Adipocyte Differentiation-Related Protein Promotes Fatty Acid Storage in Cytosolic Triglycerides and Inhibits Secretion of Very Low–Density Lipoproteins
Objective— We investigated the role of adipocyte differentiation-related protein (ADRP) in triglyceride turnover and in the secretion of very low–density lipoprotein (VLDL) from McA-RH7777 cells and primary rat hepatocytes.
Methods and Results— An increase in the expression of ADRP increased triglyceride accumulation in cytosolic lipid droplets and prevented the incorporation of fatty acids into secretable triglycerides, thereby reducing the secretion of triglycerides as well as of apolipoprotein B-100 (apoB-100) and apoB-48 VLDL. The ability of ADRP to block the secretion of apoB-100 VLDL1 decreased with increasing quantities of fatty acids in the medium, indicating a saturable process and emphasizing the importance of sequestering of fatty acids for the effect of ADRP on VLDL secretion. Knockdown (small interfering RNA) of ADRP decreased the pool of cytosolic lipid droplets but increased only the secretion of apoB-48 VLDL1. Additionally, there was an increased flow of fatty acids into β-oxidation.
Conclusions— ADRP is essential for the accumulation of triglycerides in cytosolic lipid droplets. An increase in ADRP prevents the formation of VLDL by diverting fatty acids from the VLDL assembly pathway into cytosolic triglycerides, whereas a decrease of the protein increases the sorting of fatty acids to β-oxidation and promotes the secretion of apoB-48 VLDL1.
- adipose differentiation–related protein
- cytosolic lipid droplets
- apolipoproteins B
- small interfering RNA
Cytosolic lipid droplets are ubiquitous organelles involved in the storage and turnover of neutral lipids such as triglycerides. Several proteins have been identified on these droplets, the most well known being the PAT domain proteins,1–3 including the perilipins, adipocyte differentiation-related protein (ADRP or adipophilin) and Tip 47. ADRP, which is ubiquitously expressed,4 has a central role in the formation of lipid droplets.5 These droplets are assembled at the microsomal membrane by an insulin-dependent process6 that requires phospholipase D1, extracellular signal regulated kinase 2, and the motor protein dynein.6,7 The assembly process involves the formation of small primordial droplets,7 which grow in size by a fusion process that is dependent on intact microtubules8 and dynein.6
The assembly of very–low density lipoproteins (VLDLs)9–12 starts with the cotranslational lipidation of apolipoprotein B-100 (apoB-100), forming a pre-VLDL particle. VLDL2 (Svedberg flotation [sf] units 20 to 60) is formed from pre-VLDL by additional lipidation,13 whereas VLDL1 (sf 60 to 80) is formed from VLDL2 by a mechanism that is dependent on an ADP ribosylation factor 1–controlled sorting/transport process14 and involves the addition of a bulk load of lipids to the particle.12,13 The triglycerides used in this assembly process are largely derived from triglycerides in cytosolic lipid droplets.15,16
In this article, we demonstrate that an increase in ADRP promotes the storage of triglycerides in cytosolic lipid droplets and inhibit their secretion as VLDL. Conversely, a knockdown of ADRP decreases the storage of triglycerides but channels the fatty acids mainly to β-oxidation.
Materials and Methods
Please see the online supplement, available at http://atvb.ahajournals.org.
ADRP Increases Triglyceride Storage and Decreases Its Secretion and the Secretion of VLDL in McA-RH 7777 Cells
In these experiments, McARH 7777 cells were stably transfected with ADRP in the inducible t-rex system. Induction of the McA-RH7777 cells with tetracycline gave rise to an increased expression of ADRP (supplemental Figure IA, available online at http://atvb.ahajournals.org). Such an induction resulted in a 4.0±1.5-fold increase (mean±SD; n=8) in the pool of cytosolic lipid droplets (Figure 1A and 1B). There was also a significant increase in the cellular triglyceride mass (measured after 48-hour induction) from 2.26±0.36 to 3.42±0.67 mmol (mean±SD; n=8; P<0.002; Mann–Whitney rank sum test).
There was no difference between induced and uninduced McA-RH7777 cells in the rate of accumulation of radioactive triglycerides after incubation with [3H]-palmitic acid (supplemental Figure IIA). Moreover, the induction of ADRP did not influence the total accumulation of radiolabeled triglycerides in the system (cell and culture medium; supplemental Figure IIB); however, the rate of secretion of radiolabeled triglycerides decreased by 60% after the increase in the amount of ADRP (supplemental Figure IIC). An increased cellular accumulation and a decreased secretion of triglycerides, after induction of the ADRP expression, were also observed after continuous labeling with [3H]-palmitic acid (data not shown) and after incubation with [14C]-glycerol (supplemental Figure III). Thus, an increase of ADRP increases the accumulation of triglycerides in cytosol and prevents their secretion.
The induction of the McA-RH7777 cells with tetracycline (increasing the amount of ADRP) reduced the secretion of radiolabeled apoB-100 in the VLDL1 and VLDL2 density ranges (Figure 1C). However, the ratio between VLDL1 and VLDL2 decreased from 1.4±0.2 to 0.9±0.06 (mean±SD; n=3; P=0.031; t test), indicating a larger effect on VLDL1. There was also a decrease in the secretion of radiolabeled apoB-48 in the VLDL1 density region but not of apoB-48 in the high-density lipoprotein density region (Figure 1C). These effects of ADRP overexpression were significantly greater than those on the secretion of transferrin and total protein, indicating that ADRP specifically affects the biosynthesis of VLDL.
ADRP Enhances Triglyceride Storage and Reduces Secretion of Triglycerides and VLDL in Primary Rat Hepatocytes
Infection of primary rat hepatocytes with adenovirus encoding ADRP (supplemental Figure IB) gave rise to a 1.9±0.8-fold (mean±SD; n=7) increase in the cellular pool of lipid droplets when compared with the control (adenovirus encoding Zs-green; Figure 2A and 2B). The difference between ADRP and the control (Zs-green) was less than that observed between induced and uninduced McA-RH7777 cells (Figure 1A and 1B). The expression efficiency might contribute to this difference; whereas all McA-RH7777 cells were transfected with ADRP, the efficiency of adenovirus-mediated gene transfer (measured as the expression of Zs-green) was 80%. Another reason for this discrepancy is the effect of the virus infection per se on the formation of lipid droplets. Thus, there was a 1.4±0.5-fold (mean±SD; n=8; P=0.01; Mann–Whitney rank sum test) increase in cells transfected with the adenovirus encoding Zs-green compared with the uninfected control. Finally, the amount of lipid droplets in the untreated primary cells was much higher than in uninduced McA-RH7777 cells (compare Figures 1A and 2⇓A). Decreasing the amount of glucose and insulin during the culture had a dramatic effect on the amount of lipid droplets and cellular triglycerides in the primary hepatocytes (supplemental Figure IVA). However, under these conditions, the levels of VLDL1 also decreased (supplemental Figure IVB and IVC), making the system less suitable for studies on the effect of ADRP on VLDL assembly.
There was an increased accumulation of radiolabeled triglycerides in cells infected with adenovirus encoding ADRP when compared with the Zs-green control (supplemental Figure VA), but the rate of triglyceride biosynthesis was not influenced (supplemental Figure VB). Instead, the increase in ADRP reduced the secretion of radiolabeled triglycerides (supplemental Figure VC). This decrease was much lower than that seen in McA-RH7777 cells, most likely because of the circumstances discussed above.
Finally, we demonstrated that an increased expression of ADRP significantly reduces the secretion of radiolabeled VLDL1, both with apoB-100 and apoB-48 (Figure 2C). In the case of apoB-100, this decrease was also reflected in a large decrease in the ratio between apoB-100 VLDL1 and apoB-100 VLDL2 (27.8±3.6 to 0.8±0.2 [mean±SD; n=3; P=0.04; t test]). Although the decrease in VLDL1 (both apoB-100 and apoB-48) was in the same range as that seen in McA-RH7777 cells, we found a rather large increase (percentage-wise) of apoB-100 VLDL2 in the primary cells after the increase in ADRP (compared with Zs-green). The reason for this difference is most likely to be found in the difference in the amount of apoB-100 VLDL2 that is secreted under the culture conditions used (see also supplemental Figure IV). In the McA-RH 7777 cells, a substantial amount of apoB-100 is secreted on VLDL2, whereas only a very small amount of the protein is present on VLDL2 particles when secreted from primary hepatocytes (supplemental Figure VI). It is obvious from supplemental Figure VI that the major influence of ADRP in both cell types is a decrease in the secretion of apoB-100 VLDL1. Because of the very small amount of apoB-100 VLDL2 secreted from the primary hepatocytes, we refrain from comparing the 2 cell types as to the effect of ADRP on the secretion of apoB-100 VLDL2.
The results obtained in primary hepatocytes, although less pronounced than those obtained in McA-RH7777 cells, confirm the observation that an increase in the cellular levels of ADRP increases the storage of triglycerides as cytosolic lipid droplets and reduces the secretion of both triglycerides and VLDL1. However, because the primary cells contain a larger pool of cytosolic lipid droplets at the start of the experiment and the virus treatment per se influences the accumulation of such droplets, we concluded that primary rat hepatocytes are less suitable than McA-RH7777 cells for studies on the mechanism behind the influence of ADRP on the storage and secretion of triglycerides.
ADRP Prevents the Removal of Fatty Acids From Cellular Triglycerides
Induction of the McA-RH7777 cells with tetracycline (which increased the amount of ADRP) decreased the turnover of stored radiolabeled triglycerides (Figure 3A). This observation was confirmed in primary rat hepatocytes (supplemental Figure VII). However, when the formation of acylCoA and the biosynthesis of triglycerides was inhibited by Triacsin C,8 the rate of the turnover of the stored radiolabeled triglycerides increased, approaching that observed in the uninduced cells. This indicates that the ADRP-induced decrease in the turnover of cellular triglycerides is dependent on the formation of acylCoA.
We next investigated whether the ADRP dependent inhibition of the apoB-100 VLDL1 secretion could be influenced by the addition of oleic acid to the culture medium. ADRP inhibited the secretion of radiolabeled apoB-100 VLDL1 at all concentrations of oleic acid investigated (Figure 3B). However, the inhibition was greatest at oleic acid concentrations <100 μmol/L (70% to 80%); at >100 μmol/L, it decreased almost linearly to <20% at 360 μmol/L (Figure 3B; supplemental Figure VIIIA), although increasing concentrations of oleic acid did not reduce the intracellular ADRP levels but rather tended to increase them (supplemental Figure VIIIB and VIIIC). Thus, the ADRP-induced inhibition of apoB-100 VLDL 1 secretion can be overcome by increasing the amount of fatty acids in the cell.
Increasing levels of oleic acid inhibited the secretion of radiolabeled apoB-100 VLDL2 (supplemental Results; supplemental Figure VIIID). This inhibition was suppressed when the maximal ADRP induced inhibition of the apoB-100 VLDL1 secretion was observed (supplemental Results; compare supplemental Figure VIIIA and VIIID) These observations are in line with the proposed precursor–product relationship between VLDL2 and VLDL1.13 Immunoblot experiments demonstrated that overexpression of ADRP did not affect the amount of the microsomal triglyceride transfer protein in the cell (supplemental Figure VIIIE).
Knockdown of ADRP Influences the Storage and Oxidation of Fatty Acids and the Secretion of ApoB-48 VLDL1
Small interfering RNA (siRNA) to ADRP decreased the expression of the protein when compared with a control siRNA, whereas the control proteins were unaffected (Figure 4A). Such a knockdown of ADRP gave rise to a decrease in the accumulation of lipid droplets (Figure 4B). Thus, the size of the pool of lipid droplets in cells transfected with ADRP siRNA was 66±5% of that seen in cells transfected with control siRNA (Figure 4C). ADRP siRNA did also increase the rate of β-oxidation both of radiolabeled fatty acids supplied to the cells (data not shown) and radiolabeled fatty acids present in cytosolic triglycerides (Figure 4D). A small increase in the proportion of the intracellular pool of triglycerides that was secreted was observed in cells transfected with ADRP siRNA (supplemental Figure IX).
There was a 3.1-fold increase in the rate of secretion of apoB-48 VLDL1 when the cells were transfected with ADRP siRNA, but there was no effect on the secretion apoB-100 VLDL1 (Figure 4E). The increase in apoB-48 VLDL1 differed significantly from that observed for the controls (transferrin and the total secreted proteins; Figure 4E).
This study shows that an increase of ADRP in liver cells increases the size of the pool of cytosolic lipid droplets. Conversely, a knockdown of ADRP by siRNA decreases this pool. An increase in the cellular amount of ADRP reduced the secretion of triglycerides and VLDL by preventing fatty acids from leaving the cytosolic triglycerides to be incorporated into triglycerides in VLDL. The major effect of the knockdown of ADRP was an increased β-oxidation; however, there was also an increase in the rate of secretion of apoB-48 VLDL1 after such a knockdown.
Our results confirm observations by others that ADRP promotes the formation of lipid droplets in cells.5 The effects of an increased expression of ADRP were observed in both McA-RH7777 cells (stably transfected with ADRP in the inducible t-rex system) and primary hepatocytes; however, the effect was not as pronounced in the primary cells as in the McA-RH7777 cells. Possible reasons for this discrepancy have already been discussed above. Despite these differences, the results in the primary hepatocytes confirmed the observation in McA-RH7777 cells (ie, that an increase in the amount of ADRP promotes the accumulation of lipid droplets and decreases the secretion of triglycerides and VLDL1).
The major proportion of triglycerides secreted with VLDL is formed from fatty acids derived from stored triglycerides by hydrolysis15,16 (ie, the flow of fatty acids from cytosolic lipid droplets to VLDL triglycerides is of outmost importance for the regulation of the secretion of this lipoprotein). Our results indicate that an increase in the amount of ADRP in the cell interferes with this supply of fatty acids to the VLDL triglycerides. This is based on the observation that an increase in ADRP inhibits the secretion of triglycerides and the triglyceride-rich VLDL1, an inhibition that can be overcome by increasing the availability of exogenous fatty acids. At the same time, the increase in ADRP promotes the storage of triglycerides in the cell. The last conclusion is based on the observation that an increased expression of ADRP increased the pool of cytosolic lipid droplets (and triglycerides) as well as on the observation that such an increase in ADRP expression decreased the turnover of stored triglycerides, radiolabeled in their fatty acid moiety. In the latter experiments, we compared the turnover of such labeled triglycerides in cells with a basal or increased expression of ADRP. The comparison was done in the presence or absence of Triacsin C, which prevents the activation of fatty acids and blocks their incorporation into triglycerides.8 At the basal expression of ADRP, Triacsin C did not influence the turnover of the cellular pool of radiolabeled triglycerides, indicating that the size of this pool was determined primarily by the loss of fatty acids (ie, by hydrolyzes of the triglycerides combined with a failure of the released fatty acids to be re-esterified into cytosolic triglycerides). On the contrary, when the expression of ADRP was increased (after transfection), Triacsin C treatment increased the turnover of the radiolabeled triglyceride pool (approaching that rate seen in cells with the basal expression of ADRP). This indicates that the effect of the increased expression of ADRP on the turnover of the cellular pool of triglycerides is dependent on the ability to activate fatty acids (including those fatty acids that had been released by hydrolysis) to allow them to enter into triglycerides. This, in turn, indicates that ADRP promotes the storage of newly formed triglycerides, including those formed from fatty acids released by hydrolysis.
Although an increase in the amount of ADRP gave clear effects on the secretion of both apoB-100 and apoB-48 VLDL1, the knockdown of ADRP only resulted in an increase in the secretion of apoB-48 VLDL1. This most likely reflects the very strong influence of increased amount of fatty acids on the assembly of apoB-48 VLDL1.17,18 Although the increase in the secretion of apoB-48 VLDL1 seen after ADRP knockdown was large, the amount of VLDL1 assembled by apoB-48 in these cells is only, at the most, 15% of that assembled by apoB-100. This is most likely the reason why we only detected a small increase in the secretion of triglycerides when ADRP was decreased. During the preparation of this article, an ADRP knockout mouse was published,19 showing a significant effect on the accumulation of triglycerides in the liver but not on the VLDL secretion. Where does the surplus of fatty acids disappear? We observed a large increase in the rate of the β-oxidation of fatty acids, in particular of the fatty acids that were stored in the cell. This is a plausible explanation for the loss of triglycerides from the cell when ADRP is knocked down.
Interestingly, an increase in the amount of ADRP did not influence the rate of β-oxidation (supplemental Results; supplemental Figure XA). Thus, an increase in ADRP selectively influences the assembly of VLDL, whereas a knockdown channeled fatty acids primarily into β-oxidation. This could suggest that the fatty acid pool (or pathway) that is used for β-oxidation differs from that used for VLDL assembly, and that these pools are influenced differently by variation in the cellular levels of ADRP. Thus, there may be a mechanism that secures fatty acids for β-oxidation and energy production. This needs to be addressed experimentally.
The observation that ADRP promotes the storage of triglycerides could indicate that an increase in the amount of ADRP gives rise to an increased energy consumption because the biosynthesis of triglycerides consumes energy during the activation of the fatty acids by coenzyme A and for the production of glycerol 3-P. In the case of the formation of lipid droplets, the detailed mechanism has not been established; however, the observation that it involves both the formation of a primordial droplet and a fusion of these droplets7,8 strongly suggests an energy-consuming process. Indeed, we observed that overexpression of ADRP gave rise to an increase in glucose uptake of the same magnitude as treatment with insulin20 (supplemental Results; supplemental Figure XB). This increase in glucose uptake was not just a result of the overexpression of any protein (supplemental Results; supplemental Figure XI), nor was it explained by an increased incorporation of radiolabeled glucose-carbons into triglycerides in cells overexpressing ADRP (supplemental Results; supplemental Figure XC). Based on these observations, we suggest that the ADRP-induced increase in glucose uptake is the result of an increased need for energy for the increased storage of fatty acids in lipid droplets.
In summary, increased hepatic expression of ADRP prevents fatty acids from being incorporated into triglycerides used for VLDL assembly and promotes their storage in cytosolic lipid droplets. However, together, the results from the overexpression and knockdown of ADRP point to a rather complex machinery involved in the sorting of fatty acids between storage secretion and oxidation.
Source(s) of Funding
This study was supported by grant 7142 from the Swedish Research Council/Medicine and by the Swedish Heart and Lung Foundation, the Novo Nordic Foundation, Swedish Strategic Funds (National Network and Graduate School for Cardiovascular Research), and the Söderberg Foundation.
B.M. and L.A. contributed equally to this study.
Original received October 26, 2005; final version accepted April 10, 2006.
Brasaemle DL, Barber T, Wolins N, Serrero G, Blanchette-Mackie EJ, Londos C. Adipose differentiation-related protein is an ubiquitously expressed lipid storage droplet associated protein. J Lipid Res. 1997; 38: 2249–2263.
Imamura M, Inoguchi T, Ikuyama S, Taniguchi S, Kobayashi K, Nakashima N, Nawata H. ADRP stimulates lipid accumulation and lipid droplet formation in murine fibroblasts. Am J Physiol Endocrinol Metab. 2002; 283: E775–E783.
Andersson L, Boström P, Rutberg M, Ericsson J, Magnusson B, Marchesan D, Ruiz M, Asp L, Huang P, Frohman MA, Borén J, Olofsson S-O. Phospholipase D1 and extracellular signal-regulated kinase 2 (ERK2) regulate cytosolic lipid droplet formation. J Cell Science. In press.
Marchesan D, Rutberg M, Andersson L, Asp L, Larsson T, Boren J, Johansson BR, Olofsson SO. A phospholipase D-dependent process forms lipid droplets containing caveolin, adipocyte differentiation-related protein, and vimentin in a cell-free system. J Biol Chem. 2003; 278: 27293–27300.
Bostrom P, Rutberg M, Ericsson J, Holmdahl P, Andersson L, Frohman MA, Boren J, Olofsson SO. Cytosolic lipid droplets increase in size by microtubule-dependent complex formation. Arterioscler Thromb Vasc Biol. 2005; 25: 1945–1951.
Stillemark-Billton P, Beck C, Boren J, Olofsson SO. Relation of the size and intracellular sorting of apoB to the formation of VLDL 1 and VLDL 2. J Lipid Res. 2005; 46: 104–114.
Asp L, Magnusson B, Rutberg M, Li L, Boren J, Olofsson SO. Role of ADP ribosylation factor 1 in the assembly and secretion of ApoB-100-containing lipoproteins. Arterioscler Thromb Vasc Biol. 2005; 25: 566–570.
Wiggins D, Gibbons GF. The lipolysis/esterification cycle of hepatic triacylglyecerol. Its role in the secretion of very-low-density lipoprotein and its response to hormones and sulphonylureas. Biochem J. 1992; 284: 457–462.
Wiggins D, Gibbons GF. Origin of hepatic very-low-density lipoprotein triacylglycerol: the contribution of cellular phospholipid. BiochemJ. 1996; 320: 673–679.
Borén J, Rustaeus S, Olofsson S-O. Studies on the assembly of apolipoprotein B-100- and B-48-containing very low density lipoproteins in McA-RH7777 cells. J Biol Chem. 1994; 269: 25879–25888.
Stillemark P, Borén J, Andersson M, Larsson T, Rustaeus S, Karlsson K-A, Olofsson S-O. The assembly and secretion of apolipoprotein-B48-containing very low density lipoproteins in McA-RH7777 cells. J Biol Chem. 2000; 275: 10506–10513.
Chang BH, Li L, Paul A, Taniguchi S, Nannegari V, Heird WC, Chan L. Protection against fatty liver but normal adipogenesis in mice lacking adipose differentiation-related protein. Mol Cell Biol. 2006; 26: 1063–1076.
Wolins NE, Quaynor BK, Skinner JR, Schoenfish MJ, Tzekov A, Bickel PE. S3-12, adipophilin, and TIP47 package lipid in adipocytes. J Biol Chem. 2005; 280: 19146–19155.