Voltar
Artigo Acesso aberto Revisado por pares
BIBTEX RIS

Down-regulation of Adipose Tissue Lipoprotein Lipase during Fasting Requires That a Gene, Separate from the Lipase Gene, Is Switched On

2002; Elsevier BV; Volume: 277; Issue: 14 Linguagem: Inglês

10.1074/jbc.m200325200

ISSN

1083-351X

Autores

Martin O. Bergö, Gengshu Wu, Toralph Ruge, Thomas Olivecrona,

Tópico(s)

Metabolism, Diabetes, and Cancer

Resumo

During short term fasting, lipoprotein lipase (LPL) activity in rat adipose tissue is rapidly down-regulated. This down-regulation occurs on a posttranslational level; it is not accompanied by changes in LPL mRNA or protein levels. The LPL activity can be restored within 4 h by refeeding. Previously, we showed that during fasting there is a shift in the distribution of lipase protein toward an inactive form with low heparin affinity. To study the nature of the regulatory mechanism, we determined the in vivo turnover of LPL activity, protein mass, and mRNA in rat adipose tissue. When protein synthesis was inhibited with cycloheximide, LPL activity and protein mass decreased rapidly and in parallel with half-lives of around 2 h, and the effect of refeeding was blocked. This indicates that maintaining high levels of LPL activity requires continuous synthesis of new enzyme protein. When transcription was inhibited by actinomycin, LPL mRNA decreased with half-lives of 13.3 and 16.8 h in the fed and fasted states, respectively, demonstrating slow turnover of the LPL transcript. Surprisingly, when actinomycin was given to fed rats, LPL activity was not down-regulated during fasting, indicating that actinomycin interferes with the transcription of a gene that blocks the activation of newly synthesized LPL protein. When actinomycin was given to fasted rats, LPL activity increased 4-fold within 6 h, even in the absence of refeeding. The same effect was seen with α-amanitin, another inhibitor of transcription. The response to actinomycin was much less pronounced in aging rats, which are obese and insulin-resistant. These data suggest a default state where LPL protein is synthesized on a relatively stable mRNA and is processed into its active form. During fasting, a gene is switched on whose product prevents the enzyme from becoming active even though synthesis of LPL protein continues unabated. During short term fasting, lipoprotein lipase (LPL) activity in rat adipose tissue is rapidly down-regulated. This down-regulation occurs on a posttranslational level; it is not accompanied by changes in LPL mRNA or protein levels. The LPL activity can be restored within 4 h by refeeding. Previously, we showed that during fasting there is a shift in the distribution of lipase protein toward an inactive form with low heparin affinity. To study the nature of the regulatory mechanism, we determined the in vivo turnover of LPL activity, protein mass, and mRNA in rat adipose tissue. When protein synthesis was inhibited with cycloheximide, LPL activity and protein mass decreased rapidly and in parallel with half-lives of around 2 h, and the effect of refeeding was blocked. This indicates that maintaining high levels of LPL activity requires continuous synthesis of new enzyme protein. When transcription was inhibited by actinomycin, LPL mRNA decreased with half-lives of 13.3 and 16.8 h in the fed and fasted states, respectively, demonstrating slow turnover of the LPL transcript. Surprisingly, when actinomycin was given to fed rats, LPL activity was not down-regulated during fasting, indicating that actinomycin interferes with the transcription of a gene that blocks the activation of newly synthesized LPL protein. When actinomycin was given to fasted rats, LPL activity increased 4-fold within 6 h, even in the absence of refeeding. The same effect was seen with α-amanitin, another inhibitor of transcription. The response to actinomycin was much less pronounced in aging rats, which are obese and insulin-resistant. These data suggest a default state where LPL protein is synthesized on a relatively stable mRNA and is processed into its active form. During fasting, a gene is switched on whose product prevents the enzyme from becoming active even though synthesis of LPL protein continues unabated. Lipoprotein lipase (LPL) 1The abbreviations used are: LPLlipoprotein lipase 1The abbreviations used are: LPLlipoprotein lipase plays an important physiological role in regulating the release of fatty acids from triglyceride-rich lipoproteins (1.Braun J.E.A. Severson D.L. Biochem. J. 1992; 287: 337-347Crossref PubMed Scopus (266) Google Scholar, 2.Eckel R.H. N. Engl. J. Med. 1989; 320: 1060-1068Crossref PubMed Scopus (851) Google Scholar, 3.Goldberg I.J. J. Lipid Res. 1996; 37: 693-707Abstract Full Text PDF PubMed Google Scholar, 4.Olivecrona T. Olivecrona G. Betteridge D.J. Illingworth D.R. Shepherd J. Lipoproteins in Health and Disease. Arnold, London1999: 223-246Google Scholar). The enzyme is synthesized in several extrahepatic tissues, but the highest activities are found in tissues that oxidize (e.g. heart and red skeletal muscle) or store (e.g. adipose tissue) fatty acids. After glycosylation, dimerization, and activation in the endoplasmic reticulum, the enzyme is secreted, transported through the extracellular matrix, and anchored to heparan sulfate proteoglycans on the intraluminal surface of nearby capillaries (3.Goldberg I.J. J. Lipid Res. 1996; 37: 693-707Abstract Full Text PDF PubMed Google Scholar,4.Olivecrona T. Olivecrona G. Betteridge D.J. Illingworth D.R. Shepherd J. Lipoproteins in Health and Disease. Arnold, London1999: 223-246Google Scholar). lipoprotein lipase lipoprotein lipase LPL is regulated at the level of gene expression in several physiological states: during fetal and early postnatal life, the enzyme is present in the liver but is then suppressed (5.Peinado-Onsurbe J. Staels B. Deeb S.S. Ramı́rez I. Llobera M. Auwerx J.H. Biochim. Biophys. Acta. 1992; 1131: 281-286Crossref PubMed Scopus (33) Google Scholar); in the mammary gland, the enzyme is switched on during lactation (6.McBride O.W. Korn E.D. J. Lipid Res. 1963; 4: 17-23Abstract Full Text PDF PubMed Google Scholar); in macrophages, it is switched on when the cells are activated (7.Auwerx J.H. Deeb S.S. Brunzell J.D. Peng R. Chait A. Biochemistry. 1988; 27: 2651-2655Crossref PubMed Scopus (98) Google Scholar); and in brown adipose tissue, the enzyme is switched on during cold adaptation (8.Carneheim C. Nedergaard J. Cannon B. Am. J. Physiol. 1988; 254: E155-E161PubMed Google Scholar). In white adipose tissue, LPL activity changes during the day according to the nutritional state. This appears to be mediated by posttranscriptional mechanisms. During short term fasting, LPL activity in rat adipose tissue decreases without corresponding changes in the levels of mRNA and protein mass, thereby reducing the specific activity (activity/protein mass ratio) (18.Doolittle M.H. Ben-Zeev O. Elovson J. Martin D. Kirchgessner T.G. J. Biol. Chem. 1990; 265: 4570-4577Abstract Full Text PDF PubMed Google Scholar, 20.Masuno H. Blanchette-Mackie E.J. Schultz C.J. Spaeth A.E. Scow R.O. Okuda H. J. Lipid Res. 1992; 33: 1343-1349Abstract Full Text PDF PubMed Google Scholar, 21.Park J.-W. Oh M.-S. Yang J.-Y. Park B.-H. Rho H.-W. Lim S.-N. Jhee E.-C. Kim H.-R. Biochim. Biophys. Acta Lipids Lipid Metab. 1995; 1254: 45-50Crossref PubMed Scopus (17) Google Scholar, 22.Masuno H. Blanchette-Mackie E.J. Chernick S.S. Scow R.O. J. Biol. Chem. 1990; 265: 1628-1638Abstract Full Text PDF PubMed Google Scholar). We have found that this occurs because the distribution of lipase protein shifts toward an inactive form (9.Bergö M. Olivecrona G. Olivecrona T. Biochem. J. 1996; 313: 893-898Crossref PubMed Scopus (109) Google Scholar) and that refeeding for 4 h can restore the suppressed activity (10.Bergö M. Olivecrona G. Olivecrona T. Am. J. Physiol. Endocrinol. Metab. 1996; 271: E1092-E1097Crossref PubMed Google Scholar). The signal and mechanism for these changes in the activity status of the enzyme are unknown. To delineate the regulatory mechanism, we determined the turnover rates for LPL mRNA, mass, and activity in rat adipose tissue in the fed and fasted states and explored whether the regulation during feeding, fasting, and refeeding requires synthesis of new mRNA or protein. The results point to a novel mechanism and a new perspective on the nutritional regulation of LPL in adipose tissue. Male Sprague-Dawley rats (23 days old) weighing around 60 g were bought from Möllegaard Breeding Center (Ejby, Denmark). After transport to Umeå they were allowed to acclimatize for 7–10 days; during this time, they reached a weight of ∼120 g. The rats were kept in a well ventilated room at 21 °C and 40–50% humidity and had free access to standard laboratory chow (Laktamin AB, Stockholm, Sweden) and tap water. The light in the room was on between 6 a.m. and 6 p.m. In the fasting experiments, food was withdrawn from the cages at 6 a.m., and a grid was placed at the bottom of the cages to prevent coprophagia. In the refeeding experiments, the rats were put back on standard chow at 6 a.m. after a 24 h fast, and were killed at 12 noon by cervical dislocation. The adipose depot used in all experiments was periepididymal adipose tissue. The animal ethics committee in Umeå approved all animal experiments. Actinomycin-D, cycloheximide, bovine serum albumin, aprotinin (Trasylol), and HEPES were from Sigma. Leupeptin and pepstatin were from the Peptide Institute (Osaka, Japan). Heparin was from Lövens (Malmö, Sweden). The substrate for the LPL activity assay was 3H-labeled triolein in Intralipid (10%) from Amersham Biosciences, Inc. Parker medium (Parker 199) was from SBL (Stockholm, Sweden). All other reagents were of the highest commercial grade possible. Actinomycin and cycloheximide were dissolved in 0.9% NaCl (saline) and injected intraperitoneally at concentrations of 2 and 35 mg/kg body weight, respectively, unless otherwise stated. The injected volume was 0.3 ml. LPL was extracted from tissues by homogenization in a Tris-HCl buffer (pH 8.2) containing detergents and protease inhibitors (9.Bergö M. Olivecrona G. Olivecrona T. Biochem. J. 1996; 313: 893-898Crossref PubMed Scopus (109) Google Scholar). The homogenate was centrifuged for 15 min at 3000 rpm, after which the intermediate phase (between the floating fat droplets and the pellet) for adipose tissue and the supernatant for other tissues was used for assay of LPL activity and immunoreactivity. LPL activity was measured as described previously (9.Bergö M. Olivecrona G. Olivecrona T. Biochem. J. 1996; 313: 893-898Crossref PubMed Scopus (109) Google Scholar). Briefly, 4 μl of tissue homogenate (duplicate samples) was incubated for 60 min at 25 °C with substrate in the presence of 10 μl of heat-inactivated serum from fasted rats (as source of apolipoprotein CII) and 6% bovine serum albumin. The total volume was 200 μl. After termination of lipolysis the fatty acids were extracted and counted for radioactivity. One milliunit of lipase activity represents 1 nmol of fatty acids released per minute. LPL immunoreactivity was measured with a sandwich enzyme-linked immunoadsorbent assay as described (9.Bergö M. Olivecrona G. Olivecrona T. Biochem. J. 1996; 313: 893-898Crossref PubMed Scopus (109) Google Scholar). Briefly, three different dilutions of tissue homogenate were incubated in microtiter plate wells previously coated with affinity-purified chicken anti-LPL IgG. Detection was mediated via the 5D2 monoclonal antibody (from Dr. John Brunzell, Seattle, WA) followed by a peroxidase-labeled anti-mouse IgG antibody. Absorbency at 490 nm was measured with a Spectramax microplate spectrophotometer (Molecular Devices, Sunnyvale, CA). LPL mRNA was quantified as described (10.Bergö M. Olivecrona G. Olivecrona T. Am. J. Physiol. Endocrinol. Metab. 1996; 271: E1092-E1097Crossref PubMed Google Scholar) with a ribonuclease protection solution hybridization assay modified from Durnam and Palmiter (11.Durnam D.M. Palmiter R.D. Anal. Biochem. 1983; 131: 385-393Crossref PubMed Scopus (363) Google Scholar). A 32P-labeled LPL cRNA probe synthesized from a linearized pGEM4z plasmid was incubated for 40 h with total nucleic acid samples from rat adipose tissue. Three different dilutions of each sample were assayed. The samples were precipitated, collected on Whatman filters, and counted for radioactivity. The values were compared with a standard curve and expressed relative to the amount of DNA in the sample. DNA was measured by fluorometry as described by Labarca and Paigen (12.Labarca C. Paigen K. Anal. Biochem. 1980; 102: 344-352Crossref PubMed Scopus (4550) Google Scholar). Active and inactive forms of LPL in tissue homogenates were separated on heparin-Sepharose columns (9.Bergö M. Olivecrona G. Olivecrona T. Biochem. J. 1996; 313: 893-898Crossref PubMed Scopus (109) Google Scholar). The columns were eluted with a salt gradient, and the fractions were assayed for LPL activity and immunoreactive mass. Hepatic lipase in rat postheparin plasma was measured as described (13.Olivecrona T. Olivecrona G. Rifai N. Warnick G.R. Dominiczak M.H. Handbook of Lipoprotein Testing. AACC Press, Washington DC2000: 479-497Google Scholar). The substrate was 3H-labeled triolein incorporated into an olive oil/gum arabic emulsion by sonication. LPL present in the plasma was inhibited by a high salt concentration. Statistics were performed by one-way analysis of variance. In some cases, the Bonferroni posthoc test for multiple comparisons was used. The SPSS program for Windows was used (SPSS Inc., Chicago, IL). Fasting for 24 h reduced LPL activity to 25% of fed control (p < 0.001), whereas LPL protein mass decreased only to 83% (p < 0.05) (Table I). After refeeding for 6 h, LPL activity had increased 4.3-fold to 107% of fed control, but LPL protein mass remained unchanged. In rats that had been given cycloheximide to inhibit protein synthesis, both LPL activity and protein mass dropped sharply despite refeeding (Table I). These data indicate that LPL protein is turned over rapidly and that the up-regulation of LPL during refeeding requires synthesis of new protein.Table IEffect of nutritional state on LPL activity and mass in rat adipose tissue during inhibition of transcription and/or translationTreatmentLPL activityLPL massnRefeedingCxAct-Dmu/μg DNAng/μg DNAFed94.67 ± 0.2014.81 ± 0.66Fasted91.17 ± 0.0412.33 ± 0.47Fasted4+5.00 ± 0.6212.60 ± 1.37Fasted4++0.16 ± 0.031.54 ± 0.19Fasted5++4.67 ± 0.6012.60 ± 1.49Fasted5+4.56 ± 0.1310.11 ± 0.42Fasted3+0.13 ± 0.061.38 ± 0.28Groups of rats were fasted for 24 h from 6 a.m. the morning before the experiment or were fed ad libitum. At 6 a.m. the next morning rats were injected with cycloheximide (Cx), actinomycin-D (Act-D), or saline. Some of the rats were also refed ad libitum for 6 h while the others remained fasted. At 12 noon, the rats were killed by decapitation, and adipose tissue was removed and analyzed for LPL activity and mass as described under "Materials and Methods." Data are from three separate experiments and are expressed per μg DNA (mean ± S.E.). Open table in a new tab Groups of rats were fasted for 24 h from 6 a.m. the morning before the experiment or were fed ad libitum. At 6 a.m. the next morning rats were injected with cycloheximide (Cx), actinomycin-D (Act-D), or saline. Some of the rats were also refed ad libitum for 6 h while the others remained fasted. At 12 noon, the rats were killed by decapitation, and adipose tissue was removed and analyzed for LPL activity and mass as described under "Materials and Methods." Data are from three separate experiments and are expressed per μg DNA (mean ± S.E.). To further analyze the turnover of LPL we studied fed rats, which have high LPL activity in their adipose tissue (Fig. 1). LPL activity and protein mass dropped to 40–50% of control within 2 h after the administration of cycloheximide and then decreased more slowly, reaching 20% of control at 6 h. These data demonstrate that after the block of protein synthesis by cycloheximide, the half-lives of LPL activity and protein mass in adipose tissue are less than 2 h and that LPL activity decreases as a result of protein turnover and not because the enzyme protein loses its catalytic activity. In these experiments, we used actinomycin-D, which inhibits DNA-dependent mRNA synthesis. When 24-h fasted rats were injected with actinomycin and then refed, LPL activity increased to the same level as in fed rats (Table I). LPL protein mass did not change significantly. Hence, up-regulation of LPL activity during refeeding does not require synthesis of new mRNA. Next, we determined whether down-regulation of LPL activity during fasting requires synthesis of new mRNA. Fed rats were injected with actinomycin or saline, and food was removed from the cages (Fig. 2). Eighteen h later, LPL-specific activity in adipose tissue from control rats (saline-injected) had decreased, as expected, to 30% of fed control. Surprisingly, in adipose tissue from actinomycin-injected rats, LPL-specific activity was slightly increased (140% of fed, p < 0.05). These findings show that the down-regulation of LPL-specific activity during fasting requires synthesis of new mRNA and suggest that, during fasting, a gene is switched on whose product prevents LPL from becoming active even though synthesis of LPL protein continues unabated. We predicted that if we injected actinomycin into 24-h fasted rats with low adipose tissue LPL activity, we would turn off this proposed inhibitory gene, allowing active LPL to be formed again. Our prediction was upheld; 6 h after injection of actinomycin into fasted rats, LPL activity in adipose tissue had increased almost 4-fold without refeeding (Table I), with no statistically significant change in LPL protein mass. These data indicate that turning off transcription with actinomycin could reproduce the signal that causes up-regulation of LPL activity on refeeding. To test the possibility that the increase in LPL activity by actinomycin might be caused by a paradoxical effect on the LPL protein by the actinomycin molecule itself, we performed two control experiments. First, we determined whether actinomycin affected LPL activity when added to tissue homogenates in vitro. For this, actinomycin was added at concentrations of 5, 10, 50, and 100 μg/ml either to homogenates of adipose tissue from fasted rats before assay of LPL activity or directly to the activity assay medium. In vitro addition of actinomycin had no effect on LPL activity (data not shown). Second, we performed experiments with another inhibitor of DNA-dependent RNA synthesis, α-amanitin. Three groups of rats were fasted for 24 h. Then, one group was given free access to food (refed), one was injected with α-amanitin (3 mg/kg body weight in saline), and one was injected with saline only (fasted controls). Six h later, the rats were killed, and adipose tissue LPL activity and protein mass were determined. LPL-specific activity (milliunit/mg) in adipose tissue increased from 0.14 in the fasted group to 0.40 in the refed group as expected on refeeding. In the α-amanitin-treated group, LPL specific activity also increased (0.35 milliunit/mg, p < 0.001 versus fasted controls, data not shown). These results indicate that the effect of actinomycin on LPL activity in adipose tissue was not a spurious finding; it could be reproduced with an unrelated inhibitor of transcription, α-amanitin. Previously, we showed that after 24 h of fasting, LPL activity in adipose tissue is reduced, but the level of LPL mRNA remains the same as in the fed state; however, after 60 h of fasting, the mRNA level decreases significantly to less than 50% of fed control (10.Bergö M. Olivecrona G. Olivecrona T. Am. J. Physiol. Endocrinol. Metab. 1996; 271: E1092-E1097Crossref PubMed Google Scholar). At this point, the effect of refeeding is blunted. In the following experiment, we tested whether the effect of actinomycin was similarly blunted after prolonged fasting. When rats were fasted for 60 h, LPL activity was greatly reduced (to 7.2% of fed control (Fig. 3)). When these rats were given actinomycin, the increase in LPL activity after 6 h was very modest (from 7.2 to 26% of the level in fed rats). This indicates that the magnitude of the effect of actinomycin on adipose tissue LPL activity depends on the level of LPL mRNA. A survey of the literature indicated that the most frequently used dose for actinomycin in vivo is 2 mg/kg body weight. In a dose-response experiment, LPL activity in adipose tissue increased from 1.16 ± 0.08 milliunit/μg DNA (fasted control, injected with vehicle) to 2.48 ± 0.29 milliunit/μg DNA with as little as 0.5 mg/kg 6 h after injection (p < 0.001). With a dose of 2 mg/kg, LPL activity reached 4.47 ± 0.01 milliunit/μg DNA. A further increase in the dose to 6 mg/kg did not result in an additional increase in LPL activity (p > 0.05). Hence, a dose of 2 mg/kg was used in subsequent experiments. LPL activity in adipose tissue of 24 h fasted rats was 21% of that in fed rats (Fig. 4). By 4 h after administration of actinomycin, the activity had increased to 51% of fed control, and by 6 h the activity was restored (97%). There was no further significant change through 16 h (114%). In this experiment, in the fed and fasted states and for 16 h after actinomycin administration, LPL protein mass did not change significantly. In a separate experiment, LPL mRNA was measured in the adipose tissue at 0, 6, 16, and 24 h after actinomycin administration. The level of LPL mRNA was 68.4 ± 11.1 and 72.7 ± 9.2 amol/μg DNA in fed and 24 h fasted rats, respectively (Fig. 5). After injection of actinomycin the LPL mRNA level decreased monoexponentially. The k values for the best-fit exponential function (n = N0 × e−k×t) were 0.052 ± 0.013 h−1 and 0.041 ± 0.012 h−1 for the fed and fasted states, respectively. This corresponds to half-lives of 13.3 h and 16.8 h (p > 0.05).Figure 5Turnover of LPL mRNA in adipose tissue from rats following injection of actinomycin. Fed (■) and 24 h fasted (▴) rats were injected with actinomycin. At 0, 6, 16, and 24 h, three or four rats were killed by decapitation, and epididymal adipose tissue was taken for measurement of LPL mRNA as described under "Materials and Methods." Data are expressed as amol LPL mRNA/μg DNA.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Six h after the administration of actinomycin to fasted rats, LPL activity increased slightly and inconsistently in lung, kidney, liver, and spleen. In soleus muscle and in heart, which like adipose tissue express high levels of LPL, LPL activity increased 1.5- and 1.6-fold, respectively (p < 0.01) compared with saline-injected controls. These changes were not accompanied by changes in LPL mass. The major increase in LPL activity in adipose tissue and the minor increases in some other tissues after actinomycin resulted in a 4.7-fold increase in LPL activity released to the circulating blood 10 min after a heparin injection. LPL activity in preheparin plasma increased 2-fold (p < 0.001). There were no statistically significant changes in hepatic lipase activity as measured in liver homogenate or in postheparin plasma (data not shown). Previously, we showed that there are two forms of LPL protein in tissues that can be separated by heparin-Sepharose columns. In the fasted state, the inactive form with low heparin affinity predominates. Upon refeeding, the distribution of lipase protein shifts from the inactive toward the active form (9.Bergö M. Olivecrona G. Olivecrona T. Biochem. J. 1996; 313: 893-898Crossref PubMed Scopus (109) Google Scholar), which has a high affinity for heparin. To further test whether actinomycin had an effect similar to that of refeeding, three rats were fasted for 24 h and then injected with actinomycin. Six h later, epididymal adipose tissue pieces were collected and homogenized, and the LPL protein was separated on heparin-Sepharose. Fig. 6. shows that in actinomycin-treated rats (Fig. 6A), most of the enzyme protein was in the second peak, which corresponds to the active form, whereas in the fasted rats, the first peak predominated (Fig. 6B). The ratios between the two peaks (inactive/active) were 0.38 in the actinomycin-treated rats and 2.27 in the fasted rats. Normal values for the peak ratios are 0.46 ± 0.04 and 2.34 ± 0.13 for fed and fasted rats, respectively (9.Bergö M. Olivecrona G. Olivecrona T. Biochem. J. 1996; 313: 893-898Crossref PubMed Scopus (109) Google Scholar). These data show that LPL in adipose tissue from fasted rats injected with actinomycin is present mainly in the active form, similar to LPL in adipose tissue from fed rats. We have previously shown that the mechanism that channels LPL toward the inactive form on fasting is blunted in adipose tissue from aging, obese rats (14.Bergö M. Olivecrona G. Olivecrona T. Int. J. Obes. Relat. Metab. Disord. 1997; 21: 980-986Crossref PubMed Scopus (19) Google Scholar). If actinomycin interferes with the same mechanism, its effect should also be blunted in aging rats. To test this hypothesis, rats weighing 505 ± 6 g were fasted for 24 h and then injected with actinomycin (Fig. 7). In response to actinomycin, LPL activity increased to 149% of saline-injected control in 6 h (p < 0.01). This was in sharp contrast to the young rats (126 ± 2 g), where LPL activity was up-regulated to 485% of control (p < 0.001). These data show that the response of adipose tissue LPL to actinomycin is blunted in aging rats. This study shows that in fasting rats, a gene is switched on in adipose tissue that makes the tissue produce an inactive form of LPL. Inhibition of mRNA synthesis by actinomycin completely blocked the down-regulation of LPL activity in adipose tissue during fasting. Administration of actinomycin or α-amanitin, another inhibitor of transcription, to fasted rats completely restored LPL activity within 6 h. One potential explanation for our results is that actinomycin administered in vivo changes the levels of a circulating hormone or changes nerve signaling activity to the adipose tissue and that the effect on LPL activity is secondary. Studies from the 1960s indicate that this is not the case. In these studies, actinomycin caused a dramatic increase in LPL activity in whole fat pads from fasted rats incubated ex vivo (15.Schotz M.C. Garfinkel A.S. Biochim. Biophys. Acta. 1965; 106: 202-205Crossref PubMed Scopus (15) Google Scholar, 16.Wing D.R. Robinson D.S. Biochem. J. 1968; 106: 667-676Crossref PubMed Scopus (26) Google Scholar). This clearly indicates that the effect of actinomycin is a direct action in the adipose tissue, rather than secondary to changes in nerve signaling or in circulating hormones. Several studies have shown that N-linked glycosylation and proper processing of the glycan chains are important steps in the formation of active, secretable LPL with high heparin affinity (1.Braun J.E.A. Severson D.L. Biochem. J. 1992; 287: 337-347Crossref PubMed Scopus (266) Google Scholar,17.Semb H. Olivecrona T. J. Biol. Chem. 1989; 264: 4195-4200Abstract Full Text PDF PubMed Google Scholar, 18.Doolittle M.H. Ben-Zeev O. Elovson J. Martin D. Kirchgessner T.G. J. Biol. Chem. 1990; 265: 4570-4577Abstract Full Text PDF PubMed Google Scholar, 19.Ben-Zeev O. Doolittle M.H. Davis R.C. Elovson J. Schotz M.C. J. Biol. Chem. 1992; 267: 6219-6227Abstract Full Text PDF PubMed Google Scholar, 20.Masuno H. Blanchette-Mackie E.J. Schultz C.J. Spaeth A.E. Scow R.O. Okuda H. J. Lipid Res. 1992; 33: 1343-1349Abstract Full Text PDF PubMed Google Scholar, 21.Park J.-W. Oh M.-S. Yang J.-Y. Park B.-H. Rho H.-W. Lim S.-N. Jhee E.-C. Kim H.-R. Biochim. Biophys. Acta Lipids Lipid Metab. 1995; 1254: 45-50Crossref PubMed Scopus (17) Google Scholar). Specifically, the trimming of glucose residues by glucosidases in the endoplasmic reticulum seems to be a crucial step for the attainment of activity (19.Ben-Zeev O. Doolittle M.H. Davis R.C. Elovson J. Schotz M.C. J. Biol. Chem. 1992; 267: 6219-6227Abstract Full Text PDF PubMed Google Scholar, 20.Masuno H. Blanchette-Mackie E.J. Schultz C.J. Spaeth A.E. Scow R.O. Okuda H. J. Lipid Res. 1992; 33: 1343-1349Abstract Full Text PDF PubMed Google Scholar, 21.Park J.-W. Oh M.-S. Yang J.-Y. Park B.-H. Rho H.-W. Lim S.-N. Jhee E.-C. Kim H.-R. Biochim. Biophys. Acta Lipids Lipid Metab. 1995; 1254: 45-50Crossref PubMed Scopus (17) Google Scholar). Glycosylated proteins that have been processed in the cis-Golgi have glycan chains that are resistant to cleavage by the enzyme endoglycosidase-H (Endo-H). Doolittle et al. showed that the Endo-H-resistant form of the LPL protein predominates in adipose tissue of fed rats, whereas the Endo-H-sensitive form predominates in adipose tissue of fasted rats (18.Doolittle M.H. Ben-Zeev O. Elovson J. Martin D. Kirchgessner T.G. J. Biol. Chem. 1990; 265: 4570-4577Abstract Full Text PDF PubMed Google Scholar). We have found that the catalytically active form of LPL, purified on heparin-Sepharose, is mainly resistant to Endo-H, whereas the inactive form is Endo-H-sensitive. 2M. Bergö, G. Wu, T. Ruge, and T. Olivecrona, unpublished observations. These data suggest that the mechanism for down-regulation of LPL during fasting affects the intracellular processing or trafficking of newly synthesized LPL, probably in the endoplasmic reticulum or early Golgi. Further evidence for a specific mechanism for processing of the enzyme comes from studies on mice affected by the combined lipase deficiency mutation. LPL in adipocytes from these mice is retained in the endoplasmic reticulum in an inactive form with glycan chains that are sensitive to Endo-H (22.Masuno H. Blanchette-Mackie E.J. Chernick S.S. Scow R.O. J. Biol. Chem. 1990; 265: 1628-1638Abstract Full Text PDF PubMed Google Scholar, 23.Briquet-Laugier V. Ben-Zeev O. White A. Doolittle M.H. J. Lipid Res. 1999; 40: 2044-2058Abstract Full Text Full Text PDF PubMed Google Scholar). An interesting possibility is that this genetic defect is related to the mechanism that down-regulates LPL on fasting. By injecting cycloheximide to inhibit protein synthesis, we could follow the turnover of the existing LPL molecules. LPL activity and mass in adipose tissue from both fed and fasted rats decreased rapidly and in parallel, with a half-life of less than 2 h. This means that turnover of LPL activity is directly associated with turnover of LPL protein. There are two possible pathways for this turnover: LPL protein may be degraded within the adipose tissue or the enzyme may be transported with blood for degradation in the liver (24.Vilaró S. Llobera M. Bengtsson-Olivecrona G. Olivecrona T. Am. J. Physiol. 1988; 254: G711-G722PubMed Google Scholar, 25.Chajek-Shaul T. Friedman G. Ziv E. Bar-On H. Bengtsson-Olivecrona G. Biochim. Biophys. Acta. 1988; 963: 183-191Crossref PubMed Scopus (21) Google Scholar). The turnover of LPL mRNA seems to occur slowly under most conditions. During long term fasting the half-life has been reported to be more than 24 h in white adipose tissue (10.Bergö M. Olivecrona G. Olivecrona T. Am. J. Physiol. Endocrinol. Metab. 1996; 271: E1092-E1097Crossref PubMed Google Scholar) and 40 h in brown adipose tissue (26.Mitchell J.R.D. Jacobsson A. Kirchgessner T.G. Schotz M.C. Cannon B. Nedergaard J. Am. J. Physiol. 1992; 263: E500-E506PubMed Google Scholar). During weight loss in obese Zucker rats (27.Bessesen D.H. Robertson A.D. Eckel R.H. Am. J. Physiol. 1991; 261: E246-E251Crossref PubMed Google Scholar) and during improved diabetes control in humans (28.Simsolo R.B. Ong J.M. Saffari B. Kern P.A. J. Lipid Res. 1992; 33: 89-95Abstract Full Text PDF PubMed Google Scholar), LPL mRNA turnover was also found to be slow. In the present study, the half-life of LPL mRNA was 13.3 h in the fed state. During cold adaptation in rat brown adipose tissue, LPL activity increases significantly, mediated by increased transcription of the LPL gene; upon return to warm conditions, a specific RNA-degrading enzyme is induced that accelerates the turnover of LPL mRNA (26.Mitchell J.R.D. Jacobsson A. Kirchgessner T.G. Schotz M.C. Cannon B. Nedergaard J. Am. J. Physiol. 1992; 263: E500-E506PubMed Google Scholar). Most likely this mechanism exists also in white adipose tissue since tumor necrosis factor α can accelerate the breakdown of LPL mRNA (29.Cornelius P. Enerbäck S. Bjursell G. Olivecrona T. Pekala P.H. Biochem. J. 1988; 249: 765-769Crossref PubMed Scopus (99) Google Scholar). However, this mechanism does not seem to act during fasting since the turnover of LPL mRNA did not change (half-life = 16.8 h). For this reason, mRNA was available for continuous synthesis of LPL protein even after the animals had been given actinomycin. With such a stable mRNA, it is not possible for the cell to rapidly regulate LPL activity on a transcriptional level. Instead, the regulation must be mediated on the translational or posttranslational level. Catecholamines exert a strong effect on lipid homeostasis in adipose tissue by increasing hormone-sensitive lipase activity and by decreasing LPL activity. In an elegant study, Ranganathan et al. showed that the effect of epinephrine on LPL in cultured adipocytes is mediated by a protein that binds to the LPL transcript and effectively blocks its translation (30.Ranganathan G. Vu D. Kern P.A. J. Biol. Chem. 1997; 272: 2515-2519Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar). This raises the question whether the same RNA-binding protein causes the down-regulation of LPL activity on fasting. If this were true, the positive effect of actinomycin on LPL activity could be to block the transcription of this protein. Three lines of evidence indicate that this is not the case. First, most studies indicate that the mechanism for down-regulation of LPL during short term fasting is posttranslational and that the levels of LPL protein change very little (9.Bergö M. Olivecrona G. Olivecrona T. Biochem. J. 1996; 313: 893-898Crossref PubMed Scopus (109) Google Scholar, 18.Doolittle M.H. Ben-Zeev O. Elovson J. Martin D. Kirchgessner T.G. J. Biol. Chem. 1990; 265: 4570-4577Abstract Full Text PDF PubMed Google Scholar, 20.Masuno H. Blanchette-Mackie E.J. Schultz C.J. Spaeth A.E. Scow R.O. Okuda H. J. Lipid Res. 1992; 33: 1343-1349Abstract Full Text PDF PubMed Google Scholar, 22.Masuno H. Blanchette-Mackie E.J. Chernick S.S. Scow R.O. J. Biol. Chem. 1990; 265: 1628-1638Abstract Full Text PDF PubMed Google Scholar). If a block in LPL synthesis mediated the decrease in LPL activity on fasting there would have to be an associated block in LPL degradation. The present study shows that this is clearly not the case; the turnover of LPL protein was the same in the fed and in the fasted states. Second, Wing et al. observed an increase in LPL activity in rat fat pads incubated in the presence of actinomycin (16.Wing D.R. Robinson D.S. Biochem. J. 1968; 106: 667-676Crossref PubMed Scopus (26) Google Scholar). This effect was blocked by epinephrine, most likely by inducing the LPL mRNA-binding protein. We have preliminary evidence that this occurs also in vivo. The increase in LPL activity in adipose tissue after actinomycin administration was attenuated when rats were given epinephrine, which also caused a reduction in LPL protein mass.2 This indicates that the RNA-binding protein is activated posttranscriptionally by epinephrine, since actinomycin would inhibit synthesis of its mRNA. Third, it is unlikely that a translational inhibition of LPL followed by a decreased degradation would lead to a change in the structure of glycan chains in the remaining LPL molecules. The nutritional regulation of LPL in adipose tissue has been viewed as having a default state characterized by a low, basal LPL activity that can be temporarily up-regulated after meals. Our findings in this study suggest another perspective on the regulation: the default (fed) state is characterized by a high LPL activity and mass synthesized from a stable mRNA where the newly synthesized LPL is processed mainly into the active form. The enzyme is then secreted and transported to nearby capillaries where it acts on lipoproteins. This pathway seems to predominate in most tissues that synthesize LPL (e.g., heart and skeletal muscle). During periods of caloric restriction (i.e., between meals or during fasting), LPL mRNA and protein mass remain high; however, LPL activity in adipose tissue can be suppressed by the induction of a short-lived gene product that causes newly synthesized LPL to be channeled into an inactive form. Refeeding or administration of actinomycin inhibits the expression of this putative factor, allowing active LPL to be formed. Why is there a need to down-regulate LPL during fasting? The action of adipose LPL on circulating triglycerides results in the generation of fatty acids that can either be taken up by the tissue or be released into the blood in albumin-bound form. The ratio between these two pathways varies greatly in human adipose tissue (31.Frayn K.N. Coppack S.W. Fielding B.A. Humphreys S.M. Advan. Enzyme Regul. 1995; 35: 163-178Crossref PubMed Scopus (150) Google Scholar), from predominantly storage after a large meal to predominantly release into the circulating blood during fasting. Irrespective of the fate of the generated fatty acids, the down-regulation of LPL during fasting will decrease the hydrolysis of circulating triglycerides that pass through the adipose tissue. Failure to down-regulate LPL could contribute to the excessive generation of albumin-bound fatty acids seen in individuals with insulin resistance, or it could contribute to an increased deposition of fat in adipose tissue. Consistent with this reasoning, Ong et al. found that LPL-specific activity in adipose tissue during fasting was higher in obese subjects than in lean controls (32.Ong J.M. Kern P.A. J. Clin. Invest. 1989; 84: 305-311Crossref PubMed Scopus (120) Google Scholar). Furthermore, in obese subjects, LPL-specific activity was unresponsive to a meal (refeeding). Our findings suggest that this apparent unresponsiveness to a meal may in fact reflect that LPL activity in these subjects was not down-regulated during the fasting period. If so, this would be similar to what we have found in aging, obese rats where LPL-specific activity failed to be down-regulated on fasting (14.Bergö M. Olivecrona G. Olivecrona T. Int. J. Obes. Relat. Metab. Disord. 1997; 21: 980-986Crossref PubMed Scopus (19) Google Scholar). This raises the intriguing possibility that switching on and off the gene that regulates LPL activity in adipose tissue may be deranged in obesity, insulin resistance, and related metabolic disorders. We are indebted to Ann-Sofi Jakobsson for excellent technical assistance.

Referência(s)