Linking nutritional regulation of Angptl4, Gpihbp1, and Lmf1 to lipoprotein lipase activity in rodent adipose tissue

Abstract

Background: Lipoprotein lipase (LPL) hydrolyzes triglycerides in lipoproteins and makes fatty acids available for tissue metabolism. The activity of the enzyme is modulated in a tissue specific manner by interaction with other proteins. We have studied how feeding/fasting and some related perturbations affect the expression, in rat adipose tissue, of three such proteins, LMF1, an ER protein necessary for folding of LPL into its active dimeric form, the endogenous LPL inhibitor ANGPTL4, and GPIHBP1, that transfers LPL across the endothelium.

Results: The system underwent moderate circadian oscillations, for LPL in phase with food intake, for ANGPTL4 and GPIHBP1 in the opposite direction. Studies with cycloheximide showed that whereas LPL protein turns over rapidly, ANGPTL4 protein turns over more slowly. Studies with the transcription blocker Actinomycin D showed that transcripts for ANGPTL4 and GPIHBP1, but not LMF1 or LPL, turn over rapidly. When food was withdrawn the expression of ANGPTL4 and GPIHBP1 increased rapidly, and LPL activity decreased. On re-feeding and after injection of insulin the expression of ANGPTL4 and GPIHBP1 decreased rapidly, and LPL activity increased. In ANGPTL4(-/-) mice adipose tissue LPL activity did not show these responses. In old, obese rats that showed signs of insulin resistance, the responses of ANGPTL4 and GPIHBP1 mRNA and of LPL activity were severely blunted (at 26 weeks of age) or almost abolished (at 52 weeks of age).

Conclusions: This study demonstrates directly that ANGPTL4 is necessary for rapid modulation of LPL activity in adipose tissue. ANGPTL4 message levels responded very rapidly to changes in the nutritional state. LPL activity always changed in the opposite direction. This did not happen in Angptl4(-/-) mice. GPIHBP1 message levels also changed rapidly and in the same direction as ANGPTL4, i.e. increased on fasting when LPL activity decreased. This was unexpected because GPIHBP1 is known to stabilize LPL. The plasticity of the LPL system is severely blunted or completely lost in insulin resistant rats.

Figure 1

Circadian changes. Groups of young rats (n=6) fed ad lib were sacrificed at the indicated times over a 24 hour period. Epididymal adipose tissue and blood were taken for analyses. (A) Food consumption was measured in a separate group of rats over a 3-day period. These data are expressed as % of body weight per hour. (B) Glucose concentration in blood (■) and NEFA concentration in plasma (s○). (C) mRNA abundance for LPL (●), ANGPTL4 (■) and GPIHBP1 (▲) in epididymal adipose tissue. (D) LPL activity (●) and mass (■) in the adipose tissue. Statistical comparisons were made against the values at 2:00 h.

Figure 2

Time courses for the effects of cycloheximide and actinomycin D. Young rats were fasted overnight and were then injected with cycloheximide (CHX, panels A and D-F) or ActD (panels B, C and G-I) starting at 8:00 h. Groups of rats (n=5) were sacrificed at the indicated times over a 4 or 6 hour period. Rats for the time 0 groups were not injected and were sacrificed within one hour from the start of the experiment. Epididymal adipose tissue was taken for analyses. (A) LPL activity (●) and mass (■) after CHX. (B) LPL activity (●) and mass (■) after ActD (C) mRNA levels for ANGPTL4 (■), GPIHBP1 (▲), LPL (●) and LMF1 (▼). mRNA was calculated relative to 18S mRNA. The value at time 0 was set to 1 and values at following time points calculated relative to this. (D-I) Western blots of adipose tissue as labeled in the figure. The antibodies to C-terminal ANGPTL4 used in panels E and H recognize both full-length ANGPTL4 and the fibrinogen-like C-terminal domain. The antibodies to the N-terminal domain of ANGPTL4, used in panels F and I, identifies the full-length protein but do not see the C-terminal domain.

Figure 3

Response to food deprivation. Food was withdrawn from young rats at 8:00 h. Animals were then sacrificed at the indicated time points over a six hour period (n=6 at each time). The experiment was conducted over three days. Rats for the different time points were randomly included on each day. Epididymal adipose tissue and blood were taken for analyses and the stomach with its content was cut out and weighed. (A) Weight of the stomach expressed as fraction of body weight (●) and concentration of insulin in plasma (□) (B) Concentrations of glucose (■) and NEFA (○) in blood and plasma, respectively (C) mRNA abundance for ANGPTL4 (■) and GPIHBP1 (▲). mRNA was calculated relative to 18S mRNA. The value at time 0 was set to 1 and values at following time points were calculated relative to this. (D) LPL activity (●) and mass (■).

Figure 4

Responses to re-feeding. Young rats were fasted overnight (16 h, from 16:00 h). Food was given back (at 8:00) and groups of rats (n=5) were sacrificed after one or two hours. Time = 0 min were a group of rats sacrificed within one h of continued fasting. Epididymal adipose tissue and blood was collected and analyzed. The stomach was cut out, weighed, opened and rinsed and then weighed again. The difference is the “contents in stomach” and is expressed as fraction of body weight (A) Weight of contents in stomach (●) and concentration of insulin in plasma (□) (B) Concentrations of glucose (■) and NEFA (○) in blood and plasma, respectively. (C) mRNA abundance for ANGPTL4 (■) and GPIHBP1 (▲). mRNA was calculated relative to 18S mRNA. The value at time 0 was set to 1 and values at following time points were calculated relative to this. (D) LPL activity (●) and mass (■).

Figure 5

Plasma lipids in Angptl^−/−mice. Wild-type and Angptl^−/− mice were fasted overnight or fed ad lib. Food was given back to groups of fasted mice at 6:00 h (re-fed), while others continued fasting or eating (n=6 per group). All animals were sacrificed between 9:00 h – 10:00 h. (A) FPLC analyses of pooled plasma samples representing the 6 different groups. (B) Fatty acids (NEFA) in individual plasma samples. Statistically significant (p<0.05) differences are indicated by: a when comparing fasted or re-fed animals to ad lib for the same genotype b when comparing the two genotypes for the same nutritional state.

Figure 6

Response of the LPL system to fasting and re-feeding in Angptl4^−/−mice. Epididymal adipose tissue from the same mice as described in Figure 5 was cut out and analyzed for LPL activity and mRNA abundance for LPL, ANGPTL4, and GPIHBP1. The levels of mRNA were calculated relative to cyclophilin mRNA. Statistically significant (p < 0.05) differences are indicated by: a when comparing Angptl4^−/− and WT for the same nutritional state b when comparing fasted and ad lib for the same genotype c when comparing refed and ad lib for the same genotype.

Figure 7

Effects of insulin on expression of LPL activity, ANGPTL4 and GPIHBP1. Insulin was injected i.p. to young rats fasted for 6 hours (from 8:00 h, n=6 per group). The rats were sacrificed 60 min after injection. Time = 0 min were non-injected rats sacrificed before the injected rats. LPL activity and mass and mRNA abundance for ANGPTL4 and GPIHBP1 were measured in epididymal adipose tissue.

Figure 8

Effects of age and/or obesity. Groups of rats of different ages (5 weeks, 26 weeks and 52 weeks) were either fed ad lib (solid bars) or fasted overnight for 16 h (open bars). At 8:00 h groups of fed and fasted rats of the indicated age were sacrificed (n=5 per group). The epididymal fat body was cut out and weighed. Blood was also taken for analyses of (A) glucose, (B) NEFA and (C) Insulin. mRNA levels for ANGPTL4 (D) and GPIHBP1 (E) and are expressed relative to 18S RNA. (F) LPL activity expressed per g wet weight of tissue. Statistical comparisons are made between fed and fasted animals. The numbers above some of the bars represent the relative change from the fasted to the fed state.