5 h infusion of saline (control) and then in hyperinsulinaemic (serum insulin ∼ 540 pmol/l) conditions during 8.5 h infusions of glucose and insulin to give either hyper-and normoglycaemic conditions. [3-2 H]-leucine was used as tracer and kinetic constants derived using a non-steadystate multicompartmental model. Compared to the control study, patients with NIDDM reduced VLDL1 apo B production by only 3 ± 8 % after 8.5 h of hyperinsulinaemia (701 ± 102 vs 672 ± 94 mg/day respectively, NS) in hyperglycaemic conditions and by 9 ± 21 % under normoglycaemic conditions (603 ± 145 mg/day). In contrast, in normal subjects insulin induced a 50 ± 15 % decrement in VLDL1 apo B production (p < 0.05). Direct synthesis of VLDL2 apo B in patients with NIDDM was not markedly affected by insulin. We conclude that a contributory factor to hypertriglyceridaemia in NIDDM is the inability of insulin to inhibit acutely the release of VLDL1 from the liver, despite efficient suppression of serum nonesterfied fatty acids. [Diabetologia (1997) 40: 454-462]

Mathematical modelling of hepatic lipid metabolism

“…Stable isotope tracers have been used in conjunction with compartmental modelling to elucidate the kinetics of apoB (a key structural protein in lipoproteins) and triglyceride across VLDL subfractions simultaneously. Adiels' model [2] has proven to be a particularly useful tool in determining how physiological features impact VLDL size and dyslipidemia [4,3,39,38]. A more in-depth model of VLDL assembly has been proposed by Shorten and Upreti [50] in which the lipid composition of secreted VLDLs can be determined from uptake of individual free fatty acids after elongation and desaturation by liver enzymes.…”

Postprandial VLDL-triacylglycerol secretion is not suppressed in obese type 2 diabetic men

“…Previous studies of the effect of hyperinsulinaemia on VLDL-TAG secretion in type 2 diabetes have been inconsistent with reports showing impaired [11,12] or preserved [7] suppression. This may be attributed to the prolonged and supraphysiological exposure to insulin during hyperinsulinaemic clamps, which may override the impaired suppression in type 2 diabetes at more physiological insulin concentrations.…”

“…In contrast, acute hyperinsulinaemia leads to a rapid inhibition of hepatic VLDL-TAG secretion, as evidenced from data stemming from in vitro studies, animal models and human studies [6][7][8]10]. Some previous studies have shown that diabetic patients fail to downregulate secretion of large VLDL (VLDL 1 ) in an experimental setting with high-dose insulin infusion [11,12], whereas we reported a preserved suppressive effect of hyperinsulinaemia on VLDL-TAG secretion [7]. The impact of meal-induced hyperinsulinaemia, on the other hand, is less well understood.…”

Diabetes: mellitus or lipidus?

“…The major lipoprotein abnormality in Type 2 diabetes is increased VLDL synthesis, in part due to the hepatic recirculation of NEFA outflowing in excess from adipose tissue, and in part as a result of de novo synthesis in the presence of hyperglycaemia and hyperinsulinaemia. This occurs because the hepatic VLDL synthesis, usually suppressed by insulin, continues unabated in the condition of insulin resistance [70]. There is also an impediment in VLDL and chylomicron removal due to lower than normal lipoprotein lipase (LPL) activity in adipose tissue in diabetes [71,72,73].…”