Prediction of PPAR-α ligand-mediated physiological changes using gene expression profiles

Abstract

Peroxisome proliferator-activated receptor (PPAR)-α controls the transcription of a variety of genes involved in lipid metabolism and is the target receptor for the hypolipidemic drug class of fibrates. In the present study, the molecular and physiological effects of seven different PPAR-activating drugs have been examined in a rodent model of dyslipidemia. The drugs examined were selected to display varying potencies and efficacies toward PPAR-α. To help elucidate the link between the gene regulation elicited by PPAR-α ligands and the concomitant physiological changes, we have used cDNA microarray analysis to identify smaller gene sets that are predictive of the function of these ligands.A number of genes showed strong correlations to the relative PPAR-α efficacy of the drugs. Furthermore, using multivariate analysis, a strong relationship between the drug-induced triglyceride lowering and the transcriptional profiles of the different drugs could be found. Peroxisome proliferator-activated receptor (PPAR)-α controls the transcription of a variety of genes involved in lipid metabolism and is the target receptor for the hypolipidemic drug class of fibrates. In the present study, the molecular and physiological effects of seven different PPAR-activating drugs have been examined in a rodent model of dyslipidemia. The drugs examined were selected to display varying potencies and efficacies toward PPAR-α. To help elucidate the link between the gene regulation elicited by PPAR-α ligands and the concomitant physiological changes, we have used cDNA microarray analysis to identify smaller gene sets that are predictive of the function of these ligands. A number of genes showed strong correlations to the relative PPAR-α efficacy of the drugs. Furthermore, using multivariate analysis, a strong relationship between the drug-induced triglyceride lowering and the transcriptional profiles of the different drugs could be found. ErrataJournal of Lipid ResearchVol. 46Issue 6PreviewIn the article “Substrate reduction reduces gangliosides in postnatal cerebrum-brainstem and cerebellum in GM1 gangliosidosis mice” by Kasperzyk et al., published in the April 2004 issue of the Journal of Lipid Research (Volume 46, pages 744–751), the affiliations should read as follows: Full-Text PDF Open Access Peroxisome proliferator-activated receptor (PPAR)-α is recognized as the target receptor for fibrates (1Issemann I. Green S. Activation of a member of the steroid hormone receptor superfamily by peroxisome proliferators.Nature. 1990; 347: 645-650Crossref PubMed Scopus (3059) Google Scholar), which are used in the treatment of dyslipidemic patients. Fibrates are recognized to exert their hypolipidemic action by altering lipid metabolism at multiple levels, and PPAR-α is an important regulator of both intracellular and extracellular lipid metabolism. At the intracellular level, PPAR-α regulates a number of genes that encode enzymes involved in the oxidation of fatty acids. These include upregulations of acyl-CoA oxidase (2Dreyer C. Krey G. Keller H. Givel F. Helftenbein G. Wahli W. Control of the peroxisomal beta-oxidation pathway by a novel family of nuclear hormone receptors.Cell. 1992; 68: 879-887Abstract Full Text PDF PubMed Scopus (1214) Google Scholar, 3Tugwood J.D. Issemann I. Anderson R.G. Bundell K.R. McPheat W.L. Green S. The mouse peroxisome proliferator activated receptor recognizes a response element in the 5′ flanking sequence of the rat acyl CoA oxidase gene.EMBO J. 1992; 11: 433-439Crossref PubMed Scopus (804) Google Scholar) the rate-limiting enzyme of peroxisomal β-oxidation, medium-chain acyl-CoA dehydrogenase (4Gulick T. Cresci S. Caira T. Moore D.D. Kelly D.P. The peroxisome proliferator-activated receptor regulates mitochondrial fatty acid oxidative enzyme gene expression.Proc. Natl. Acad. Sci. USA. 1994; 91: 11012-11016Crossref PubMed Scopus (491) Google Scholar), a central enzyme of mitochondrial β-oxidation, and cytochrome P450 genes (5Muerhoff A.S. Griffin K.J. Johnson E.F. The peroxisome proliferator-activated receptor mediates the induction of CYP4A6, a cytochrome P450 fatty acid omega-hydroxylase, by clofibric acid.J. Biol. Chem. 1992; 267: 19051-19053Abstract Full Text PDF PubMed Google Scholar) involved in microsomal ω-hydroxylation of fatty acids. The induction of fatty acid oxidation, combined with the upregulation of the fatty acid transport protein (6Martin G. Schoonjans K. Lefebvre A.M. Staels B. Auwerx J. Coordinate regulation of the expression of the fatty acid transport protein and acyl-CoA synthetase genes by PPARalpha and PPARgamma activators.J. Biol. Chem. 1997; 272: 28210-28217Abstract Full Text Full Text PDF PubMed Scopus (479) Google Scholar), causes a shift in hepatic fatty acid metabolism with decreased triglyceride synthesis and increased catabolism. These intracellular changes are accompanied by regulation of extracellular lipid metabolism and transport. Induction of the lipoprotein lipase by PPAR-α (7Schoonjans K. Peinado-Onsurbe J. Lefebvre A.M. Heyman R.A. Briggs M. Deeb S. Staels B. Auwerx J. PPAR alpha and PPAR activators direct a distinct tissue-specific transcriptional response via a PPRE in the lipoprotein lipase gene.EMBO J. 1996; 15: 5336-5348Crossref PubMed Scopus (1026) Google Scholar) increases the lipolysis of triglycerides in chylomicrons and VLDL particles. This lipolysis generates precursors to HDL particles. The expression of the two major apolipoproteins of HDL particles, apolipoproteins A-I and A-II, is also upregulated by PPAR-α (8Vu D.N. Schoonjans K. Laine B. Fruchart J.C. Auwerx J. Staels B. Negative regulation of the human apolipoprotein A-I promoter by fibrates can be attenuated by the interaction of the peroxisome proliferator-activated receptor with its response element.J. Biol. Chem. 1994; 269: 31012-31018Abstract Full Text PDF PubMed Google Scholar, 9Vu D.N. Schoonjans K. Kosykh V. Dallongeville J. Fruchart J.C. Staels B. Auwerx J. Fibrates increase human apolipoprotein A-II expression through activation of the peroxisome proliferator-activated receptor.J. Clin. Invest. 1995; 96: 741-750Crossref PubMed Scopus (366) Google Scholar). These events are important factors in fibrate-induced increases of HDL levels. The increased HDL levels may, along with increased cholesterol efflux from peripheral cells through upregulation of the ATP binding cassette transporter A1 transport protein (10Chinetti G. Lestavel S. Bocher V. Remaley A.T. Neve B. Torra I.P. Teissier E. Minnich A. Jaye M. Duverger N. Brewer H.B. Fruchart J.C. Clavey V. Staels B. PPAR-alpha and PPAR-gamma activators induce cholesterol removal from human macrophage foam cells through stimulation of the ABCA1 pathway.Nat. Med. 2001; 7: 53-58Crossref PubMed Scopus (1011) Google Scholar), facilitate the reverse cholesterol transport from peripheral tissues back to the liver. Furthermore, apolipoprotein C-III (apoC-III), a major lipoprotein of chylomicrons and VLDL particles, is transcriptionally downregulated by PPAR-α (11Staels B. Vu D.N. Kosykh V.A. Saladin R. Fruchart J.C. Dallongeville J. Auwerx J. Fibrates downregulate apolipoprotein C-III expression independent of induction of peroxisomal acyl coenzyme A oxidase. A potential mechanism for the hypolipidemic action of fibrates.J. Clin. Invest. 1995; 95: 705-712Crossref PubMed Google Scholar). In the present study, rats fed a high-cholesterol diet (HCD) were used as a dyslipidemic model (12Paigen B. Morrow A. Brandon C. Mitchell D. Holmes P. Variation in susceptibility to atherosclerosis among inbred strains of mice.Atherosclerosis. 1985; 57: 65-74Abstract Full Text PDF PubMed Scopus (491) Google Scholar) to study the physiological and transcriptional changes occurring as the result of treatment with various PPAR-α activators. Seven different PPAR-α activators with varying efficacies were selected and tested in the rat model. Using cDNA microarrays, the induced transcriptional changes in the liver were examined. The transcriptional changes of genes encoding apoC-III and peroxisomal enoyl-CoA:hydrotase-3-hydroxyacyl-CoA bifunctional enzyme (bifunctional enzyme) correlated very well to the in vitro-measured PPAR-α efficacies of the drugs. Thus, these genes may be useful as markers for the in vivo efficacy of PPAR-α activators. Furthermore, using multivariate analysis, we identified a set of 19 genes capable of predicting the triglyceride-lowering capacity of each of the drugs. All cDNA inserts were PCR-amplified using the T7 and M13 reverse primers. PCR products (100 μl each) were purified on the Qiagen 96-well PCR purification system and subsequently checked for size and purity on agarose gels. The purified PCR products were mixed 1:1 in DMSO (Sigma, St. Louis, MO) and rearranged in 384-well plates. cDNA targets were spotted in duplicate onto silanized glass slides (Amersham Biosciences, Buckinghamshire, UK) at 55% relative humidity using the Molecular Dynamics Gen. III microarray spotter. All slides were UV cross-linked (50 mJ) in a Stratagene UV cross-linker. Cy-3- and Cy-5-labeled single-stranded cDNA probes were prepared from 15 μg of total RNA from the test and control tissues, respectively. The RNA was incubated with 1 μg of anchored oligo-dT primer at 70°C for 10 min and subsequently chilled on ice for 30 s The primer-annealed RNA was incubated with reaction mixture [1× Superscript II buffer (Life Technologies, Taastrup, Denmark), 10 mM DTT, 200 μM deoxy-guanosine, -adenosine, -thymidine triphosphate, 100 μM dCTP, 100 μM Cy-3/Cy-5-dCTP (Amersham Biosciences), and 200 U of Superscript II reverse transcriptase (Life Technologies)] at 42°C for 2.5 h in a final volume of 20 μl. The RNA template was removed by alkaline denaturation (incubation with 2 μl of 2.5 M NaOH at 37°C for 15 min). Labeled probes were purified on GFX spin columns (Pharmacia Biotech, Uppsala, Sweden). The Cy-3- and Cy-5-labeled cDNAs were mixed 1:1 in hybridization buffer version 2 (Amersham Biosciences) and subsequently 1:1 in formamide (Sigma). The mixed probes were injected into an automated slide processor (Molecular Dynamics, Sunnyvale, CA) and hybridized at 42°C for 16 h. After hybridization, the slide was washed in 1× SSC, 0.2% SDS at 55°C for 10 min and subsequently in 0.1× SSC, 0.2% SDS at 55°C for 10 min. The slide was scanned in a Gen. III microarray laser scanner (Molecular Dynamics). Image analyses were performed with All were using The and were to the of of the in each RNA was from tissues using the to the of the with two of 20 μg of total RNA was on a 20 mM mM and 1 mM The RNA was to a (Amersham Biosciences) using and was subsequently by UV cDNA were using the and hybridization and the buffer was were using a (Molecular and with (Molecular Dynamics). cDNA was prepared from 1 μg of total RNA from each of the treatment using and CA) to the PCR was performed on from each treatment of using PCR on an and for lipoprotein lipase fatty acid transport protein acyl-CoA and were as for these were as follows: acyl-CoA oxidase and were using SDS and expression levels for and acyl-CoA oxidase were to the levels. cells were in were on 96-well the to a of at A total of μg of μg of μg of μg of and μg of was well using to the were to protein for h by the of and cDNAs were by PCR using cDNA by reverse transcription of from human and respectively. cDNAs were into and The binding of each PPAR was by PCR acid to the acid to the acid to the and to the binding of the transcription by in into the I. B. P. M. for expression in 1992; PubMed Scopus Google Scholar), the and were by The was by an encoding of the sequence sequence of the gene of PubMed Scopus Google Scholar) into the the was from and was from All PPAR-α ligands were in DMSO and to the were tested in in from to were with for h by was tested in at test was and 100 μl of 1 mM and was to each The was performed using the to the was by on a To a μl from each was to a were performed on the using a from and read in a The were used to and for the The of a is as induction with an each the efficacy is as relative with for for and for The is the of were via using The are as rats of were fed a Research for 10 The cholesterol induced a and and the of of were for the These have been to induce efficacy in A of rats fed was in A total of were were from the into of and triglycerides were of treatment on the on was at the of the and h were on All analyses were performed on a are as with of the and was from and was from The D. The synthesis of novel and Med. 1994; Scopus Google Scholar), P. I. K. J. J. H.B. K. V. PPAR with hypolipidemic and Med. Chem. PubMed Scopus Google Scholar), J. P. P. S. and I. acid and their in the treatment of and Scholar), J. P. P. S. and I. acid and their and Scholar), and J. P. P. S. and I. acid and their in the treatment of and Scholar) were to published The to analysis was performed with Sweden). The from the 19 genes in 2 in both and were used as All were and to and the analysis was identified to be in by bifunctional P450 enzyme dehydrogenase for binding binding fatty acyl CoA response protein for element binding to synthetase acyl-CoA transporter PCR analyses of PPAR target genes that were in the microarray analyses present on the acid transport protein and acyl-CoA PCR analyses of PPAR target genes that were in the microarray analyses present on the acid transport protein and acyl-CoA PCR analyses of PPAR target genes that were in the microarray analyses present on the acid transport protein and acyl-CoA acid transport gene expression was using a cDNA microarray The the of genes that are by at of the PPAR-α are as expression levels from high-cholesterol with expression levels from high-cholesterol changes in and from 10 of from for as in and PCR analyses of PPAR target genes that were in the microarray analyses present on the acid transport protein and acyl-CoA Open in a gene expression was using a cDNA microarray The the of genes that are by at of the PPAR-α are as expression levels from high-cholesterol with expression levels from high-cholesterol changes in and from 10 of from for as in and In receptor with the binding of each of the PPAR receptor were used to the of PPAR activation capacity of both the and the test To the efficacy of between and were used as in the and respectively. activation with the with in the PPAR-α with in the and with in the was as The in PPAR-activating of the used in study are in of different efficacies are as relative to for peroxisome proliferator-activated for and for of Open in a efficacies are as relative to for peroxisome proliferator-activated for and for of is from the were selected to display varying PPAR-α The was as a PPAR-α The of the selected are in a model for rats were fed a and (12Paigen B. Morrow A. Brandon C. Mitchell D. Holmes P. Variation in susceptibility to atherosclerosis among inbred strains of mice.Atherosclerosis. 1985; 57: 65-74Abstract Full Text PDF PubMed Scopus (491) Google Scholar) for drug The changes in triglyceride total cholesterol and in with are in The in 1 were subsequently to the for were at to induce efficacy for the of The levels of total and were on the of The induced changes are in All a of triglyceride and total The reduction of triglycerides from to The of total cholesterol was with HDL levels were increased by and were for the of the PPAR-α are to peroxisomal in of increases in liver were also from treatment liver as of total increased the with of and increases were also for and with and in liver To into the of transcriptional changes induced by the PPAR cDNA microarray analyses were performed on RNA from liver. RNA was prepared from and were from each of the A cDNA microarray genes was This gene identified by display T. Holmes E. K. J. genes by the PubMed Scopus Google Scholar) along with 20 identified PPAR-α Furthermore, encoding enzymes of the major and and transcription factors were in the selected the of gene All were performed as using the and to the and test respectively. were with in the of test and A total of genes were to be by at of the that were to be at and of To on the transcriptional changes that were to be a direct of the PPAR-α genes that were by at were identified and are in To the regulation of the genes in 2 the in efficacies of the the of each gene was as a function of the relative the genes encoding peroxisomal bifunctional cytochrome P450 binding binding protein and very correlations between the expression changes and the relative in efficacies of the ligands were found. for these genes are To the of the microarray and the to the in analysis were on these The for apoC-III and bifunctional enzyme the microarray and the correlations to the relative in efficacies and were also to be on the the to relative efficacies was between and the on the microarray from the were to the for which been to be by PPAR-α in rat liver (6Martin G. Schoonjans K. Lefebvre A.M. Staels B. Auwerx J. Coordinate regulation of the expression of the fatty acid transport protein and acyl-CoA synthetase genes by PPARalpha and PPARgamma activators.J. Biol. Chem. 1997; 272: 28210-28217Abstract Full Text Full Text PDF PubMed Scopus (479) Google Scholar). In the of genes expression to of the physiological each gene of the identified gene set was tested for of the identified genes on their very well to the changes in total cholesterol apoC-III reduction was to to the increase in HDL levels is that the in is which of the PPAR-α target genes The increases in liver of the different treatment correlated to of the physiological to the relative in efficacies of the PPAR-α apoC-III to be a useful for HDL increase and the relative in PPAR-α efficacy also to a to the PPAR-α efficacies in to the physiological changes very as the combined expression profiles from multiple may be a of predicting physiological on gene expression In an to the changes as in vivo by using of the transcriptional changes, multivariate analysis was a multivariate model to two and S. M. a of 2001; Scopus Google Scholar). Using analysis, of the gene expression changes were as and the were as multiple can correlated using the for the genes from and control in both and were used as All were and to and using was to a model to the in triglyceride with and was the that was the the model for triglyceride by the the 2 the between the and decreased from to This that the model for triglyceride lowering is and In using the set of gene expression was to that could total cholesterol changes HDL In the was as well by the expression of In the present study, rats fed a (12Paigen B. Morrow A. Brandon C. Mitchell D. Holmes P. Variation in susceptibility to atherosclerosis among inbred strains of mice.Atherosclerosis. 1985; 57: 65-74Abstract Full Text PDF PubMed Scopus (491) Google Scholar) were used as a model for to physiological and transcriptional changes treatment with seven different PPAR-α activators. These the and along with and The were in their relative PPAR-α efficacies from to and were as well for a study of transcriptional and physiological The PPAR-α activation of the help to genes from gene All of the PPAR-α a lowering on the levels of triglyceride and total cholesterol The reduction in levels of triglyceride from for to for of triglycerides is the of fibrates in both and and drugs as R. K. D. R. E. of to in 1994; PubMed Scopus Google Scholar), A. M. E. of a in of rat a with 1996; Full Text Full Text PDF PubMed Scopus Google Scholar), and G. changes in J. 2001; PubMed Scopus Google Scholar) have been to triglycerides in the of in cholesterol is also a that been to be by fibrates in rats of of for hypolipidemic a for Full Text Full Text PDF PubMed Scopus Google Scholar). In the present study, in total cholesterol were with and as the drugs. These can be by the very increase in total The drug-induced increase in is also very in In drugs as and have S. T. of fibrates to changes in hepatic apolipoprotein C-III a potential for their of Lipid 1995; Full Text PDF PubMed Google Scholar). in rats fed a and have been to increase increases both A-I and to lipid in Full Text PDF PubMed Scopus Google Scholar, D. C. R. of and on liver and lipoprotein synthesis in and Full Text PDF PubMed Scopus Google Scholar). In the present study, an increase in was for and an increase in was also for to be to factors activation of and which are PPAR-α increase HDL HDL in the of The upregulation of HDL levels, the of the HDL upregulation by the PPAR-α is an through PPAR-α of the genes that were to be are involved in lipid metabolism and have been to be by PPAR These include bifunctional enzyme N. Green S. a peroxisome proliferator response element of the bifunctional enzyme PubMed Scopus Google Scholar), S. I. E. induction of cytochrome PCR of human and mouse J. PubMed Scopus Google Scholar), V.A. The of PPAR-alpha drug action in an in vivo model by gene expression 2001; PubMed Scopus Google Scholar), acyl-CoA dehydrogenase F. F. C. J. beta-oxidation enzyme gene expression in the rat J. Scholar), fatty acyl-CoA synthetase acid K. M. H. A. Krey G. Wahli W. P. Staels B. T. Auwerx J. Induction of the A synthetase gene by fibrates and fatty is by a peroxisome proliferator response element in the Biol. Chem. 1995; Full Text Full Text PDF PubMed Scopus Google Scholar), acid M. G. J. K. S. in the promoter of the rat gene encoding the 1996; PubMed Scopus Google Scholar), apoC-III (11Staels B. Vu D.N. Kosykh V.A. Saladin R. Fruchart J.C. Dallongeville J. Auwerx J. Fibrates downregulate apolipoprotein C-III expression independent of induction of peroxisomal acyl coenzyme A oxidase. A potential mechanism for the hypolipidemic action of fibrates.J. Clin. Invest. 1995; 95: 705-712Crossref PubMed Google Scholar), J.C. G. D. Peroxisome proliferator-activated receptor mediates induction of the mitochondrial gene by fatty Biol. Chem. 1994; 269: Full Text PDF PubMed Google Scholar), and enzyme for fatty acid H. T. Moore D.D. The peroxisome proliferator activated receptor regulates enzyme Biol. Chem. 1994; 269: Full Text PDF PubMed Google Scholar). also been that the liver a PPAR response element H. J. S. and of the peroxisomal proliferator response element in rat PubMed Scopus Google Scholar). In in which were for 10 with and gene were to be very to in for a toward gene expression changes of for a number of which transcriptional regulation treatment In to the transcriptional the used may also have the expression profiles of a number of the genes in a microarray study of the transcriptional to a genes as in of E. and target genes identified by microarray analysis in liver of Lipid Full Text Full Text PDF PubMed Scopus Google Scholar). downregulated by the cholesterol of the diet and These were also in the present study and are as PPAR target Thus, the regulation of these genes was by both the of total cholesterol and the direct transcriptional This the of the of diet for gene expression and that predictive gene sets may be used in to the in which were for correlations between the in PPAR-α efficacies for each and the expression profiles of the a number of very were identified The on from both and microarrays, were for apoC-III and bifunctional This is to the that a relationship between the gene expression profiles from a of different drug and the efficacy of the been These genes may be useful in the in vivo of expression profiles of genes can to different that multiple gene expression into may a of the physiological response of This may be the for drugs with multiple target as fibrates and we used a multivariate model and the expression profiles of the genes as and the as The which was using the hybridization from the was triglyceride lowering the triglyceride lowering the by the a very was 2 A very of the genes used as encode enzymes involved in triglyceride metabolism. In to the very of triglyceride the of the total cholesterol and the HDL changes were the the for these the were for the in total cholesterol and for the in HDL levels. is very important to and model be for by the that should an between and to In the present study, seven different PPAR-α activators with varying efficacies were in rats fed a have that gene expression in vivo can useful that very well to both the of the drugs in and the in the of the of multivariate as to be a very useful of gene expression in a physiological Using the from 19 a very of the triglyceride-lowering of the drugs could be The from the present study be useful the potential of PPAR-α activators in rats and the of gene expression in the of drug The and for for the and and for and of the acyl-CoA binding protein binding apolipoprotein C-III peroxisomal enoyl-CoA:hydrotase-3-hydroxyacyl-CoA bifunctional enzyme cytochrome P450 high-cholesterol diet to peroxisome proliferator-activated receptor