Europ.J.clin.Invest. 5, 521-526 (1975)

Effect of Lipids on Insulin, Growth Hormone and Exocrine Pancreatic Secretion in Man * S. Raptis, H.C. Dollinger, L. von Berger, J. Kissing, K.E. Schroder, U. Klor and E.F. Pfeiffer Department of Internal Medicine Endocrinology and Metabolism, the Section of Gastroenterology and the Department of Nutrition and Metabolism Center of Internal Medicine and Paediatrics, University of Ulm, Ulm/Donau, Germany Received: May 30, 1974, and in revised form: January 22, 1975

Abstract. Influences of fat on release of insulin, growth hormone and pancreatic enzyme secretion were studied in 35 metabolically healthy subjects. A fat solution containing 40 g of soy bean oil was administered, I.V., orally and intraduodenally. In all cases there was a similar increase of insulin but the rise in serum insulin after oral or intraduodenal fat administration was not related to the changes in plasma free fatty acids, free glycerol and triglyceride levels. Blood sugar responded according to insulin secretion. The route of fat administration may possibly influence growth hormone secretion. Following intraduodenal fat administration volume and bicarbonate contents of the duodenal juice rose slightly whereas trypsin and bilirubin content increased considerably. These results suggest that insulin secretion after oral or intraduodenal administration of fat is influenced by intestinal factors. Cholecystokinin-pancreozymin and gastric inhibitory polypeptide are qualified to serve as such factors. Key words: Insulin, growth hormone, exocrine pancreatic secretin, secretion, cholecystokinin-pancreozymin, free fatty acids, free glycerol, triglycerides, blood glucose, intestinal hormones.

It has been shown that small amounts of glucose given intrav'enously cause glycosuria ( I ) , but that oral administration of much larger amounts does not (2). Oral glucose seems to activate substances in the gastrointestinal tract which accelerate utilization of sugar. The existence of such factors is further supported by the fact that insulin secretion increases more after oral glucose than I.V. administration of an equal amounts of sugar (3, 4, 5). This difference in insulin secretion was also found after administration of amino acids in animals (6) and in man ( 7 , 8). In both animals and man it has been shown that intestinal factors such as secretin, cholecystokinin-pancreozymin (CCK/PZ), gastric inhibitory polypeptide (GIP) , gastrin, glucagon, serotonid and duodenal extracts can stimulate insulin secretion (9, 10, l l , 12). Furthermore increases in insulin secretion (13, 14, 15, 16, 17, 18, 19) and diminished secretions of growth hormone (20, 21, 22, 23) were found in

* Presented in abstract form at the annual meeting of the European Association for the Study of Diabetes, Madrid, September 6 - 8, 1972, and in part at the 8th Congress of the International Diabetes Federation, Brussels, July 15 20, 1973. Parts of this paper are contained in J. Kissing's doctoral thesis.

-

Supported by Deutsche Forschungsgemeinschaft, Bad Godesberg, Germany. SFB 87/D 8 Endocrinology, Ulm.

v i t r o and in v i v o following administration of

lipids. The purpose of this study was therefore to find out whether there is a difference in insulin and growth hormone secretion after oral and intravenous administration of fat in man, and whether this difference can be explained by the release of intestinal hormones.

Material and Methods Thirty-five metabolically healthy volunteers were examined. In 10, 200 ml of commercially available 20 percent Intralipid(R) Vitrum (Deutsche Kabi GmbH, 8 Miinchen 15) containing a total of 40 g of fractionated soy bean oil, 2.4 g of fractionated lecithin from egg yolk and 5 g glycerol was administered by constant intravenous infusion of 120 min. duration. On another day the subjects ingested a similar amount of the lipid preparation within 5 min. The solution was isotonic with blood and the fat particles corresponding in size to the natural chylomicrons, were eliminated from the blood stream at the same rate (24). Blood samples were taken for determination of triglyceride (TG) , free glycerol (FG) and free fatty acids (FFA), lipoprotein electrophoresis, blood sugar (BS), inarmnologically measurable insulin (IMI) and growth hormone (HGH) at certain intervals. The TG were determined according to the method of Eggstein and Kreutz (25), FG to the method of Wie-

S. Raptis e t a l . : Effects of Lipids on Insulin

522 land (26), FFA to that of Dole and Meinertz (27). The lipoprotein electrophoresis was ascertained after Rapp and Kahlke (28). The BS was measured with the hexokinase method of Schmidt (29), the IMI by that of Melani and his associates (30) and HGH according to Glick and his colleagues ( 3 1 ) . In 10 additional subjects duodenal content was aspirated during a constant secretin (0.5 U per kilogram of body weight) infusion, and the volume as well as the bicarbonate, trypsin and bilirubin content determined. Secretin was obtained from GIH, Research Unit, Karolinska, Stockholm (batch No. 15181). The bicarbonate content of the aspirated duodenal juice was determined according to Preshaw and Grossman (321, and trypsin activity according to Willig and Kijrber (33). The bilirubin content of the aspirate was ascertained with a method modified from that of Jendrassik and GrSf(34). Secretin was administered intravenously for two hours; during the second hour the fat solution was intraduodenally infused. For this test a special three-lumen tube was used, with one lumen serving to aspirate the gastric juice, the second lumen to aspirate the duodenal contents, and the third to instill the fat solution at a site distal to the point of aspiration. An inflatable rubber balloon was placed between the second and third tube openings to prevent reflux of the instilled fat solution. Fifteen subjects were given a normal saline solution intravenously, orally or intraduodenally to serve as a control group. Statistical evaluation was carried out using the Student paired t-test. The results given represent mean values 2

m.

Results

Upon the I . V . infusion of fat, TG, FG and FFA increased significantly in the plasma within 20 - 50 min. The maximum of the three parameters was attained after 120 min. (Fig. 1). The plasma level of TG was increased T-fold, that of FG 12-fold, and the FFA level was doubled. Upon the oral administration of fat all three parameters also rose significantly, NF and FFA within 140 - 160 minutes, FG being elevated after 80 minutes. The plasma level of TG following oral administration was doubled, that of FG rose only slightly, and the FFA level was also doubled (Fig. 1 ) . In the lipoprotein electrophoresis, 10 min. after the fat infusion, a strong chylomicron band was found, which disappeared 150 min. later. The pre-B-lipoproteins indicated a continuous increase after 20 min., reaching a maximum after 130 min. An evidence of tailing was always present in the pre-B-lipoprotein band following the administration of fat, as well as an extended weaker band. The oral administration of fat induced traces of chylomicrons after 45 min. A weak chylomicron band was found from the 90th min. to the end of the experiment. From 75 min. to the final stage of the test, the pre-@-lipoprotein band was only slightly intensified as compared with normal bands, with an extremely slight tailing only. An extended weak pre-B-band as seen in intravenous administration could not be found. Twenty minutes after the maximal increase of the various fat parameters, i.e. 140 min. after initiating the I.V. infusion of fat, a definite

FFA

I

,

,

..-.I

L.........,.,.,...., 60 100 U0 180mX03#) FFA

' Min.

360

r.

Fig. 1 . Triglycerides (TG), free fatty acids (FFA) and free glycerol (FG) after intravenous and oral (p.0.) administration of a fat solution containing 40 g soy bean o i l (the upper part represents intravenous and the lower part oral administration, S represents the beginning significance)

S. Raptis e t aZ.: Effects of Lipids on Insulin

523

rise in insulin secretion was observed (Fig. 2 ) . In the same subjects a significant insulin secretion (Fig. 2) was seen as early as 10 minutes following fat ingestion, this being at a time when TG, FG and FFA in the plasma had not yet risen and no changes were indicated by lipoprotein electrophoresis. In congruence with the rise

40

/MI ( n =lo)

! I ,

0 M

-10

in insulin secretion, BS fell significantly (Fig. 3 ) . The HGH secretion after I.V. infusion showed a definite decline synchronous with the maximal increase of the fat parameters in plasma. Following'oral administration of fat, we found a multiphase course of HGH (Fig. 4 ) .

cb $0 6b

I

I

I

I

UO

100

I

1

180

pU I m l

I

1

220

NO'p 160'p 170'p

I

260

1

1

I

< 0.05 < 0.05

1

360 Min.

300

I MI ( n =lo)

A

< 0,025

1M'p < 0.0025

35

n7'p < 0.05 2 w p < 0.005 30'p c 0.0025 45'p c 0.025 60'p < 0.05

25

75'p < 0.05

15

i n

0

Lb

6;,

'

I

80 l b o

I

I

140

1

1bO

1

1

1

240

1

1

280

1

3M

1

1

360 M i n .

Fig. 2 . Immunologically measurable insulin (IMI) in 10 healthy subjects after intravenous and oral (p.0.) administration of fat mg/ 100 ml

130'p