Expedited Processing RequiredSupgn?sor's lniliab CBIC CODWG SHEET 000 INITIAL: AZ CASE N0. mm: PHONETEMATL- CONSOLIDATEIJTNO. or ms: - CONTACT- Jam sum nms: DOCUMENT TYPE DOCUMENT TYPE SUPPORT T28 128 Follow Up Amendmew 13 13 Follow Up Substauahon 21 21 Follow Up CommunS 6 6 Follow Up SuspensTorLEUer 8A Spec Pair 8A FolIOw Up 80 8 Follow Up 8E SE Follow Up AR (a AR . Follow Up EONAFIDE BONAFIDE Follow Up DOCKET {t DOCKET FolIOw ETI ETI Follow Up FYI FYI Fouow Up COMMIT: IUR Up (i?m ?65?13?7773 LCREX LOFEX ?chow Up 4'00 0131 RF EJSCXE - Low VOLUME LVE Up 313.? LOW VOLUMEMODIFICATION Em Follow Up 17%: ?arm NOC Fo?ow Up (mauve. n0 PC PC Follow Up '11 FMN PMN Follow Up PMNI PMNI Follow Up j. POLYMER EXEMPT Follow Up 1: TMEA TMEA FollowUp i XL, XL FolIOw 1 1 u: 08' El MCANS MCANS Follow Up INVENTORY INVENTORY FollowUp .1 Follow Up Please use the Secuon 4 Codmg Shea! for Secuan 4 Datumenu '05 [Jul Jean B. Sweeney Vice President 3M Environmental, Health and Safety Operations 3M Center, Building 0224-05-W-03 St. Paul, MN 55144;-1-000 651 737 3569 ,s 8EHQ-0210-00373DH MR#324971 February 16,2010 CERTIFIED MAlL NO CBI Document Processing Center EPA East - Room 6428 Attn: Section 8(e) Office of Pollution Prevention and Toxics, U.S. EPA 1200 Pennsylvania Avenue NW Washington, DC 20460-000 1 Re: TSCA 8(e) Substantial Risk Notice: Sulfonate-based and Carboxylic-based Fluorochemicals, Docket 8EHQ-0598-373 - Results from a mechanistic investigation of the effect of PFBS, PFHS, and PFOS on lipid and lipoprotein metabolism in transgenic mice Dear Sir or Madam: 3M is submitting this notice to supplement previous submissions on sulfonyl and carboxylicbased fluorochemicals (FCs), and more specifically our July 28,2006 and January 8,2007 submittals concerning data generated by TNO Laboratories in Leiden, Netherlands. These data suggest an effect of perfluorooctanesulfonate (PFOS) and perfluorohexanesulfonate (PFHS) on body weight, food consumption, and serum cholesterol in 15% dietary fat fed E3 Leiden transgenic mice. Enclosed is a final report for follow up investigation to the aforementioned study. The current study was conducted at TNO Laboratories in Leiden, Netherlands using APOE*3LiedenhuCETP transgenic mice. Results help to explain the mechanism of action by which perfluorobutanesulfonate (PFBS), PFHS, and PFOS alter lipid and lipoprotein metabolism in animals treated with these chemicals. An effect noted in this study that has not been clearly demonstrated in other studies on these chemicals was a decrease in high density lipoprotein (HDL) in animals treated with PFOS and PFHS. If you have any questions, please contact Deanna Luebker at (65 1) 737-1374 or djluebker@mmm.com. Sincerelv. V Jean B. Sweeney Vice President Environmental, Health and Safety Operations Enclosure 3M#03 Mechanism of different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 TNO-Report for the study entitled: Mechanism of the effect of different PFAS's (PFBS vs PFHS vs PFOS) on lipid and lipoprotein metabolism in APOE*3-Leiden/huCETP transgenic mice TNO Biosciences Gaubius Laboratory Zernikedreef 9, 2333 CK Leiden The Netherlands General Phone + 3 1 7 1 518 1818 Specific Fax + 3 1 7 1 518 1 9 0 1 Drafted by: Dr. Hans M.G. Princen Elsbet Pieterman B.Sc. I n assignment of: 3M, Medical department Project number: 031.12685 Study number: 3M#03 Date: 17 November 2009 Final version Previous version: Draft 2 October 2009 Number of pages: 153 CONFIDENTIAL Signature and date: J.W.A. van den Hoorn M.C.E. Maas E.J. Pieterman BSc. Principal Investigator Dr. J.R.O. Hanemaaijer Dr. H.M.G. Princen Team Manager Study Director All rights reserved. No part of this report may be reproduced and/or published i n any form by print, photoprint, microfilm or any other means without the previous written permission from TNO. All information which is classified according t o Dutch regulations shall be treated by the recipient in the same way as classified information of corresponding value in his own country. No part of this information will be disclosed to any third party. I n case this report was drafted on instructions, the rights and obligations of contracting parties are subject to either the Standard Conditions for Research Instructions given to TNO, or the relevant agreement concluded between the contracting parties. Submitting the report for inspection to parties who have a direct interest is permitted. O 2009 TNO 3M#03 Mechanism of different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 Summary Introduction and Aim Perfluorinated alkyl sulphonates are fully fluorinated amphiphilic organic molecules with strong surface-tension reducing properties. They are stable to environmental and metabolic degradation. Perfluorooctanesulfonate (PFOS) is widely dispersed in humans, fish-eating wildlife, and surface waters. Toxicological studies in rats and mohkeys have shown a reduction in serum cholesterol after treatment with PFOS; however; such reductions have not been observed among exposed workers. I n the present study the focus is put on a further elucidation of the mechanism responsible for the observed changes in plasma triglycerides and cholesterol levels in the atherogenic apoB-containing (VLDL, IDL, LDL) lipoproteins and to investigate the effect o f PFBS, PFHS and PFOS on HDL metabolism. The' APOE*3-Leideh(E3L)IhuCETP transgenic mouse model was used to study these effects. The 'aim 'of the study' is further to emphasize the differences in biological effects between PFBS, PFHS and PFOS and t o publish these data. Material and methods The study, which was subdivided in three experiments,, had the following design: after a run-in period of 4 weeks on a Western type diet (containing .14 OO/ beef tallow, 1 O/O corn oil, 0.25 O/O 'cholesterol), mice received a Western type diet (control) or a western type diet containing 0.03 OO/ PFBS (31.8 mgjkg body weightlday), 0.006% PFHS (6.0 mgjkg body weightlday), 0.003 O/O PFOS (3.1 mg/kg body weightlday) or as a positive control 0.03% fenofibrate (31.1 mg/kg body weight /day). I n all experiments body weight, food intake, plasma cholesterol, HDL cholesterol and triglycerides were measured after 4 weeks of treatment. I n experiment 1 lipolytic activity (LPL and HL activity) was measured from post-heparin plasma after 5 weeks of treatment and feces were collected for the determination of bile acids, neutral sterols and fatty acids. At the end of the experiment (6 weeks of treatment) VLDL-triglyceride and de novo apoB production was measured and lipid composition of VLDL was determined. I n experiment 2 after 4 weeks of treatment at the end of the experiment the gall bladder was cannulated and during 45 minutes bile flow, biliary cholesterol, phospholipids and bile salt output were measured. Directly hereafter the in vivo clearance of VLDL-like triglycerides-rich particles was determined. I n experiment 3, next to plasma ApoAl and CETP activity and mass determinations, the in vivo clearance of autologous labeled HDL was measured after 4 weeks of treatment. From this experiment livers were collected and a part was used for liver histology analysis, a part was used for liver lipid analysis and a part was used for microarray analysis. Mechanism of action of PFBS PFBS reduced plasma cholesterol and triglyceride levels by about 25% and 4596, respectively. The data from the physiological experiments indicate that the decreases in lipid levels are caused by increased clearance of VLDL-TG and VLDL-CE and mildly reduced VLDL-particle production. PFBS had no clear effect on HDL-cholesterol and apoAl levels. PFBS showed no effect on genes involved in HDL metabolism, in line with the unchanged HDL levels. PFBS mildly increased liver weight, but had no effect on ALT and development of hepatosteatosis. Based on mRNA signals PFBS appears to have mild PPARa-agonistic activity ((3-oxidation increased and liver size increased). The present data indicate that PFBS has no increased CVD risk profile. Mechanism of action of PFHS PFHS reduced plasma cholesterol and triglyceride levels by about 60% and 75%, respectively. The data from the physiological experiments supported by hepatic mRNA levels indicate that the decreases in lipid levels are caused by increased lipolysis and clearance of VLDL-TG and VLDL-CE, and strongly reduced VLDL-TG and VLDL-particle production. I n line with the higher lipolytic activity in plasma, the hepatic mRNA signal for LPL (4.3-fold) was increased. PFHS strongly decreased HDL-cholesterol (about -75%) and apoAl (about -75%) levels. Based on mRNA signals we*conclude that' PFHS reduces HDL levels by down-regulation of apoAl synthesis and HDL maturation (ABCal, LCAT). These adverse changes are most likely the result of PXRagonistic activity. Increased remodeling (PLTP) and decreased uptake (SR-B1) are suggested t o be responsible for the formation of larger HDL particles. PFHS increased liver weight, ALT, and resulted in hepatosteatosis, as observed by biochemical and histological measures. 3M#03 Mechanism o f different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 Based on mRNA signals PFHS has strong PPARa-agonistic (lipolysis increased, @-oxidation increased, FA uptake increased, liver size increased) and PXR-agonistic activity (FA uptake increased, FA synthesis increased and HDL synthesis and maturation decreased, liver size increased). Mechanism of action of PFOS: PFOS reduced plasma cholesterol and triglyceride levels by about 65% and 70°/o, respectively. A similar mechanism of action is active with PFOS as with PFHS. The data from physiological experiments supported by hepatic mRNA levels indicate that the decreases in lipid levels are caused by increased .lipolysis and clearance of VLDL-TG, and strongly reduced VLDL-TG and VLDL-particle production. I n line with the higher lipolytic activity in plasma, the hepatic mRNA signal for LPL (2.1-fold) was increased. PFOS increased mRNA levels of cholesterol synthesizing enzymes and cholesterol esterification genes. Moreover, the gene coding for the major rate-limiting enzyme in the bile acid synthetic pathway, and genes involved in biliairy cholesterol excretion were decreased by PFOS. Inhibition of cholesterol metabolism and excretion may form an explanation for increased hepatic cholesterol levels found with PFOS. PFOS strongly decreased HDL-cholesterol (about -65%) and apoAl (about -80%) levels. Based on mRNA signals we conclude that PFOS reduces HDL levels by down-regulation o f apoAl synthesis and HDL maturation. These adverse changes are most likely the result of PXR-agonistic activity. PFOS increased liver weight, ALT, and resulted in pronounced hepatosteatosis and liver cholesterol accumulation, as observed by biochemical and histological measures. Based on mRNA signals PFOS has strong PPARa-agonistic (lipolysis increased, @-oxidation increased, FA uptake increased, liver size increased) and PXR-agonistic activity (FA uptake increased, FA synthesis increased and HDL synthesis and maturation decreased, liver size increased). Involvement of other nuclear transcription factors in the regulation of lipid and lipoprotein metabolism by PFHS and PFOS Involvement of CAR and LXR in the changes in lipid and lipoprotein metabolism caused by PFHS and PFOS cannot be fully excluded, but is less likely. Little is know about the role of CAR in lipid metabolism. CAR has been shown t o decrease P-oxidation genes as CPTl and enoyl CoA hydratase. The latter genes were, however, increased in the present experiments. LXR increases fatty acid synthesis by induction of SREBPlc expression, which was 2-fold decreased, however. I n addition, LXR induces expression of CETP mRNA, whereas in the present experiments a decrease in CETP activity was found. It cannot be excluded that this is caused by a strongly decreased acceptor pool for CE transfer. Involvement of RXR in the observed effects cannot be excluded, since RXR forms a heterodimer together with a larger number of nuclear transcription factors like PPARa, PXR and LXR. However, direct activation of RXR, for instance with bexarotene leads t o opposite effects with increased levels of triglycerides and apoB-containing lipoproteins. Mechanism of action of fenofibrate Fenofibrate reduced plasma cholesterol and triglyceride levels by about 40% and 7O0/0, respectively. The data from the physiological experiments supported by hepatic mRNA levels indicate that the decreases in lipid levels are caused by strongly increased lipolysis and clearance of VLDL-TG and VLDL-CE, despite the increased VLDL-TG production rate. I n line with the higher lipolytic activity in plasma, the hepatic mRNA signal for LPL (4.6-fold) was increased. LDLR mRNA as marker for increased uptake of VLDL remnant particles was enhanced by 1.5-fold. Fenofibrate paradoxically increases VLDL-TG production despite reducing plasma TG, which may be caused by enhanced hepatic free fatty acid uptake resulting from strongly accelerated peripheral LPLmediated lipolysis of VLDL or by increased de novo hepatic TG synthesis. Fenofibrate increased HDL-cholesterol (+5O0/0) and formation of large HDL-1 particles, and had no effect on apoAl. Fenofibrate strongly decrease in CETP activity, which was found majorly responsible for the increased HDL levels upon treatment with fenofibrate and PPARa,y-agonists. Fenofibrate increased liver weight, without effects on ALT and hepatosteatosis, and decreased liver cholesterol content. Based on mRNA signals fenofibrate has strong PPARa-agonistic activity (lipolysis increased, FA uptake increased, @-oxidation increased; HDL remodeling decreased). Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Contents Summary 2 contents 1 IntrOdUCtion aim 1.1 Aim of the study 8 2 Materials and methodsQ 2.1 Test substances reference substance 9 2.2 Mice 9 2.3 Animal welfare 9 2.4 Diets . 9 2.5 Compound administration 9 2.6 Treatment groups 10 2.7 Study design 10 2.8 Measurements experiment 1 13 2.9 Measurements experiment 2 14 2.10 Measurements experiment 3 14 2.11 Statistical analysis 16 3 Deviations from the protocoll? 4 18 4.1 Results study 1 18 4.1.1 Markers of general well?being 18 4.1.2 Body weight 18 4.1.3 Liver and perigonadal fat weight 19 4.1.4 Food intake 21 4.1.5 Plasma ALT . 22 4.1.6 Plasma cholesterol 23 4.1.7 Plasma HDL?cholesterol 24 4.1.8 Plasma triglycerides 25 4.1.9 Lipoprotein profiles 27 4.1.10 Plasma free glycerol 28 4.1.11 Plasma free fatty acids 29 4.1.12 Post- -heparin LPL and HL activity 30 4.1.13 Fecal lipids 32 4.1.14 ?triglycerides and de novo ApoB production 38 4.2 Results study 2 43 4.2.1 Markers of general well-being 43 4.2.2 Body weight 43 4.2.3 Liver and perigonadal fat weight 44 4.2.4 Food intake 46 4.2.5 Plasma ALT 47 4.2.6 Plasma cholesterol 47 4.2.7 Plasma HDL-cholesterol 49 4.2.8 Plasma triglycerides 50 4.2.9 Lipoprotein profiles 51 4.2.10 Biliary bile acids, cholesterol and phosphoiipids production 52 4.2.11 In vivo clearance of VLDL?like TG?rich particles and uptake in tissues 57 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 4.3 Results study 3 64 4.3.1 Markers of general well?being 64 4.3.2 Body weight 64 4.3.3 Liver and perlgonadal fat weight 65 4.3.4 Food intake 67 4.3.5 Plasma ALT 6 8 4.3.6 Plasma cholesterol 68 4.3.7 Plasma HDL?cholesterol.. 69 4.3.8 Plasma triglycerides 71 4.3.9 Lipoprotein profiles 72 4.3.10 Plasma ApoAl 73 4.3.11 Plasma CETP mass 74 4.3.12 Plasma CETP activity 7 5 4.3.13 In vivo clearance of autologous HDL A 76 4.3.14 Liver microsomal DGAT activity 79 4.3.15 Liver lipid analysis 80 4.3.16 Liver histology 82 4.3.17 Liver microarray analysis 85 5 Conclusions and comment589 6 References 7 Appendices.. Appendix I Body weight 94 Appendix II Tissue weight 97 Appendix Food intake 100 Appendix IV Plasma choiesterol 103 Appendix Plasma 106 Appendix VI Plasma triglycerides 1 09 Appendix VII Plasma free glycerol 112 Appendix Plasma free fatty acids 113 Appendix IX Plasma ApoAl 114 Appendix Plasma CETP mass 115 Appendix XI Plasma CETP activity 116 Appendix XII Post heparin LPL and HL activity 117 Appendix Fecal lipids 118 Appendix XIV VLDL-triglycerides and de novo ApoB production 123 Appendix XV Biliary bile acids, cholesterol and phospholipids 126 Appendix XVI In vivo clearance of VLDL-like TG-rich particles and uptake in liver 130 Appendix XVII In vivo clearance of autologous HDL 140 Appendix Liver mlcrosomal DGAT activity 141 Appendix XIX Liver lipids 142 Appendix XX Liver microarray analysis 143 Appendix XXI Summary table 153 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Sponsor of the study 3M, Medical Department 3M Center, Building 0220-02-E-02 St.Paui, Minnesota 55144?1000 USA Study contact John L. Butenhoff, CIH, DABT Corporate Scientist Medical Department 3M Center, Building 0220-02-E-02 St.Paui, Minnesota 55144-1000 USA Phone: +1.651.733.1962 Fax: +1.651.733.1773 E-mail: Testing facility TNO, Business Unit Biosciences Mail address: P.0.Box 2215, 2301 CE Leiden Delivery address: Zernikedreef 9, 2333 CK Leiden The Netherlands Responsible personnel Study director Dr. Hans M.G. Princen Phone: +31 71 518 1471 Fax: +31 71 518 1901 Eemail: hans.princen@tno.ni Technical staff Marian Bekkers Simone Droog Annemarie Maas Elsbet Pieterman Karin Toet Marijke Voskuilen Dr. Marjan van Erk Dr. Anita van den Hoek Dr. Jos? van der Hoorn Dr. Patrick Rensen Advisors Prof. Dr. Louis M. Havekes E-mail: louis.havekes@tno.n Dr. Patrick Rensen E?mail: p.c.n.rensen@lumc.n 3M#03 1 Mechanism o f different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 Introduction & aim Perfluorinated alkyl sulphonates are fully fluorinated amphiphilic organic molecules with strong surface-tension reducing properties. They are stable to environmental and metabolic degradation. Perfluorooctanesulfonate (PFOS) is widely dispersed in humans, fish-eating wildlife, and surface waters. Toxicological studies in rats and monkeys have shown a reduction in serum cholesterol after treatment with PFOS; however, such reductions have not been observed among exposed workers. I n order to obtain more insight into the possible mechanism of action the APOE*3-Leiden transgenic mouse, a well-recognized animal model for hyperlipidemia and atherosclerosis, has been used in a previous study to investigate the in vivo effects of three perfluoro-alkyl-sulphonates with different chain length, PFBS (C4), PFHS (C6) and PFOS (C8), on plasma lipids and lipoproteins and bile acid metabolism. Of the three PFAS, PFHS and PFOS have been withdrawn from the market by 3M in the beginning of 2000 because of environmental issues, whereas PFBS is marketed in various industrial applications. APOE*3Leiden transgenic mice exhibit elevated plasma cholesterol and triglyceride levels, mainly confined t o the VLDVLDL sized lipoprotein fraction (1). Extensive previous research showed that, in contrast t o wild-type mice, APOE*3Leiden transgenic mice are highly responsive to fat and cholesterol feeding as far as the effects on plasma VLDL and chylomicron levels are concerned (2, 3). I n addition, we have found that drugs and dietary compounds influencing either the chylomicron and VLDL production and/or the hepatic clearance of lipoproteins exert relatively strong effects on plasma cholesterol and triglyceride levels (4-11, see for review ref. 12). I n contrast, in normal wild-type mice the plasma cholesterol and triglyceride levels are very low and (almost) not responsive to diet and hypolipidemic drugs. This animal model has been proven t o be representative for the human situation regarding plasma lipoprotein levels, lipoprotein profiles, its responsiveness to hypolipidemic drugs (like statins, fibrates etc.) (4-8, 10) and nutrition (9, 11). I n addition, depending on the level of plasma cholesterol APOE*3Leiden mice develop atherosclerotic lesions in the aorta resembling those found in humans with respect t o cellular composition and morphological and immunohistochemical characteristics (3). TNO in collaboration with the Leiden University Medical Center has recently developed the APOE*3Leiden(E3L)IhuCETP transgenic mouse, which has proven t o be very suitable for testing the effects of drugs and nutritional factors on plasma HDL and triglyceride levels, atherosclerosis and metabolic syndrome. I n the newly generated mouse, human cholesterol ester transfer protein (huCETP) under control of its natural flanking regions is introduced into the APOE*3-Leiden mouse resulting in a more human-like lipoprotein profile with transfer of cholesterol ester from HDL to the apoB-containing lipoproteins in exchange for triglycerides. As a result of this adverse lipoprotein distribution and the higher amount of atherogenic apoB-containing lipoproteins, the E3L.CETP transgenic mice develop increased atherosclerosis on a Western-type diet as compared to E3L transgenic mice (13). The E3L.CETP transgenic mice respond to treatment with (registered) drugs as fibrates (14), statins (IS), niacin (16) the CETP inhibitor torcetrapib (7) at similar dosages and in a similar way t o humans, with decreases in the apoB-containing lipoproteins and an increase in HDL levels. From the previous study in APOE*3-Leiden transgenic mice, it was concluded that PFHS and PFOS have strong cholesterol and triglycerides lowering effects. Lipoprotein profiles of PFHS and PFOS treated APOE*3-Leiden mice also showed the formation of a 'large HDL" particle, presumably an apoE-, cholesteryl ester-rich HDL-1 particle. Moreover, PFHS and PFOS treated animals showed increased plasma ALAT levels and liver size, decreased bile acid synthesis and cholesterol-7ahydroxylase activity, increased fatty acid oxidation, and decreased body weight, epididymal fat and increased energy expenditure via burning of fat, all suggesting PPARa agonist activity for these two chemicals. PFBS differed markedly from PFHS and PFOS by having no or only minor effects on the measured parameters, whereas PFHS appeared t o have intermediate effects taking into account the higher dose applied as compared to PFOS. I n the present study the focus is put on a further elucidation of the mechanism responsible for the observed changes in plasma triglycerides and cholesterol levels in the atherogenic apoB-containing (VLDL, IDL, LDL) lipoproteins and to investigate the effect of PFBS, PFHS and PFOS on HDL metabolism. Therefore, in this study the more appropriate APOE*3-Leiden(E3L)IhuCETP transgenic 3M#03 Mechanism o f different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 mouse model was used. The aim of the study is further t o emphasize the differences in biological effects between PFBS, PFHS and PFOS and t o publish these data. 1.1 Aim of the study To elucidate the mechanism of action responsible for, the observed changes in plasma triglycerides and cholesterol levels in the atherogenic apoB-containing (VLDL, IDL, LDL) lipoproteins by PFHS and PFOS and to investigate the effect of PFBS, PFHS and PFOS on HDL metabolism in E3L.CETP transgenic mice. To emphasize the differences in biological effects between PFBS, PFHS and PFOS and to publish these data. - . 3M#03 Mechanism of different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 2 Materials and methods 2.1 Test substances & reference substance Test substances: Reference substance: PFBS (perfluorobutanesulfonate): L-7038, lot 2 (1999) PFHS (perfluorohexanesulfonate): 127498-80, lot L9051 PFOS (perfluorooctanesulfonate): FC-95, lot 217 Fenofibrate: F6020, lot 117K1486 PFBS, PFHS and PFOS were provided by 3M Medical Department (sent to TNO on 10-Jun-04). Fenofibrate was purchased from Sigma (St. Louis, USA). 2.2 Mice Based on the large difference in half-life between males (17h) and females (3h) in CD-1 mice, we have used males in the proposed experiments. One-hundred-twenty male heterozygous APOE*3Leiden.CETP mice, 7-10 weeks of age, from the SPF breeding stock a t TNO-Biosciences (Leiden) have been used, and housed during the experiment in macrolon cages (maximal 4 mice per cage), in clean-conventional animal rooms at TNO-Leiden (relative humidity 50-60°/o, temperature ~ 2 1 ' ~light , cycle 7 am t o 7 pm). Individual animals are marked by ear punch-holes. Mice were supplied with food and acidified tap water ad libitum. 2.3 Animal welfare Experiments were performed conform to the rules and regulations set forward by the Netherlands Law on Animal Experiments. Experiments had been approved by the Animal Experiment Committee of TNO under registration number 2483. The study director was entitled t o terminate experiments in case of serious unexpected animal discomfort. 2.4 Diets Mice received Western type diet (WTD; semi synthetic diet containing 14% beef tallow, 1°/o corn oil and 0.25% of cholesterol), purchased from ABdiets (Woerden, The Netherlands). This resulted in total cholesterol (TC) plasma levels of about 8 mmol/L and triglyceride levels of about 2 mmol/L. Diets were renewed once per week. 2.5 Compound administration The required quantities of PFBS, PFHS and PFOS were supplied by 3M (sent t o TNO on 10-Jun-04). Fenofibrate (Sigma, St. Louis) was used as a positive control. All compounds were administered orally as admix t o the WTD. Preparation of all diets (with different compounds) were performed according to: Operation procedure number 14 of "Laboratorium procedures Lipiden" entitled "Preparation of diet chunks from powdered food". The lyophilized diet chunks were stored in vacuum bags in an alarm-secured -20°C room. 3M#03 2.6 Mechanism of different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 Treatment groups Group 1: WTD diet Group 2: WTD 0.03 O/O fenofibrate (w/w) or 31.8* mg/kg body weightlday (positive control) Group 3: WTD 0.03 O/O PFBS (w/w) or 31.1* mg/kg body weightlday Group 4: WTD 0.006 O/O PFHS (w/w) or 6.0* mglkg body weightlday Group 5: WTD + 0.003 O/O PFOS (wlw) or 3.1* mglkg body weightlday * Average intake of compounds based on average body weight and food intake in the three indicated experiments + + + 2.7 Study design Design experiment 1 1 Weeks of treatment:' Group 1: diet WTD Group 2: WTD + 0.03 % fenofibrate Group 3: WTD + 0.03 % PFBS Group 4: WTD + 0.006 % PFHS Group 5: WTD + 0.003 % PFOS b c r i f i c e EDTA-plasma, serum, liver, . -41 -31 -21 -1 I0I1 1 x x x b X 2 1 31 4 1 5 1 6 ---------- X-X-.-.-.-.-.-.-, X b . -X X--------, )( -.-. -.-.-., I Experiment 1was performed according to the scheme as outlined above. I n short: A run-in period of 4 weeks was started with 54 male E3L.CETP mice on a semi synthetic western type diet (containing 14 010 beef tallow, 1 O/O corn oil, 0.25 O/O cholesterol). I n week 0 the animals were randomized on body weight, plasma cholesterol, HDL-cholesterol and triglycerides (after 4h fasting) in 5 groups of 8 animals and the 6-weeks treatment was started. Body weight, food intake, plasma cholesterol, HDL-cholesterol, triglycerides, glycerol and free fatty acids (after 4h fasting) were measured at t=O weeks and 4 and 6 weeks after start of treatment. Plasma ALAT as a parameter for liver damage and lipoprotein profiles were measured in pooled samples per group a t t=O, 4, 6 weeks (ALAT) and t=4, 6 weeks (lipoprotein profiles). Extra plasma (without radioactivity) was collected at 6 weeks for measurements of plasma levels of the compounds by 3M. After 5 weeks of treatment lipolytic activity (LPL and HL activity) was measured from postheparin plasma from fasted mice (4h fasting) and feces was collected for the determination of bile acids, neutral sterols and fatty acids. At the end of the experiment VLDL-triglyceride and de novo apoB production was measured and lipid composition of VLDL was determined. After sacrifice EDTAplasma, serum and perigonadal adipose tissue and liver was collected. Both tissues were directly frozen in liquid nitrogen and stored below - 70 OC. The run-in period of study 1started on 25 March 2008, mice were sacrificed on 6 June 2008. 3M#03 Mechanism of different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 Design experiment 2 Weeks of treatment: Group 1: diet WTD Group 2: WTD + 0.03 % fenofibrate Group 3: WTD + 0.03 % PFBS Group 4: WTD + 0.006 % PFHS Group 5: WTD + 0.003 % PFOS ISacrifice EDTA-plasma, serum, liver, 1 -41 -31 -21 -1 I0 I112 1 3 1 - 1 4 -.-.-.-.-, XX- x X------- X- x------, X---------, X - . - . - . + I Experiment 2 was performed according to the scheme as outlined above. I n short: A run-in period of 4 weeks was started with 38 male E3L.CETP mice on a semi synthetic western type diet (containing 14 OO/ beef tallow, 1 O/O corn oil, 0.25 O/O cholesterol). I n week 0 the animals were randomized on body weight, plasma cholesterol, HDL-cholesterol and triglycerides (after 4h fasting) in 5 groups of 6 animals and the 4-weeks treatment was started. Body weight, food intake, plasma cholesterol, HDL-cholesterol and triglycerides (after 4h fasting) were measured at t=O and 4 weeks. Plasma ALAT as a parameter for liver damage and lipoprotein profiles was measured in pooled samples per group at t=O and 4 weeks (ALAT) and t=4 weeks (lipoprotein profiles). Extra plasma (without radioactivity) was collected at 4 weeks for measurements of plasma levels of the compounds by 3M. After 4 weeks of treatment at the end of the experiment the gall bladder was cannulated and during 45 minutes bile flow, biliary cholesterol, phospholipids and bile salt output were measured. Directly hereafter the in vivo clearance of VLDL-like triglycerides-rich particles was determined. After sacrifice EDTA-plasma, serum and perigonadal adipose tissue and liver was collected. The perigonadal adipose tissue and liver were directly frozen in liquid nitrogen and stored below -70 O C . The run-in period of study 2 started on 16 June 2008, mice were sacrificed on 15 August 2008. The extra mice from study 3, used for bile cannulation were sacrificed on 2 October 2008. 3M#03 Mechanism of different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.I2685 Design experiment 3 Weeksoftreatment: Group 1: diet WTD Group 2: WTD + 0.03 % fenofibrate Group 3: WTD + 0.03 % PFBS Group 4: WTD + 0.006 % PFHS Group 5: WTD + 0.003 % PFOS ISacrifice EDTA-plasma, serum, liver, intestine, 1-41-31-21-IIOI1121314 x---- X- I I ..I Experiment 3 was performed according to the scheme as outlined above. I n short: A run-in period of 4 weeks was started with 38 male E3L.CETP mice on a semi synthetic western type diet (containing 14 O/O beef tallow, 1 O/O corn oil; 0.25 O/O cholesterol). I n week 0 the animals were randomized on body weight, plasma cholesterol, HDL-cholesterol and triglycerides (after 4h fasting) in 5 groups of 7 animals and the 4-weeks treatment was started. Body weight, food intake, plasma cholesterol, HDL cholesterol and triglycerides and plasma ApoAl, CETP activity and mass (after 4h fasting) was measured at t=O and 4 weeks. ALAT as a parameter for liver damage and lipoprotein profiles was measured in pooled samples per group at t=O and 4 weeks (ALAT) and t=4 weeks (lipoprotein profiles). Extra plasma (without radioactivity) was collected at 4 weeks for measurements of plasma levels of the compounds by 3M. After 3 weeks of treatment one mouse of each group was sacrificed by COz and serum was collected from the retro-orbital vein. A part of the liver was directly frozen in liquid nitrogen. HDL was isolated and radiolabeled with 3H-cholesteryl oleyl ether. The in vivo clearance of autologous HDL was determined in 5 mice per group for 24 hr after an injection of 3~-cholesteryloleyl ether labeled autologous HDL. After sacrifice EDTA-plasma, serum and small intestine, perigonadal adipose tissue and liver were collected. The small intestine was directly frozen in liquid nitrogen and stored below -70 OC for mRNA isolation. The liver was cut in four parts; two parts of the liver were directly frozen in liquid nitrogen and stored below - 70 OC for mRNA isolation and liver lipid analysis, one part was used fixed in buffered formalin for liver histology and a part was used for the isolation of microsomes. Perigonadal adipose tissue was directly frozen in liquid nitrogen and stored below - 70 OC.'One mouse per group was used for an extra bile cannulation at t=5 weeks. The runLin period of study 3 started on 28 July 2008, mice were sacrificed on 25 September 2008, except the extra mice from study 3. These mice, used as extra mice for the bile cannulation experiment for study 2, were sacrificed on 2 October 2008. 3M#03 Mechanism of different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031 .I2685 The last measurement (CETP activity) was performed on 7 September 2009. Liver histology analysis was performed in week 38, 2009. 2.8 Measurements experiment 1 Body weight and food intake (per cage) at 0, 4 and 6 weeks. Plasma cholesterol, HDL cholesterol, triglycerides, glycerol and free fatty acids (after 4h fasting) at 0, 4 and 6 weeks. Plasma cholesterol and triglycerides were determined using kit "Cholesterol CHOD-PAP" and kit "Triglycerides GPO-PAP" both from Roche (Mannheim, Germany). Plasma glycerol was determined using "the free glycerol determination kit" from Sigma (St. Louis, USA). Plasma free fatty acids were determined using kit 'NEFA C" from WAKO (Neuss, Germany). Plasma HDL-cholesterol was determined using kit "Cholesterol CHOD-PAP" from Roche (Mannheim, Germany) after the precipitation of apoB containing lipids by MnClz (0.2 mol/L) and heparin (500 IU/mL). Lipoprotein profiles at group level at 4 and 6 weeks. Lipoproteins were separated by FPLC analysis using an AKTA apparatus. Analyses were performed in freshly obtained pooled samples per group. Cholesterol and phospholipid profiles were measured in thg fractions using kit "Cholesterol CHOD-PAP" from Roche (Mannheim, Germany) ,and kit "Phospholipids" from Instruchemie (Delfzijl, the Netherlands), respectively. Plasma ALAT at group level at 0, 4 and 6 weeks. Plasma ALAT was measured using the spectrophotometric assay of the Roche Reflotron plus system (Mannheim, Germany). ~ipoproteinlipase and hepatic lipase activity at 5 weeks. Lipolytic activity (lipoprotein lipase and hepatic lipase activity) was determined as described previously (18). I n short: Postheparin plasma from fasted mice (4h) was collected from the tail vein at 20 minutes after intraperitoneal injection of heparin (0.5 IU/g body weight). Posth.eparin plasma triacylglycerol hydrolase activity was determined in the presence or absence of 1 mol/L NaCl t o estimate both the lipoprotein lipase (LPL) and hepatic lipase (HL) activity. LPL activity was calculated as the portion of total lipase activity inhibited by 1 mol/L NaCI. VLDL-triglyceride and VLDL-apoB production and lipid composition of nascent VLDL at 6 weeks. The rate of hepatic VLDL-triglyceride production, de novo apoB secretion and lipid composition of nascent VLDL was determined in 4 h r (or overnight) fasted mice as described previously (18, 19). I n short: Mice were anesthetized with fluanisone-fentanylmidazolam intraperitoneally and injected intravenously with 0.1 m l phosphate-buffered b e l Biomedicals, Irvine, USA) t o measure de novo saline containing 100 pCi ~ r a n ~ ~ s - l a(ICN apoB synthesis. After 30 mint the animals received a Triton WR1339 injection (500 mg/kg body weight), which virtually completely inhibited VLDL clearance by blocking LPL mediated lipolysis. Blood samples were drawn 0, 15, 30, 60 and 90 min after Triton WR1339 injection and plasma triglycerides concentrations were measured. After 90 minutes mice were killed and blood was collected by heart punction for isolation of VLDL and subsequent determination of de novo apoB synthesis and VLDL composition. For that purpose VLDL particles (density < 1.019) were separated from other lipoproteins by density gradient ultracentrifugation. The protein content of the VLDL fraction was determined by a Lowry protein determination, and triglycerides and total cholesterol concentrations were determined using kit "Cholesterol CHOD-PAP" and kit "Triglycerides GPO-PAP" both from Roche (Mannheim, Germany). Phospholipid and free cholesterol concentrations were determined using kit "Phospholipids" and kit "Free cholesterol C" both from Instruchemie (Delfzijl, The Netherlands). The 3 5 ~ - a p content o~ of VLDL was measured after selective precipitation of apoB with isopropanol. The VLDL-TG production rate was calculated by curve fitting (trendlines with the equation: plasma TG=a*time b, in which a is the calculated production rate). Lipid composition of VLDL was calculated per newly synthesized ApoB and expressed as pmolldpm ApoB. Fecal excretion of bile acids, neutral sterols and fatty acids a t 5 weeks (n= 6 per group). Fecal bile acids, neutral sterols and fatty acids were determined in feces collected during a 48-72 h time period in 3 subgroups at 2 consecutive time points during week 5, by gas chromatographic (GC) analysis as described previously (19, 20). After sacrifice at 6 weeks (end of VLDL-triglyceride and VLDL-apoB production experiment) EDTA-plasma (also needed for isolation of VLDL and subsequent determination of de novo apoB synthesis and VLDL composition), serum, perigonadal adipose tissue and liver. EDTAplasma and serum were collected from heart punction. The liver and perigonadal adipose tissue was weighed and directly frozen in liquid nitrogen and stored below -70 OC. + 3M#03 Mechanism o f different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 Plasma taken at 6 weeks after 4h fasting was sent to 3M for measurements of plasma levels of compounds. Samples were sent on dry ice by Fedex courier on December '1 2008 and arrived at the 3M center on December 2"d 2008. Measurements experiment 2 Body weight and food intake (per cage) at 0 and 4 weeks. Plasma cholesterol, HDL cholesterol and triglycerides (after 4h fasting) at 0 and 4 weeks. Plasma cholesterol and triglycerides were determined using kit 'Cholesterol CHOD-PAP" and kit "Triglycerides GPO-PAP" both from Roche (Mannheim, Germany). Plasma HDLcholesterol was determined using kit "Cholesterol CHOD-PAP" from Roche (Mannheim, . o,, f ' a p . o ~containirig .. lipids by MnClz (0.2 mol/L) and heparin Germany) after the precipitation (500 IU/mL). Lipoprotein profilesat group level at 4 weeks. Lipoproteins were separated by FPLC analysis using an AKTA apparatus. Analyses were performed in freshly obtained pooled samples per group. Chol'esterol and phospholipid profiles were measured in the fractions using kit "Cholesterol CHOD-PAP" from Roche (Mannheim, Germany) and kit "Phospholipids" from Instruchemie (Delfzijl, the Netherlands) respectively. Plasma ALAT at group level at 0 and 4 weeks. Plasma ALAT was measured using the spectrophotometric assay of the Roche Reflotron plus system (Mannheim, Germany). Bile cannulation, bile flow, biliary cholesterol, phospholipids, bile acids at 4 weeks. Determination of biliary lipid secretion and bile flow was determined as described previously (20). I n short: The common bile duct of anesthetized mice was ligated, and the gallbladder was cannulated. Bile was collected in 15 minutes intervals for 45 minutes. Bile flow was measured. Cholesterol and phospholipid concentrations in bile were determined determined using kit 'Cholesterol CHOD-PAP" and kit "Phospholipids" from Roche (Mannheim, Germany) and Instruchemie (Delfzijl, The Netherlands), respectively. Total bile acids were determined in bile using kit 'total bile acids assay" from Lucron Bioproducts (Milsbeek, the Netherlands) Clearance of VLDL-like particles and uptake in liver at 4 weeks. The in vivo clearance of VLDL-like triglycerides-rich particles was adapted from what was described previously (19). I n short: Directly after bile cannulation experiment VLDL-like emulsion particles containing 200 pCi 3 ~ - t r i o l e i nand 20 pCi 14C-cholesteryl oleate were injected into the tail vein at a dose of 1 mg triglycerides per mouse. At 2, 5, 10, 20 and 30 min blood samples (50 pl) were taken from the tail vein. 3H and I4C activities were counted in 10 pl serum and corrected for total serum volume. At the end of the experiment liver, heart, perigonadal fat, spleen and muscle (femoralis) were collected. All tissues were weighed. Lipids were extracted from the tissues by an overnight incubation at 60 OC in 500 pl Solvable (PerkinElmer, Wellesley, USA) and radioactivity was measured. The clearance rate of VLDL-like particles was calculated by curve fitting (trendlines with the equation: O/O of injected d~se=b*e~'"'~/~, in which a is the calculated half life of the labeled VLDL-like particles). After sacrifice at 4 weeks (end of in vivo VLDL clearance experiment) collection of EDTAplasma, remainder liver and perigonadal adipose tissue. EDTA-plasma was collected from heart punction. The remainder of the livers and perigonadal fat tissues were directly frozen in liquid nitrogen and stored below -70 OC. 15 p1 o f (non-radioactive) plasma taken at 4'weeks after 4h fasting was sent to 3M for measurements of plasma levels of compounds. Samples were sent on dry ice by Fedex courier on December lSt 2008 and arrived at the 3M center on December 2nd 2008. Measurements experiment 3 Body weight and food intake (per cage) at 0 and 4 weeks. Plasma cholesterol, HDL-cholesterol and triglycerides (after 4h fasting) at 0 and 4 weeks. Plasma cholesterol, HDL-cholesterol and triglycerides were determined using kit "Cholesterol CHOD-PAP" and kit "Triglycerides GPO-PAP" both from Roche (Mannheim, Germany). Plasma HDL-cholesterol was determined using kit 'Cholesterol CHOD-PAP" from Roche (Mannheim, Germany) after the precipitation of apoB containing lipids by MnClz (0.2 mol/L) and heparin (500 IU/mL). Lipoprotein profiles at group level at 4 weeks. Lipoproteins were separated by FPLC analysis using an AKTA apparatus. Analyses were performed in freshly obtained pooled samples per group. Cholesterol and phospholipid profiles were measured in the fractions using kit 3M#03 Mechanism o f different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 "Cholesterol CHOD-PAP" from Roche (Mannheim, Germany) and kit "Phospholipids" from Instruchemie (Delfzijl, the Netherlands), respectively. Plasma ALAT at group level a t '0 and 4 weeks. Plasma ALAT was measured using the spectrophotometric assay of the Roche Reflotron plus system (Mannheim, Germany). Plasma CETP activity (17) and mass (16) at 0 and 4 weeks. CETP activity was determined using kit "Roar CETP Activity assay kit" from Roar Biomedical Inc (New York, USA). CETP mass was determined using the CETP ELISA from Daiichi Pure Chemicals Co. (Tokyo, Japan). Plasma ApoAl concentration at 0 and 4 weeks. Plasma ApoAl concentrations were determined using a sandwich ELISA (16). Rabbit anti-mouse ApoAl polyclonal antibody from Abcam PIC. (Cambridge, UK) was coated overnight (3 pg/mL) onto Costar strips (New York, USA) at 4OC and was incubated with diluted mouse plasma (dilution 1:400,000) for 90 min at 37OC. Subsequently, goat anti-mouse ApoAl antibody from Rockland Immunochemicals Inc. (Gilbertsville, USA; dilution 1:3,000) was added and incubated for 90 min at 37OC. Finally, HRP-conjugated rabbit anti-goat IgG antibody from Rockland Immunochemicals Inc. (Gilbertsville, USA; dilution 1:15,000) was added and incubated for 90 min at 37OC. HRP was detected by incubation with tetramethylbenzidine from Organon Teknika (Boxtel, The Netherlands). Purified mouse ApoAl from Biodesign International (Saco, USA) was used as a standard. . I n vivo clearance of autologous HDL at 8 weeks. The radiolabeling of auto logo"^ HDL and in vivo clearance of autologous HDL was performed as described previously (14). Iri short: For the radiolabeling of autologous HDL, one mouse from each experimental group was euthanized by C02, and blood was drawn from the retro-orbital vein. (One piece of the liver was collected and directly frozen in liquid nitrogen and stored below -70 OC for mRNA isolation and liver lipids determination). Serum was collected and HDL was isolated after density ultracentrifugation. HDL (0.32 pmol HDL-cholesterol) was radiolabeled by incubation (37,OC, 24 h) with 3~-cholesteryloleyl ether labeled egg yolk phosphatidylcholine vesicles (40 pCi, 0.5 mg of phospholipid) in the presence of lipoprotein-deficient serum ( 1 m,L) from E3Leiden.CETP mice. Subsequently, HDL was reisolated after density ultracentrifugation. For the in vivo clearance of autologous HDL, mice were injected via the tail vein with a trace of autologous 3~-cholesteryloleyl ether labeled HDL (0.1 pCi in PBS) a t 8 am. Blood was collected at 1, 2, 4, 8 and 24 h after injection and 3 ~ - a c t i v i t ywas counted in plasma samples to determine the plasma decay of 3H-cholesteryl oleyl ether. HDL clearance was calculated by curve fitting timeframe 0-8 h (trendlines with the equation O/O o f injected d o s e = 1 0 0 * e ~ * ~in ~ ~which ~, a is the calculated fractional catabolic rate of HDL). After sacrifice at 4 weeks (end of in vivo HDL clearance experiment) collection of EDTAplasma, serum, liver, small intestine and perigonadal adipose tissue. EDTA-plasma and serum was collected from hea,rt punction. The liver was weighed and cut in 4 parts, two parts and the small intestine were directly frozen in liquid nitrogen and stored below -70 OC. One part was fixed in phosphate buffered formalin (lOO/o) for liver histology. The remainder of the liver was used t o isolate microsomes for DGAT activity measurements (per liver microsomes were isolated) Perigonadal adipose tissue was weighed and directly frozen in liquid nitrogen and stored below -70 OC. 15 p1 of (non-radioactive) plasma taken a t 4 weeks after 4h fasting was sent t o 3M for measurements of plasma levels of compounds. Samples were sent on dry ice by Fedex courier on December lS 2008 and arrived at the 3M center on December 2nd 2008 DGAT2 activity in liver microsomes. The DGAT activity in liver microsomes was determined as described previously (7, 18) in the presence of 5 mmol/L or 100 mmol/L MgCI2. Since DGAT2 activity is inhibited a t higher concentrations of MgC12 (100 mmol/L), DGAT2 activity can be calculated by subtracting the DGAT activity in the presence of 100 mmol/L MgC12 (only DGATl activity) from the DGAT activity in the presence of 5 mmol/L MgC12 (DGATl and DGAT2 activity). Micro array analysis livers: RNA was isolated from collected liver pieces. After quality control of the RNA, microarray analysis was carried out a t Service XS B.V. (Leiden, the Netherlands) using the AfFymetrix technology platform and Affmetrix Genechip@ mouse genome MOE430-2.0 arrays. Data were sent to TNO Zeist for gene expression data analysis (21). Liver lipids (FC, CE, TG and PL) in 6 samples per group and liver histology in 5 samples per group. Liver lipids were determined by a HPTLC analysis as previously described (22). 3M#03 2.11 Mechanism of different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 Statistical analysis Significance of differences between the groups was calculated non-parametrically, using the computer program SPSS. A Kruskall-Wallis test for several independent samples was used, followed by a Mann-Whitney U-test for independent samples. A P-value < 0.05 was considered statistically significant. The resulting p-values were given in separate tables in the report. 3M#03 Mechanism o f different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 3 Deviations from the protocol Experiment 1 - Lipolytic activity determination: Post-heparin plasma was collected in week 5 instead of week 4 of treatment. Instead of 1 I U heparinlg body weight 0.5 I U heparinlg body weight heparin was injected intraperitoneally, 20 minutes before collection of post-heparin plasma. - Feces were collected during a 48 h and a consecutive 72 h time period in 3 subgroups instead of 2 consecutive time points of 48 hours. - We decided not t o collect liver pieces for histology analysis, liver lipid analysis and mRNA isolation and subsequent micro array analysis in experiment 1 but in experiment 3, since injection of Triton WR-1339 (used to block VLDL clearance) could have an effect on liver gene expression and liver lipids. Intestine was collected in experiment 3 instead of experiment 1. - At sacrifice no serum was collected, only EDTA-plasma. All of the plasma was used for VLDL isolation (via ultracentrifugation). - Lipid composition of VLDL is expressed per dpm 3 5 ~ - ~ instead p o ~ of per mg protein, since the protein concentration was very low in the PFHS and PFOS treatment, and could not be measured properly. Experiment 2 - 5 extra mice from experiment 3 (one mouse per group) were used t o perform extra bile cannulations for determination of bile flow, biliary cholesterol, phospholipids, bile acids. These extra mice were used since some mice from experiment 2 had t o be excluded from analysis because of a very low bile production. The excluded mice cooled off too much during the cannule placement operation, and the mice were not warm enough during the bile collection, which affected bile production drastically. - Clearance of VLDL-like particles and uptake in liver at 4 weeks: We decided to only measure uptake in liver of VLDL-like particles at the end of the VLDL-clearance experiment, instead of taking liver biopts at every blood sampling point, to minimize loss of mice. Next t o the liver, uptake of VLDL-like particles was measured in heart, muscle (fernoralis), perigonal fat and spleen. - At sacrifice no serum was collected, only EDTA-plasma. No intestine was collected and stored in liquid nitrogen. Experiment 3 - Treatment was started with 5 groups o f 7 animals instead of 5 groups of 6 animals. The extra mouse per group was used for an extra bile cannulation (see also deviations in experiment 2). - Instead of 0.4 yM HDL per group, 0.32 yM HDL was used for radiolabeling. The reason for this was that less than 0.4 pM HDL was isolated from the PFHS and PFOS treatment groups. - We decided not t o collect liver pieces for histology analysis, liver lipid analysis and mRNA isolation and subsequent micro array analysis in experiment 1 but in experiment 3, since injection of Triton WR-1339 (used t o block VLDL clearance) could have an effect on liver gene expression and liver lipids. Intestine was collected in experiment 3 instead of experiment 1for that same reason. - After consultation with the Sponsor (see email correspondence of 28-Jan-2009), we decided t o perform micro array analysis on the livers instead of liver and intestinal mRNA analysis of some selected genes. - Microsomes were isolated per liver instead of combining 2 livers; per group 5 remainders of livers were used for microsomes isolation. Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 4 Results 4.1 Results study 1 4.1.1 Markers of general well-being During the study one mouse (mouse 22) was put in a separate cage because of an open wound from fighting. Another mouse (mouse 15) showed some scabs in the belly area. No other speci?c clinical signs were observed during the study. At sacrifice, besides the livers no macroscopic differences were observed (between the groups. Livers were visibly somewhat enlarged in the fenofibrate group and strongly enlarged in the PFHS and PFOS group. 4.1.2 Body weight Values are absolute values (grams) and are means i S.D. from 8 mice per group. Individual body weights are given in appendix I. Body weight (9) time (weeks) 0 1 4 6 Control avg 26.3 27.5 28.6 29.0 sd 2.4 2.5 2.6 2.9 Fenofibrate avg 26.8 27.6 28.0 28.8 sd 2.9 2.3 2.3 2.6 PFBS avg 26.1 27.2 28.0 28.4 sd 1.7 1.8 1.8 1.8 PFHS avg 27.3 28.5 29.1 28.9 sd 2.2 2.2 2.1 2.0 PFOS avg 27.2 28.0 28.2 28Control vs: Fenofibrate 0.574 0.798 1.000 0.959 5 . PFHS 0.382 0.328 0.574 0.878 PFOS 0.382 0.645 0.721 0.721 Fenofibrate vs PFBS 0.279 0.382 0.382 0.505 PFHS 0.87 0.7 1 0.721 .798 PFOS 0.959 0.878 0.798 0.382 vs PFOS 0.279 0.505 1.000 0.721 VS During the study an increase in body weight was seen. Between groups no significant differences were observed. 18 Mechanism of di??erent on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Figure 4.1.2 Body weight 40 *Control Fenofibrate PFBS ?new PFHS 35 Body weight '0 UI I Time (weeks) p<0.05 vs. control 4.1.3 Liver and perigonadal fat weight Values are absolute values (grams) and are means 5.0. from 8 mice per group. Individual tissue weights are given in appendix II. Tissue weight (9) Liver Perigonadal fat Control avg 1.27 0.62 sd 0.18 0.24 Feno?brate avg 1.83 0.50 5d 0.27 0.11 PFBS avg 1.45 0.51 5d 0.17 0.15 PFHS avg 3.32 0.45 sd 0.23 0.14 PFOS avg 3.23 0.38 sd 0.29 0.10 weight: P?value Liver Perigonadal fat een rou . . Control vs: Feno?brate 0.001 0.234 PFHS <0.001 0.083 PFOS 1 7 Feno?brate vs PFBS 0.003 0.798 PFHS <0.001 0.442 PFOS <0.001 0.065 PFOS <0.001 0.130 19 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Livers were significantly enlarged in all treatment groups. Fenofibrate and PFBS treatment increased liver weights by respectively +44% and The PFHS and PFOS groups showed a robust induction in liver weights, respectively by +162% and +155 which was also significantly higher as compared to the PFBS group. Although a slight decrease was seen in all treatment groups, only PFOS significantly decreased perigonadal fat weight (by 39%, Figure 4.1.33 Liver weight 5 RControl ?Fenofibrate 4 II PFBS t. e33 PFHS week 6 p<0.05 vs. control Figure 4.1.3b Perigonadal fat weight 1.2 Control I Fenofibrate 1_o . a PFBS 333: PFHS I PFOS 9 co I Perigonadal fat weight In a: I week 6 p<0.05 vs. control 20 Mechanism of di??erent on lipid and iipoprotein metabolism in mice TNO project number 031.12685 4.1.4 Food intake Values are absolute values (gram/mouse/day) and are means S.D. from 3-4 cages per group. Food intake values per individual cage are given in appendix Food intake (g/day/mouse) time (weeks) ?1-0 0-1 3-4 5-6 Control avg 2.8 2.8 2.9 sd 0.2 0.2 0.4 0.5 Fenofibrate avg 2.8 2.7 3.3 3.3 5d 0.2 0.2 0.2 0.3 PFBS avg 2.8 2.9 3.0 3.4 sd 0.2 0.3 0.1 0.6 PFHS avg 2.8 2.8 2.9 2.9 sd 0.2 0.4 0.3 0.3 PFOS avg 2.8 2.8 3.0 2.9 sd 0.2 0.2 0.4 0.4 Food intake: P-value time . . . Control vs: Fenofibrate 1.000 0.114 0.400 PFBS PFHS 1.000 0.700 1.000 PFOS 0.700 .400 1. Fenofibrate vs PFBS 0.400 0.114 1.000 PFHS PFOS 0.629 0.400 0.229 vs . . PFOS 1.000 0.700 0.229 vs All groups showed a similar food intake. Figure 4.1.4 Food intake 4 Control 1 - *Fenofibrate ?an PFHS 0 I I I -1 -0 0-1 3-4 5-6 Time (weeks) Food intake (glmouseiday) p<0.05 vs. control 21 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 4.1.5 Plasma ALT Values are absolute values from measurements in pooled plasma samples per group (8 mice per group) at t=0, 4 and 6. roup t=0 t=4 wee t=6 weeks 96 156 1 390 Although no statistical analysis could be performed on pooled ALT levels, PFOS and to a lesser extend PFHS seem to increase plasma ALT levels. Figure 4.1.5 Plasma ALT 1000 +Control -0- Fenofibrate 800 Time (weeks) 22 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 4.1.6 Plasma cholesterol Values are absolute values (mmol/L) and are means 5: 5.0. from 8 mice per group. Individual plasma cholesterol levels are given in appendix IV. Cholesterol (mmoI/L) time (weeks? 0 4 6 Control avg 7.9 8.1 9.0 sd 1.5 1.9 1.1 Fenofibrate avg 7.8 5.0 5.6 sd 1.5 0.6 1.1 PFBS avg 7.9 6.2 6.8 sd 1.5' 0.9 1.4 PFHS avg 8.1 3.2 3.6 sd 1.5 0.7 0.6 PFOS avg 8Control vs: Fenofibrate 1.000 <0.001 PFBS 1. 0.01 PFHS 0.959 <0.001 I Feno?brate vs PFBS 0.959 0.065 . 1 PFOS 0.798 0.002 PFOS 0.878 <0.001 All treatment groups showed a significant reduction in plasma cholesterol at t=4 and 6 weeks of treatment. Fenofibrate decreased plasma cholesterol by -38% at both time points. PFBS, PFHS and PFOS treatment reduced plasma cholesterol levels respectively by 60% and ?63% after 4 weeks of treatment and by -60% and ?64% at 6 weeks of treatment. PFHS and PFOS decreased cholesterol levels to a larger extent than PFBS (p<0.001 at both time points). PFHS and PFOS showed a similar inhibition. 23 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Figure 4.1.6 Plasma cholesterol 14 Control Fenofibrate PF BS PF HS wo- PFOS ?l I 0 Plasma cholesterol (mmollL) Time (weeks) p<0.05 vs. control 4.1.7 Plasma HDL-cholesterol Values are absolute values (mmol/L) and are means SD. from 8 mice per group. Individual plasma HDL-cholesterol levels are given in appendix V. HDL-Cholesterol (mmol/L) time (weeks? 0 4 6 Control avg 0.94 1.36 1.50 sd 0.17 0.21 0.41 Fenofibrate avg 0.96 2.06 2.31 5d 0.19 0.32 0.54 PFBS avg 0.93 1.52 1.46 . sd 0.23 0.24 0.50 PFHS avg 1.05 0.85 0.69 sd 0.32 0.16 0.08 PFOS avg 0.94 0.62 0.44 sd 0.25 0.12 0.13 . Control vs: Fenoflbrate 0.721 0.001 0.007 PFHS 0.328 0.001 <0.001 PFO .9 1 Feno?brate vs PFBS . 0.574 0.003 0.010 PFOS 0.574 <0.001 <0.001 . PFOS 0.959 <0.001 <0.001 24 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Plasma-HDL levels were increased at t=4 and 6 weeks in the fenoflbrate group (respectively by +52% and PFBS treatment showed no significantly different plasma HDL?cholesterol levels as compared to the control group, but both PFHS and PFOS significantly decreased HDL- cholesterol after 4 and 6 weeks of treatment (PFHS respectively by ?38% and PFOS by -54% and Compared to PFBS treatment, PFHS and PFOS also significantly decreased HDL-cholesterol levels. When comparing PFHS and PFOS treatment, PFOS significantly decreased HDL?cholesterol to a larger extent (p=0.028 and p=0.001 at t=4 and 6 weeks respectively). Figure 4.1.7 Plasma HDL-cholesterol 9" ?O?Control 3?5 *Fenofibrate meprBs we~PFHs l? 01 I I Plasma HDL-cholesterol (mmolILTime (weeks) p<0.05 vs. control 4.1.8 Plasma triglycerides Values are absolute values (mmol/L) and are means S.D. from 8 mice per group. Individual plasma triglycerides levels are given in appendix VI. Triglycerides (mmol/L) time (weeks) 0 4 6 E?trol avg 2.24 1.91 1.91 sd 0.75 0.43 1.01 Fenofibrate avg 2.19 0.77 0.52 sd 0.76 0.47 0.29 PFBS avg 2.08 1.08 1.02 sd 0.43 0.42 0.63 PFHS avg 1.97 0.59 0.41 sd 0.81 0.18 0.13 PFOS avg 2.17 0.74 0.52 5d 0.70 0.09 0.12 25 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 4 . . . Fenofibrate 1.000 0.002 0.001 .7 PFHS 0.721 <0.001 <0.001 PFOS 0.798 <0.001 1 Fenofibrate vs PFBS 0.959 0.028 0.015 PFHS 0.878 0.798 0.234 PFOS 1.000 0.195 0.234 vs PFOS 0.959 0.015 0.002 vs All treatment groups showed significant reductions in plasma triglycerides at t=4 and 6 weeks of treatment. Fenofibrate decreased plasma triglycerides by -60% and respectively. PFBS, PFHS and PFOS treatment reduces plasma triglycerides levels respectively by ?69% and ?61% at t=4 weeks of treatment and by -79% and -73% at t=6 weeks of treatment. PFHS and PFOS decreased triglycerides levels to a larger extent than PFBS (respectively p=0.003 and p=0.015 at t=4 weeks and p=0.001 and p=0.002 at t= 6 weeks). When comparing the PFHS and PFOS group, at t=4 weeks a borderline significance was seen (p=0.05) towards lower triglycerides levels in the PFHS group, but at t=6 weeks no significant differences were seen between groups. Figure 4.1.8 Plasma triglycerides 4 *Control ?WFenofibrate New PFHS 3 -o-Pl=os Plasma triglycerides (mmollLTime (weeks) p<0.05 vs. control 26 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 4.1.9 Lipoprotein profiles Lipoproteins were separated on a superose column. Values are absolute values (mM) from cholesterol (left column) and phospholipids (right column) measurements in pooled plasma per group (with 8 mice per group) at t=4 and 6. We consider fractions 4-8 as 9-14 as 13-16 as large-HDL (HDL1) and 17-24 as HDL. After 4 and 6 weeks of treatment, plasma cholesterol levels were decreased in the LDL peak in all groups as compared to the control group. Fenofibrate and PFHS showed the strongest and a similar reduction, PFOS seemed to decrease VLDL to a lesser extend than PFHS. PFBS decreased VLDL-LDL to a lesser extent than PFHS and PFOS. HDL was increased in the fenofibrate group-Whereas PFBS showed no clear effects on HDL, both PFHS and PFOS strongly decreased the HDL peak. PFHS seemed to decrease HDL to a lesser extent than PFOS and in the profile of group HDL1 (large HDL) was increased. The changes in the lipoproteins (apoB?containing lipoproteins (VLDL-LDL) and HDL) are in line with changes in plasma cholesterol and HDL. Figure 4.1.9 Lipoprotein profiles 3.0 1 2 t=4 weeks W?Control t=4 weeks Feno?brate was hisansfraction 3.0 1.2 t=6 weeks ?o?conuol t=6 weeks 2 5 ?*Feno?brala 1 *Fano??brate TEE: 12:Cholesterol (mmoliL) Phospholipids (mmoliL) a! fraction fraction 27 Mechanism of di??erent on lipid and Iipoprotein metabolism in mice TNO project number 031 . 12685 4.1.10 Plasma free glycerol Values are absolute values (mmol/L) and are means S.D. from 8 mice per group. Individual plasma free glycerol levels are given in appendix VII. ree 9 me 0.03 0.03 .20 0.1 0.05 0.03 0.21 0.16 0.03 0.04 0.1 .09 0.03 0.01 0. 5 .09 0.03 0.03 wee Control vs: Fenofibrate 0.878 PFHS 0.721 PFOS . 59 Fenofibrate vs PFBS 0.382 PFOS 1.000 PFOS 0.505 Fenofibrate, PFHS and PFOS showed significant reductions in free glycerol levels at 4 and 6 weeks of treatment (respectively by -51% and -43% at t=4 weeks and by 52% and ~54% at t=6 weeks). PFBS showed a transient decrease at t=4 weeks (by but at t=6 weeks plasma free glycerol levels were not significantly different from the control group. PFHS and PFOS decreased free glycerol levels to a larger extent than PFBS (respectively p=0.001 and p=0.001 at t=4 weeks and p<0.001 and p=0.002 at t=6 weeks). No significant differences were seen between the PFHS and PFOS group. 28 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Figure 4.1.10 Plasma free glycerol 0.5 Control Fenofibrate PFBS 0.4 - mew PFHS -O- PFOS I Plasma free glycerol (mmolIL) Time (weeks) p<0.05 vs. control 4.1.11 Plasma free fatty acids Values are absolute values (mmol/L) and are means :h 5.0. from 8 mice per group. Individual plasma free fatty acid levels are given in appendix Free fatty acids (mmol/L) time (weeks? 0 4 6 Control avg 0.99 0.69 0.74 sd 0.27 0.10 0.07 Fenofibrate avg 0.98 0.72 0.68 sd 0.17 0.11 0.10 PFBS avg 0.90 0.66 0.67 5d 0.14 0.06 0.12 PFHS avg 1.02 0.41 0.39 sd 0.37 0.09 0.06 PFOS avg 1.02 0.44 0.29 sd 0.37 0.06 0.03 . Control vs: Feno?brate 0.721 0.721 0.382 PFBS .721 . 4 0.279 PFHS 0.959 <0.001 <0.001 PFOS 0. 9 .001 <0.001 Feno?brate vs PFBS 0.382 0.382 0.798 0.798 <0.001 <0.001 PFOS 0.798 <0.001 <0.001 PFOS 0.505 <0.001 <0.001 29 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031 . 12685 Free fatty acid levels were similar in the control, fenofibrate and PFBS groups. Both PFHS and PFOS significantly decreased free fatty acid levels after 4 and 6 weeks of treatment (PFHS respectively by ~41% and PFOS respectively by -37% and After 4 weeks of treatment no significant differences were seen between the PFHS or PFOS group but after 6 weeks the PFOS group showed significantly lower FFA levels than the PFHS group Figure 4.1.11 Plasma free fatty acids 1 .4 *O-Control Fenofibrate 1.2 - PFOS O) I I I Plasma free fatty acids (mmoilLTime (weeks) p<0.05 vs. control 4.1.12 Post-heparin LPL and HL activity Values are absolute values (pmol and are means SD. from 8 mice per group. Individual post-heparin LPL and HL activity values are given in appendix XII. Lipolytic activity (umol HL LPL activity activity Control avg 7.4 12.0 sd 0.9 2.5 Fenofibrate avg 12.3 25.1 sd 1.4 3.1 PFBS avg 5.3 14.3 sd 2.1 2.4 PFHS avg 8.5 20.8 sd 3.0 5.1 PFOS avg 9.0 18.5 sd 1.4 2.9 30 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 a r0 I Control vs: Fenofibrate <0.001 <0.001 PFHS 0.798 0.001 21 0. Feno?brate vs PFBS <0.001 <0.001 0. PFOS <0.001 0.001 vs . PFOS 0.003 0.010 VS Post-heparin Iipoprotein lipase activity was strongly induced after fenofibrate treatment (by Both PFHS and PFOS increased LPL activity to a lesser extent (respectively by +74% and Hepatic lipase activity was significantly increased in the feno?brate group and the PFOS group (respectively by +67% and In contrast PFBS significantly decreased hepatic lipase activity by ?28% PFBS had significantly lower levels of LPL and HL activity as compared to PFHS (0.007 and 0.05, respectively) and PFOS (0.010 and 0.003, respectively). Figure 4.1.12 Post heparin LPL and HL activity 30 I Control I Fenofibrate 25 - I PFBS PFHS I PFOS I 10" Lipolytic activity (pmol a HL LPL t=5 weeks p<0.05 vs. control 31 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 4.1.13 Fecal lipids Values are absolute values (umoi/loo gram mouse/day) or relative values of total neutral sterols, phytosterols, bile acids and fatty acids, respectively) and are means SD. from 3 (combined) cages per group at 2 consecutive time points. Values per (combined) cage per time point are given in appendix Fecal lipids Total neutral Total Total Total (umol/IOO gram mouse/day) sterols phytosterols bile acids fatty acids Control avg 36.3 3.5 5.5 257 - sd 10.7 0.7 1.6 222 Fenofibrate avg 44.5 3.4 3.4 410 sd 9.8 0.6 0.8 121 PFBS avg 38PFHS avg 39.1 2.9 3.3 159 sd 5.2 0.2 0.9 128 PFOS avg 34.0 3.4 2.8 188 sd 12.9 0.6 0.3 86 ota sterols osterois bile acids Control vs: Fenofibrate 0.180 0.937 0.015 0.180 PFBS .310 1.000 0.589 0.4 5 PFHS 0.310 0.065 0.004 0.394 Fenofibrate vs PFBS 0.394 0.485 0.093 0.009 PFOS 0.180 0.818 0.240 0.009 vs . PFOS 0.937 0.937 0.002 0.240 vs . .1 Nor fecal neutral sterols (figure 4.1.13a), neither fecal phytosterols (figure 4.1.13b) were affected by fenofibrate, PFBS, PFHS or PFOS treatment. Total bile acid excretion was significantly lower in the fenofibrate, PFHS and PFOS groups (figure 4.1.13c). Fecal bile acids were decreased by -41% and respectively. Although the fenofibrate group showed a somewhat higher fatty acid excretion, none of the treatment groups were significantly different from the control group (figure 4.1.13d). 32 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Figure 4.1.13a Total neutral sterol excretion 70 I Control I Fenofibrate 60 - PFBS ire PFHS PFOS Neutral sterol excretion (pmoll100 gram mouse/day) 0 :3 I .- t=5 weeks p<0.05 vs. control Feno?brate PFHS Fenoflbrate vs PFBS PFOS V5 PFOS VS The fecal neutral sterol composition of the fenofibrate group was somewhat changed, the percentage excreted cholesterol was but significantly increased as compared to the control group (94.8% vs The other treatment groups showed the same composition as the control group. 33 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Figure 4.1.13b Total phytosterol excretion 6 IControl Fenofibrate 5_ (6 IPFOS =>222 - t=5 weeks p<0.05 vs. control Only PFOS treatment affected fecal phytosterol composition as compared to the control groum'PFOS showed an increased stigmasterol (14.8 vs 9.2 and a decreased b- sitosterol and campesterol content (61.0% vs 64.7% and 24.2% vs 26.1%, respectively). 34 Mechanism of di??erent on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Figure 4.1.13c Total bile acid excretion 10 I Control a Fenofibrate a PFBS 8 PFHS PFos a: I Bile acid excretion I (pmoll100 gram mouselday) I - t=5 weeks p<0.05 vs. control Feno?brate vs Fecal bile acid composition was significantly changed after treatment with fenofibrate, PFBS and PFOS. Feno?brate increased the percentage lithochate and hyodeoxycholate/ ursocholate (16.6% vs 10.6 and 16.4% vs respectively) and decreased w- muricholate content (18.5% vs PFBS only increased hyodeoxycholate/ ursochoiate content signi?cantly (12.5% vs while PFOS decreased the percentage ofiithocholate and b?muricholate vs 10.6% and 14.1% vs 25.5% respectively). 35 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Figure 4.1.13d Total fatty acid excretion 600 I Control I Fenofibrate PFBS 500 - tr PFHS I PFOS .b I Fatty acid excretion (pmoll100 gram mouselday) on 100 - t=5 weeks p<0.05 vs. control Fenofibrate 0.093 0.310 0.093 PFHS 0.818 0.065 0.937 1. 1 . 0.093 0.818 0.485 0.589 0.065 0.240 0.589 0.002 Although changes in fatty acid composition were small, PFBS decreased the percentage of (0.62% vs 0.75%) and PFHS increased the percentage of vs significantly. No effects of fenofibrate or PFOS were seen. 36 Mechanism of di??erent on lipid and iipoprotein metabolism in mice TNO project number 031 . 12685 Figure 4.1.13e Fatty acid balance 10000 aControl Fenofibrate II PFBS 8000 '5 I PFOS Q. -. 1- 0 a: :5 6000 - s: 4000 8 8 g- i? a 2000 - . t=5 weeks p<0.05 vs. control Fatty acid balance Fatty acid Symol/lOO gram mouse/day) input-output Control avg 5492 5d 326 Fenofibrate avg 6114 sd 436 PFBS avg 6146 sd 910 PFHS avg 5565 sd 209 PFOS avg 5593 sd 458 atty a in ut ut een . Control vs: Fenofibrate 0.015 PFHS 0.589 .589 Fenofibrate vs PFBS 0.589 .0 PFOS 0.240 VS PFOS 0.310 V5 The fatty acid balance (fatty acid input output) was significantly increased by fenofibrate treatment (by 11% The perfluor alkyl sulfonate compounds were not significantly changed. 37 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 4.1.14 VLDL-triglycerides and de novo ApoB production Values are absolute values (mmol/L for plasma triglycerides, pmoI/h for production rate, 104 dpm/ml/h for de novo ApoB pmol/li)4 for T6 production per de novo ApoB and pmol/lo4 for lipid composition per de novo ApoB) and are means S.D. from 7-8 mice per group. Individual values are given in appendix XIV. Control vs: Fenofibrate Feno?brate vs VS VS production rate (pmol/h) production rate Control ang 6.3 sd 1.4 Fenofibrate Fenoflbrate avg 9.3 5d 1.7 PFHS PFBS avg 6.9 sd 1.5 PFHS avg 1.7 sd 0.7 vs PFOS avg 0.9 5d 0.4 VS At t==0 min (before Triton WR1339 injection) fenofibrate, PFHS, PFOS and to a lesser extent PFBS treatment decreased plasma triglycerides levels (in line with the plasma samples, taken at 4 weeks and 6 weeks of treatment). The increase in plasma triglycerides was more or less the same in the control group and the PFBS group. Fenofibrate increased VLDL- triglycerides to a higher extent than the control group. Both PFHS and PFOS showed a significantly reduced VLDL triglycerides production. In figure 4.1.14b the VLDL?triglycerides production rate (=slope figure 4.1.14a) is shown. Fenofibrate significantly increased the production rate by PFHS and PFOS strongly decrease TG production rate, respectively by -74% and whereas PFBS showed no effects. PFHS and PFOS decreased production rate as compared to PFBS (p<0.001 for both treatments). 38 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Figure 4.1.14a Plasma triglycerides after Triton WR1339 injection 14 Control Fenofibrate 4-- PFBS PFHS -O- PFOS .Jam, 23 2 3 ?w ?w ?my?,mwmm ?aw-man? Plasma triglycerides (mmoIIL100 Time after Triton WR1339 injection (min) p<0.05 vs. control Figure 4.1.14b VLDL-triglycerides production rate 16 II Control 14 . I Feno?brate a PFBs PFHS 12 - I PFOS VLDL-triglycerides production rate (pmollh) p<0.05 vs. control 39 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 De Novo ApoB ApoB (104 dpm/mL/h) Control avg 6.1 sd 1.0 Feno?brate Fenofibrate avg 6.4 .. 5:2: PFBS avg 5.1 . sd 03 Feno?brate vs PFBS PFHS avg 1.5 PFOS sd 0.3 PFOS avg 0.8 PFOS sd 0.2 TS production par de novo TG production per novo ApoB (umol/104 dpm) per ApoB nthesized P-value Control avg 0.80 sd 0.09 vs: Feno?brate Fenofibrate avg 1.12 5d 0.19 PFHS PFBS avg 1.09 . PFOS sd 0.21 Fenoflbrate vs PFBS PFHS avg 0.86 PFOS sd 0.24 PFOS avg 0.86 PFOS sd 0.20 Fenofibrate showed no effects on de novo ApoB rate (figure 4.1.14c). Since fenofibrate increased VLDL?triglycerides production, the newly VLDL- triglycerides per ApoB were plotted in figure 4.1.14d to examine TG content of newly VLDL. Fenofibrate showed triglycerides-rich newly ApoB particles (increase in triglycerides of +40% as compared to the control). PFBS showed a decreased ApoB (by and an increase in triglycerides-rich newly particles (increase in triglycerides of +36% as compared to the control). Both PFHS and PFOS strongly inhibited ApoB (reSpectively by -76% and resulting in newly VLDL particles with similar triglycerides content as the control group. PFHS and PFOS had a significantly reduced de novo ApoB rate and TG production per ApoB as compared to PFBS. 4o Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Figure 4.1.14c De novo ApoB 12 a Control Feno?brate 10 - PFBS is PFHS I PFOS 3 .. (1 o4 0) De novo ApoB rate A I p<0.05 vs. control Figure 4.1.14d TG production per ApoB 2.5 IControl IFeno?brate IPFBS 2-0 germs a IPFOS p<0.05 vs. control 41 Mechanism of di??erent on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Lipid composition VLDL Total Free Cholesterol Triglycerides Phospholipids (umol/loq ApoB) cholesterol cholesterol ester Control avg 0.65 0.18 0.47 0.79 0.23 sd 0.18 0.04 0.14 0.19 0.05 Feno?brate avg 0.36 0.15 0.21 0.99 0.22 50 0.07 0.03 0.05 0.15 0.04 PFBS avg 0.75 0.21 0.54 1.08 0.29 5d 0.55 0.12 0.43 0.20 0.08 PFHS avg 0.46 0.20 0.25 0.80 0.22 0.15 0.06 0.11 0.22 0.04 PFOS avg 1.61 0.49 1.12 1.09 0.39 50 1.15 0.23 0.92 0.18 0.12 ree cholesterol cholesterol ester Fenofibrate 0.001 0.152 0.001 PFHS 0.029 0.613 0.006 . . Feno?brate vs PFBS 0.001 0.130 <0.001 PFOS <0.001 <0.001 <0.001 VS PFOS 0.007 0.003 0.065 VS Lipid composition of isolated VLDL (isolated VLDL is the sum of newly VLDL and VLDL present in plasma at t=0 min), calculated as pmol lipid per newly ApoB, was significantly different from control in all treatment groups. Fenofibrate decreased cholesterolester and increased triglycerides content, respectively by ?55% and PFBS significantly increased triglycerides content by 37%, PFHS showed a decreased cholesterolester composition by PFOS treatment increased free cholesterol, triglycerides and phospholipids content in VLDL, respectively by +164%, +38% and When compared to the PFHS group, the PFOS group showed an isolated VLDL with increased free cholesterol, triglycerides and phospholipids, indicative for larger particles. A compared to PFBS, PFHS induced smaller particles and PFOS larger particles. Figure 4.1.14e Lipid composition VLDL 4 a Control I Fenofibrate a PFBS PFHS I PFOS I Lipid composi ion VLDL per ApoB dpm) Total cholesterol Free cholesterol Cholesterol ester Triglycerides Phospholipids p<0.05 vs. control 42 Mechanism of different on lipid and Iipoprotein metabolism in mice 4.2 Body weight . time (weeks) 0 1 4 Control . avg 26.4 26.7 28.01 sd 1.3 1.1 1.4 Fenofibrate avg 26.1 26.3 27.9 sd 1.2 1.2 1.6 PFBS avg 26.6 26.6 27.8 sd 2.3 2.3 2.3 avg 26.6 26.9 28.2 - - sd 2.4 2.4 2.2 PFOS avg 26.2 26.3 27.2 sd 1.8 1.6 1.6 TNO project number 031 . 12685 Results study 2 4.2.1 Markers of general well-being No specific clinical signs were observed during the study. At sacrifice, besides the livers no macroscopic differences were observed between the groups. Livers were visibly somewhat larger in the fenofibrate group and strongly enlarged in the PFHS and PFOS group. One mouse (mouse 21) showed some white spots in the liver and bile stones in the biadder. 4. 2.2 Body weight Values are absolute values (9) and are means 5.0. from 6 mice per group. Individual values are given in appendix I. me een . . Control vs: Fenofibrate 0.818 0.937 1. PFHS 0.937 0.937 0.818 0.485 Feno?brate vs PFBS 0.937 0.818 PFOS 0.937 0.589 vs . PFOS 0.818 0.699 VS During the study mice gained body weight in all groups. There were no significant differences between groups. 43 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Figure 4.2.2 Body weight 40 a Control I Fenofibrate PFBS PFHS 35 PFOS Body weight Time (weeks) p<0.05 vs. control 4.2.3 Liver and perigonadai fat weight Values are absolute values (9) and are means S.D. from 6 mice per group. Individual values are given in appendix II. Tissue weight (9) Liver Perigonadal fat Control avg 1.43 0.50 sd 0.14 0.17 Fenofibrate avg 1.76 0.58 sd 0.16 0.12 PFBS avg 1.63 0.58 sd 0.28 0.09 PFHS avg 3.01 0.40 sd 0.35 0.11 PFOS avg 2.97 0.41 sd 0.26 0.12 SSUG WE ti een ro Control vs: Fenofibrate 0.009 . 1 PFHS 0.002 PFOS 2 Feno?brate vs PFBS 0.394 PFHS PFOS 0.002 PFOS 0.002 1. 44 Mechanism of different PFASs on lipid and iipoprotein metaboiism in mice TNO project number 031.12685 Liver weights were significantly increased in the fenofibrate, PFHS and PFOS group. After 4 weeks of treatment PFBS treatment did not result in significantly increased livers. Fenofibrate increased liver weights by The PFHS and PFOS groups showed a robust induction in liver weights, respectively by +110% and +107%. Although a slight decrease in perigonadal fat weight was seen after PFOS and PFHS treatment, this was not significantly different from the control group. Liver weights in the PFHS and PFOS groups were significantly higher as compared to the PFBS group, whereas weights of perigonadal fat were significantly decreased in the PFHS and PFOS groups. Figure 4.2.3a Liver weight 5 Control I Fenofibrate 4 PFBS PFHS ?p<0.05 vs. control Figure 4.2.3b Perigonadal fat weight 1.2 lControl I Fe nofibrate 1-0 IIPFBS ta PFHS g, I PFOS 'week 4 p<0.05 vs. control 45 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031 . 12685 4.2.4 Food intake Values are absolute values (g/day/mouse) and are means SD. or range from 2-3 cages per group. Individual values per cage are given in appendix Food intake (g/day/mouse) time (weeks) -1-0 0-1 3-4 Control avg 2.8 2.8 . 2.8 sd/range 0.1 0.1 0.1 Fenofibrate avg 2.8 2.9 2.8 sd/range 0.1 0.1 0.1 PFBS avg 2.8 2.7 2.7 sd/range 0.1 0.2 0.2 PFHS avg 2.8 2.7 2.8 sd 0.1 0.4 0.4 PFOS avg 2.8 2.7 2.7? sd 0.1 0.2 0.2 Food intake: P-value time weeks 0-1 3-4 Between 0.865 0.920 i vs: Fenofibrate 1.000 0.667 PFHS 0.800 0.800 PFOS 0.400 .800 Fenofibrate vs PFBS 0.667 0.667 PFOS 0.800 0.800 PFBS vs 1. . PFOS 1.000 0.800 PFHS vs PFOS 1.000 1.000 All groups showed a similar food intake during the study. Figure 4.2.4 Food intake 5 ?Control a Fenofibrate i? a PFBS in 4 - 3 PFHS I PFOS -1-0 0-1 3-4 Time (weeks) 46 Mechanism of di?erent on lipid and iipoprotein metabolism in mice TNO project number 031.12685 4.2.5 Plasma ALT Values are absolute values from measurements in pooled plasma samples per group (6 mice per groupt=0 weeks the control group showed a higher ALT level as compared to the other groups. After 4 weeks of treatment, the control group still had the highest plasma ALT level, but PFHS and PFOS showed, with respect to t=0 levels, a higher increase in plasma ALT. Figure 4.2.5 Plasma ALT 400 IControl I Fenofibrate I PFBS PFHS 300 I PFOS Time (weeks) 4.2.6 Plasma cholesterol Values are absolute values (mmol/L) and are means d: S.D. from 6 mice per group. Individual values are given in appendix IV. Cholesterol (mmol/L) time (weeks) 0 4 Control avg 8.4 7.6 sd 1.3 1.2 Feno?brate avg 8.4 4.9 sd 2.9 0.5 PFBS avg 8.7 6.3 sd 1.9 1.7 PFHS avg 8.5 2.5 sd 2.7 0.7 PFOS avg 8.6 3.0 5d 2.3 0.7 47 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 P-va een rou Control vs: Fenofibrate PFBS PFHS PFOS Fenofibrate vs PFBS PFOS VS PFOS VS As was seen in study 1, all treatment groups showed a significant reduction in plasma cholesterol at t=4 weeks of treatment. Fenofibrate decreased plasma cholesterol by PFBS, PFHS and PFOS treatment reduced plasma cholesterol levels respectively by -67% and ?60% after 4 weeks of treatment. PFHS and PFOS decreased cholesterol levels to a larger extent than PFBS PFHS and PFOS showed a similar inhibition. Figure 4.2.6 Plasma cholesterol 12 i Control 5 Fe nofibrate 10 - a PFBS 223:: PFHS a PFOS Plasma cholesterol (mmolIL) Time (weeks) p<0.05 vs. control 48 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 4.2.7 Plasma HDL-cholesterol Values are absolute values (mmol/L) and are means :t 5.0. from 6 mice per group. Individual values are given in appendix V. HDL-Cholesterol (mmol/L) time (weeks) 0 4 Control avg 0.96 1.22 sd 0.47 0.29 Fenofibrate avg 1.12- 2.39 sd 0.68 0.24 PFBS avg 1.19 1.72 sd 0.53 .. 0.37 PFHS avg 1.03 0.80 sd 0.37 0.12 PFOS avg 1.07 0.58 sd 0.43 0.10 me wee . Control vs: Fenofibrate 0.937 1 PF HS 0.937 PFOS 0.937 Fenofibrate vs PFBS 0.937 PFOS 0.937 PFOS 1.000 HDL-cholesterol levels were increased after 4 weeks treatment in the fenofibrate group, by In contrast to experiment 1, PFBS treatment showed a significantly increased plasma HDL-cholesterol, by Both PFHS and PFOS significantly decreased HDL?cholesterol after 4 weeks of treatment, by -34% and respectively. When comparing PFHS and PFOS treatment, PFOS significantly decreased HDL-cholesterol to a larger extent (p20.009). PFHS and PFOS significantly decreased HDL as compared to PFBS. 49 Mechanism of different on lipid and ifpoprotein metabolism in mice TNO project number 031 . 12685 Figure 4.2.7 Plasma HDL-cholesterol 4 I Control a Fenofibrate PFBS 3 8% PFHS PFOS Plasma HDL-cholesterol (mmollL) Time (weeks) p<0.05 vs. control 4.2.8 Plasma triglycerides Values are absolute values (mmol/L) and are means i: S.D. from 6 mice per group. Individual values are given in appendix VI. Triglycerides (mmoI/L) time (weeks) 0 4 Control avg 2.26 1.37'" sd 0.71 0.21 Fenofibrate avg 2.27 0.48 sd 0.79 0.12 PFBS avg 2.54 0.87 sd 1.02 0.25 PFHS avg 2.25 0.56 sd 0.96 0.17 PFOS avg 2.39 0.68 sd 0.76 0.21 rou . Control vs: Fenofibrate 1.000 PFHS 1.000 PFOS 0.699 F?nofibrate vs PFBS 0:818 PFHS 0.937 PFOS 0.818 PFOS 0.937 50 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 All treatment groups showed significant reductions in plasma triglycerides at t=4 weeks of treatment. Fenofibrate decreased plasma triglycerides by PFBS, PFHS and PFOS treatment reduced plasma triglycerides levels respectively by -59% and -50% after 4 weeks of treatment. PFHS decreased triglycerides levels to a larger extent than PFBS PFOS showed no significantly different triglycerides levels as compared with PFHS and PFBS. Figure 4.2.8 Plasma triglycerides 4 I Control I Fenofibrate a PFBS PFHS I PFOS Plasma triglycerides (mmolIL) Time (weeks) p<0.05 vs. control 4.2.9 Lipoprotein profiles Lipoproteins were separated on a superose column. Values are absolute values (mM) from cholesterol (left column) and phospholipids (right column) measurements in pooled plasma per group (with 6 mice per group) at t=4. We consider fractions 4-8 as 9?14 as 13-16 as large-HDL (HDL1) and 17-24 as HDL. The profiles are more or less the same as in experiment 1. After 4 weeks of treatment, plasma cholesterol levels were decreased in the VLDL-LDL peak in all groups as compared to the control group. Feno?brate and PFHS showed the strongest and a similar reduction, PFOS seemed to decrease VLDL to a lesser extent than PFHS. PFBS decreased VLDL to a lesser extend than PFHS and PFOS. HDL was increased in the fenofibrate group, with the appearance of a large HDL1 particle. Whereas PFBS seemed to increase HDL somewhat, both PFHS and PFOS strongly decreased the HDL peak. PFHS seemed to decrease a lesser extent than PFOS and in the profile of the PFHS group HDL1 (large HDL) was increased. The changes in the lipoproteins (ApoB-containing proteins (VLDL-LDL) and HDL) are in line with the changes in plasma cholesterol and HDL. 51 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Figure 4.2.9 Lipoprotein profiles 3.0 1.2 t=4 weeks ~4?Control t=4 weeks 2 5 ?Os-Fenofibrate 1 0 mam-Feno?brate .. 12:fraction fraction 4.2.10 Biliary bile acids, cholesterol and phospholipids production Values are absolute values (uL/min/kg mouse for bile flow, nmol/min/kg mouse for biliary bile acids, phospholipids and cholesterol) and are means SD. from 5?7 mice per group. Individual values are given in appendix XV. Bile flow (pl/min/kg mouse) time 15 min 30 min 45 min total Control avg 43.2 38.7 41.3 40.3 sd 7.8 7.1 7.1 6.8 Fenofibrate avg 50.5 47.4 45.2 47.7 sd 5.5 6.6 9.5 4.8 PFBS avg 43.7 38.8 41.2 41.2 5d 3.6 12.2 13.8 9.4 PFHS avg 63.9 58.0 54.9 58.9 sd 13.6 8.9 11.0 10.4 PFOS avg 51.3 40.8 39.9 44.0 sd 15.3 19.7 16.6 17.0 me on - ro . Control vs: Fenofibrate 0.329 0.394 - .4 PFHS 0.017 0.093 PFOS .343 . 0.6 Fenofibrate v5 PFBS 0.126 0.429 PFHS 6 . PFOS 1.000 0.445 vs PFOS 0.530 0.755 VS Although bile flow was somewhat increased in the fenofibrate group at all collection points, this was not significantly different from the control group. However, fenofibrate did significantly increase the resulting total bile flow (by 18%, 52 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 PFHS treatment increased bile flow, this was significantly different from the control group for bile collected 0?15 min and 15-30 min after the cannula was placed (by +48 and respectively). The resulting total bile flow was also significantly increased in the PFHS group (by PFBS and PFOS showed no effects on bile flow. Figure 4.2.103 Bile flow 120 IControl I Fenofibrate Tn? 100 - IPFBS a: PFHS IPFOS 0-15 min 15-30 min 30-45 min total Time bile collection p<0.05 vs. control Bile acid flow time (nmol/min/kg mouse) 15 min 30 min 45 min total Control avg 614 539 615 573 sd 262 119 189 179 Feno?brate avg 468 436 469 457 sd 43 48 140 60 PFBS avg 610 500 603 571 sd 137 173 255 156 PFHS avg 625 535 514 558 sd 123 96 101 85 PFOS avg 543 459 483 495 sd 100 130 201 136 Feno?brate PFHS PFOS Feno?brate vs PFBS PFOS PFOS 53 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Although fenofibrate and to a lesser extent PFHS and PFOS seemed to decrease bile acid flow, this was not significantly different from the control group. PFBS showed the same bile acid flow as the control group. Figure 4.2.10b Bile acid flow 1200 lControl a Fenofibrate 1000 - .l PFHS 52? I PFOS 800 - .0-15 min 15-30 min 30-45 min total Time bile collection p<0.05 vs. control Phospholipid ?ow time Elmol/min/kg mouse) 15 min 30 min 45 min total Control avg 330 310 400 346 139 79 139 105 Feno?brate avg 451 467 457 459 ed 132 123 180 136 PFBS avg 393 372 448 405 sd 69 109 192 93 PFHS avg 620 553 576 583 sd 166 153 148 144 PFOS avg 463 422 443 443 5d 214 176 181 178 osp Between Control vs: Fenofibrate PFHS Fenofibrate vs PFBS PFOS PFOS 54 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Although biliary phospholipid flow was somewhat increased in the fenofibrate group at all collection points, this was only different from the control group at t=15-30 min (by +51% The calculated total phospholipid flow, however, was not significantly changed. PFHS treatment increased phospholipid flow, this was significantly different from the control group for bile collected 0-15 min and 15-30 min after the cannula was placed (by +88 and +78% respectively. The resulting total phospholipid flow was also significantly increased in the PFHS group (by PFBS and PFOS showed no effects on biliary phospholipid flow. Figure 4.2.10c Phospholipid flow 1200 IControl ?Fenofibrate 1000 - EPFBS 3 IPFOS 800 - 1.1 0-15 min 15-30 min 30-45 min total Time bile collection p<0.05 vs. control Choiesterol time (nmol/min/kg mouse) 15 min 30 min 45 min total Control avg 33.6 33.0 38.1 35.4 sd 15.7 16.1 13.7 13.0 Feno?brate avg 34.3 38.3 37.8 36.8 sd 10.8 9.9 11.6 9.0 PFBS avg 32.5 29.4 38.8 33.6 sd 9.8 13.7 14.5 10.4 PFHS avg 41.8 41.6 39.6 41.0 sd 7.9 11.5 4.4 6.6 PFOS avg 44.9 42.4 41.2 42.8 sd 15.1 21.3 12.7 14.4 55 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 co - min - 5 Between 0.284 0.386 0.984 0.398 Control vs: Fenofibrate 0.792 0.429 1.000 0.589 . 1 00 . PFHS 0.247 0.177 0.818 0.240 0.2 0.432 . 31 0.295 Feno?brate vs PFBS 0.931 0.126 1.000 0.429 PFOS 0.234 0.945 0.628 0.295 PFOS 0.202 0.3.43 0.755 0.268 5 For biliary cholesterol flow, no significant'changes were seen between groups. Figure 4.2. 10d Cholesterol flow 100 wControl I Fenofibrate ?3 80 3 IPFOS 0-15 min 15-30 min 30-45 min total Time bile collection p<0.05 vs. control 56 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 4.2.11 In vivo clearance of TG-rich particles and uptake in tissues Values are relative values of injected dose for plasma decay and uptake in different tissues) and absolute values (min for half life) and are means SD. from 4-6 mice per group. Individual values are given in appendix XVI. decay In plasma ofI dose H]?trlolein decay in plasma: P-value Feno?brate Fenofibrate vs VS VS half-life Half-life -triolein half-life: P-value Half-life (min) Control avg 21.0 rou 5d 6.2 vs: Feno?brate Fenofibrate avg 6.8 sd 0.9 PFHS PFBS avg 10.2 5d 3 3 no?brate vs PFHS avg 8.1 sd 1.3 PFOS avg 10.4 5d 1.9 The clearance of VLDL?like [3H]-triolein labeled particles in plasma (as a measure for TG clearance) was significantly increased in all treatment groups (figure 4.2.11a). Fenofibrate showed the strongest clearance, followed by PFHS, whereas PFOS and PFBS showed a comparable clearance. In figure 4.2.11b the [3H]?triolein half-life (=1/slope figure 4.2.11a) is shown. Fenofibrate significantly decreased TG half life by PFBS and PFOS both reduced [3H]-trioleln half? life by -51% and PFHS showed a decreased of which was a significantly stronger decrease as the PFOS group. The half-life of did not significantly differ between PFBS on the one hand and PFHS and PFOS on the other hand. 57 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Figure 4.2.11a [3H]-triolein decay in plasma 100 WControl - WFenofibrate Time after injection (min) p<0.05 vs. control Figure 4.2.11b [3H1-triolein half-life 40 [Control I Fenofibrate ?ii? PFHS 30 IPFOS half-life (min) p<0.05 vs. control Total liver uptake was increased in all treatment groups. Fenofibrate, PFHS and PFOS showed a significant induction of respectively +149% and PFBS showed a trend towards a higher uptake 58 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 PFHS showed a significantly decreased [3H]-triolein uptake in the heart and the spleen (by - 52% and -53% respectively), PFOS increased uptake in gonadal white adipose tissue by 164%) and fenofibrate showed a higher uptake in muscle and gonadal fat (by 65% and 162% respectively). [3H]-triolein tissue uptake liver heart spleen muscle of injected dose) Control avg 8.7 1.99 0.77 4.82 5d 1.8 0.61 0.18 1.66 0.21 Fenofibrate avg 17.3 1.26 0.53 7.96 2.04 Sd 3.7 0.45 0.17 1.79 0.87 PFBS avg 13.4 1.81 0.57 6.38 1.45 sd 2.6 0.98 0.14 2.16 1.79 PFHS avg 21.8 0.96 0.36 6.97 1.82 5d 4.2 0.37 0.13 1.22 1.41 IPFOS avg 16.1 1.46 0.47 5.37 2.05 Sd 3.6 0.75 0.21 1.43 1.77 ?tri0lein tissue upta P-value heart spleen muscle I vs: Fenoflbrate 0.151 .095 0.032 PFHS 0.017 0.009 0.052 .24 Fenoflbrate vs PFBS 0.413 0.730 0.413 PFOS 0.662 0.009 0.004 PFOS 0.762 0.610 0.914 Figure 4.2.11c Total liver [3H]-triolein uptake 40 IControl lFenofibrate CD if, grams A 30 IPFOS p<0.05 vs. control 59 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Figure 4.2.11d Total tissue [3H]-triolein uptake 12 I Control I Feno?brate 10 - I PFBS 33 PFHS I PFOS Total tissue [3H]-triolein uptake of injected dose) 0} heart spleen muscle p<0.05 vs. control The clearance of VLDL?like [14CJ-cholesteryl oleate particles (as a measure for VLDL- choiesterolester clearance) was significantly increased in the fenofibrate group and the PFBS and PFHS groups (figure 4.2.11e). PFOS showed a similar decay as the control group. Fenofibrate showed the strongest increase, whereas PFBS and PFHS showed a comparable decay. In figure 4.2.11f the [14C]-cholestery oleate half-life (=1/slope figure 4.2.11e) is shown. Fenofibrate significantly decreased CE half life by PFBS and PFHS reduced half-life by -60% and ~54% resoectively. PFOS was not significantly different from the control group and the PFBS group. [?C1-cholesteryl oleate decay in ma of dose 60 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031 . 12685 oleate decay P-value Control vs: Feno?brate vs oleate haif?life Half-life half Lmin) P?value Control avg 555 25.2 vs: Feno?brate Feno?brate avg 1 1.0 5d 1.7 PFHS 22. FBS avg 5 Feno?brate vs PFBS sd 6.1 PFHS avg 26_3 PFQS 5d 6.7 sd 84.2 Figure 4.2.11e [14C1-cholesteryl oleate decay 100 .92. I 8 '5 . - *Control 9 - +Fenofibrate ween-Time after injection (min) p<0.05 vs. control 61 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Figure 4.2.11f oleate half-life 200 I Control I Fenofibrate - II PFBS is; PFHS 15? PFOS 100 - 50- [14C]-cholesteryl oleate half-life (min) o- p<0.05 vs. control Total liver oleate was increased in all treatment groups, except PFOS. Fenofibrate, PFBS and PFHS showed a significant induction of respectively +142%, +74% and PFHS showed a signi?cantly decreased ['14C1-cholesteryl oleate uptake in the spleen (by - other tissues were not significantly affected by any treatment. oleate tissue liver heart spleen muscle uptake of injected dese) Control avg 26.3 1.40 1.04 1.02 0.02 sd 8.1 0.32 0.23 1.42 0.03 Fenofibrate avg 63.8 1.45 0.79 1.95 0.39 sd 9.7 0.64 0.20 0.59 0.17 PFBS avg 45.9 1.50 0.88 1.70 0.18 sd 3.5 1.04 0.27 1.56 0.30 PFHS avg 47.1 1.05 0.50 1.91 0.33 sd 10.9 0.45 0.19 1.20 0.32 PFOS avg 19.1 0.96 0.63 0.55 0.41 sd 5.0 0.37 0.30 0.70 0.64 ]-cho estery oleate tissue liver rt spleen muscle ke: P-val . ntrol vs: Fenoflbrate 0.008 0.310 0.421 . . 3 PFHS 0.017 0.009 0.429 0.126 . 0.126 0. PFBS 0.016 0.556 0.413 PFOS 0.004 0.537 0.017 PFOS 0.010 0.257 0.114 62 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Figure 4.2.119 Total liver [14C1-cholesteryl oleate uptake 100 I Control 8 Fenofibrate a PFBS 30 PFHS I PFOS a: I Total liver oleate uptake of injected close) p<0.05 vs. control Figure 4.2.11h Total tissue [14C1-cholesteryl oleate uptake 6 Control I Fenofibrate 5 .. a PFBS if; PFHS I PFOS 00 I I Total tissue oleate uptake of injected dose) heart spleen muscle p<0.05 vs. control 63 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031 . 12685 4.3 Results study 3 4.3.1 Markers of general well-being No specific clinical signs were observed during the study. At sacrifice, besides the livers no macroscopic differences were observed between the groups. Livers were visibly somewhat increased in the fenofibrate group and largely increased in the PFHS and PFOS group. One mouse (mouse 23) showed a granular liver. 4.3.2 Body weight Values are absolute values (9) and are means SD. from 6?7 mice per group. Individual values are given in appendix 1. Body weight (9) time (weeks) 0 1 4 Control avg 27% 27.8 29.8 5d 1.3 1.3 1.6 Fenofibrate avg 27.0 27.2 29.0 sd 2.9 2.9 2.7 PFBS avg 27.6 27.9 30.1 sd 2.0 2.3 2.7 PFHS avg 27.4 27.3 28.7 sd 1.1 0.8 0.9 PFOS avg 27.1 27.4 28Control vs: Fenofibrate 0.383 PFHS 0.456 0. Fenofibrate vs PFBS 0.456 PFOS 0.456 PFOS 0.902 During the study mice gained body weight in all groups. In contrast to study 1 and 2, the PFOS group showed a small but significantly decreased body weight after 4 weeks of treatment 64 Mechanism of different on iipid and iipoprotein metabolism in mice TNO project number 031.12685 Figure 4.3.2 Body weight 40 Control a Feno?brate a PFBS 35 PFHS I PFOS Body weight on I 20- Time (weeks) p<0.05 vs. control 4.3.3 Liver and perigonadal fat weight Values are absolute values (9) and are means S.D. from 6 mice per group. Individual values are given in appendix II. Tissue weight (9) Liver Perigonadal fat Control avg 1.61 0.56 sd 0.13 0.16 Feno?brate avg 1.94 0.50 sd 0.16 0.14 PFBS avg 1.72 0.52 5d 0.20 0.18 PFHS avg 2.93 0.36 sd 0.52 0.05 PFOS avg 2.95 0.39 sd] 0.27 0.13 fat Control vs: Fenofibrate 0.485 PFHS 0.009 Feno?brate vs PFBS 0.937 PFOS 0.132 vs . PFOS 0.180 vs Livers were significantly increased in the fenofibrate, PFHS and PFOS group. Fenofibrate increased liver weights by The PFHS and PFOS groups showed a robust induction in liver weights, respectively by +82% and Liver weight in the PFHS and PFOS groups was significantly higher as compared to the PFBS group. 65 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Although a decrease in perigonadal fat weight was seen after PFOS and PFHS treatment, this was only significantly different in the PFHS group (by Figure 4.3.3a Liver weight 5 I Control I Fenofibrate 4 . . a PFBS 3e PFHS I PFOS Liver weight week 4 p<0.05 vs. control Figure 4.3.3b Perigonadal fat weight 1.2 ?Control a Fenofibrate 1.0 - CD IPFOS ?5 0.8 - 3 .272 0.0 p<0.05 vs. control 66 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 4.3.4 Food intake Values are absolute values (g/day/mouse) and are means 5.0. from 3 cages per group. Individual values per cage are given in appendix Food intake (Q/day/mouse) time (weeks) -1?0 0-1 3-4 Control avg 3.1 2.9 2.8 sd 0.3 0.1 0.2 Fenofibrate avg 3.1 2.7 2.8 sd 0.3 0.4 0.3 PFBS avg 3.1 2.9 2.9 sd 0.3 0.3 0.3 PFHS avg 3.1 2.8 2.8 0.3 0.3 0.3 PFOS avg 3.1 2.8 2.6 5d 0.3 0.3 0.2 Food intake: P-value time weeks een Control vs: Fenofibrate 1.000 1.000 1 .00 1 PFHS 0.700 1.000 PFOS Fenofibrate VS PFBS 0.700 1.000 PFHS 1.000 1.000 PFOS 1.000 0.400 PFOS 0.700 0.200 VS All groups showed a similar food intake during the study. Figure 4.3.4 Food intake 5 Control I Fenofibrate I PFBS 4 PFHS I PFOS Food intake (gldaylmouse) -1 -0 0-1 3-4 Time (weeks) p<0.05 vs. control 67 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 4.3.5 Plasma ALT Values are absolute values from measurements in pooled plasma samples per group (6?7 mice per groupAfter 4 weeks from treatment'PFHS and PFOS showed higher ALT levels than the control group. In contrast to studies 1 and 2 PFBS treatment also seem to increase plasma ALT. Figure 4.3.5 Plasma ALT 500 IControl i Fenofibrate 400 . PFBS .-. PFHS IPFOS Time (weeks) 4.3.6 Plasma cholesterol Values are absolute values (mmol/L) and are means 5.0. from 5-7 mice per group. Individual values are given in appendix IV. Cholesterol (mmol/L) time (weeks) 0 4 Control avg 7.3 7.9 sd 1.0 0.8 Fenofibrate avg 7.7 5.1 sd 1.2 0.1 PFBS avg 7.6 6.1 sd 1.9 1.1 PFHS avg 7.5 2.4 sd 1.7 0.5 PFOS avg 7.7 2.6 5d 1.4 0.6 68 Mechanism of different on lipid and lipoprofein metabolism in mice TNO project number 031.12685 een . . Control vs: Fenofibrate 0.535 0.002 PFHS 0.535 0.002 Fenofibrate vs PFBS 1.000 0.015 PFOS 1.000 0.002 VS . PFOS 0.710 0.002 VS As was seen in experiments 1 and 2,'al treatment groups showed a signi?cant reduction in plasma cholesterol at t=4 weeks of treatment. Fenofibrate decreased plasma cholesterol by PFBS, PFHS and PFOS treatment reduced plasma cholesterol levels respectively by -69% and -67% after 4 weeks of treatment. PFHS and PFOS decreased cholesterol levels to a larger extent than PFBS PFHS and PFOS showed a similar inhibition. Figure 4.3.6 Plasma cholesterol 12 I Control I Fenofibrate 10 I PFBS a PFHS I PFOS Plasma cholesterol (mmollL) Time (weeks) p<0.05 vs. control 4.3.7 Plasma HDL?cholesterol Values are absolute values (mmol/L) and are means S.D. from 6-7 mice per group. Individual values are given in appendix V. 69 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 HDL-Cholesterol (mmol/L) time (weeks) 0 4 Control avg 1.19 1.07"" sd 0.26 0.24 Fenofibrate avg 1.20 1.89 sd 0.40 0.19 PFBS avg 1.34 1.19 sd 0.66 0.53 PFHS avg 1.07 0.40 5d 0.41 0.20 PFOS avg 1.21 0.28 sd 0.33 0.13? me Fenofibrate 0.902 PFHS 0.620 PFOS .0 Fenofibrate vs PFBS 1.000 PFOS 1.000 VS PFOS 1.000 VS HDL-cholesterol levels were increased after 4 weeks treatment in the feno?brate group, by Both PFHS and PFOS significantly decreased HDL?cholesterol after 4 weeks of treatment, by ?62% and 44%, respectively. PFBS treatment did not result in significantly altered HDL?cholesterol levels. When compared to PFBS, PFHS and PFOS treatment showed a decreased plasma HDL- cholesterol, when comparing PFHS and PFOS treatment, no significant changes were seen. Figure 4.3.7 Plasma HDL-cholesterol 2.5 I Control I Fenofibrate a PFBS 2-0 PFHS I PFOS 1.5- 1.0- 0.5 - Plasma HDL-cholesterol (mmolIL) 0.0 '1 Time (weeks) p<0.05 vs. control 70 Mechanism of di??erent on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 4.3.8 Plasma triglycerides Values are absolute values (mmol/L) and are means 5.0. from 6-7 mice per group. Individual values are given in appendix VI. Triglycerides (mmol/ L) time (weeks) 0 4 Control avg 1.617. 1.74 sd 0.39 0.42 Fenofibrate avg 1.70 0.38 sd 0.67 0.05 PFBS avg 1.74 0.95 sd 0.82 0.32 PFHS avg 1.81 0.54 5d 0.61 0.29 PFOS avg 1.31 0.49 5d 0.46 0.14 ro . Control vs: Fenofibrate 0.710 0. PFHS 0.620 1.000 Fenofibrate vs PFBS 0.710 PFOS 0.710 PFOS 0.902 All treatment groups showed significant reductions in plasma triglycerides at t=4 weeks of treatment. Fenofibrate decreased plasma triglycerides by PFBS, PFHS and PFOS treatment reduced plasma triglycerides levels respectively by -69% and ?72% after 4 weeks of treatment. Both PFHS and PFOS decreased triglycerides levels to a larger extent than PFBS PFOS showed no significantly different triglycerides levels as compared to PFHS. 71 Cholesterol (mmollL) Mechanism of di??erent on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Figure 4.3.8 Plasma triglycerides 4 Control I Fenofibrate PFBS 3 laziPFl-lS I PFOS Plasma triglycerides (mmollL) Time (weeks) p<0.05 vs. control 4.3.9 Lipoprotein profiles Lipoproteins were separated on a superose column. Values are absolute values (mM) from cholesterol (left column) and phospholipids (right column) measurements in pooled plasma per group (with 6- 7 mice per group) at t=4 We consider fractions large? (HDL1) and 17- 24 as HDL. The profiles are more or less the same as in experiment 1 and 2. After 4 weeks of treatment, plasma cholesterol levels were decreased in the VLDL-LDL peak in all groups as compared to the control group. Fenofibrate showed the strongest reduction, PFOS seemed to decrease VLDL to a lesser extent than PFHS. PFBS decreased VLDL-LDL to a lesser extend than PFHS and PFOS. HDL was increased in the fenofibrate group, with formation of a HDL1 particle. Whereas PFBS showed no clear effects on HDL, both PFHS and PFOS strongly decreased the HDL peak. PFHS seemed to decrease HDL to a lesser extend than PFOS. In contrary to the profiles in experiment 1 and 2 HDL1 (large HDL) was not increased in the PFHS group. The changes in the lipoprotein profiles are in line with the changes in plasma lipids. Figure 4.3.9 Lipoprotein profiles 1.6 0.8 t=4 WESRS ?On-Control t=4 Weeks +Control 1_4 - mFeno?brate ?MFenoflbrate A 1-2 ?mm-is 0.6 - +PFos 1fraction fraction Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 4.3.10 Plasma ApoA1 Values are absolute values (mg/mL) and are means :t S.D. from 6?7 mice per group. Individual values are given in appendix IX. ApoAl time weeks) 0 4 Control avg 1.7? 2.28 5d 0.40 0.50 Fenofibrate avg 1.73 2.65 sd 0.70 0.23 PFBS avg 1.94 2.30 sd 0.82 1.00 PFHS avg 1.54 0.54 sd 0.60 0.26 PFOS avg 1.73 0.43 sd 0.30 0.08 -va ro Control vs: Fenofibrate 0.710 1.0 PFHS 0.620 0.80 Fenofibrate vs PFBS 0.902 PFOS 0.710 VS PFOS 1.000 VS Whereas PFHS and PFOS stongly inhibited plasma ApoAl (by -76% and v81% respectively, PFBS and fenofibrate showed not significant effects on ApoAl levels. Figure 4.3.10 Plasma ApoA1 4 I Control I Fenofibrate I PFBS PFHS 3 I PFOS Plasma ApoA1 Time (weeks) p<0.05 vs. control 73 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 4.3.11 Plasma CETP mass Values are absolute values (pg/mL) and are means i S.D. from 5-7 mice per group. Individual values are given in appendix X. WP mass (pg/mL) time _(weeks) 0 4 Control avg 13.7 14.7 sd 1.6 1.4 Fenofibrate avg 13.1 9.1 2.5 0.6 PFBS avg 12.8 11.8 sd 2.7 1.1 PFHS avg 13.8 9.4 sd 3.7 1.9 PFOS avg 12.6 9.1 5d 1.8 1.3 mass: P- me rou . . Control vs: Fenofibrate 0.710 0.004 PFHS 1.000 0.004 PFOS 0.383 0.002 Fenofibrate vs PFBS 0.902 0.004 0. 2 PFOS 0.805 0.662 PFOS 1.000 0.015 Plasma CETP levels were significantly decreased in all treatment groups. Fenofibrate decreased plasma CETP concentration by PFBS, PFHS and PFOS treatment reduced plasma CETP levels respectively by -36% and -38% after 4 weeks of treatment. PFHS and PFOS decreased levels to a larger extent than PFBS (p=0.052 and 0.015 respectively). PFOS and PFHS showed similar CETP levels. 74 Mechanism of o'i??erent on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Figure 4.3.11 Plasma CETP mass 20 a Control I Fenofibrate I PFBS ?35 PFHS PFOS Plasma CETP mass (uglmL) _Time (weeks) p<0.05 vs. control 4.3.12 Plasma CETP activity Values are absolute values (pmol/h) and are means i 5.0. from 4-7 mice per group. Individual values are given in appendix XI. wee Control vs: Fenofibrate PFHS PFOS Fenofibrate vs PFBS PFOS PFOS 75 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031 . 12685 CETP activity levels were significantly decreased after fenofibrate and PFOS treatment. Fenofibrate decreased plasma CETP activity by Although PFBS, PFHS and PFOS treatment reduced plasma CETP activity, this was not significantly different in the PFBS grOUp A trend toward lowered CETP activity was seen in the PFHS group (by PFOS decreased CETP activity by Figure 4.3.12 Plasma CETP activity 350 [Control A I Fenofibrate 300 i names 0 a PFHS 250 - I PFOS Time (weeks) p<0.05 vs. control 4.3.13 In vivo clearance of autologous HDL Values are relative values of injected dose for plasma decay of labeled autologous HDL) and absolute values (pools for fractional catabolic rate of HDL and mM for catabolic rate of HDL) and are means i 5.0. from 5 mice per group. Individual values are given in appendix XVII. ]-cholesl:ery oieyl ether decay 76 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 ]?cholesteryl oleyl ether decay in sma: P-value rou VS: Feno?brate vs VS VS Fractional Catabolic Rate HDL FCR (pools Control avg 0.21 sd 0.02 Feno?brate Fenofibrate avg 0.16 sd 0.01 PFHS PFBS avg 0'24 Feno?brate vs PFBS sd 0.03 PFHS avg 0.32 PFOS sd 0.04 PFOS avg 0.33 PFOS sd 0.03 Catabolic rate HDL 5R (mM Control avg 0.23 5d 0_04 Feno?brate 0.056 Fenofibrate avg 0.31 5.. 0.05 PFHS 3-33; PFBS avg 0.25 . sd 0'09 Fenofibrate vs PFBS 0.222 PFHS aV9 0'12 PFOS 0.008 sd 0.06 vs PFOS avg 0-08 PFOS 0.008 sd 0.04 vs The clearance of HDL (measured with a trace of autologous labeled [3H]?cholestery oleyl ether) in plasma was significantly increased in the PFHS and the PFOS groups. Fenofibrate decreased HDL clearance. PFBS showed an increased HDL clearance at two time points (4 and 24 h, figure 4.3.13a). In figure 4.3.13b the fractional catabolic rate of HDL (=slope 0-8 figure 4.3.13a) is shown. Fenofibrate significantly decreased the fractional catabolic rate by PFHS and PFOS showed a similarly increased fractional catabolic rate (by 50% and 54%, respectively). However, calculation of the catabolic rate (figure 4.3.13c), which takes into account the different pool sizes of HDL-cholesterol after the various treatments, showed that the clearance of HDL was actually decreased by PFHS and PFOS treatment (by -48% and ~65% respectively). The catabolic rate was somewhat increased by fenofibrate, although this was not significantly different from the control group PFBS showed no effects. 77 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Figure 4.3.13a [3H]-cholesteryl oleyl ether decay in plasma 100 .. ?."Control .5. +Fenofibrate 3 8 5 1; 3 1: +PFos .3 a: '5 .93Time after injection p<0.05 vs. control Figure 4.3.1313 Fractional catabolic rate of HDL 0.6 IControl lFeno?brate 0.0 p<0.05 vs. control 78 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Figure 4.3.13C Cataboiic rate of HDL 0.6 IControl I Fenofibrate 0.5 - I PFBS _l IPFOS E?p<0.05 vs. control 4.3.14 Liver microsomal DGAT activity Values are absolute values (nmol/min/mg protein) and are means 5.0. from 5 mice per group. Individual values are given in appendix DGAT activity DGAT-Z (nmol/min/mg protein) Control avg 2.2 1.6 sd 0.8 0.4 Fenofibrate avg 2.1 2.8 sd 0.5 1.0 PFBS avg 2.6 1.7 sd 0.6 0.7 PFHS avg 2.6 2.4 sd 0.9 0.8 PFOS avg 3.4 2.8 sd 0.5 0.9 Fenofibrate 0.841 0.056 PFHS 0.421 0.151 .056 0. Fanofibrate v5 PFBS 0.222 0.151 1 . PFOS 0.016 1.000 PFOS 0.095 0.056 79 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Hepatic microsomai DGAT-Z activity (which is primarily involved in VLDL assembly in the liver) was significantly increased in the PFOS treatment group (by Fenofibrate showed a trend towards significance in a higher DGAT-Z activity (by p= 0.056). PFBS and PFHS showed no significant effects. None of the treatment groups affected DGAT-1 activity (primarily involved in T6 storage in the liver) significantly, although PFOS showed a trend towards significance and DGAT-2 activities tended to be higher in the PFOS group as compared to the PFBS treated mice. Figure 4.3.14 Liver microsomal DGAT activity 6 I Control I Fenofibrate 5 . I PFBS a: PFHS I PFOS Liver microsomal DGAT activity (nmollminlmg protein) p<0.05 vs. control 4.3.15 Liver lipid analysis Values are absolute values (pg/mg protein) and are means :t S.D. from 6 mice per group. Individual values are given in appendix XIX. Liver lipids FC CE TC TG (Hg/mg protein) Control avg 14.3 24.5 38.9 721.5 Sd 1.4 3.5 3.9 17.5 Fenofibrate avg 11.7 11.9 23.6 59.9 5d 0.8 1.5 2.3 15.9 PFBS . avg 11.6 15.9 27.5 61.8 5d 2.0 6.4 8.3 30.2 PFHS avg 13.1 24.2 37.3 113.5 Sd 1.6 5.7 7.0 24.8 PFOS avg 16.6 47.6 64.1 217.2 5d 1.5 10.3 11.5 47.5 80 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 een Control vs: Fenofibrate PFBS PFHS PFOS Fenofibrate vs PFBS PFOS PFOS Hepatic triglycerides were significantly increased after PFHS and PFOS treatment (by +52% and +192%, respectively). PFOS showed a significantly higher increase than PFHS Both fenofibrate and PFBS seem to decrease hepatic triglycerides somewhat, but this was not significantly different from the control. Hepatic free cholesterol and cholesterolester levels were affected by both fenofibrate and PFBS, free cholesterol was reduced by ?18% and ?19% and cholesterolester was decreased by -52% and respectively. PFHS did not have an effect on hepatic cholesterol levels, but PFOS showed increases in both cholesterol and cholesterol ester (by +16% and respectively). PFOS increased hepatic triglycerides and cholesterol levels to a significantly higher extent as compared to PFBS. The same holds true for triglycerides with there was a tendency towards increased cholesterolester levels. Figure 4.3. 15 Liver lipids 300 1 Control I Fenofibrate a PFBS 31% PFHS 200 . PFOS 250 - 150 - 100 - liver lipids protein) 50- p<0.05 vs. control 81 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 4.3.16 Liver histology HPS stained liver slides were examined histopathologically (based on necrosis with inflammatory cells, proteinaceous droplets, hepatocellular hypertrophy and hepatocellular microvacuolation). Per group 5 livers were analyzed. Table 4.3.16 Histopathologically examination of livers naceous microvacuolation Very slight MF multifocal moderate diffuse Control: Only 1 out of the 5 control CETP mice, mouse 2, showed hepatocellular hypertrophy and hepatocellular microvacuolation. The liver histology of the other 4 mice appeared to be quite normal (figure 4.3.16a, table 4.3.16). Mouse 1 3 appeared to have more glycogen in the liver than mouse 4 6. In all control mice, proteinaceous droplets could be detected in the cytoplasm of the hepatocytes (deposition of ApoE3Leiden proteins, ref). Figure 4.3.16a HPS stained control livers Control (mouse 2, 100x) with hepatocellular and mmrovathi B: Control (mouse 1, 100x) with more glycogen (white cytoplasm) C: Control (mouse 4, 100x) 82 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031 . 12685 Feno?brate: The addition of fenofibrate to the diet resulted in fewer proteinaceous droplets. The hepatocytes were hypertrophic and had a fine granular eosinophilic cytoplasm, especially in zone around the central vein (?gure 4.3.16b, table 4.3.16). Mouse 10 showed moderate hepatocelluiar vacuolation. Figure 4.3.16b HPS stained livers, treated with fenofibrate A: Drawing of liver Iobule, showing zone A, and (100x) B: Fenofibrate treatment (mouse 8, 100x) C: Fenofibrate treatment (mouse 10, 100x) CV: central vein PFBS: The addition of PFBS to the diet resulted in fewer proteinaceous droplets. Three out of five mice showed moderate hepatocellular microvacuolation (figure 4.3.16c, table 4.3.16). Figure 4.3.16c HPS stained livers, treated with PFBS A: PFBS treatment (mouse 16, 100x) B: PFBS treatment (mouse 17, 100x) C: PFBS treatment (mouse 18, 100x) 83 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 PFHS: The addition of PFHS to the diet resulted in fewer proteinaceous droplets. The hepatocytes were hypertrophic and had a fine granular eosinophilic cyt0plasm, especially in zone around the central vein (figure 4.3.16d, table 4.3.16). One mouse (M23) showed moderate diffuse hepatoceilular microvacuolation. The other four mice only showed focally, very slight microvacuolation. Focal aggregates of inflammatory cells/necrotic hepatocytes could be detected in the livers of 3 out of 5 mice. Figure 4.3.16d HPS stained livers, treated with PFHS A: PFHS treatment (mouse 22, 100x) B: PFHS treatment (mouse 23, 100x) C: PFHS treatment (mouse 27, 200x). Aggregates of inflammatory cells (arrows) PFOS.- The addition of PFOS to the diet resulted in hypertrophic hepatocytes with fine microvacuolation of the cytoplasm (table 4.3.16). Figure 4.3.16e, in which stainings of PFOS, PFHS and control livers were put next to each other, clearly showed the increased lipid filled vacuoles in the hepatocytes after PFOS treatment. Figure 4.3.16e PFOS treatment vs PFHS treatment vs control A: PFOS treatment (Mouse 29, 100x) B: PFHS treatment (Mouse 22, 100x) C: Control (Mouse 4, 100x) 84 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 4.3.17 Liver microarray analysis Liver microarray analysis and subsequent gene expression data analysis was performed on 6 mice per group. The total transcriptome analysis report is included in appendix XX. In this paragraph tables of gene expression data of selected pathways with transcription factors and genes involved (fold change vs control, with q-values) are shown. Q-values are p? values corrected for multiple tested. Triglyceride metabolism Lipoiysis In line with the higher lipolytic activity in plasma, hepatic signals for LPL (fenofibrate 4.6-fold, PFHS 4.3?fold and PFOS 2.1 fold) were increased. Fatty acid/triglyceride Although a major regulator of fatty acid was decreased with PFBS, PFHS and PFOS several genes in the fatty acid pathway were increased, suggesting that regulation does not take place via activation of LXR but probably via PXR. Several ACS genes and were increased with fenofibrate, PFHS and PFOS. SCDZ was increased by PFHS and PFOS. Table 4.3.16a Gene expression data of triglyceride metabolism pathway with transcription factors and genes involved Control 85 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 B-oxidation Genes involved in B-oxidation of fatty acids and 1b, ACO, enoyl hydratase, thiolase (acetyl-Coenzyme A and ACS were increased with fenofibrate, PFHS and PFOS. Fatty acid uptake, binding Genes involved in fatty acid uptake and binding, most prominently CD36 and FATP, and to a lesser extent FABP genes were increased with fenofibrate, PFHS and PFOS. Triglyceride uptake Themajor receptor involved in uptake of VLDL remnants and LDL, LDLR, was increased with fenofibrate. showed no effect on genes involved in triglyceride metabolism except for mild increases in B?oxidation genes enoyl hydratase and ACS and fatty acid transporter CD36. 86 Mechanism of different PFAS's on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Cholesterol metabolism Uptake The major receptor involved in uptake of VLDL remnants and LDL, LDLR, was increased with fenofibrate, in line with decreased liver cholesterol levels. and storage In line with decreased liver cholesterol levels also cholesterol genes and reductase and squalene were increased with fenofibrate. and squalene were also increased with PFHS. Cholesterol esterification genes and ACATZ were increased with fenofibrate, PFHS and PFOS. - Bile acid metabolism and biliairy cholesterol excretion The gene coding for the major rate-limiting enzyme in the bile acid pathway, was strongly decreased with PFHS and PFOS. Also genes involved in bile acid uptake, NTCP, and biliairy excretion, BSEP, were decreased. Genes involved in biliairy cholesterol excretion, were decreased by PFHS and PFHS. Inhibition of cholesterol metabolism and excretion may form an explanation for increased hepatic cholesterol levels found with PFOS. PFBS showed no effect on genes involved in cholesterol metabolism except for a decrease in which is not understood well. However, regulation by takes place on the protein level, which was not measured. Table 4.3.16b Gene expression data of cholesterol metabolism pathway with transcription factors and genes involved 87 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 HDL metabolism maturation and remodeling Genes involved in HDL apoAl the major protein of HDL, and HDL maturation, and LCAT, were decreased by PFHS and PFOS. PLTP a gene involved in remodeling of HDL, leading to larger particles was increased by PFHS and fenofibrate. Uptake The major gene involved in uptake of HDL-cholesterol esters, was decreased by fenofibrate, PFHS and PFOS. The latter mechanism together'with the strong decrease in CETP activity is responsible for the increased HDL levels with fenofibrate. However, this mechanism cannot explain the strong decrease in HDL levels Observed with PFHS and PFOS. The decreased HDL levels most likely result from decreased HDL and maturation (decreased gene expression of apoAl, and LCAT). PFBS showed no effect on genes involved in HDL metabolism, in line with unchanged HDL levels Table 4.3.16c Gene expression data of HDL metabolism pathway with transcription factors and genes involved Table 4.3.16d Gene expression data of transcription factors and safety parameters 88 3M#03 Mechanism o f different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 5 Conclusions and comments I n appendix XXI a table is included, in which the effects PFBS, PFHS, PFOS an fenofibrate on all measured parameters (except for the micro-array data) are summarized. Mechanism of action of PFBS PFBS reduced plasma cholesterol and triglyceride levels by about 25% and 45%, respectively. The data from physiological experiments indicate that the decreases in lipid levels are caused by increased clearance of VLDL-TG and VLDL-CE-and mildly reduced VLDL-particle production. PFBS showed no effect on genes involved i,n triglyceride metabolism except for mild increases in poxidation genes enoyl CoA hydratase and ACS and fatty acid transporter CD36. PFBS also showed no effect on genes involved in cholesterol metabolism except for a decrease in SREBPla, which is not understood well. However, regulation'by SREBPlaIc takes place on the protein level, which was not measured. PFBS had no effect on HDL-cholesterol and apoAl levels. PFBS showed no effect on genes involved in HDL metabolism, in line with the unchanged HDL levels. PFBS mildly increased liver weight, but had no effect on ALT and development of hepatosteatosis. Based on mRNA signals PFBS appears to have mild PPARa-agonistic activity (p-oxidation increased and liver size increased) The present data indicate that PFBS has no increased CVD risk profile. Mechanism of action of PFHS PFHS reduced plasma cholesterol and triglyceride levels by about 60% and 75%, respectively. The data from physiological experiments supported by hepatic mRNA levels indicate that the decreases in lipid levels are caused by increased lipolysis and clearance of VLDL-TG and VLDL-CE, and strongly reduced VLDL-TG and VLDL-particle production. I n line with the higher lipolytic activity in plasma, the hepatic mRNA signal for LPL (4.3-fold) was increased. Although a major regulator of fatty acid synthesis, SREBPlc, was decreased with PFBS, PFHS and PFOS, several genes in the fatty acid synthesis pathway were increased, suggesting that regulation does not take place via activation of LXR but probably via PXR. Several ACS genes, SCD2 and DGATl were increased with PFHS. Increased hepatic fatty acid synthesis together with reduced VLDL-TG secretion may form an explanation for the accumulation of triglycerides in the liver. I n addition, increased uptake and binding of fatty acids (increased mRNA levels of CD36 and FATP, and t o a lesser extent FABP genes) which result from enhanced lipolysis may contribute t o development of hepatosteatosis. This occurred, despite the observation that genes involved in poxidation of fatty acids (CPTlb, ACO, enoyl CoA hydratase, thiolase (acetyl-Coenzyme A acyltransferase) and ACS were increased with PFHS. PFHS strongly decreased HDL-cholesterol (about -75%) and apoAl (about -75%) levels. Although HDL catabolic rate and the major gene involved in uptake of HDL-cholesterol esters, SR-B1 were strongly decreased, we conclude that based on mRNA signals PFHS reduces HDL levels by downregulation of apoAl synthesis and HDL maturation (ABCal, LCAT). The latter adverse changes are most likely the result of PXR-agonistic activity (24). Increased remodeling (PLTP) and decreased uptake (SR-81) are suggested to be responsible for the formation of larger HDL particles. PFHS increased liver weight, ALT, and resulted in hepatosteatosis, as observed by biochemical and histological measures. Based on mRNA signals PFHS has strong PPARa-agonistic (lipolysis increased, p-oxida'tion increased, FA uptake increased, liver size increased) and PXR-agonistic activity (FA uptake increased, FA synthesis increased and HDL synthesis and .maturation decreased, liver size increased). 3M#03 Mechanism o f different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 Mechanism of action of PFOS PFOS reduced plasma cholesterol and triglyceride levels by about 65% and 7O0/0, respectively. A similar mechanism of action is active with PFOS as with PFHS. The data from physiological experiments supported by hepatic mRNA levels indicate that the decreases in lipid levels are caused by increased lipolysis and clearance of VLDL-TG, and strongly reduced VLDL-TG and VLDL-particle production. I n line with the higher lipolytic activity in plasma, the hepatic mRNA signal for LPL (2.1-fold) was increased. Although a major regulator of fatty acid synthesis, SREBPlc, was decreased with PFBS, PFHS and PFOS, several genes in the fatty acid synthesis pathway were increased, suggesting that regulation does not take place via activation of LXR but probably via PXR. Several ACS genes, SCD2 and DGATl were increased with PFOS. Increased hepatic fatty acid synthesis together with reduced VLDL-TG secretion may form an explanation for the accumulation of triglycerides in the liver. Also increased uptake and binding of fatty acids (increased mRNA levels of CD36 and FATP, and to a lesser extent FABP genes) which result from enhanced lipolysis may contribute to development of hepatosteatosis. This occurred, despite the observation that genes involved in (3-oxidation of fatty acids (CPTlb, ACO, enoyl CoA hydratase, thiolase (acetyl-Coenzyme A acyltransferase) and ACS were increased with PFHS. PFOS increased mRNA levels of the cholesterol synthesizing enzymes HMGCoA synthase and squalene synthase and cholesterol esterification genes ACATl and ACAT2. Moreover, the gene coding for the major rate-limiting enzyme in the bile acid synthetic pathway, CYP7a1, and genes involved in biliairy cholesterol excretion, ABCg5/g8 were decreased by PFOS. Inhibition of cholesterol metabolism and excretion may form an explanation for increased hepatic cholesterol levels found with PFOS. PFOS strongly decreased HDL-cholesterol (about -65%) and apoAl (about -80%) levels. Although HDL catabolic rate and the major gene involved in uptake of HDL-cholesterol esters, SR-B1 were decreased, we conclude that based on mRNA signals PFOS reduces HDL levels by down-regulation of apoAl synthesis and HDL maturation (LCAT). The latter adverse changes are most likely the result of PXR-agonistic activity (24). PFOS increased liver weight, ALT, and resulted in pronounced hepatosteatosis and liver cholesterol accumulation, as observed by biochemical and histological measures. Based on mRNA signals PFOS has strong PPARa-agonistic (lipolysis increased, (3-oxidation increased, FA uptake increased, liver size increased) and PXR-agonistic activity (FA uptake increased, FA synthesis increased and HDL synthesis and maturation decreased, liver size increased). Involvement of other nuclear transcription factors in the regulation of lipid and lipoprotein metabolism by PFHS and PFOS. Involvement o f CAR and LXR in the changes in lipid and lipoprotein metabolism caused by PFHS and PFOS cannot be fully excluded, but is less likely. Little is know about the role of CAR in lipid metabolism. CAR has been shown t o decrease poxidation genes as CPTl and enoyl CoA hydratase. The latter genes were, however, increased in the present experiments. LXR increases fatty acid synthesis by induction of SREBPlc expression, which was 2-fold decreased, however. I n addition, LXR induces expression of CETP mRNA, whereas in the present experiments a decrease in CETP activity was found. It cannot be excluded that this is caused by a strongly decreased acceptor pool for CE transfer. Involvement of RXR in the observed effects cannot be excluded, since RXR forms a heterodimer together with a larger number of nuclear transcription factors like PPARa, PXR and LXR. However, direct activation of RXR, for instance with bexarotene (25) leads to opposite effects with increased levels of triglycerides and apoB-containing lipoproteins (25). Using the miceroarray database, we suggest to study the involvement of the above and other transcription factors (ArH) involved in the metabolism of xenobiotics in more detail with respect t o changes in other metabolic processes and pathways like glucose metabolism, inflammation and immuneresponse. 3M#03 Mechanism o f different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 Mechanism of action of fenofibrate Fenofibrate reduced plasma cholesterol and triglyceride levels by about 40% and 70°/o, respectively. The data from physiological experiments supported by hepatic mRNA levels indicate that the decreases in lipid levels are caused by strongly increased lipolysis and clearance of VLDLTG and VLDL-CE, despite the increased VLDL-TG production rate. I n line with the higher lipolytic activity in plasma, the hepatic mRNA signal for LPL (4.6-fold) was increased. LDLR mRNA as marker for increased uptake of VLDL remnant particles was enhanced by 1.5-fold. Several ACS genes and DGATl mRNA, involved in fatty acid and triglyceride synthesis, were increased with fenofibrate, whereas DGAT2 activity primarily responsible for VLDL secretion tended t o be increased by 73% (p=0.056). I n addition, genes involved in fatty acid uptake and binding, most prominently CD36 and FATP, and t o a lesser extent FABP genes were increased. Several genes involved in @-oxidationof fatty acids (CPTla and lb, ACO, enoyl CoA hydratase, thiolase (acetyl-Coenzyme A acyltransferase) and ACS were strongly. increased with fenofibrate. Taken together the net effect of these changes i n metabolic pathways in fatty metabolism is no effect of fenofibrate on liver triglyceride levels. I n conclusion, fenofibrate paradoxically increases VLDL-TG production despite reducing plasma TG, which may be caused by enhanced hepatic free fatty acid uptake resulting from strongly accelerated peripheral LPL-mediated lipolysis of VLDL or by increased de novo hepatic TG synthesis. Fenofibrate reduced liver cholesterol content resulting in enhanced LDLR mRNA levels and mRNA levels of cholesterol synthetic enzymes (HMGCoA synthase and reductase and squalene synthase) and decreased mRNA levels of cholesterol metabolizing enzymes (CYP7al), the latter leading t o reduced fecal bile acid excretion. Fenofibrate increased HDL-cholesterol (+50°/o) and formation of large HDL-1 particles, and had no effect on apoAl. Although the major gene involved in uptake of HDL-cholesterol esters, SR-B1, was decreased, the catabolic rate was not significantly decreased. Fenofibrate strongly decrease in CETP activity, which was found majorly responsible for the increased HDL levels upon treatment with fenofibrate and PPARa,y-agonists (14,26). Fenofibrate increased liver weight, without effects on ALT and hepatosteatosis, and decreased liver cholesterol content. Based on mRNA signals fenofibrate has strong PPARa-agonistic activity (lipolysis increased, FA uptake increased, @-oxidationincreased; HDL remodeling decreased). , 3M#03 Mechanism o f different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 6 References Van den Maagdenberg AMJM, Hofker MH, Krimpenfort PJA, de Bruijn IG, van Vlijmen B, van der Boom H, Havekes LM, Frants RR. Transgenic mice carrying the apolipoprotein E3Leiden gene exhibit hyperlipoproteinemia. J Biol Chem 1993; 268: 10540-10545. Van Vlijmen B, van den Maagdenberg AMJM, Gijbels MJJ, van der Boom H, HogenEsch H, Frants RR, Hofker MH, Havekes LM. Diet-induced hyperlipoproteinemia and atherosclerosis n 1994; 93: 1403-1410. in apolipoprotein E3-Leid'en transgenic mice': J ~ l i Invest Groot PHE, van Vlijmen BJM, GM, Hofker MH, Schiffelers R, Vidgeon-Hart M, Havekes LM. Quantitative assessment of aortic atherosclerosis in apoE3Leiden transgenic mice and its relationship t o serum cholesterol' exposure. Arterioscler Thromb Vasc Biol 1996; 16: 926-933. Kleemann R, Princen HMG, Emeis JJ, Jukema JW Fontijn RD, Horrevoets AJG, Kooistra T, Havekes LM. Rosuvastatin reduces atherosclerosis development beyond and independent of its plasma cholesterol-lowering effect in apoE*3Leiden transgenic mice: Evidence for antiinflammatory effects of rosuvastatin. Circulation 2003; 108: 1368-1374. Delsing DJM, Jukema JW, van de Wiel MA, Emeis JJ, van der Laarse A, Havekes LM, Princen HMG. Differential effects of amlodipine and atorvastatin treatment and their combination on atherosclerosis in ApoEs3-Leiden transgenic mice. 3 Cardiovasc Pharmacol 2003; 42: 6370. Verschuren L, Kleemann R, Offerman EH, Szalai AJ, Emeis 31, Princen HMG, Kooistra T. Effect of low dose atorvastatin versus diet-induced cholesterol lowering on atherosclerotic lesion progression and inflammation in apolipoprotein E*3-Leiden transgenic mice.Arterioscler Thromb Vasc Biol 2005; 25: 1-7. Delsing DIM, Post SM, Groenendijk M, Solaas K, vanderBoom H, vanDuyvenvoorde W, dewit ECM, Bloks VW, Kuipers F, Havekes LM and Princen HMG. Rosuvastatin reduces plasma lipids by inhibiting VLDL production and enhancing hepatobiliary lipid excretion in ApoE*3-Leiden mice. J Cardiovasc Pharmacol 2005; 45: 53-60. Kooistra T, Verschuren L, van der Weij 3, Koenig W, Toet K, Princen HMG, Kleemann R. Fenofibrate reduces atherogenesis in APOE*3Leiden mice: evidence for multiple antiatherogenic effects besides lowering plasma cholesterol Arterioscler Thromb Vasc Biol 2006; 26: 2322-2330. Van Vlijmen BJM, Mensink RP, van 't Hof HB, Offermans RFG, Hofker MH, Havekes LM. Effects of dietary fish oil on serum lipids and VLDL kinetics in hyperlipidemic apolipoprotein E*3-Leiden transgenic mice. J Lipid Res 1998; 39: 1181-1188. Delsing DJM, Offerman EH, van Duyvenvoorde W, van der Boom H, de Wit ECM, Gijbels MU, Van der Laarse A, Jukema JW, Havekes LM, Princen HMG. The acyl-CoA:cholesterol acyltransferase (ACAT)-inhibitor avasimibe reduces atherosclerosis additionally to its cholesterol-lowering effect in apoE*3Leiden mice. Circulation 2001; 103: 1778-1786. Volger OL, van der Boom H, de Wit ECM, van Duyvenvoorde W, Hornstra G, Plat 3, Havekes LM, Mensink RP, Princen HMG. Dietary plant stanol esters reduce VLDL-cholesterol secretion and bile saturation in apoE*3Leiden transgenic mice. Arterioscler Thromb Vasc Biol 2001; 21: 1046-1052. Zadelaar ASM, Kleemann R, Verschuren L, de Vries-van der Weij J, van der Hoorn 3, Princen HM, Kooistra T. Mouse models for atherosclerosis and pharmaceutical modifiers. Arterioscler Thromb Vasc Biol 2007; 27: 1706-1721. Westerterp MI Van der Hoogt CC, de Haan W, Offerman EH, Dallinga-Thie GM, Jukema JW, Havekes LM, Rensen PCN. Cholesteryl ester transfer protein decreases HDL and severely aggravates atherosclerosis in APOE*3Leiden mice. Arterioscler Thromb Vasc Biol 2006; 26: 2552-2559. Van der Hoogt CC, de Haan W, Westerterp M, Hoekstra MI Dallinga-Thie GM, Romijn JA, Princen HMG, Jukema JW, Havekes LM, Rensen PCN. Fenofibrate increases HDL cholesterol by reducing cholesteryl ester transfer protein expression. J Lipid Res 2007; 48: 1763-1771. De Haan W, van der Hoogt CC, Westerterp M, Hoekstra M, Dallinga-Thie GM, Princen HMG, Romijn JA, Jukema JW, Havekes LM, Rensen PCN. Atorvastatin increases HDL cholesterol by reducing CETP expression in cholesterol-fed APOEU3-Leiden.CETP mice. Atherosclerosis 2008; 197: 57-63. Van der Hoorn JWA, de Haan W, Berbee JFP, Havekes LM, Jukema JW, Rensen PCN, Princen HMG. Niacin increases HDL by reducing hepatic expression and plasma levels of ens on 3M#03 Mechanism o f different PFAS's on lipid and lipoprotein metabolism in ApoE3L-CETP mice TNO project number 031.12685 cholesteryl ester transfer protein in APOE*3-Leiden.CETP mice. Arterioscler Thromb Vasc Biol 2008; 28: 2016-2022. De Haan W, de Vries-van der Weij J, van der Hoorn JWA, Gautier T, van der Hoogt CC, Westerterp M, Romijn JA, Jukema JW, Havekes LM, Princen HMG, Rensen PCN. Torcetrapib does not reduce atherosclerosis beyond atorvastatin and induces more pro-inflammatory lesions than atorvastatin. Circulation 2008; 117: 2515-2522. Post SM, Groenendijk M, Solaas K, Rensen PCN, Princen HMG. Cholesterol 7a-hydroxylase deficiency in mice on an apoE*3Leiden background impairs very-low-density lipoprotein production. Arterioscler Thromb Vasc Biol 2004; 23: 768-774. Duivenvoorden I,Voshol PJ, Rensen PCN, Duyvenvoorde W, Romijn JA, Emeis 33, Havekes LM, Nieuwenhuizen WF. Dietary sphingolipids lower plasma cholesterol and triacylglycerol and prevent lever steatosis in apoE*3Leiden mice. Am J Clin Nutr 2006; 84: 312-321. Post SM, de Crom R, van Haperen R, van Tol A, and Princen HM. Increased fecal bile acid excretion in transgenic mice with elevated expression of human phospholipid transfer protein. Arterioscler Thromb Vasc Biol 2003; 23: 892-897. Kleemann R, Verschuren L, van Erk MI, Nikolsky Y, Cnubben NH, Verheij ER, Smilde AK, Hendriks HF, Zadelaar S, Smith GJ, Kaznacheev V, Nikolskaya T, Melnikov A, Hurt-Camejo E, van der Greef 3, van Ommen 6, Kooistra T. Atherosclerosis and liver inflammation induced by increased dietary cholesterol intake: a combined transcriptomics and metabolomics analysis. Genome Biol 2007; 8: R200. Havekes LM, de Wit EC and Princen HM. Cellular free cholesterol in Hep G2 cells is only partially available for down-regulation of low-density-lipoprotein receptor activity. Biochem J 1987; 247: 739-746. Mensenkamp AR, van Luyn MJ, van Goor H, Bloks V, Apostel F, Greeve J, Hofker MH, Jong MC, van Vlijmen BJ, Havekes LM, Kuipers F. Hepatic lipid accumulation, altered very low density lipoprotein formation and apolipoprotein E deposition in apolipoprotein E3-Leiden transgenic mice. J Hepatol 2000 Aug; 33: 189-98. De Haan W, de Vries-van der Weij 3, Mol IM, Hoekstra M, Romijn JA, Jukema JW, Havekes LM, Princen HMG, Rensen PCN. PXR agonism decreases HDL levels in APOE*3-Leiden.CETP mice. Biochim Biophys Acta 2009; 1791: 191-197. de Vries-van der Weij 3, de Haan W, Hu L, Kuif M, Oei HL, van der Hoorn JW, Havekes LM, Princen HM, Romijn JA, Smit JW, Rensen PC. Bexarotene induces dyslipidemia by increased very low-density lipoprotein production and cholesteryl ester transfer protein-mediated reduction of high-density lipoprotein. Endocrinology 2009; 150: 2368-75. Van der Hoorn, Jukema JW, Havekes LM, Lundholm E, Camejo G, Rensen PCN, Princen HMG. The dual PPARa/y agonist tesaglitazar blocks progression of pre-existing atherosclerosis in APOE*3-Leiden.CETP transgenic mice. Brit J Pharmacol 2009; 156: 10671075. Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 7 Appendices Appendix I Body weight am Group Mouse# Cage# Earta ht . week 0 week 1 week 4 week 6 29.7 31.0 32.4 3231.2 32.2 33.2 26.9 .3 . . 25.3 27.0 28.1 29.9 7 23.9 25.4 26.4 26.6 23.6 .4 25 26.3 27.5 28.6 29.0 28.5 28.0 28.9 29.4 26.5 26.3 26.7 26.6 26.8 27.9 27.9 28.8 .0 21. 22.8 23. .7 27.3 28.9 2 . 2.3 16 27.0 2. Week 5: mouse 22 put in separate cage (10A), because of fighting wounds 94 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 amaze week 4 7 95 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 EMILE Group Mouse# Cage# week 1 26.9 96 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Appendix II Tissue weight emu Body weight, liver weight and perigonadai fat weight at sacrifice. Mouse 15: scabs on belly (from fighting?) Mouse 22: open wound at back from fighting Mouse 32: white spot on liver Mouse 38,39,40: pale liver 97 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Body weight, liver weight and perigonadal fat weight at sacrifice. 98 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Mi Body weight, liver weight and perigonadal fat weight at sacrifice. Mouse 23: granular liver 99 Mechanism of different on lipid and IipOprotein metabolism in mice TNO project number 031.12685 Appendix Food intake M1. Food intake mo week 0-1 week -4 week 5100 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Studz 2 wee -1 Cage# Food intake 101 Mechanism of different PFAS's on lipid and iipoprotein metabolism in mice TNO project number 031. 12685 Studz 3 Food intake mo da Cage# 102 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Appendix IV Plasma cholesterol Stud! 1: Group Mouse# Cage# Plasma cholesterol t=0 weeks t=4 weeks t=6 weeks 8.41 8.17 8.8.99 .75 5.32 .02 6.32 9.15 1 8.07 7.97 .02 . 1 8.05 9.02 1. 1. 5.47 7.19 4.89 4.77 4.84 4.79 .84 5.3 6.10 4 .5 1.06 5.80 5.40 6.63 9.30 7.14 103 Mechanism of di 'erent on lipid and iipoprotein metabolism in mice TNO project number 031.12685 5mg! g; 1 i 104 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 105 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Appendix Plasma HDL-cholesterol 5mm Group Mouseit Cage# Plasma HDL?cho t=0 t=4 1 1 0.67 1.34 0.75 0.92 0.96 1.46 .46 106 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 ?air?2i t=4 weeks 107 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Group Mouse# Cage# Plasma HDL-cholesterol t=0 =4 1.52 1.39 1.44 1.02 1.24 0.89 1.29 108 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Appendix VI Plasma triglycerides 109 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031 . 12685 t=4 weeks 48 1.2 1. .21 110 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 EMILE 111 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Appendix VII Plasma free glycerol 112 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Appendix Plasma free fatty acids 2mm; 113 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Appendix IX Plasma ApoA1 114 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031 . 12685 Appendix Plasma CETP mass 115 Mechanism of di?l?erent on lipid and iipoprotein metaboiism in mice TNO project number 031.12685 Appendix XI Plasma CETP activity QEIE g?igig 116 Mechanism of different on iipid and iipoprotein metabolism in mice TNO project number 031 . 12685 Appendix XII Post heparin LPL and HL activity 24.15 21.64 14.77 20.79 117 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Appendix Fecal lipids Group Cage# total neural sterois excretion of I mol/ 100 mouse! choiesterol cholestanol Eathosterol 33.54 . 93.21 3.93 2.31.08 94.06 3.21 2.62 42.23 94.69 1.30 3.9 23. 93 0 36.31 93.6 0.6 32.00 . 94.08 43.36 93.96 49.60 95.64 53. 94. 2 .76 10+ 10A 118 Mechanism of different on iipid and iipoprotein metabolism in mice TNO project number 031.12685 total phytosterols excretion 100 9 +1 119 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 120 Mechanism of different on lipid and ifpoprotein metabolism in mice TNO project number 031 . 12685 Fecal fatty acids 121 Mechanism of di??erent on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Fatty acid balance roup excretion 122 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Appendix XIV VLDL-triglycerides and de novo ApoB production - i 1' Mouse# Body weight Plasma VLDL-TG production t=0 min t=15 min t=30 min t=60 min t=3.975 3.591 412 3.618 3.564 123 Mechanism of di??erent on lipid and lipoprotein metabolism in mice TNO project number 031 . 12685 Mouse# Body welgh?c pl serum used mi VLDL 355 1n ApoB de novo ApoB TG production 10? 1oq 1 32.3 200 .197 4756.4 4.74 0.917 4 30.3 280 10320.5 7.47 0.908 124 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 125 Mechanism of different on lipid and iipoprotein metabolism in mice 1 TNO project number 031 . 12685 i 1 Appendix XV Biliary bile acids, cholesterol and phospholipids Bile Group volume Bile ?ow 30 min 45 min 30 min total .50? 13.40 17 8. 1 10 mouse moose arm mouse 2 canula hardly any flow, after 30 min adjusted mouse 7 mouse not warm enough mouse 13 mouse not warm enough mouse 18 mouse died during anaesthesia mouse 19 mouse not arm enough 126 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031 . 12685 Bile acid ?ow Group Bile acid cone in bile (mM) Bile acid flow (nmol/min/kg mouse) min mm 425 . 2 12.38 . 15.3 11.76 2.38 3.98 . 1 mouse mouse not warm enoug mouse 2 canula wrong, hardiy any ?ow, after 30 min adjusted mouse 7 mouse not warm enough mouse 13 mouse not warm enough mouse 18 mouse died during anaesthesia mouse 19 mouse not warm enough 127 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Phosgholigid flow mouse 1 mouse not warm eno mouse 2 canuia wrong, hardly any ?ow, after 30 min adjusted mouse 7 mouse not warm enough mouse 13 mouse not warm enough mouse 18 mouse died during anaesthesia mouse 19 mouse not warm enough 128 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 5mm 45 min 1 30 min 45 .12 0.27 1 . mouse mouse not warm mouse 2 canula wrong, hardly any flow, after 30 min adjusted mouse 7 mouse not warm enough mouse 13 mouse not warm enough mouse 18 mouse died during anaesthesia mouse 19 mouse not warm enough 129 Mechanism of different on lipid and lipOprotein metabolism in mice TNO project number 031.12685 Appendix XVI In vivo clearance of VLDL-like TG-rich particles and uptake in liver Plasma DECAY SEM 100.00 . 100.00 85.62 . 77.79 42. . 55.72 SERUM DECAY 3H (t2 100%) control 3 4 5 6 SEM SERUM HALF-0.053 0.040 0.031 0.023 0.04 . 0.91 0.01 5 13.03 17.42 22.29 29.75 22.50 (35215701 ?0.25 2.79 5 MASS (9) control ORGAN DISTRIBUTION 3H control 13D Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Plasma DECAY 3H Groupa?Z-Ja. fen'ofihrate . :SEM 100. 85.15 68.26 24. 6.60 SERUM DECAY 3H (t2 100%) fenofibrate Mouse] ORGAN MASS (9) fenofibrate SEM spleen 0.076 ORGAN DISTRIBUTION 3H DOSIS) feno?brate 7 131 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Plasma DECAY 3H 1314 15 16 17 13 SEM mouse died SERUM DECAY 3H (t2 100%) PFBS SERUM HALF-LIFE 3H 14 15 16 0.063 0.097 0.085 0.01 4 10.93 7.18 8.15 1.66 4 MASS PFBS 13 14 15 16 17 18 SEM ORGAN DISTRIBUTION 3H DOSIS) PFBS 13 14 15 15 17 18 SEM spleen muscle 132 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Piasma DECAY 3H 100. 19 68.94 21.18 20 100.00 52.84 16.26 SERUM DECAY 3" (t2 100%) PFHS 19 SERUM HALF-LIFE 3H 20 100. 21 8. 21 55.16 22 100.00 74.06 25.07 22 23 100.00 54.76 11.0.078 0.077 0.102 0.070 0.102 0.01 6 1/2 8.86 8.99 6.83 9.96 6.78 0.53 6 MASS PFHS 19 20 21 22 23 24 SEM 0.155 ORGAN DISTRIBUTION 3H DOSIS) PFHS 19 20 21 22 23 24 SEM spleen 133 Mechanism of di??erent on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Plasma DECAY 3H Group 5 PFOS 25 26 27 28 0 100. . 100.00 2 SEM 63. . . 61.35 10 20 27.7 . 19.28 SERUM DECAY 3H 1 PFOS 25 SERUM HALF-LIFE 3H 25 MASS PFOS 25 .1 0. 0.101 spleen ORGAN DISTRIBUTION 3H DOSIS) PFOS 25 2.6 134 Mechanism of different PFAS's on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Plasma DECAY 14C 1 2 3 4 5 6 SEM SERUM DECAY 14C (t2 1000/0) control 94.81 54. SERUM HALF-LIFE 14C 2 3 4 5 0.023 0.015 0.011 0.007 0.00 5 30.40 45.01 66.01 97.63 11.74 5 ORGAN MASS (9) control ORGAN DISTRIBUTION 14C DOSIS) 135 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Plasma DECAY 14c Group 22SEM SERUM DECAY 14C (t2 100%?) fenofibrate SERUM 14C 9 10 11 12 0.054 0.076 0.055 0.070 51. 0.00 5 12.88 9.10 12.70 9.97 I 0.75 5 ORGAN MASS fenofibrate ORGAN DISTRIBUTION 14C 00515) fenofibrate muscle 136 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Plasma DECAY 14c 13 14 15 1e 17 18 SEM 96.32 63.21 47.30 SERUM DECAY 14C (t2 1000/0) PFBS 65.62 49.11 SERUM HALF-LIFE 14C 14 15 16 0.024 0.040 0.040 0.00 4 28.41 17.16 17.29 3.06 4 MASS PFBS spleen ORGAN DISTRIBUTION 14C DOSIS) PFBS spleen 137 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 Plasma DECAY 14C 76. 64.90 63.52 35. 45.05 SERUM DECAY 14C (t2 100%2) PFHS . 2 74.4 83.33 40.23 59.10 SERUM HALF-LIFE 14C 0.032 0.018 0.039 0:026 0.027 0.025 A 003%] 0.00 6 t1/2 21.46 37.67 17.96 26.46 25.67 28.29 26.25 66973] 2.75 6 MASS PFHS ORGAN DISTRIBUTION 14C DOSIS) PFHS spleen 138 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Plasma DECA 14C Group 5 PFOS - 25 79.76 78.30 73.97 77.83 67.32 65.65 SERUM DECAY 14C (t2 100%) SERUM HALF-LIFE 14C PFOS 98.1 920.003 0.005 0.016 0.018 0.020 0.00 6 1/2 239.02 150.68 43.32 39.61 35.36 34.39 6 MASS PFOS ORGAN DISTRIBUTION 14C DOSIS) PFOS spleen 139 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Appendix XVII In vivo clearance of autologous HDL . Group# Mouse# Isolated HDL Quantity pmol HDL Needed for 0.32 pmol NB: Groups PFHS and PFOS not enough for 0.4 pmol. Therefore all groups 0.32 pmol HDL labeled, in stead of 0.4 pmol. 3H after Group# Mouse# 10 I needed 200000 via 2 18179.9 16939147.3 1 .8 2261 . Group Mouse# of dose in time 1 1 74.9 . 40.7 3 71.4 . 39.2 61.9 140 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 Appendix Liver microsomal DGAT activity Group Mouse# protein microsomes activity DGAT-2 activity mg/m nmol/min/ rotein nmoE/mi in 4.9 1.73 1.29 7.4 3.53 .7 1.62 1.85 7. . 2.08 mouse on average mouse 16 DGAT-Z is based on 1 measurement CY 141 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031 . 12685 Appendix XIX Liver lipids Group Mouse# Cage# Ea Liver I ds rotein CE TC 20.4 33.5 26.0 42.3 23. 39.3 25.9 38.4 21. 36.0 24.6 38.9 .5 3. 12.1 23.7 1 2 . 9.9 20.6 10 1. 13.0 25.5 1 . 142 Mechanism of different on Iipid and lipoprotein metabolism in mice Appendix XX TNO project number 031.12685 Liver microarray anaiysis THO Quality of Life twwenmumnmwe?k Memek 4' Wm'm Otrmnam': For mm mm Mew-h 1'45. TNO report Transcripzome anaiysis of effects of 3M compomlds in liver of mice Date 5 Jame 2009 Rumor?) Mn) 321 1113 Ext Copv no cog? No ofcoptes no ofcopw 31:11th of pages 20 Number of appegdxces . numbet or'appendtres Customer 3M Prejccmame Pmectname Progecmumbcr Pietermumbet All ?ghts mm ed No part of Ihis pubhcattou may be :epsoduced and at 1313135233th by pom. pitotopnm. mtcm?lm 01' any exile! means turnout the 13121 101?; tmtten consent of T520 In case report was mm on insmmuons, the tights and obhgmons of contractmg games subject to whet the ?tanciard C?oztdmons for Research Instructions given to TKO or the tele'. ant agreement condude? bem?een the contraetmg pat-us. Submitting the report for mspectzon to patties Who have a direct interest 13 penttitted i? 200? INC 143 Utrechtseweg 43 Box 360 3700 A.) Zeast The Netherlands I'll 3069314344 +3i 30 695 ?2 24 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 ?mo repon 2 i 19 Contents Infrutiuction 3 Methods 4 2 2.1 RNA anti nucroalmy .. 4 2.2 Gene expiressmn data anaiyszs . 4 3 Results . 5 3. Diffez'ezlimlly expressed genesLipid metabolism. .. .. . .. 6 3.2.1 Pre-selecred 1151 ofgenes2.2 facmr analyns of genes involved .. .. .. .. . . .. 3 4 i0 144 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 TNO repon 3 I 10 1 Introduction {Vascular and Metabolic Diseases) team is conducung a study for 3M. 1m esngaiing the effects of 3 compounds 11: CETP mute and companng these e?ects to the e?ects of feno?brate (FF). is E?PARalpha activator. Study compounds {Pl-?38. PFHS. PFOS) are per-t?luox?o-alkyleulfouates. These compounds accumulate 1:1 humans and nature, Two of the study compounds were taken off the market. one 15 use. 3M aims to :m-eshgale e?ecte of these compounds light of polennal health mks. Fu?hermore. this knowledge will help them to develop compounds that are less harmful of the will be peniormecl to unesezgate effects of the compounds on a paws-selected set of genes and on lipid and hpopl?o'rein metabohsm pathways. and to compale these effects to the effects of PPAR-alpha ?mix-moi" feuo?bmte. 145 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031 . 12685 TND report 4! 1O 2 Methods RNA isolation and microarray hybridization RNA was isolated from liters of muze After quality contml of the RNA. microan'ay analysis was carried out at Sex-woe XS BX. (Leiden. the Netherlands.) using the Affymetrix technology platform and Affmettix Gene-Clnp??: mouse genome arrays. Data were sent to TNO. 2.2 Gene expression data analysis Quality comm} of microamty data was perfomted using Bto??onductor packages (an sisnpleaffy and All samples pasted the QC. Raw signal littenmies CEL- files) were normalized using the GCRMA algmithm (go-mm slow) For annotation of probes and of sxgnals fiom plobes tepiesentmg one gene the custom MNBI was used (based on EntiezGene. t?erstou 11.0.2) bieinarraymbni ht in). This resulted in expression values for 16331 genes. :epresented by umque Entrez gene identi?ers. Genes were filtered for expression above 5 5 01? more samples. resulting in a set of 11587 genes that was used for analysis. Gene expreealon data were log- transformed [base 2). Statistical analy?s was mm; the moderated t-teat (Lintma Imp. bloinf.? ehi.etlu.aul1uuna with COITECUOII for multiple testing. Cut-off for statistically significant changes was set at q-mlue (-0.65 (q-x-alue p-anlue conected for itiultiple testing). factor analysis ?'35 performed in Biblioephere soft? are v7.21 {Genomatix Software (31an. Munich. Gummy). Fm each compound. genes weie selected that were signi?cantly diflereutially expressed and that were mtolt ed in lipid metabolism. Involvement in lipid metabolism was determined based on Gene Ontology annotation in at least one of the following biological! processes: lipid lipzd catabolism, lipid kiosyutheeis. fatty acid fatty acid metaboliom. fatty acid beta oxidation! cholesterol metabolism. cholesterol biosymliesis. cholesterol oatabolism. hpoprotem ntetaliolism. lipid dampen. cholesterol transport. In addition. I-pro?ler was performed using expression values corrected for mean expression in the control group. This resulted in scoxes {t-scores) and signi?cance values for pathways and biological processes. A positive score means that the 1mj01ity of the genes in the patlm 3y are up-zegulated; a negatwe score means that the majonty of the genes in the pathway are down-regulated. Pathways and biological processes with signi?cant scores (M in 5 or 6 annuals per group were selected. A hiem?chical clustermg of these pathways and biological processes and then scores in all samples was generated in Ge:1ePattem{Broad Institute. MIT. USA). 146 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 TNO report 5 i to 3 Results 3.1 Differentially expressed genes in the statistical analysis. the feno?brate (FF), PFBS. PFHS and PFOS groups were emanated to the comm] group. PP resulted in 2924 differentially expressed gene. PFBS resuited in 43S dif?'?erentially expressed genes. PEI-IS resulted in 4230 differentially expressed genes and PFOS resulted in 3986 differentialiy expresseci genes (?gure 1). significant changes 4500 i PFHS PFOSVBCI Figure l. Nlunber of differentially expreswd genes each of the inten?enuoms compared to control ?gure 2 moraines the overlap in differentially expressed genes for FF and each of the three 3M compoimds? Overall. 43-54% of the genes dxfferentially expressed response to the 5M compounds are also regulated by FF. Furtheunore. there is considerable overlap 111 genes regulated by PFHS and PFOS. as shown in ?gure 3. In total. 2935 genes are regulated by both compounds. in addmon. 362 of these are also regulated. by PEBS. 147 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 mo repo? 61 1D FF PFBS PFHS 11 0) CO (0 PFOS . Figure 2. Venn dxagrams showing 0? erlap in sets of differennaily expx?esse? genes :11 response to FF and PFBS. PEI-I and PFOS rape-actively. PFHS PFOS 2e age F1gme 3 Venn dxegmm showing ovexlap been-eel: genea reguiated PFOS and PFBS. 3.2 Lipid metabolism 3 2 A Pie-seiecred fist ofgenes A set of genes of mterest was de?ned. ?1th a fume on metabohsm aud choiestemi metabolism The expresswn changes as .1 mini! of the 4 interveuuone are lasreci the excel ?le added to this report. Of the 87 genes in the IISL 2'2 were expressed i and 111:: nmjomy was expiessed 111 response :0 one 01 more of the ante-mentions Expressaon of seven genes was net measured (genes were not pzesem an I the microan'ayj and expression of :0 genes was below detectmn FF mtewentmn resuhed in 3] nit-regulated genee and 7 down-regi?amd pre-eelec?red genes Of the 3] genes upaegulated by PF. 25 ?3313 also ezgm?camly up-reguiated b} PFHS. 18 were also up-regulmed by PPOS and 3 wexe also up-zegulated by The selected see of 148 Mechanism of different on lipid and iipoprotein metabolism in mice TNO project number 031.12685 TNO report 7110 genes was grouped into smaller sen of genes related to triglyceride metabolism {lipolysis fatty acidt'triglyceride beta oxidation. fatty acid YLDL as?sentblaget?fonnation. PL excretion). cholesterol (cholesterol storage. uptake. metabolism. excretion) and metabolism (HDL ?brillation. HDL maturation. HDL modelling ?tlestabtlisation. HDL uptake}. 3.3.2 factor ?trait-'52: quettes mt oh ed in item? ittemboitsm in addition to focusing on the pie-selected list of genes. we aimed to identify transcription {stators relevant for regulating the exoression of genes involved in lipid metabolism in response to the different compounds Genes involved in lipid metabolism were identi?ed from Gene Ontology annotation in at leoet one of the following biological processes: lipid cataboiism. lipid fatty and fatty acid metabolism. fatty and beta oxidation. cholesterol metabolism. cholesterol cholesterol oatabolisni. lipoprotein item. lipid transport. cholestenol transport. Next. for each compound. the significantly differentially exptessed genes involved in lipid metabolism were selected: 143 genes for feno?brate. 172 genes for PFHS. 162 genes fox PFOS and 23 genes for PFBS. These sets of genes were submitted to transcription ?ctor analysis in the Bibltosphere software by Genomatix. Au excel ?le containing the results of this analysis is added to the report. The results ?le lists factors that are expected to play in 201:: in regulating expression of the signi?cantly differentially expressed genes involved in lipid metabolism based 0n piesence of a binding site in the promoter region combined with eta-citation it: literattnet or based on co-citation in literature only (fox transcription cofactors). The top three of transcription factors is equal for PFHS. PFOS and PFBS: PPAR-alplia. PPAR-gammo and LXR-beta. 3.3 Pathway anal} sis Figure 4 gives an overview of functional gene eets (based on gene ontology annotation) tegulated by one or mote interventions. ?ielected functional gene sets were Signi?cantly tegulated in at least 5 animals of one or more treatment groups. The shows a number of groups of gene sets that Show a similar pro?le of response The ?rst cluster consists of gene sets related to transcription. which are most strongly down-regulated by PFHS and P505 The second cluster of gene sets related to in?ammation. F1: results in down-regulation of these gene sets. However. PFBS results in tip-regulation of some gene sets. most clearly of acute-phase response and in?ammatory tesponse. The third cluster consists of gene sets related to lipid metabolism and energy metabolism (mitochondnou). FF most strongly upvregnlates lipid metabolism. the effect of PFHS and PFOS 15 similar but less pronounced. A small set of gene sets related to hydrolysis or metabolism are strongly tip-regulated by PF. PFHS and PFOS. These effects could be related to activation. Lastly. gene sets related to glutatliione transferase and monooxygennse activity are speci?cally down-regulated by and PFDS. The effects. in these clusters of gene sets are also Visualized in figure 5. 149 Mechanism of different on iipid and iipoprotein metabolism in mice TNO project number 031 . 12685 TNO report 8 1 1D at was; My ?313$?an a: an pm?: wean-sine 01 WW mm mm" RNA was? LI. Inga efamu oi mmw?m Mme names: mi mom {a namip?tion WW M9th manna! ma.? uespome .n woundm mm: mmVrenpa-wo edlareta?sr awn Wm 00% we :erm? a Wm W?bt?c' m?ammat?m ?mm? a! mum [10mins cum Whammy maximise \Ir' ind cease! WM and grouse my m?mx process gm f? Mum'm mm Him-m alum am pm mid enemy, metabolism 1 1 03A class My gi-?rz-CUAC ?aw: aw cumwmrasc ?mm adm: Emma?: win-Mnla. 11ml Jam-yum 13mg; ?the" medal? w?l?fh? faking. pe?a "273- cwaym in Win: 57am gnwa?a??e?u whit? an add metabuiism rim? lfll?v?cll: cud-v ?1er 11W Figure 4. luster of scores for functional gene groups (beset? on Gene Oxlmlogy) thai are signi?can?y affected in at least 5 annuals of a treamaem gm}? Red indicates positive score (Le. majority of genes up-regulated) and biue indxcates negame score (Le. majmuty of genes damn-reguiated} 150 Mechanism of different on lipid and Iipoprotein metabolism in mice TNO project number 031.12685 1N0 repon "109 was ages . .2er 9'39 mm? mum-ta?: ma *xuwn . w?was -2a+n1kmm rum- .me 7 mm ?mm M??m LWW no: . . if was 9H: 7-95 nan ~26Figure 5. Response pm?les of eiusters of gene sets that Show a sim?ar response to the treatments. 151 Mechanism of di??erent on lipid and Iipoprotein metabolism in mice TNO project number 031 . 12685 TNO repon 10 i 10 4 Conclusions - PFBS $1219 a more limiter? effect on 13f ln'er in nnce than and - Illa-re lsk?a gonsxderable ore: lap genes regulated by FF and the se :egulated by PFHS mfpros - Overall gem set analysn imiicates that fatty acid 111etabohsm. hydz?olase activity and metabohsm are most up-regulated by PF. :bllowed by PFHS and PFOS FF tap-regulates inicl and 13pm transport. thew gene sets are not signi?cantly affected by the 3M cempouncl?s FF dour? regulates In?ammation and immune response. PFBS resulted Lip-regulation of acute inflammatory response. Speca?c effects of PFHS and PFOS are up- regulation at" gluta?none u?ansferaee and monooxygenase activity and down- regulanon of 152 Mechanism of different on lipid and lipoprotein metabolism in mice TNO project number 031.12685 Appendix XXI Summary table Parameter Study week Of weight Body Weight, Food Intake intake and tissue ht Liver weight fat weight Plasma cholesterol Plasma triglycerides HDL-cholesterol Plasma parameters Lipoproteln pro?les (pooled) (pooled) production 8i de novo ApoB composltion Apo clearance Plasma CETP mass HDL clearance Fecal lipids Blle ?ow and biliary lipids Liver measurements lipids 153 $10.35 FEB 17 201$ us POSTAGE was? cmsa mam mam 55mm 1 .32..) CERTIFIED UDUH EELS Deanna Luebker, General Of?cm 2 3M Center 651 737 03 St. Paul. 55l-l4-llm?