Impaired Plasma Nonesterified Fatty Acid Tolerance Is an Early Defect in the Natural History of Type 2 Diabetes

Abstract

Context: Abnormal plasma nonesterified fatty acid (NEFA) metabolism may play a role in the development of type 2 diabetes.

Objectives: Our objectives were to demonstrate whether there is a defect in insulin-mediated suppression of plasma NEFA appearance (Ra_NEFA) and oxidation (Ox_NEFA) during enhanced intravascular triacylglycerol lipolysis early in the natural history of type 2 diabetes, and if so, to determine whether other mechanisms than reduced insulin-mediated suppression of intracellular lipolysis are involved.

Design: These are cross-sectional studies.

Setting: The studies were performed at an academic clinical research center.

Participants: Nine healthy subjects with both parents with type 2 diabetes (FH+) and nine healthy subjects with no first-degree relatives with type 2 diabetes (FH−) with similar anthropometric features were included in the studies.

Interventions: Pancreatic clamps and iv infusion of stable isotopic tracers ([1,1,2,3,3-²H₅]-glycerol and [U-¹³C]-palmitate or [1,2-¹³C]-acetate) were performed while intravascular triacylglycerol lipolysis was simultaneously clamped by iv infusion of heparin plus Intralipid at low (fasting) and high insulin levels. Oral nicotinic acid (NA) was used to inhibit intracellular lipolysis.

Main Outcome Measures: Ra_NEFA and Ox_NEFA were determined.

Results: During heparin plus Intralipid infusion at high plasma insulin levels, and despite similar intravascular lipolytic rates, FH+ had higher Ra_NEFA and Ox_NEFA than FH− (Ra_NEFA: 17.4 ± 6.3 vs. 9.2 ± 4.2; Ox_NEFA: 4.5 ± 1.8 vs. 2.3 ± 1.5 μmol/kg lean body mass/min), independent of NA intake, gender, age, and body composition. In the presence of NA, insulin-mediated suppression of Ra_NEFA was still observed in FH−, but not in FH+.

Conclusions: Increased Ra_NEFA and Ox_NEFA during intravascular lipolysis at high insulin levels occur early in the natural history of type 2 diabetes.

Healthy offspring with two type 2 diabetes parents display an increase in plasma non-esterified fatty acid appearance and oxidation during stimulated intravascular triacylglycerol lipolysis and hyperinsulinemia, suggesting that impaired fatty acid storage in adipose tissues occur early in the natural history of type 2 diabetes and may contribute to ectopic fat accumulation in these individuals.

Excess exposure of nonadipose tissues to high plasma nonesterified fatty acids (NEFAs) decreases insulin-mediated glucose disposal (1), reduces glucose transport into muscle (2), stimulates endogenous glucose production (3), reduces hepatic insulin clearance (4), and impairs glucose-stimulated insulin secretion (5,6,7,8,9). The offspring of two parents with type 2 diabetes have a very high risk of developing overt type 2 diabetes (10). They also have increased postprandial triacylglycerol (TG) levels and blunted early postprandial lowering of NEFA (11). Thus, it has been suggested that in subjects with a family history of type 2 diabetes, impaired postprandial reduction of NEFA by insulin could be involved in the development of insulin resistance and type 2 diabetes through increased exposure of nonadipose tissue to NEFAs (5).

The effect of insulin on postprandial NEFA metabolism is complex. In the postprandial state, the suppression of NEFA by insulin could be altered early on in the natural history of type 2 diabetes, although this has not been previously addressed to our knowledge. In adipose tissue, insulin stimulates lipoprotein lipase (LPL)-mediated lipolysis of chylomicron and very low density lipoprotein (VLDL) TG in the circulation (i.e. stimulates intravascular TG lipolysis) (12), suppresses intracellular adipose tissue lipolysis (13), and may stimulate esterification of NEFA in adipose tissue. Thus, insulin may contribute to adipose tissue uptake of NEFAs generated from intravascular TG lipolysis (13,14), i.e. insulin may reduce NEFA spillover.

Our primary aim was to demonstrate the early occurrence in the natural history of type 2 diabetes of a defect in insulin-mediated suppression of plasma NEFA appearance rate (Ra_NEFA) and plasma NEFA oxidation rate (Ox_NEFA) during enhanced intravascular TG lipolysis. If present, we also aimed to establish whether this defect was independent of insulin-mediated regulation of both LPL and intracellular adipose tissue lipolysis, i.e. whether reduced insulin-mediated adipose tissue uptake of NEFA generated during intravascular TG lipolysis occurs early on in the natural history of type 2 diabetes.

Subjects and Methods

Subjects

Nine healthy subjects with both parents with type 2 diabetes [subjects with both parents whose diabetes started after age 30, but not requiring insulin therapy at diagnosis (FH+)] and nine healthy subjects with no first-degree relatives with type 2 diabetes (FH−) participated in the studies (Table 1). All subjects tested negative for diabetes based on a 75-g oral glucose tolerance test (15). No subject was taking any chronic medication other than stable thyroid hormone replacement therapy, had no active medical condition known to affect lipid levels or insulin sensitivity, and had no cardiovascular disease. All female subjects were premenopausal and participated during the follicular phase of their menstrual cycle. Using the questionnaire of Sallis et al. (16), habitual energy expenditure was similar in both groups (Table 1). Written informed consent was obtained before participation, in accordance with the Declaration of Helsinki and all applicable laws and regulations. Experimental protocols were duly approved by the Research Ethics Institutional Review Board at the Centre hospitalier universitaire de Sherbrooke.

Table 1.

Characteristics of the study participants

	FH+ (n = 9)		FH− (n = 9)		P value
	Mean ± se	Range	Mean ± se	Range
Age (yr)	41.2 ± 11.7	24–56	34.7 ± 10.8	21–47	0.25
Gender (M/F)	4/5		6/3		0.64
Height (cm)	166.7 ± 10.5	154.0–185.0	170.5 ± 9.9	152.0–185.5	0.44
Weight (kg)	70.1 ± 9.3	55.5–84.2	71.0 ± 11.4	59.1–93.0	0.85
BMI (kg/m2)	25.2 ± 3.0	20.0–28.6	24.4 ± 3.3	20.0–30.9	0.59
WC (cm)	82 ± 9	69–95	84 ± 6	78–98	0.64
LBM (kg)	52.4 ± 14.1	36.7–74.1	56.3 ± 10.8	38.9–74.4	0.52
Non-LBM (kg)	17.6 ± 7.8	5.0–29.3	14.6 ± 11.4	4.9–36.4	0.53
Estimated EE (kcal/d)	2417 ± 381	1935–2932	2713 ± 522	1902–3468	0.19
TG (mmol/liter)	1.70 ± 1.59	0.48–5.06	1.09 ± 0.42	0.58–1.86	0.28
TC (mmol/liter)	4.92 ± 0.99	3.57–6.55	4.34 ± 0.51	3.84–5.26	0.14
HDL-C (mmol/liter)	1.32 ± 0.45	0.79–2.07	1.23 ± 0.27	0.90–1.77	0.60
LDL-C (mmol/liter)	2.82 ± 0.75	1.64–3.86	2.62 ± 0.54	1.64–3.35	0.53

EE, Energy expenditure; F, female; HDL-C, high-density lipoprotein cholesterol; LBM, lean body mass; LDL-C, low-density lipoprotein cholesterol; M, male; TC, total plasma cholesterol; WC, waist circumference. 

Experimental protocols

All subjects participated in four separate study protocols (A/B and C/D), each conducted 3–4 wk apart. These protocols were modeled on our previously described protocol, which produces a sustained and similar elevation of intravascular TG lipolysis at different plasma insulin levels (17). On arrival, body weight and height were measured, and lean body mass was determined by bioelectrical impedance (Hydra ECF/ICF; Xitron Technologies, San Diego, CA). A 4-h period of intravascular TG lipolysis was achieved using an iv infusion of heparin (250 U/h Hepalean; Organon Teknika, Scarborough, Ontario, Canada) and Intralipid 20% (40 ml/h; Baxter, Mississauga, Ontario, Canada) under normoinsulinemic (Novolin R; Novo Nordisk, Bagsværd, Denmark; 0.05 mU.kg⁻¹·min⁻¹ continuous infusion; protocols A and C) or hyperinsulinemic (Novolin R, 0.8 mU/kg primed, 1.2 mU.kg⁻¹·min⁻¹ continuous infusion; protocols B and D) conditions with infusion of octreotide acetate and human GH (17). Although glucagon is suppressed by octreotide acetate, it was not replaced to avoid insulin secretion breakthrough at low insulin infusion rates (18), and because of its minimal effect on NEFA metabolism in humans (19).

During protocols A and B, a constant infusion of [U-¹³C]K palmitate (0.01 μmol/kg/min in 100 ml 25% human serum albumin; Cambridge Isotopes Laboratories Inc., Andover, MA) was started at time zero. In protocols C and D, [1,2-¹³C]sodium acetate (0.08 μmol/kg/min; Cambridge Isotopes Laboratories Inc.) was infused in place of the [U-¹³C]K palmitate (17). Mean variation of plasma glucose, insulin, and NEFA levels between protocols to determine fractional acetate recovery and those to determine plasma NEFA metabolism were less than 5%. The choice of [U-¹³C]palmitate as the NEFA tracer was based on the following: 1) palmitate, oleate, and linoleate are the most prevalent NEFAs, both in human plasma and in Intralipid, and they have similar plasma clearance rates in humans (20); 2) palmitate was previously used to trace total NEFA turnover in humans after oral fat intake (13,21); and 3) during iv infusion of heparin plus Intralipid (HI) in humans, we have shown that NEFA appearance rates determined with [U-¹³C]palmitate predict very well the rates determined using simultaneous palmitate and linoleate tracers (r = 0.90; P < 0.001) (22). To quantify plasma glycerol flux as a measure of whole body lipolysis, a primed continuous (1.6 μmol/kg; 0.11 μmol/kg/min) infusion of [1,1,2,3,3-²H₅]glycerol (Cambridge Isotopes Laboratories Inc.) was also administered during protocols A and B (23). To prime the bicarbonate pool, a bolus infusion of sterile [1-¹³C]NaHCO₃ 1.2 μmol/kg (Cambridge Isotopes Laboratories Inc.) was administered at time zero in all four protocols (24). All tracers were pretested for sterility and nonpyrogenicity.

During the last 2 h of the protocols, nicotinic acid (NA) was given orally (100 mg at 120 and 200 min, and 150 mg at 150 and 180 min) because it has been previously shown to profoundly suppress intracellular adipose tissue lipolysis for up to 4 h (25). As also previously described (17), this allowed us to discriminate between insulin-mediated inhibition of intracellular lipolysis as opposed to insulin-mediated enhancement of adipose tissue uptake of NEFA derived from intravascular TG lipolysis.

Each protocol started with 30-min bed rest, after which blood samples were taken at 10-min intervals during baseline and steady states, i.e. during the last 30 min of the first 2 h (without NA), and again during the last 30 min of the 4-h clamp period (with NA). Blood and breath samples were collected, and urine nitrogen excretion, oxygen consumption (VO₂), and CO₂ excretion (VCO₂) were measured as previously described (17).

Laboratory determinations

Glucose, insulin, glucagon, GH, total plasma NEFA, and TG were measured as previously described (17). Plasma glycerol and [1,1,2,3,3-²H₅]glycerol enrichment were measured by gas chromatography-mass spectrometry. Individual plasma NEFA (palmitate, linoleate, oleate) and [U-¹³C]palmitate enrichment in plasma were measured by liquid chromatography-mass spectrometry (17). Intraassay and interassay coefficients of variation were under 6.1% for all assays. Breath ¹³CO₂/¹²CO₂ ratio was determined by isotope ratio mass spectrometry (Sercon Ltd., Crewe, Cheshire, UK).

Calculations

The plasma palmitate appearance rate (Ra_palmitate) was calculated from the C16:0 mass +16 enrichment of plasma palmitate as previously described (17). Ra_NEFA was calculated by multiplying Ra_palmitate by the ratio of NEFA to palmitate concentration. The plasma glycerol appearance rate (Ra_glycerol) was determined from plasma glycerol enrichment (mass +5) corrected for infusion rate of free glycerol contained in the Intralipid infusate (17). Fractional palmitate oxidation (Fox_palmitate) was determined as described (26) with correction for the fractional acetate recovery assessed in each individual (17). The plasma palmitate oxidation rate (Ox_palmitate) was then calculated by multiplying Fox_palmitate by Ra_palmitate. Ox_NEFA was calculated by multiplying Ox_palmitate by the ratio of NEFA to palmitate concentration (22). Net total body carbohydrate and fatty acid oxidation (CHOox and FATox, respectively) were estimated by indirect calorimetry (27). Insulin sensitivity was assessed at high insulin level by dividing the glucose infusion rate by plasma insulin level.

Statistical analysis

Data at 10-min intervals were averaged for baseline and the last 30-min clamp with and without NA at fasting and high insulin, and were expressed as mean ± sd unless otherwise specified. Within-group comparisons were performed by ANOVA for repeated measures with Scheffe’s post hoc test. ANOVA was used for between-group comparisons at baseline, with and without NA at pancreatic clamp, at fasting plasma insulin level (INS_LOW), and during pancreatic clamp at hyperinsulinemia (INS_HIGH). Gender distribution between FH+ and FH− was compared by Fisher’s exact test. Multivariate linear regression was used to assess the effect of FH+ or FH− on NEFA appearance during INS_LOW and INS_HIGH, both before and after correcting for potential confounding variables in the models. A two-tailed P value less than 0.05 was considered significant. All analyses were performed with the SAS software for Windows, version 9.1.3 (SAS Institute Inc., Cary, NC).

Results

Plasma glucose, insulin, TG, GH, and glucagon levels (Table 2)

Table 2.

Plasma levels of glucose, insulin, TG, GH, and glucagon

	Group	Low insulin level protocol			High insulin level protocol
		Baseline	INSLOW	INSLOW NA	Baseline	INSHIGH	INSHIGH NA
Glucose (mmol/liter)	FH−	5.0 ± 0.6	5.2 ± 0.9	5.4 ± 0.9	5.1 ± 0.3	5.1 ± 0.6	5.5 ± 0.9
	FH+	5.3 ± 0.6	6.1 ± 0.6a	6.0 ± 0.6	5.5 ± 0.6	5.6 ± 0.6	5.4 ± 0.6
Insulin (pmol/liter)	FH−	59 ± 30	56 ± 15	62 ± 21	74 ± 30	468 ± 129b,c	441 ± 123b,d
	FH+	72 ± 39	57 ± 12	57 ± 15	80 ± 42	474 ± 87b,c	443 ± 78b,d
TG (mmol/liter)	FH−	0.66 ± 0.24	1.36 ± 0.39b	1.20 ± 0.36b	0.68 ± 0.24	1.33 ± 0.39b	1.16 ± 0.30b
	FH+	1.23 ± 1.02	2.54 ± 2.01b	2.39 ± 2.04b	1.42 ± 1.02a	2.66 ± 1.92b	2.59 ± 1.92b
GH (μg/liter)	FH−	0.5 ± 0.6	0.8 ± 0.3	0.8 ± 0.3	0.7 ± 0.6	0.6 ± 0.3	0.7 ± 0.3
	FH+	1.4 ± 1.8	0.6 ± 0.3a	1.0 ± 1.2	1.2 ± 1.8	0.6 ± 0.3	0.9 ± 0.9
Glucagon (ng/liter)	FH−	70 ± 30	36 ± 15b	35 ± 18b	68 ± 21	38 ± 15b	29 ± 15b
	FH+	68 ± 18	31 ± 15b	29 ± 12b	70 ± 18	28 ± 12b	27 ± 15b

Data are mean ± sd. 

^aP < 0.05 vs. FH− by ANOVA. 

^bP < 0.05 vs. baseline by ANOVA with Scheffe’s test. 

^cP < 0.05 vs. INS_LOW by ANOVA with Scheffe’s test. 

^dP < 0.05 vs. INS_LOW NA by ANOVA with Scheffe’s test. 

During INS_LOW, plasma glucose did not significantly change from baseline in the FH− group but increased in the FH+ group (P = 0.01). By design, plasma insulin was elevated from baseline during INS_HIGH, a change not affected by NA in either group. Plasma TG tended to be higher in the FH+ group (significant only at baseline for INS_HIGH and at INS_HIGH NA). As expected, plasma TG significantly increased during HI infusion in both groups and was unaffected by NA. Plasma GH was largely unaffected by the protocols. Plasma glucagon was similarly reduced from baseline in both groups and was not affected by NA. Insulin sensitivity was not significantly different between the two groups (0.067 ± 0.041 vs. 0.102 ± 0.058 μmol/kg lean body mass/min per pmol/liter in FH+ and FH−, respectively; P = 0.16).

Plasma concentration of NEFA and glycerol (Table 3)

Table 3.

Plasma levels of NEFA, palmitate, oleate, linoleate, and glycerol

	Group	Low insulin level protocol			High insulin level protocol
		Baseline	INSLOW	INSLOW NA	Baseline	INSHIGH	INSHIGH NA
NEFA (μmol/liter)	FH−	477 ± 171	1041 ± 210a	807 ± 159a,b	416 ± 192	470 ± 147b	440 ± 102c
	FH+	483 ± 117	1324 ± 297a,d	1027 ± 267a,b,d	480 ± 144	683 ± 195b,d	710 ± 213c,d
Palmitate (μmol/liter)	FH−	102 ± 33	164 ± 33a	125 ± 24b	92 ± 36	66 ± 21b	62 ± 15
	FH+	109 ± 30	186 ± 51a	134 ± 36b	102 ± 30	81 ± 27b	84 ± 24c,d
Oleate (μmol/liter)	FH−	207 ± 66	283 ± 69a	197 ± 51b	181 ± 69	105 ± 36b	91 ± 24a,c
	FH+	206 ± 63	323 ± 93a	215 ± 66b	180 ± 54	131 ± 48b	131 ± 48c,d
Linoleate (μmol/liter)	FH−	92 ± 93	443 ± 93a	407 ± 90a	84 ± 33	203 ± 75a,b	209 ± 54a,c
	FH+	91 ± 36	510 ± 210a	430 ± 162a	89 ± 33	264 ± 108a,b	282 ± 102a,c
Glycerol (μmol/liter)	FH−	67 ± 15	170 ± 15a	173 ± 18a	62 ± 9	149 ± 9a	155 ± 18a
	FH+	61 ± 12	186 ± 30a	166 ± 27a	69 ± 15	170 ± 36a	170 ± 30a

Data are mean ± sd. 

^aP < 0.05 vs. baseline by ANOVA with Scheffe’s test. 

^bP < 0.05 vs. INS_LOW by ANOVA with Scheffe’s test. 

^cP < 0.05 vs. INS_LOW NA by ANOVA with Scheffe’s test. 

^dP < 0.05 vs. FH− by ANOVA. 

Total NEFA, and plasma palmitate, oleate, and glycerol levels were similar at baseline in both groups. At INS_LOW, HI infusion significantly increased total NEFA, palmitate, oleate, and linoleate levels from baseline in both groups. During both INS_LOW and INS_HIGH, total NEFAs were significantly higher in the FH+ group. As expected after the HI infusion during INS_LOW, the increase in total NEFA was significantly reduced by NA to a similar extent in both groups. Increases in plasma palmitate and oleate, but not linoleate, were also significantly blunted by NA at INS_LOW in both groups. At INS_HIGH, total NEFA, palmitate, and oleate did not significantly change from baseline, whereas plasma linoleate increased significantly in both groups. Neither total NEFA, nor palmitate, oleate, or linoleate were significantly affected by NA at INS_HIGH in either group. As expected during HI infusion, in both groups, total NEFA, palmitate, oleate, and linoleate were significantly lower at INS_HIGH vs. INS_LOW. During INS_HIGH plus NA, plasma total NEFA, palmitate, and oleate were significantly higher in the FH+ group. In both groups, plasma glycerol was significantly increased from baseline to the same extent during INS_LOW or INS_HIGH, with or without NA.

Indirect calorimetry and net substrate oxidation rates (Table 4)

Table 4.

Indirect calorimetry and net substrate oxidation rates

	Group	Low insulin level protocol			High insulin level protocol
		Baseline	INSLOW	INSLOW NA	Baseline	INSHIGH	INSHIGH NA
VO2a	FH−	3.72 ± 0.57	3.74 ± 0.54	3.79 ± 0.63	3.74 ± 0.54	2.88 ± 0.57	3.85 ± 0.63
	FH+	4.20 ± 1.11	4.23 ± 0.99	4.37 ± 1.20	4.01 ± 0.90	4.27 ± 1.08	4.43 ± 1.23b
VCO2a	FH−	3.07 ± 0.45	3.05 ± 0.48	3.13 ± 0.51	3.15 ± 0.39	3.46 ± 0.48b,c	3.53 ± 0.45b,d
	FH+	3.49 ± 0.84	3.39 ± 0.72	3.47 ± 0.87	3.56 ± 0.87	3.88 ± 1.11	4.12 ± 1.41b,d
RQ	FH−	0.83 ± 0.06	0.82 ± 0.06	0.83 ± 0.06	0.85 ± 0.03	0.89 ± 0.06c	0.92 ± 0.03b,d
	FH+	0.84 ± 0.03	0.81 ± 0.03	0.80 ± 0.03	0.89 ± 0.06	0.91 ± 0.06c	0.92 ± 0.06d
CHOoxe	FH−	11.4 ± 4.2	10.4 ± 4.5	11.5 ± 4.2	13.0 ± 2.4	18.2 ± 5.4b,c	20.5 ± 3.6b,d
	FH+	13.5 ± 3.3	10.5 ± 3.3	9.9 ± 2.7	18.4 ± 7.2f	22.0 ± 10.2c	25.1 ± 14.7d
FAToxe	FH−	3.8 ± 1.2	4.0 ± 1.2	3.9 ± 1.2	3.4 ± 1.2	2.5 ± 1.5	1.9 ± 1.5d
	FH+	4.1 ± 1.8	4.8 ± 1.8	5.2 ± 2.1	2.6 ± 1.2	2.3 ± 1.5c	1.8 ± 1.8d

Data are mean ± sd. 

^aExpressed in ml/kg lean body mass/min. 

^bP < 0.05 vs. baseline by ANOVA with Scheffe’s test. 

^cP < 0.05 vs. INS_LOW by ANOVA with Scheffe’s test. 

^dP < 0.05 vs. INS_LOW NA by ANOVA with Scheffe’s test. 

^eExpressed in μmol/kg lean body mass/min. 

^fP < 0.05 vs. FH− by ANOVA. 

VO₂ did not change from baseline during any of the experimental protocols in the FH− group but significantly increased from baseline at INS_HIGH plus NA in FH+ subjects. VCO₂ and respiratory quotient (RQ) increased at INS_HIGH in both groups, especially with NA. At INS_LOW, net CHOox did not change with HI infusion, with or without NA, but, in both groups, was significantly higher than baseline at INS_HIGH. VO₂, VCO₂, RQ, and net CHOox and FATox were not significantly different between groups.

Glycerol and NEFA metabolic rates (Fig. 1)

Figure 1.

Ra_glycerol (A), Ra_palmitate (B), and oxidation rate (Ox_palmitate) (C). Plasma NEFA appearance rate (Ra_NEFA) (D) and oxidation rate (Ox_NEFA) (E). †, P < 0.05 vs. INS_LOW by ANOVA with Scheffe’s test. *, P < 0.05 vs. FH− by ANOVA. FH− (open bars). FH+ (closed bars). Data are mean ± se. LBM, Lean body mass.

Plasma glycerol enrichment (mass +5 tracer to tracee ratio) was similar throughout the experimental protocols in both groups (data not shown). Ra_glycerol corrected for exogenous free glycerol contained in Intralipid was also similar in both groups and was not significantly affected by either insulin or NA (Fig. 1A). Expressing Ra_glycerol per unit of nonlean body mass did not affect these results (data not shown).

In the FH− group, plasma palmitate enrichment (mass +16 tracer to tracee ratio) was 0.012 ± 0.006 on average and was similar throughout the experimental protocols (P = 0.19). In the FH+ group, plasma palmitate enrichment was lower during INS_LOW without NA compared with the other experimental conditions (0.006 ± 0.003, 0.008 ± 0.003, 0.009 ± 0.003, and 0.008 ± 0.003 in INS_LOW, INS_LOW NA, INS_HIGH, and INS_HIGH NA respectively; P = 0.002). During INS_HIGH, plasma palmitate enrichment was lower in FH+ (with or without NA; P = 0.07 and P = 0.05, respectively). At INS_LOW, Ra_palmitate was similar in both groups and tended to be lower after NA. At INS_HIGH, Ra_palmitate was significantly higher in FH+ with or without NA, and was lower than at INS_LOW (Fig. 1B). Fox_palmitate was not significantly changed by insulin or NA and was similar in both groups (data not shown). During INS_LOW, Ox_palmitate was similar in both groups but tended to be lower with NA. During INS_HIGH, Ox_palmitate was significantly lower than during INS_LOW, whereas NA blunted this difference in both groups. Ox_palmitate was significantly higher in FH+ subjects, but only during INS_HIGH (Fig. 1C). In FH− subjects, Ra_NEFA and Ox_NEFA tended to be reduced by NA at INS_LOW, and were significantly reduced by insulin with and without NA. Neither NA nor insulin significantly changed Ra_NEFA or Ox_NEFA in FH+ subjects. Ra_NEFA and Ox_NEFA were significantly higher in FH+ subjects during INS_HIGH, but not during INS_LOW (Fig. 1, D and E).

Effect of family history of type 2 diabetes on NEFA metabolism with and without adjustment for other potential confounding variables

During INS_LOW, FH+ subjects had significantly higher NEFA (P = 0.03), but not higher Ra_NEFA or Ox_NEFA, an effect that was no longer significant after accounting for age or gender. During INS_HIGH, the FH+ group had significantly higher NEFA (P = 0.02), higher Ra_NEFA (P = 0.005), and higher Ox_NEFA (P = 0.01) compared with the FH− group. These associations remained significant after correction for age, gender, body mass index (BMI), nonlean body mass, lean body mass, or waist circumference. Partial mismatch in gender between FH+ and FH− was unlikely to be responsible for the differences we observed at high insulin levels because plasma NEFA levels, Ra_NEFA and Ox_NEFA were not significantly different between men and women (data not shown). However, we found a significant inverse correlation between plasma NEFA level (r = −0.52; P = 0.03), Ra_NEFA (r = −0.51; P = 0.03), and Ox_NEFA (r = −0.47; P = 0.05), and insulin sensitivity determined during INS_HIGH.

Comparison of the effects of insulin with and without NA on Ra_NEFA (Fig. 2)

Figure 2.

NA-mediated change (INS_LOW NA minus INS_LOW) (A), insulin-mediated change (INS_HIGH minus INS_LOW) (B), and insulin-mediated change during NA intake (INS_HIGH NA minus INS_LOW NA) (C) in Ra_palmitate. NA-mediated change (D), insulin-mediated change (E), and insulin-mediated change during NA intake (residual) (F) in plasma NEFA appearance rate (Ra_NEFA). FH− (open bars). FH+ (closed bars). P values are from comparison of the two groups of participants by ANOVA. Data are mean ± se. LBM, Lean body mass.

Reduction of Ra_palmitate and Ra_NEFA (Fig. 2, A and D, respectively) by NA at INS_LOW was not significantly different between the FH+ and FH− groups. In the absence of NA, insulin-mediated reductions of Ra_palmitate or Ra_NEFA (INS_LOW minus INS_HIGH) were also not significantly different between the two groups (Fig. 2, B and E, respectively). In contrast in FH−, insulin reduced Ra_palmitate and Ra_NEFA in the presence of NA (INS_LOW NA minus INS_HIGH NA), but these effects were absent in FH+ (P = 0.03 and P = 0.06 for difference in change of Ra_palmitate and Ra_NEFA, respectively, between FH+ vs. FH− subjects) (Fig. 2, C and F).

Discussion

We found that impaired suppression of systemic Ra_NEFA by insulin during enhanced intravascular TG lipolysis occurs before the development of overt hyperglycemia and hyperinsulinemia in the natural history of type 2 diabetes, and is independent of obesity. Our experimental protocol induced similar increases in intravascular TG lipolysis in both groups, but FH+ subjects displayed significantly higher NEFA levels during high insulin infusion, an effect due to an increase in Ra_NEFA. Increased Ra_NEFA was associated with higher Ox_NEFA independently of age, gender, or body composition. During INS_LOW, although NEFA levels were also higher in FH+, the difference in Ra_NEFA and Ox_NEFA between FH+ and FH− was not significant. Thus, our results suggest that postprandial elevation of NEFA observed in FH+ subjects (11) may occur, at least in part, from failure of insulin to fully suppress systemic appearance of NEFA, a defect that appears to be independent of intravascular TG lipolysis. Ra_NEFA is elevated in the postprandial state in obese patients with type 2 diabetes compared with younger and leaner healthy individuals. Our results demonstrate that impaired suppression of systemic Ra_NEFA by insulin during enhanced intravascular TG lipolysis occurs before the development of overt hyperglycemia and hyperinsulinemia in the natural history of type 2 diabetes, and is independent of obesity. The results of the present study apply to individuals with BMIs ranging from 20.0–30.9 kg/m².

Insulin-mediated reduction of Ra_NEFA during HI infusion can be totally accounted for by inhibition of intracellular adipose tissue lipolysis in FH+ but not in FH− subjects. Even after NEFA suppression by NA, a further insulin-mediated decline in Ra_NEFA still occurred in the FH− group. NA resulted in profound suppression of plasma NEFA levels (25) and intracellular lipolysis (17). Furthermore, NA-induced suppression of Ra_NEFA did not differ significantly between FH+ or FH− subjects. NA reduced Ra_NEFA through a mechanism independent of and competing with that of insulin (28). We have also shown previously that this HI iv infusion protocol resulted in steady-state plasma NEFA levels and kinetics between 90 and 240 min during infusion (22). Thus, we believe that suppression of intracellular lipolysis together with steady stimulation of intravascular lipolysis was achieved in the present protocol. Our results suggest that in the presence of experimentally increased intravascular TG lipolysis, some component of insulin-mediated reduction of Ra_NEFA in FH− subjects results from a mechanism other than suppression of adipose tissue lipolysis. Moreover, our data suggest that the observed increase in Ra_NEFA during hyperinsulinemia and HI infusion in FH+ subjects resulted, at least in part, from failure of insulin to act on this NEFA-suppressing mechanism. Direct adipose tissue uptake of NEFA was recently demonstrated in vivo in humans (21,29). As previously suggested (5), insulin-facilitated uptake of NEFA from intravascular TG lipolysis in the local microcirculation of adipose tissue (14) could possibly be impaired early on in the natural history of type 2 diabetes. Therefore, we propose that NEFA uptake by adipocytes is probably defective in FH+.

The increased oxidative disposal of NEFA in the FH+ group observed in the present study is also consistent with higher exposure for organs known to oxidize NEFA (e.g. liver, heart, and skeletal muscle) in subjects at high risk for developing type 2 diabetes. Increased postprandial deposition of dietary fatty acids in liver and skeletal muscle was previously reported in subjects with type 2 diabetes (30). In healthy subjects, dietary fatty acids start to appear in plasma TG within hours of oral intake (21,31,32). Indeed, a fraction of plasma NEFA derived from intravascular TG lipolysis is available in the systemic circulation during both the fasting and postprandial state (21,33,34). One potential mechanism by which this increased NEFA availability may occur is reduced insulin-mediated suppression of NEFA spillover from intravascular TG lipolysis during the postprandial state.

Despite higher Ox_NEFA in FH+, net FATox in our protocol was not different between the two groups. Thus, a larger fraction of the fatty acids oxidized in FH+ subjects during the postprandial state may come from the circulation, with consequent sparing of lean tissue intracellular stores of TG. Over time, this NEFA overload from plasma could potentially contribute to intramyocellular lipid accumulation, typical of insulin-resistant individuals and seen early in the offspring of type 2 diabetics (35). Such a defect in NEFA disposal in adipose tissues could contribute to the gradual development or maintenance of insulin resistance in muscle, and to the worsening of β-cell failure and glucose tolerance over time (5). It might also contribute to enhanced susceptibility to the lipid-induced β-cell failure observed in the offspring of type 2 diabetics (7). Others have shown that the insulin-resistant offspring of type 2 diabetics display impaired skeletal muscle oxidative metabolism during fasting (36), an effect associated with impaired muscle mitochondrial substrate oxidation (37). Thus, both impaired muscle oxidative metabolism together with altered Ra_NEFA regulation by insulin during the postprandial state possibly occur in the offspring of type 2 diabetics and predispose them to ectopic fat accumulation.

Partial mismatch of gender between FH+ and FH− subjects is a potential limitation of our study. Our heparin-Intralipid-insulin infusion model with hyperinsulinemia may also have caused a further increase in LPL activity in the adipose tissue microcirculation, an effect that could have been different in the FH+ vs. FH− subjects (38). However, we found no difference in Ra_glycerol after correction for exogenous glycerol infusion (a marker of total TG lipolysis) in FH+ or FH− subjects during the experimental protocols. We cannot totally exclude that higher plasma TG levels in FH+ subjects may have caused more intravascular TG lipolysis in the microcirculation of other tissues that can use glycerol, i.e. skeletal muscle, heart, or liver, an effect that would increase NEFA appearance in the face of similar glycerol appearance rates. However, this seems unlikely due to the net spillover of NEFA from adipose tissue, but not from muscle, in the postprandial state in healthy humans (21), suggesting that the most important source of NEFA spillover is adipose tissue. Nevertheless, this point suggests that lean tissue in FH+ subjects metabolizes more circulating NEFA despite possible overexposure to fatty acids produced from intravascular TG lipolysis in the local microcirculation.

The metabolic abnormalities observed in FH+ participants in our study were subtle and observed only at high insulin levels during iv infusion of HI. Of note, the FH+ subjects in our study tended to be more insulin resistant. The significant inverse correlations between plasma NEFA levels, Ra_NEFA, Ox_NEFA, and insulin sensitivity suggest that plasma NEFA intolerance occurs concomitantly with early decrease in insulin-mediated glucose disappearance. Whether NEFA intolerance occurs at this early stage under more physiological conditions (i.e. during the postprandial state) will require more studies.

In conclusion, under experimental hyperinsulinemia, we observed that nondiabetic offspring of type 2 diabetics have significantly increased Ra_NEFA and Ox_NEFA during stimulated intravascular TG lipolysis. Our results suggest that defective suppression of systemic NEFA spillover from intravascular TG lipolysis by insulin occurs early on in the natural history of type 2 diabetes and may contribute to ectopic fat storage in these individuals.