Expression of uncoupling proteins-1, -2 and -3 mRNA is induced by an adenocarcinoma-derived lipid-mobilizing factor

Abstract

The abnormalities of lipid metabolism observed in cancer cachexia may be induced by a lipid-mobilizing factor produced by adenocarcinomas. The specific molecules and metabolic pathways that mediate the actions of lipid-mobilizing factor are not known. The mitochondrial uncoupling proteins-1, -2 and -3 are suggested to play essential roles in energy dissipation and disposal of excess lipid. Here, we studied the effects of lipid-mobilizing factor on the expression of uncoupling proteins-1, -2 and -3 in normal mice. Lipid-mobilizing factor isolated from the urine of cancer patients was injected intravenously into mice over a 52-h period, while vehicle was similarly given to controls. Lipid-mobilizing factor caused significant reductions in body weight (-10%, P=0.03) and fat mass (-20%, P<0.01) accompanied by a marked decrease in plasma leptin (-59%, P<0.01) and heavy lipid deposition in the liver. In brown adipose tissue, uncoupling protein-1 mRNA levels were elevated in lipid-mobilizing factor-treated mice (+96%, P<0.01), as were uncoupling proteins-2 and -3 (+57% and +37%, both P<0.05). Lipid-mobilizing factor increased uncoupling protein-2 mRNA in both skeletal muscle (+146%, P<0.05) and liver (+142%, P=0.03). The protein levels of uncoupling protein-1 in brown adipose tissue and uncoupling protein-2 in liver were also increased with lipid-mobilizing factor administration (+49% and +67%, both P=0.02). Upregulation by lipid-mobilizing factor of uncoupling proteins-1, -2 and -3 in brown adipose tissue, and of uncoupling protein-2 in skeletal muscle and liver, suggests that these uncoupling proteins may serve to utilize excess lipid mobilized during fat catabolism in cancer cachexia.

Figure 1

Twelve per cent SDS polyacrylamide gel electrophoresis of human LMF. Lane 1, molecular weight markers; Lane 2, human LMF purified from the urine of cachectic cancer patients, showing expected molecular weight of 43 kDa. Detection was by Coomassie Blue staining.

Figure 2

Body weight changes in LMF-treated and vehicle-treated control mice. Arrows indicate time-point of LMF injection. *P<0.01; LMF vs controls. Data are mean±s.e.m. for six mice per group.

Figure 3

Effect of LMF treatment on mRNA levels of UCP-1, -2 and -3 in BAT, and of UCP-2 and -3 in skeletal muscle. Mice were injected with LMF and tissues were removed at 52 h; control mice received vehicle injections. Total RNA was extracted from mouse BAT or skeletal muscle, and the mRNA levels of the UCPs were measured using Northern blotting with chemiluminescence detection. Data are mean±s.e.m. for six mice per group, expressed as percentage of controls (mean=100%).

Figure 4

(A) Example of RNA quantification by competitive RT–PCR. Lanes 1 to 6, the initial concentrations of the competitor at 100, 50, 25, 12.5, 6.25 and 3.13 pg. The band density ratios between the target and competitor were determined after photographing the agarose gel and plotted vs the initial amount of competitor added in the RT–PCR reaction. At the equivalence point (ratio=1), the initial amount of target RNA corresponds to the initial amount of competitor. (B) Effect of LMF treatment on UCP-2 mRNA in liver. Mice were injected with LMF and liver was removed at 52 h; control mice received vehicle injections. Data are mean±s.e.m. for six mice per group.

Figure 5

(A) Western blot showing UCP-1 concentrations in mitochondrial preparations of brown adipose tissue from vehicle-treated and LMF-treated mice. (B) UCP-1 protein content in control and LMF-treated mice (mean±s.e.m. of the signals shown in A).

Figure 6

(A) Western blot showing UCP-2 content in mitochondrial preparations of liver from vehicle-treated and LMF-treated mice (format as in Figure 5).

Figure 7

Histology of mouse liver. Fresh frozen tissue from five mice per group was sectioned on a cryostat and stained with Oil Red O. There was marked intracytoplasmic lipid accumulation in the liver from LMF-treated mice (B) but little in the vehicle-treated mice (A). Hepatocytes from all zones of the hepatic lobule were affected equally.