Glucose-Dependent Insulinotropic Polypeptide May Enhance Fatty Acid Re-esterification in Subcutaneous Abdominal Adipose Tissue in Lean Humans

Eight healthy males with a mean age of 30 years ± 6 (means ± SD) and BMI of 23.2 ± 1.4 were studied. The ethics committee in Copenhagen Municipality approved the protocol (H-d-2008-043).

Each subject participated in randomized order in four different experiments separated by approximately 3 weeks. On two of the occasions, the subjects underwent a hyperinsulinemic (150–200 pmol/l)–hyperglycemic (6.5–7 mmol/l) (HI-HG) clamp with continuous infusion of either GIP (1.5 pmol⁻¹ · kg⁻¹ · min⁻¹) or saline during 300 min. On two other occasions, GIP or saline was infused alone during the study.

The subjects' habitual dietary intakes were recorded for 1 day before the first experiment and were replicated before subsequent experiments. The subjects arrived at 8:00 a.m. after having fasted for at least 12 h. The investigations were performed with the subjects in a supine position in a room kept at 24°C. A catheter was inserted into an antecubital vein for the infusion of GIP, glucose, insulin, or saline. The subjects were further catheterized in a subcutaneous vein on the anterior abdominal wall (9) and in a radial artery. After three baseline measurements, a continuous infusion of either GIP or saline alone or in combination with an HI-HG clamp was initiated at time = 0.

Synthetic human GIP (1–42) (Polypeptide Laboratories, Wolfenbüttel, Germany) was dissolved in sterilized water containing 2% human serum albumin (Human Albumin; CSL Behring, Germany) and sterile filtrated. Vial content was tested for sterility and bacterial endotoxins (European pharmacopoeia 2.6.14, Method C, turbidimetric kinetic method). The peptide was demonstrated to be >97% pure and identical to the natural human peptide by high-performance liquid chromatography, mass, and sequence analysis.

Insulin (Actrapid Human; Novo Nordisk, Denmark) was infused at a continuous rate of 10 μU · m⁻² · min⁻¹. During concomitant GIP infusion, the insulin infusion rate was adjusted according to the expected endogenous insulin secretion in response to GIP aiming toward plasma insulin levels comparable with those obtained during insulin infusion alone. Therefore, insulin was infused at a lower rate (7 μU · m⁻² · min⁻¹) under these conditions.

Adipose tissue metabolism was measured by Ficks principle (10). Adipose tissue blood flow (ATBF) was measured continuously by the ¹³³Xenon washout technique (11).

Arterial and venous blood samples were drawn simultaneously at time −30, −15, 0, and hereafter every 30 min until discontinuation of the infusion. Glucose concentrations were measured every 10 min for clamp adjustments.

Blood samples were analyzed for TAG, glycerol, FFA, and glucose by enzymatic methods modified to run on a Hitachi 912 automatic analyzer (Boehringer Mannheim). In addition, arterial blood was analyzed for GIP (12) and insulin and C-peptide by autoDELPHIA automatic fluoroimmunoassays (Wallac, Turku, Finland).

All results are presented as means ± SEM. Changes in concentrations with time and between the four experiments were compared using mixed linear models with repeated measures where time, treatment, and interaction time × treatment were included in the model as fixed factors, and the subjects were included as random effects. The t test for paired data was used for comparing GIP and saline experiments. A significance level of 0.05 was chosen. The Statistical Analysis Software, version 9.1, was used for statistical calculations.

During GIP alone and GIP in combination with the HI-HG clamp experiment, intact GIP concentrations reached physiological postprandial levels after ∼60 min at 68.3 ± 1.5 and 66.5 ± 1.6 pmol/l, respectively (Fig. 1 A). During infusion of saline alone and saline in combination with the HI-HG clamp experiment, GIP concentrations remained at fasting levels (14.7 ± 1.6 and 15.8 ± 1.8 pmol/l, respectively; P = not significant).

During the HI-HG clamp experiment with and without GIP, insulin concentrations increased rapidly during the first 30 min to a mean plateau of 169 ± 4 and 167 ± 2 pmol/l, respectively (P < 0.0001) (Fig. 1,C). C-peptide concentrations increased as well during the HI-HG clamp experiment with and without GIP from fasting levels at 493 ± 7 and 409 ± 9 pmol/l, respectively (P = not significant) (Fig. 1,D) to a mean plateau of 1,108 ± 27 and 635 ± 17 pmol/l, respectively (P < 0.0001). During GIP alone, insulin and C-peptide concentrations increased slightly but significantly during the first 60 min compared with saline (P = 0.02) (Fig. 1 C and D), while no difference occurred during 60–300 min.

During the HI-HG clamp experiments with and without GIP, glucose levels increased toward target levels during the first 60 min (6.7 ± 0.1 and 6.8 ± 0.2 mmol/l, respectively) (Fig. 1,B) and remained constant and similar during 60–300 min (6.6 ± 0.1 and 6.7 ± 0.1 mmol/l). There was no significant difference between the amounts of glucose infused during these two experiments (118 ± 11 and 110 ± 8 g). During GIP infusion alone, a decrease in glucose concentrations occurred during the first 60 min compared with saline infusion (P = 0.01) as a result of stimulation of endogenous insulin secretion during GIP infusion (Fig. 1 B). No significant differences in plasma glucose concentrations were seen during 60–300 min.

The arterial concentrations of the metabolites during the four experiments are given in Fig. 2. During the HI-HG clamp experiments with and without GIP, arterial TAG, FFA, and glycerol concentrations decreased significantly with no differences seen between the two experiments. During GIP or saline alone, arterial TAG concentrations remained constant and no differences were seen between the two experiments. Arterial FFA and glycerol concentrations did not differ significantly during saline and GIP infusions alone and increased steadily during the 300 min.

Figure 3 shows the subcutaneous abdominal adipose tissue blood flow during the four experiments. During the GIP and HI-HG clamp experiment, ATBF increased significantly during the first 30 min, reached a plateau of 3.9 ± 0.4 versus 1.4 ± 0.1 ml · min⁻¹ · (100 g tissue)⁻¹ after 90 min, and differed significantly from the saline and HI-HG clamp experiment (1.4 ± 0.1 ml · min⁻¹ · [100 g tissue]⁻¹, P < 0.0001). During the saline and HI-HG clamp experiment, ATBF remained virtually constant with a tendency toward a slight decrease at the end of the experiment. During GIP alone, a slight but significant ATBF increase was seen during the first 60 min when compared with saline (1.6 ± 0.02 vs. 1.1 ± 0.01 ml · [100 g tissue]⁻¹ · [60 min]⁻¹, respectively; P = 0.04) concomitant with elevated insulin levels. After 60 min, ATBF was similar during the two experiments and increased steadily and differed significantly from ATBF during the HI-HG clamp experiment without GIP during the last hour (P < 0.05).

Figure 4 shows net fluxes of TAG, FFA, glycerol, and glucose in adipose tissue. Baseline TAG, FFA, glycerol, and glucose fluxes were similar in the four experiments (Fig. 4,A, C, E, and G). TAG hydrolysis (P = 0.009) (Fig. 4,B) and glucose uptake (P < 0.03) (Fig. 4,H) increased and FFA output (P = 0.04) (Fig. 4,D) decreased significantly during the GIP and HI-HG clamp experiment compared with the HI-HG experiment without GIP. In addition, a decreased FFA-to-glycerol output ratio was seen during the GIP and HI-HG clamp experiment (P = 0.02) (Fig. 4 J) compared with the HI-HG clamp alone. No differences were seen in TAG, FFA, glycerol, and glucose fluxes between the GIP and saline experiments.

The major finding in the present study is that GIP in combination with hyperinsulinemia and hyperglycemia increased blood flow, glucose uptake, TAG hydrolysis, and FFA re-esterification, resulting in increased TAG deposition in the anterior, abdominal, subcutaneous adipose tissue. Under fasting conditions, GIP did not affect adipose tissue lipid metabolism.

Since the possible metabolic effects elicited via GIP in adipose tissue has not been described previously, we used a prolonged HI-HG clamp technique with plasma glucose and insulin concentrations similar to those found after ingestion of a carbohydrate-rich meal. It was found that during the HI-HG clamp experiment in combination with GIP, TAG hydrolysis increased significantly compared with the HI-HG clamp experiment without GIP. This effect was brought about concomitantly with a significant increase in the subcutaneous ATBF. ATBF increases significantly postprandially, and this increase appears to be of particular importance in the regulation of lipid metabolism by facilitating transport and deposition of lipids in adipose tissue (7). Insulin per se does not seem to affect ATBF, however insulin may indirectly stimulate ATBF. Recently, a study reported that ATBF was markedly higher after oral glucose than during the intravenous insulin-glucose infusions (13). The present findings suggest that GIP has vasoactive effects in adipose tissue, although the design of the study cannot rule out that other substances may play a role. C-peptide has been found to have dose-dependent vascular effects in skeletal muscle in the concentration range between 0–1 nmol/l (14). In the present experiments, the C-peptide concentration increased from about 0.4 to 0.7 nmol/l in the HI-HG clamp experiments without a concomitant increase in adipose tissue blood flow. In the HI-HG clamp experiment with GIP, the C-peptide concentration increased to about 1 nmol/l. However, in light of the missing flow increase in GIP HI-HG clamp experiment, it seems unlikely that the flow increase found in clamp experiments with GIP primarily is elicited via C-peptide.

Simultaneously with the increase in adipose tissue blood flow, an increase in adipose tissue TAG hydrolysis took place, probably reflecting an increased substrate supply to lipoprotein lipase (LPL). Samra et al. (15) have demonstrated that when adipose tissue blood flow increased by infusion of adrenaline, TAG hydrolysis increased exactly in parallel with increased blood flow, implying that TAG hydrolysis is normally limited by substrate delivery, which is consistent with our present study.

While TAG hydrolysis was significantly higher during the HI-HG clamp with GIP compared with the clamp without GIP (Fig. 4), the FFA release was lower in the clamp experiment with GIP. This suggests that FFA derived from LPL-mediated hydrolysis of the circulating TAG was directed into the adipose tissue probably to be esterified and stored, similar to what has been shown previously in subjects examined in the fed state (8). Concomitantly with the FFA uptake, there was an increase in glucose uptake during GIP in combination with insulin. Taken together, these results indicate that GIP directs FFA released by LPL-mediated TAG hydrolysis toward tissue uptake instead of release to the circulation.

In vitro studies have shown that GIP stimulates lipolysis (16,17). However, McIntosh et al. (16) found that GIP-stimulated lipolysis in 3T3 cells was inhibited by insulin, suggesting that the lipolytic effect of GIP is weaker than the antilipolytic effect of insulin. In the present study, we could not demonstrate any lipolytic effect of GIP under any of the experimental circumstances.

In conclusion, GIP in combination with hyperinsulinemia and light hyperglycemia increased adipose tissue blood flow, increased adipose tissue glucose uptake, increased FFA re-esterification, and thus resulted in increased adipose tissue TAG deposition.

No potential conflicts of interest relevant to this article were reported.

M.A. researched the data, contributed to the discussion, and wrote the manuscript. L.S. researched the data, contributed to the discussion, and reviewed the manuscript. S.M. and B.S. reviewed the manuscript. J.J.H. contributed to the discussion and reviewed the manuscript. J.B. researched the data, contributed to the discussion, and wrote the manuscript.

The authors gratefully acknowledge the technical assistance of Marianne Falck and Jannie Møller Larsen, Department of Clinical Physiology, Bispebjerg Hospital, and Jeppe Bach and Regitze Kraunsøe, Department of Biomedical Sciences, Copenhagen University.