Scientific paper Modulation of Hepatic Transcriptome in the Poloxamer P-407 Hyperlipidemia Mouse Model Drago Kuzman,1* Klementina Fon Tacer,2 M. Černe,1 Tadeja Režen,2 Jure Acimovic,2 Urška Čegovnik,3 Darko Kocjan,1 Uroš Urleb1 and Damjana Rozman2 1 Lek Pharmaceuticals d.d., Ljubljana, Slovenia 2 Centre for Functional Genomics and Bio-Chips, Faculty of Medicine, University of Ljubljana, Slovenia 3 University Clinic for Pulmonary and Allergic Diseases Golnik, Golnik, Slovenia * Corresponding author: E-mail: drago.kuzman@sandoz.com Received: 25-07-2008 Abstract The poloxamer P-407 hyperlipidemia mouse is a genetically unaltered, nondiet-induced model of atherosclerosis which is useful in studying effects of lipid lowering drugs. While the model is relatively well described on the biochemical level, the knowledge regarding the global gene expression was lacking so far. As expected, the P-407 elevates plasma triglycerides and cholesterol. Using microarray analysis and RT-PCR we show for the first time that poloxamer triggers a concerted modulation of hepatic genes involved in multiple steps of atherogenesis, such as lipid metabolism, initiation of atherogenesis, lesion formation, and atheroma progression. 106 hepatic genes were upregulated and only 40 downre-gulated, suggesting that upregulation of atherogenic genes precedes the atheroma formation. Modulated expression of circadian genes has also been observed in this type of hyperlipidemia but further studies are needed to elucidate whether this effects the rhythm of animals. Keywords: Poloxamer, hyperlipidemia, atherogenesis, circadian 1. Introduction Cardiovascular diseases are the leading cause of death in the developed world.12 Abundant data link hypercholesterolemia to atherogenesis. It has been appreciated recently that dyslipidemia and atheroma formation are coupled by inflammatory mechanisms.3 One of the important unsolved questions is how good are different animal hyperlipidemia models for reflecting the situation in humans. As alternative to knock-down and high fat diet, application of chemical substances can also lead to hyper-lipidemia.2 Poloxamer P-407 is a block copolymer comprised of repeated poly(oxyethylene)-poly(oxypropilene)-poly(oxyethylene) units,4 that induces a dose-controlled dyslipidemia in mice.5 A single injection causes a dose-dependent hyperlipidemia in rats and mice, increasing both plasma triglyceride and cholesterol.6 The hypercho-lesterolemia seems to be associated with an up-regulation of both the protein expression and activity of the key regulatory enzyme of cholesterol synthesis, the HMG-CoA re-ductase, and a decrease in expression of the LDL-recep-tor, responsible for cholesterol uptake.5 The aim of our work was to evaluate by biochemical parameters and expression profiling the poloxamer P-407-induced chronic hyperlipidemia in the mouse liver. Liver has been chosen as a major metabolic organ and the large reservoir of macrophages in the body. In addition to disbalance of lipid metabolism, the poloxamer P-407 model reveals modulation of transcriptome markers of different stages of atherogenesis. It shows also for the first time that poloxamer-based hyperlipidemia modulates the expression of some circadian transcription factors. This knowledge can be translated to humans and applied in studying the effects of novel hypolipidemic drugs. 2. Material and Methods 2. 2. Animal Treatment and Tissue Collection The work on animals was performed in compliance with all relevant EU legislation, in particular with the Council Directive 86/609/EEC and the European Convention no. 123, the Slovenian Protection of Animals Act (98/99) and Instructions on the conditions for the authori- zation of experiments on animals for scientific and research purposes (40/85)/Amsterdam protocol on animal welfare and approved by Veterinary Administration of the Republic of Slovenia. Seven-week-old female mice (C57BL/6, Harlan, Italy) were housed under a 12:12h light:dark schedule with lights on at 7AM. They were fed the standard, low-fat chow diet (4.4% fat, diet 3430, Provimi Kliba AG, Kai-seraugst, Switzerland). They were randomly assigned into two groups. 24 mice were administered 0.5 g/kg poloxa-mer P-407 by intraperitoneal (i.p.) injection every third day (poloxamer P-407 group) and a control group (16 mice) was injected normal saline. Animals were sacrified in the early light phase, between 7-9 AM. Blood was collected at regular intervals by retro-orbital sampling technique or from the heart. Livers were collected, weighted, frozen in liquid nitrogen and stored at -80 °C or in RNAlater (Qiagen) for subsequent analyses. 2. 2. Determination of Plasma Cholesterol and Triglycerides and the Liver Cholesterol Blood was transferred into EDTA-containing tubes (BD Biosciences-Pharmingen) and plasma collected by centrifugation. Total cholesterol and triglyceride analysis was determined on Cobas Mira analyzator by Roche Diagnostics kits. Results for each subgroup are presented as a median of subgroup measurements. Liver cholesterol was measured at days 96 and 111. For each time/treatment point samples of 4 animals have been pooled into 2 biological replicas, frozen, lyophilized and weighed. Sterols were extracted and analyzed by reverse phase HPLC as described previously.7 Quantity of sterols was normalized on tissue weight and is represented in ppm (mass of sterols in micrograms per dry mass of tissue sample in grams). The level of liver cholesterol is calculated as the average of two biological replicas. 2. 3. Microarray Experiment and Bioinformatic Analysis Liver samples were homogenized in TRI reagent (Sigma, Taufkirchen, Germany), the total RNA isolated and subjected to RNeasy Cleanup (Qiagen, Valencia, CA). RNA concentration and quality was determined by Agilent 2100 Bioanalyzer (Agilent Technologies, CA, USA). RNA from 12 poloxamer P-407 treated mice was pooled into 3 biological replicas on day 96. Reference total RNA has been obtained by pooling 4 mice from control group day 96. For cDNA arrays 10 g of total RNA was used for labelling. Synthesis of Cyanine 3- and Cyanine 5-labeled cDNA was performed by Fluorescent Direct label Kit Protocol (Agilent Technologies, CA, USA). For oligonucleo-tide arrays, 0.5 g of total RNA was labelled using the Low RNA input fluorescence linear amplification kit (Agilent Technologies, CA, USA). After hybridization, the slides were scanned by TECAN LS200 scanner (Tecan Group Ltd., Maennedorf, Switzerland). Image data were analyzed by ArrayPro4.5 software (Media Cybernetics). Features where the Cy3 and Cy5 signals were less than 2.6 times higher than the local background, were filtered out. Remaining data were loess normalized (span 20%). Gene expression ratio (M) was calculated as the median of independently repeated microarray experiments. Statistical significance of differential gene expression was evaluated according to the Local Pooled Error test.8 Threshold for significance was set to p <0.01 and to a minimum of 2.0 fold change. Genes were classified into functional categories based on Gene Ontology (GO) with modifications in Orange software9 and upgraded by expert-based biological classification. The data comply with the MIAME standard and are submitted to the National Centre for Biotechnology Information Gene Expression Omnibus (accession number GSE6305). 2. 4. Quantitative Real-time PCR (Q-RT-PCR) Equal amounts of RNA from 4 mice of each group were pooled, treated with DNAse I (Sigma) and converted to first strand cDNA. Q-RT-PCR was preformed by standard procedures using either by TaqMan technology (Assay-on-demand for Hmgr, Hmgs, Sc4mol, 8,7-iso, Dhcr7-Applied Biosystem, PE) or SybrGreen technology (Vcam, Lpl, Srebp1a, Srebp1c, Srebp2, Saa3, Pon1, Fpps, Fabp5, Tim1, Clock, Cry1, Bmal1, Cyp51, Cyp7a1, Insig1, Cyp8b1, Insig2a, DBP) on an Applied Biosystems Prism 7900 HT sequence detection system. Reactions were performed in triplicate. The relative amount of all mRNAs were calculated by the comparative Ct (cycle number at threshold) method (User Bulletin No. 2, Applied Biosystems, PE) using 18S RNA as the internal control. Relative mRNA levels were determined by expressing the m-RNA amount of the poloxamer P-407 treated-animals relative to normal fed controls. Standard deviation and significance level of fold change comparison between the po-loxamer P-407 treated and control group was evaluated as described.10 The treatment of RNA isolated from Hepa1 cells was the same as it is written for RNA isolated from mice liver. 3. Results 3. 1. Lipid Profiles of Poloxamer P-407chronic Hyperlipidemia Mouse The poloxamer P-407 chronic hyperlipidemia was assessed as described.11 Figure 1 shows the time course of plasma total cholesterol (A) and triglycerides (B). Poloxa-mer increased plasma cholesterol above normal level a) b) Figure 1: Time course of plasma cholesterol (A) and triglycerides (B) in the poloxamer P-407 treated (solid line) and control mice (dashed line). Time represents duration of the poloxamer P-407 treatment. Medians of lipid levels of P-407 treated mice are shown by black circles, medians of lipid levels of control mice are designated with white circles. Number of animals per treated and control groups were n = 8 (days 0, 27, 60, 93), n = 4 (days 111, 124, 144) and n = 12 for treated group (day 96) and n = 4 for control group (day 96). Error bars represent an average of 1 standard deviation of measured lipid values for poloxamer P-407 treated mice which is more than 20 times larger as standard deviation of lipids measurements for saline treated mice (not shown). nearly 8-times at day 111. Increase in cholesterol is accompanied by the increase in triglycerides, which is in accordance with previous reports.2511 Surprisingly, the liver a) b) Figure 2: Plasma (A) and liver (B) cholesterol levels in poloxamer P-407 induced hyperlipidemia model. Cholesterol levels are presented as mean s.d. for poloxamer P-407 treated group (black bars) and for control group (grey bars); * p <0.05, ** p <0.01, n = 4 per group. cholesterol at days 96 and 111 remains in the normal range (Fig. 2). Comparison to the published high fat diet models shows an inverse relation between plasma and liver cholesterol,1213 suggesting that high fat diets result primarily in the increase of liver cholesterol while poloxamer P-407 application leads primarily to the increase of plasma cholesterol. 3. 2. Poloxamer P-407 Based Hyperlipidemia Activates Atherogenic Pathways of The Hepatic Transcriptome The hepatic transcriptome has been evaluated at day 96 as a core data set for biological interpretation. A list of genes with statistical significance (p <0.01 and minimal 2.0 fold change) was created and divided into physiologically relevant groups by (Table 1) the gene ontology widget of Orange.9 A representative set of genes has been confirmed by real time PCR (Fig. 3). The poloxamer P-407 based chronic hyperlipidemia results in 106 upregulated and 40 downregulated genes. The vast majority of modulated genes is connected to various stages of atherogenesis (Fig. 4, Tab. 1), such as lipid metabolism, initiation of atherogenesis, lesion formation, and atheroma progression. Interestingly, major circadian oscillator genes are also modulated. Disturbance of the lipid metabolism is reflected in modulated cholesterol and fatty acid homeostasis. This includes repression of genes involved in cholesterol synthesis, modulation of genes from bile acid synthesis, lipoprotein metabolism and fatty acid transport (Table 1, Fig. 3). Among 146 differentially expressed genes related to atherosclerosis (Table 1), 74 are linked to inflammation processes, initiation of atherosclerosis, lesion formation and atheroma progression. Nearly all genes, related to different stages of atherogenesis are overexpressed in the po-loxamer model in comparison to control livers. Table 1: Differential expression of genes in the 96-day poloxamer P-407 hyperlipidemia model. Genes that were found differentially expressed also in the 84-day high fat diet model published by Recionos et al. [12] are printed in bold; genes that were previously linked to atherogenic processes are labelled by +. Gene Gene Bank Gene name Fold P value Athero- Symbol ID Change genesis LIPID METABOLISM Lpl NM_008509 lipoprotein lipase 2.01 >1.E-15 Fabp5 NM_010634 fatty acid binding protein 5, adipocytes, macrophages 2.01 Fabp7 NM_021272 fatty acid binding protein 7, brain 1.35 1.0E-09 Apobec1 NM_031159 apolipoprotein B editing complex 1 1.15 5.0E-08 Fabp4 NM_024406 fatty acid binding protein 4, adipocytes, macrophages 1.13 7.0E-07 Stard4 NM_133774 StAR-related lipid transfer (START) domain containing 4 -1 1.0E-06 Cyp51 NM_020010 cytochrome P450, 51 -1.05 4.0E-07 Hmgcs1 AK045094 3-hydroxy-3-methylglutaryl-Coenzyme A synthase 1 -1.11 3.0E-05 Cyp39a1 NM_018887 cytochrome P450, family 39, subfamily a, polypep. 1 -1.12 3.0E-07 Acas2 NM_019811 acetyl-Coenzyme A synthetase 2 (ADP forming) -1.16 3.0E-07 Sc4mol NM_025436 sterol-C4-methyl oxidase-like -1.22 3.0E-06 Idi1 AK029302 isopentenyl-diphosphate delta isomerase -1.29 2.0E-07 Hmgcr M62766 3-hydroxy-3-methylglutaryl-Coenzyme A reductase -1.37 3.0E-10 Fdps NM_134469 farnesyl diphosphate synthetase -1.39 9.0E-08 Sqle NM_009270 squalene epoxidase -1.39 6.0E-10 INITIATION OF ATHEROSCLEROSIS Monocyte adhesion Vcam1 NM_011693 vascular cell adhesion mol 1 1 2.07 >1.E-15 + Itgax NM_021334 integrin alpha X 1.26 3.0E-10 Iqgap1 NM_016721 IQ motif containing GTPase activating protein 1 1.26 5.0E-09 Sdcbp NM 016807 syndecan binding protein Chemoattractions and monocyte migration Anxa2 NM_007585 annexin A2 Anxa3 NM_013470 annexin A3 Ccl6 NM_009139 chemokine (C-C motif) ligand 6 Capg NM_007599 capping protein (actin filament), gelsolin-like Ccl4 NM_013652 chemokine (C-C motif) ligand 4 Cotl1 NM_028071 coactosin-like 1 (Dictyostelium) Pkib NM_008863 protein kinase inhibitor beta, cAMP dependent Tmsb4x NM_021278 thymosin, beta 4, X chromosome Evl NM_007965 Ena-vasodilator stimulated phosphoprotein Ccl2 NM_011333 chemokine (C-C motif) ligand 2 Ptpns1 AI835480 protein tyrosine phosphatase, non-recept. type substrate 1 1 1.42 1.34 1.84 1.76 1.35 1.31 1.26 1.21 1.2 1.14 1.04 1.0E-05 5.0E-10 1.0E-09 2.0E-16 2.0E-16 3.0E-11 2.0E-08 1.7E-10 6.0E-07 5.4E-08 8.0E-08 3.5E-07 LESION FORMATION Macrophage activation Ms4a7 Lyzs Cd68 Cd63 Mpeg1 Csf2rb2 Icb-1 Cd5l Csf2rb1 Glipr1 Cd9 Gdf15 Csf2ra Cd53 AF237917 NM_017372 NM_009853 NM_007653 L20315 NM_007781 BC013712 NM_009690 BF134814 BC025083 NM_007657 NM_011819 NM_009970 NM 007651 membrane-spanning 4-domains, subfamily A, member 7 2.05 >1.E-15 lysozyme 1.98 >1.E-15 CD68 antigen 1.63 7.0E-14 Cd63 antigen 1.61 7.0E-13 macrophage expressed gene 1 1.57 1.1E-11 colony stimulating factor 2 receptor, beta 2, 1.55 2.0E-13 low-affinity (granulocyte-macrophage) cDNA sequence BC013712 1.39 2.6E-10 antigen-like 1.38 3.0E-07 colony stimulating factor 2 receptor, beta 1, 1.26 2.5E-09 low-affinity (granulocyte-macrophage) GLI pathogenesis-related 1 (glioma) 1.21 4.8E-10 CD9 antigen 1.14 5.0E-07 growth differentiation factor 15 1.1 2.0E-07 colony stimulating factor 2 receptor, alpha, low-affinity 1.08 1.0E-06 CD53 antigen 1.06 5.0E-07 + + + + + + Gene Gene Bank Gene name Fold P value Athero- Symbol ID Change genesis Emr1 X93328 EGF-like module containing, mucin-like, hormone receptor-like sequence 1 1.03 5.8E-06 Laptm5 NM_010686 lysosomal-associated protein transmembrane 5 1.03 2.0E-06 Spic NM_011461 Spi-C transcription factor (Spi-1/PU.1 related) 1 1.5E-06 Fut2 NM_018876 fucosyltransferase 2 -1.07 3.0E-06 Cd163 NM_053094 CD163 antigen, SRCR family class B -1.72 2.0E-16 + Inflammatory response Clecsf12 NM_020008 Ca2+-dependent lectin-like receptor gene, DECTIN-1 2.66 >1.E-15 Bcl2a1a NM_007534 B-cell leukemia/lymphoma 2 related protein Ala 2.17 0.0E+00 Egr1 NM_007913 early growth response 1 2.13 0.0E+00 + Blnk Y17159 B-cell linker 2 >1.E-15 Tnfaip2 NM_009396 tumor necrosis factor, alpha-induced protein 2 1.72 2.0E-14 Tyrobp NM_011662 TYRO protein tyrosine kinase binding protein 1.7 4.0E-14 Ear11 BC027557 eosinophil-associated, ribonuclease A family, memberll 1.66 7.0E-14 Earl NM_007895 eosinophil-associated, ribonuclease A family, member 1 1.6 9.0E-14 Pld3 NM_011116 phospholipase D3 1.52 1.5E-11 Vim NM_011701 vimentin 1.38 5.0E-10 Fcerlg NM_010185 Fc receptor, IgE, high affinity I, gamma polypeptide 1.37 2.0E-10 Nfil3 NM_017373 nuclear factor, interleukin 3, regulated 1.35 5.0E-10 Clec2 NM_019985 C-type lectin-like receptor 2 1.24 5.0E-08 Igsf7 AY048685 immunoglobulin superfamily, member 7 1.18 1.1E-07 Cd48 X53526 CD48 antigen 1.14 5.0E-08 Clqg NM_007574 complement component 1, q subcomponent, gamma polypeptide 1.12 1.0E-06 Tbxasl NM_011539 thromboxane A synthase 1, platelet 1.11 2.0E-07 + Cyba AK021200 cytochrome b-245, alpha polypeptide 1.08 8.0E-07 Phlda1 NM_009344 pleckstrin homology-like domain, family A, member 1 1.04 4.0E-06 + Tslpr NM_016715 thymic stromal-derived lymphopoietin, receptor 1.01 8.0E-06 Lgmn NM_011175 legumain 1.01 9.0E-06 + Ear6 NM_053111 eosinophil-associated, ribonuclease A family, member 6 1.01 6.0E-07 Pim3 BC017621 proviral integration site 3 -1.18 7.0E-07 Atheroprotection Gp49b NM_013532 glycoprotein 49 B 1.59 8.0E-13 Serpinb6c AF425084 serine (or cysteine) proteinase inhibitor, clade B, member 6c 1.23 5.2E-08 Il18bp NM_010531 interleukin 18 binding protein 1.1 6.0E-07 + Il10rb NM_008349 interleukin 10 receptor, beta 1.02 6.0E-06 + ATHEROMA PROGRESSION AND COMPLICATION Ctss NM_021281 cathepsin S 1.96 1.0E-14 + Ctsb AK083393 cathepsin B 1.31 3.0E-06 + Clpx NM_011802 caseinolytic protease X (E.coli) 1.31 3.0E-08 Crip1 BC030406 cysteine-rich protein 1 (intestinal) 1.3 9.6E-09 Rgs2 NM_009061 regulator of G-protein signaling 2 1.2 6.0E-08 + Oas1f AK037025 2'-5' oligoadenylate synthetase IF 1.09 6.5E-07 + Ctsd NM_009983 cathepsin D 1.01 4.0E-05 + Plk3 U21392 polo-like kinase 3 (Drosophila) -1.48 2.0E-11 Calcification Gpnmb NM_053110 glycoprotein (transmembrane) nmb 4.62 0.0E+00 Trem2 NM_031253 triggering receptor expressed on myeloid cells 2 4.42 >1.E-15 S100a9 NM_009114 S100 calcium binding protein A9 (calgranulin B) 2.04 >1.E-15 + Adam8 NM_007403 a disintegrin and metalloprotease domain 8 1.66 1.0E-15 S100a8 NM_013650 S100 calcium binding protein A8 (calgranulin A) 1.48 2.0E-13 + CIRCADIAN REGULATION Bmal1 NM_007489 aryl hydrocarbon receptor nuclear translocator-like 1.21 4.0E-10 Per2 NM_011066 period homolog 2 (Drosophila) -1.47 2.0E-11 Ccrn4 U70139 CCR4 carbon catabolite repression 4-like (S. cerevisiae) -2.37 >1.E-15 Dbp NM_016974 D site albumin promoter binding protein -2.81 >1.E-15 Figure 3: Comparison of real time PCR and microarray gene expression assessments Relative gene expression of poloxamer P-407 treated animals (day 96) vs control group is presented as mean s.d. of biological replicas (n = 3 per group). In microarray results both oligo and cDNA microarray measurements are considered. Significantly modulated genes (p <0.01) are designated by the symbols X (measurement obtained by Q-RT-PCR) and + (measurement obtained by DNA microarray). Gene encoding DBP enzyme was downregulated over the presented graph range (DBP fold change: -8.07 ± 3.72). 3. 3. Poloxamer P-407 Based Hyperlipidemia Modulates Circadian Genes Major circadian oscillator genes and clock-controlled downstream genes are among the most modulated genes detected by microarray analysis (Table 1, Fig. 3). While Bmal is overexpressed in the early light phase in hyperlipidemia compared to normolipidemic controls, Ccm4, Dbp and Per2 are at the same time downregulated. This opposite type of regulation is in accordance to the circadian pattern of expression of these genes. While Ccm4 (nocturin), Dbp and Per2 peak close to the time of light offset (zeitgeber time ZT 12), Bmal is highest at the start of the light phase (ZT 0).14,15 4. Discussion The poloxamer P-407 chronic hyperlipidemia mouse model is a genetically unaltered, nondiet-induced mouse model of atherosclerosis that was studied regarding kinetics of hyperlipidemia, atherosclerosis, treatment with Figure 4: Modulated genes involved in different atherogenic processes. Classified 93 genes are listed in Table 1. statins and the lipid metabolism related enzyme activi-ties.5,16 The model shows an intense modulation of the at-herogenesis-related transcriptome, which was similar to the atherosclerosis-prone genetically compromised LDLR -/- mice,17 but changes are more significant as detected in the high fat diet models.12,13,17 Our study shows for the first time that poloxamer P-407 triggers a concerted modulation of a variety of hepatic genes involved in multiple steps of atherogenesis (Table 1). Some of them have already been described in connection to the development of atherosclerosis either at the level of transcriptome12,13,17,18 or as protein markers.19-21 It is important to note that the gene modulation that was measured at day 96 of the poloxamer P-407 treatment seems to precede the observable atheroma formation since observable atherosclerotic lesions have been described at day 120.11 The poloxamer P-407 treated mice reach 8 times higher plasma lipid levels compared to the diet induced models while liver cholesterol is comparable to controls. Hypercholesterolemia seems to be associated with upre-gulation of both the protein expression and the activity of HMG-CoA reductase, the key regulatory enzyme of cholesterol synthesis.5 The downregulation of Hmgcr and other genes of cholesterol synthesis might reflect an adaptation mechanism aimed to protect cells from synthesizing cholesterol. Repression of genes of cholesterol synthesis with no significant repression of SREBP-2 is expected since "cholesterol sensing" as a part of cholesterol feedback regulation works at the post-transcriptio-nal level.22,12 A significant decrease in the activity of Cyp7a1 is combined with no change in the Cyp7a1 m-RNA and a reduced expression of Cyp39 from the alternative pathway.2 Reduced cholesterol elimination seems to represent an important factor leading to hyperlipide-mia. Additionally, the poloxamer P-407 inhibits the lipo- protein lipase protein,24 while an upregulated expression is observed in the liver (Table 1). It seems that for some proteins that are affected by poloxamer P-407, adaptive mechanisms lead to expression aimed to minimize the effect of this pathological stimulus. The overlap between transcriptomes of the chronic poloxamer and cholesterol diet models12 is low. Only Lpl from the lipid metabolism, S100a9, Cd53 and Spic from lesion formation, Iqgapq and Crip1 from initiation of atherosclerosis and Chka from cell stress/defence response mechanisms are over 2fold upregulated in both models. Studies describing the proteome biomarkers of athe-rogenesis show fatty acid binding proteins FABP 4 and 5, cathepsins D and B, S100 calcium binding protein A14 and annexin 5 proteins19 significantly upregulated in ma-crophages after treatment with oxidized LDL. We observed upregulation of genes from the same families (Table 1), indicating that they represent true markers of atheroge-nesis. All proteome studies19-21 underline the importance of CCl family proteins, especially the chemokine Ccl2 (Mcp1) from initiation of atherogenesis, that are overex-pressed in our model. Interestingly, while the downregula-ted genes in diet and poloxamer P-407 models represent genes of cholesterol synthesis,121317 no similar observation has been reported at the proteome level. An unexpected observation was the modulation of circadian oscillators (Clock, Bmal1, Per2) and circadian output genes (Dbp, CCr4). Circadian rhythms are daily cycles of physiology and behaviour that are driven by an endogenous oscillator with a period of about one day. The suprachiasmatic nucleus represents the master circa-dian synchronizer but most peripheral cells circadian oscillators that are based on interconnected transcriptiontranslation feedback loops.25 CLOCK and BMAL1 transcription factors activate genes encoding negative components of the major feedback loop, such as genes of the period (Per) or chryptochrome (Cry) families.26 Recent data on the conditionally knock out clock -/- mice challenge a central feature of the current mammalian cir-cadian clock model regarding the necessity of CLOCK:BMAL1 heterodimers for clock function.27 Our study indicates that in the early light phase Clock and Bmal1/Arntl1 are upregulated in livers of poloxamer P-407 treated mice while Per2, Dbp and CCr4 are among the few downregulated genes. A broader investigation is needed to determine through which mediators hyperlipi-demia affects the circadian gene expression and whether this affects also the circadian behaviour of these hyperli-pidemic animals. 5. Acknowledgments The authors are grateful to dr. A. Stalc for introducing of poloxamer P-407 hypelipidemia model. The work was supported by funds of the Lek Pharmaceuticals, d.d and the Slovenian Research Agency grants J1-6713, P1-0527, P1-104 and L1-6707. K. Fon Tacer and T. Režen were supported by the fellowship from the Slovenian Research Agency. Work has been generated in the context of the STEROLTALK project, funded by the European Community as contract No. LSHG-CT-2005-512096 under 6'h Framework Programme for Research and Technological Development in the thematic area of Life sciences, geno-mics and biotechnology for health. This work reflects only the author's views and the European Community is not liable for any use that may be made of the information contained therein. 6. References 1. S. Shefer, G. Salen, A. Honda, A. K. Batta, L. B. Nguyen, G. S. Tint, Y. A. loannou, R. Desnick, J. Lip. Res. 1998, 39, 2471-2476. 2. T. P. Johnston, J. Cardiovasc. Pharmacol. 2004, 43, 595-606. 3. P. Libby, Nature 2002, 420, 868-942. 4. W. Loh, A. E. Beezer, J. C. Mitchell, Langmuir 1994, 10, 3431-3434. 5. C. Leon, K. M. Wasan, K. Sachs-Barrable, T. P. Johnston, Pharm. Res. 2006, 23, 1597-1607. 6. M. A. Johnson, T. L. Macdonald, Biochem. Biophys. Res. Commun. 2004, 324, 446-50. 7. K. Fon Tacer, S. Kalanj-Bognar, M.R. Waterman, D. Rozman, J. Steroid. Biochemist. Molecular Biology 2003, 85, 429-438. 8. N. Jain, J. Thatte, T. Braciale, K. Ley, M. O'Connell, J. K. Lee, Bioinformatics 2003, 19, 1945-51. 9. T. Curk, J. Demsar, Q. Xu, G. Leban, U. Petrovic, I. Bratko, G. Shaulsky, B. Zupan, Bioinformatics 2005, 21, 396-8. 10. K. J. Livak, T. D. Schmittgen, Methods 2001, 25, 402-408. 11. T. P. Johnston, J. C. Baker, D. Hall, S. Jamal, W. K. Palmer, E. E. Emeson, Atherosclerosis 2000, 149, 303-13. 12. S. Kim, I. Sohn, J. I. Ahn, K. H. Lee, Y. S. Lee, Gene 2004, 340, 99-109. 13. K. N. Maxwell, R. E. Soccio, E. M. Duncan, E. Sehayek, J. L. Breslow, J. Lip. Res. 2003, 44, 2109-19. 14. T. Kudo, E. Nakayama, S. Suzuki, M. Akiyama, S. Shibata, Am. J. Physiol. Endocrinol. Metab. 2004, 287, E644-51. 15. T. Yamamoto, Y. Nakahata, H. Soma, M. Akashi, T. Mamine, T. Takumi, BMC Mol Biol 2004, 5, 18. 16. K. M. Wasan, R. Subramanian, M. Kwong, I. J. Goldberg, T. Wright, T. P. Johnston, J. Pharm. Pharm. Sci. 2003, 6, 18997. 17. A. Recinos, 3rd, B. K. Carr, D. B. Bartos, I. Boldogh, J. R. Carmical, L. M. Belalcazar, A. R. Brasier, Physiol. Genomics 2004, 19, 131-42. 18. P. Puddu, E. Cravero, G. M. Puddu, A. Muscari, Int. J. Clin. Pract. 2005, 59, 462-72. 19. E. M. Fach, L. A. Garulacan, J. Gao, Q. Xiao, S.M. Storm, Y. P. Dubaquie, S. A. Hefta, G. J. Opiteck, Mol. Cell Proteo-mics 2004, 3, 1200-10. 20. J. Y. Park, J. K. Seong, Y. K. Paik, Proteomics 2004, 4, 51423. 21. R. Tabibiazar, R. A. Wagner, A. Deng, P. S. Tsao, T. Querter-mous, Physiol. Genomics 2006, 25, 194-202. 22. M. S. Brown, J. L. Goldstein, Proc. Natl. Acad. Sci. USA 1999, 96, 11041-11048. 23. R. B. Rawson, Nat. Rev. Mol. Cell Biol. 2003, 4, 631-640. 24. J. S. Millar, D. A. Cromley, M. G. McCoy, D. J. Rader, J. T. Billheimer, J. Lipid. Res. 2005, 46, 2023-8. 25. J. A. Ripperger, U. Schibler, Curr. Opin. Cell Biol. 2001, 13, 357-62. 26. N. Gekakis, D. Staknis, H. B. Nguyen, F. C. Davis, L. D. Wilsbacher, D. P. King, J. S. Takahashi, C. J. Weitz, Science 1998, 280, 1564-9. 27. J. P. Debruyne, E. Noton, C. M. Lambert, E. S. Maywood, D. R. Weaver, S. M. Reppert, Neuron 2006, 50, 465-77. Povzetek Mišji model hiperlipidemije povzročene s poloksamero P-407 je primeren za proučevanje ateroskleroze in vpliva zdravil za zniževanje lipidov. Zanj je značilno, da hiperlipidemija ni povzročena z genskimi spremembami živali, oziroma z energetsko bogato dieto. Model je relativno dobro poznan na nivoju biokemije, medtem ko še ni bil obravnavan z globalno analizo transkriptoma. Po pričakovanjih P-407 povzroči dvig triglicerdov in holesterola v plazmi miši. Z uporabo DNA biočipov in RT-PCR prvič pokažemo, da P-407 povzroči usklajene spremembe v izražanju jetrnih genov, ki so vpleteni v različne stopnje aterogeneze, kot so lipidni metabolizem, iniciacija aterogeneze, formiranje lezij in napredovanje aterome. V jetrih se je izražanje povečalo 106-im, zmanjšalo pa 40-tim genom, kar napeljuje na domnevo, da povečanje izražanja aterogenih genov napoveduje formiranje aterome. Zaznali smo tudi spremembo izražanja cirkadičnih genov, vendar bodo potrebne dodatne analize za potrditev povezave z časovnim ritmom živali.