Glycosylation and secretion of lipoprotein lipase by 3T3-L1 adipocytes: effects of brefeldin A

Increased effect of fucoidan on lipoprotein lipase secretion in adipocytes

AIMS

Fucoidan, a sulfated polysaccharide extracted from brown seaweed (F. vesiculosus) is recognized as an effective anticoagulant but its anti-lipidemic potency has not been well defined. We investigated the effect of fucoidan on lipoprotein lipase (LPL) secretion by human adipocytes.

MAIN METHODS

LPL mRNA and protein expressions were measured using semi-quantitative RT-PCR, ELISA and immunohistochemistry in cultured adipocytes with or without fucoidan treatment. LPL enzyme activity was determined by a fluorometric assay.

KEY FINDINGS

In cultured adipocytes, fucoidan induced LPL secretion in a dose- and time-dependent manner. An initial increase in LPL was maintained at a significant level but much slower than that in heparin-treated cells. Fucoidan also dose-dependently induced a cofactor of LPL, the apolipoprotein C-II (ApoC-II) secretion. In fucoidan-treated cells, LPL mRNA was time-dependently increased and LPL protein expression was also inceased. Treatment with both heparin and fucoidan showed no further increase in media LPL activity compared to heparin alone. In the conditioned medium from fucoidan-treated cells followed for 4 h, LPL activity decayed exponentially with half-life of about 180 min. In addition, the extracellular LPL mass in cycloheximide (a protein synthesis inhibitor) and fucoidan-treated cells did not change markedly, but LPL shifted significantly from active to inactive form.

SIGNIFICANCE

These results suggest that fucoidan acts like heparin by releasing LPL in addition to increasing the intracellular transport and decreasing the degradation of LPL in the medium. Furthermore, LPL and ApoC-II secretion induced by fucoidan may be involved in regulating plasma triglyceride lowering clearance.

Synthesis of active high mannose-type lipoprotein lipase in human adipose tissues

Lipoprotein lipase (LPL) activity in the retroperitoneal adipose tissue of a patient with Cushing's syndrome and in the subcutaneous adipose tissue of a patient with aseptic necrosis of the femoral head was higher than that in the corresponding tissues of the control subjects. The amount of [35S]methionine incorporated into LPL was also higher in these patients than in control subjects. However, the ratio of activity and amount of radioactivity in the LPL of patients was identical to that of control subjects, indicating that LPL synthesized in the adipose tissues of patients had a normal specific activity. LPL with Mr = 57000 was composed of two types of subunits: one type was partially endo H-sensitive, yielding a product with Mr = 55000, and the other was totally endo H-sensitive, yielding a product with Mr = 52000. Both retroperitoneal and subcutaneous adipose tissues of control subjects contained nearly equal amounts of partially and totally endo H-sensitive subunits. In the retroperitoneal adipose tissue of a patient with Cushing's syndrome, 8% of subunits were partially endo H-sensitive and 92% were totally endo H-sensitive. In the subcutaneous adipose tissue of a patient with aseptic necrosis of the femoral head, 21% of subunits were partially endo H-sensitive and 79% were totally endo H-sensitive. The 24-h treatment of subcutaneous adipose tissue of a control subject with 1 mM 1-deoxymannojirimycin (dMM) caused the synthesis of active, but totally endo H-sensitive, LPL. Thus, in human adipose tissue, the processing of one oligosaccharide chain of an LPL subunit to a complex type chain in the trans Golgi was not necessary for the expression of activity.

Combined lipase deficiency (cld/cld) in mice affects differently post-translational processing of lipoprotein lipase, hepatic lipase and pancreatic lipase

Lipoprotein lipase (LPL) and hepatic lipase (HL), which act on plasma lipoproteins, belong to the same gene family as pancreatic lipase. LPL is synthesized in heart, muscle and adipose tissue, while HL is synthesized primarily in liver. LPL is also synthesized in liver of newborn rodents. The active form of LPL is a dimer, whereas that of HL has not been established. Combined lipase deficiency (CLD) is an autosomal recessive mutation (cld) in mice which impairs post-translational processing of LPL and HL. Cld/cld mice have very low LPL and HL activities (< 5% of normal), yet normal pancreatic lipase activity. They develop massive hypertriglyceridemia and die within 3 days after birth. The CLD mutation allows synthesis, glycosylation and dimerization of LPL, but blocks activation and secretion of the lipase. Thus, dimerization per se does not result in production of active LPL. Immunofluorescence studies showed that LPL is retained in endoplasmic reticulum (ER) in cld/cld cells. Translocation of Golgi components to ER by treatment with brefeldin A (BFA) enabled synthesis of active LPL in cultured cld/cld brown adipocytes. Thus, production of inactive LPL in cld/cld cells results from inability of the cells to transport LPL from ER. The CLD mutation allows synthesis and glycosylation of HL, but blocks activation of the lipase. Immunofluorescence studies located HL mostly outside of cells in liver, liver cell cultures and incubated adrenal tissue of normal and cld/cld mice and mostly inside of cells in liver cell cultures and adrenal tissues treated with monensin (to block secretion of protein). These findings demonstrate synthesis and secretion of HL by both liver and adrenal cells of normal and cld/cld mice. Thus, the CLD mutation allows secretion of inactive HL by liver and adrenals. However, it does not block synthesis or secretion of active pancreatic lipase. Our findings indicate that LPL, HL and pancreatic lipase, although closely related, are processed differently.

Effect of long-term treatment of 3T3-L1 adipocytes with chlorate on the synthesis, glycosylation, intracellular transport and secretion of lipoprotein lipase

Lipoprotein lipase (LPL) is synthesized and glycosylated in the endoplasmic reticulum (ER), transported through the Golgi to the cell surface, and finally secreted. To examine the role of heparan sulphate proteoglycans (HSPG) in the synthesis, activity, intracellular transport and secretion of LPL, 3T3-L1 adipocytes were cultured for 7 days in the presence of 20 mM chlorate, an inhibitor of sulphation of HSPG. Treatment of cells with 20 mM chlorate for 7 days caused a 55% decrease in LPL activity in the intracellular compartment and a 79% decrease in the cell-surface compartment. The synthetic rate of LPL in chlorate-treated cells was identical with that in control cells as determined by biosynthetic labelling. The study with endoglycosidase H (endo H) showed that the treatment with chlorate increased the proportion of LPL subunits which were totally endo H-sensitive. The study with a heparin-Sepharose column showed that 3T3-L1 adipocytes contained three forms of LPL. The first form, accounting for 35% of the LPL, did not bind to the heparin-Sepharose column and had little or no activity; the second form, accounting for 32%, bound to the column and was eluted with 0.4-0.75 M NaCl but had no activity; the third form, accounting for 33%, bound to the column and was eluted with 0.8-1.2 M NaCl and had activity. In chlorate-treated cells, the first form accounted for 66% of the LPL, the second form 15% and the third form 19%. When cells were incubated for 1 h with brefeldin A, which translocates Golgi proteins to the ER [J. Lippincott-Schwartz, L.C. Yuan, J.S. Banifacino and R.D. Klausner (1989) Cell 56, 801-813; J. Lippincott-Schwartz, J. Glickman, J.E. Donaldson, J. Robbins, T.E. Kreis, K.B. Seamon, M.P. Sheetz and R.D. Klausner (1991) J. Cell Biol. 112, 567-577], the chlorate-induced decrease in cellular LPL activity was restored. These findings indicate that LPL synthesized in chlorate-treated cells can be processed to be fully active, but chlorate-treated cells are unable to transport LPL to the Golgi and accumulate inactive LPL with a lower affinity for heparin in the ER. The treatment with chlorate decreased the proportion of LPL subunits that were endo H-resistant, indicating that the processing of oligosaccharide chains of LPL in the trans-Golgi was impaired in chlorate-treated cells. The amount of 35S-labelled LPL secreted by chlorate-treated cells was identical with that secreted by control cells, whereas the level of LPL activity in the medium of chlorate-treated cells was 25% of that in the medium of control cells, indicating that most of the LPL secreted by chlorate-treated cells was inactive.

Brefeldin A enables synthesis of active lipoprotein lipase in cld/cld and castanospermine-treated mouse brown adipocytes via translocation of Golgi components to endoplasmic reticulum

Brown adipocytes cultured from newborn combined-lipase-deficient (cld/cld) mice and castanospermine (CST)-treated 3T3-L1 adipocytes synthesize lipoprotein lipase (LPL) which is inactive and retained in the endoplasmic reticulum (ER) [Masuno, Blanchette-Mackie, Chernick and Scow (1990) J.Biol. Chem. 265, 1628-1638; Masuno, Blanchette-Mackie, Schultz, Spaeth, Scow and Okuda (1992) J. Lipid Res.33, 1343-1349]. Brefeldin A (BFA), which is known to block protein transport from ER and translocate Golgi components to ER, was used here to study the effect of translocated Golgi enzymes on LPL retained in ER of cld/cld and CST-treated mouse brown adipocytes. Brown adipocytes cultured from newborn normal mice contained 3000-5000 m-units of LPL activity/mg of DNA and secreted 35 m-units of LPL activity/mg of DNA per h. BFA at 10 micrograms/ml doubled LPL activity in normal cells within 2 h as it stopped completely secretion of active LPL. LPL in mouse cells has two N-oligosaccharide chains per subunit. Analyses with SDS/PAGE and immunoblotting showed that about one-third of LPL subunits in untreated normal cells were totally endo-beta-N-acetylglucosaminidase (endo H)-resistant, one-third were partially endo H-resistant, and one-third were totally endo H-sensitive. BFA decreased to zero the proportion of subunits which were totally endo H-resistant, while it increased the proportion which were partially endo H-resistant. Thus, BFA blocked processing of one oligosaccharide chain per subunit to endo H-resistance. Sucrose-gradient centrifugation studies showed that BFA increased the proportion of LPL subunits in normal cells which were present as active dimers. LPL activity in cld/cld adipocytes was 120 m-units/mg of DNA and that in normal adipocytes treated with CST was 430 m-units/mg of DNA. Most LPL subunits in such cells were totally endo H-sensitive and some were partially endo H-resistant, but none were totally endo H-resistant. Some of the subunits, in both cld/cld and CST-treated cells, were present as inactive LPL dimers. BFA increased LPL activity in cld/cld cells to 2100 m-units/mg of DNA and that in CST-treated cells to 2600 m-units/mg of DNA within 2 h. BFA increased in both groups the proportion of LPL subunits which were partially endo H-resistant. BFA also increased the proportion which were present as active dimers. Immunofluorescence studies in normal and cld/cld adipocytes showed that BFA caused retention of LPL in large tubular and spherical structures and in ER, but not in Golgi. When BFA was withdrawn and protein synthesis was blocked with cycloheximide, LPL in normal cells was transferred to Golgi within 30 min and disappeared within 60 min, whereas LPL in cld/cld cells was retained in large vesicles and ER. The findings indicate that BFA enabled synthesis of active LPL in cld/cld and CST-treated cells via translocation of Golgi components to ER. Also, cld/cld cells synthesized LPL which could be processed to active lipase and the enzymes needed for activation of the lipase were present in Golgi of such cells. Production of inactive LPL in cld/cld adipocytes probably results from their inability to transport LPL from ER to Golgi.