Document 6K4GjYEEgeLdap56zKNLpda1

TNO-report CONFIDENTIAL THE EFFECT OF 3 PERFLUORINATED ALKYL SULPHONATED ON CHOLESTEROL/BILE ACID METABOLISM IN 15 %-FAT FED E3-LEIDEN TRANSGENIC MICE IN VIVO AND ON FATTY ACID CONVERSION INTO CHOLESTEROL IN RAT HEPATOCYTES IN VITRO. Project/ study number: 031.10074 / 31-02 TNO Quality of Life TGaNubOiusBilaobosrcatioerynces Z0erkBeocxr2o2e195!. 2T3h0e1NCetEheLroiiadnedns PFahxone 4+33117711551186 11920018 ALukt.horCso:hen 1HoPeegteeer-mdaenNobel I3nM,asMseidgaimaelntDeop:artment SJtouhdnyLm_eBnuttoern:hof, Ph.D, CIH, DABT CMeodripcoarlatDeeSpcairetntmiesntt 3StMPaCualn,taMri.nBnueislodtiang505212440--10020-002 UPrsoane: +1651733.1962 FEmaaxil: fout+e1n6h5o1f7i@3m3m1m77c3om SOrt.udLyoudiisreHct.oCr:ohen Status and Date Final version: 04 December 2006 PrDervafito:us1v8eOrcstioobnesr2008 DDrraafft2s:: 3208 ONcotvoebmebre2r0028006 Numobfpaegers: 35 Authorization by Dr. P.HA. Quax, Technology manager of Vascular and Metabolic Diseases: / /[ Signature and date: = &nd6 Au == H 10135 TNO-report of study 3M-02 Contents PFAS in ApoE3Leiden mice Final report Testing facility Contributors 1. Introduction 2. Study design 3. Results of in vivo experiment (ApoE3Leiden mice) 3.1. Body weight and food intake measurements 3.2. Plasma measurements 3.3. Measurements in metabolic cages 3.3.1 Food and drink intake 3.3.2 Moving (X) activity 3.3.3 O2-consumption, CO2-production and respiratory exchange ratio (RER) 3.3.4 Energy expenditure 3.4 Fecal bile acid, sterol, and fatty acid measurements 3.5 Intestinal cholesterol absorption 3.6 Perigonadal fat tissue weight 3.7 Liver weight and hepatic lipid contents 3.8 Liver enzyme activities: HMG-CoA reductase, Cholesterol 7-hydroxylase and sterol 27-hydroxylase 4. Results of in vitro experiment (rat hepatocytes) 5. Summary and conclusions 5.1. Effects of PFBS in ApoE3Leiden mice 5.2. Effects of PFHS in ApoE3Leiden mice 5.3. Effects of PFOS in ApoE3Leiden mice 04-12-2006 page 3 page 3 page 3 page 4 page 6 page 6 page 7 page 13 page 13 page 15 page 15 page 17 page 18 page 22 page 23 page 23 page 25 page 30 page 32 page 32 page 32 page 33 2 of 35 TNO-report of study 3M-02 Testing facility Business Unit Biosciences TNO Quality of Life Gaubius Laboratory Zernikedreef 9 P.O. Box 2215 2301 CE Leiden Phone: +31 71 5181209 PFAS in ApoE3Leiden mice Final report fax: +31 71 5181901 Contributors Study director: Dr. Louis H. Cohen Phone: +31 71 5181469 E-mail: louis.cohen@tno.nl Experimental work: Ing. Christa van Thiel Ing. Bep Hoegee Ing. Anita van Nieuwkoop Ing. Angela Koudijs Ing. Wim van Duyvenvoorde Ing. Koen van der Mark Ing. Ronald van der Sluis Dr. Peter Voshol Ing. Elsbet Pieterman Advisor : Dr. Hans Princen Prof. Dr. Louis M. Havekes 04-12-2006 1. INTRODUCTION 3M has observed that the compounds of interest have strong cholesterol lowering effects in animal studies. In order to obtain more insight into the possible mechanism of action the effects of the compounds on cholesterol and bile acid synthesis in primary culture of rat hepatocytes and on bile acid uptake in hamster ileum pieces in vitro have been investigated in a previously performed study. From this study it was concluded that the compounds possibly increased hepatic bile acid synthesis, however indirect effect on sterol synthesis might be involved as well. In the present study the in vivo effect of the test compounds on cholesterol and bile acid metabolism have been investigated in ApoE3Leiden transgenic mice. This animal model shows a humanlike lipoprotein pattern and, unlike wild type mice, is sensitive for lipid lowering drugs like statins. Furthermore, the plasma cholesterol levels can be elevated by adjustment of the diet. Additionally, in the present project information on the effects of the test compounds on hepatic cholesterol synthesis have been obtained via an in vitro study, in which the conversion of long and short chain fatty acids into cholesterol in the presence of the compounds have been investigated. 3 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 2. STUDY DESIGN The design for the performance of this study has been described in the final study protocol of 08-05-2006. Deviations from this protocol have been mentioned below. Approval for this study by the TNO Animal Experiment Committee has been obtained on 06-062006 under number DEC#2125. The final study design and the concomitant measurement points have been depicted in scheme 1. week of treatment -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 group 1 (n=10); 15%-fat (HF) diet; control group 2 (n=10); HF + 0.01% PFBS x group 3 (n=10); HF + 0.03% PFBS x group 4 (n=10); HF + 0.006% PFHS x group 5 (n=10); HF + 0.02% PFHS x group 6 (n=10); HF + 0.003% PFOS x group 7 (n=10); HF + 0.01% PFOS x Body weight and food intake Plasma TC, TG Plasma ketone bodies Plasma lipoprotein profiles (TC, PL) Plasma ALAT Plasma samples for cmpd determination Intestinal cholesterol uptake Fecal lipid excretion (sterols, BA, FA) Metabolic cage analysis Liver weight, lipids (TG, FC, CE) Liver microsomes; enzyme determinations Perigonadal fat weight xxxxxxxxxxxxx xx x x x x p p pp p p pp x x x xx xxx 4 x x x (4 = 4 mice/group 1, 3, 4, 6) (p = pooled per group) Scheme 1. Experimental design of the study After a 3 weeks run-in period on a semi synthetic high-fat (HF; 14% lard, 1% corn oil and 0.25% cholesterol) diet, 75 male ApoE3Leiden mice (11-12 weeks of age at start of run-in) were randomized on basis of body weight, plasma cholesterol and triglycerides (measured after 4 h fasting). Then, the groups received respectively HF-diet (control group 1) and 0.01%, 0.03% of PFBS (groups 2 and 3), 0.006% PFHS (group 4) and 0.003% PFOS (group 6) in the HF- diet for the following 10 weeks. Groups 5 and 7 received respectively 0.02% PFHS and 0.01% PFOS in the diet, but after 2 weeks of treatment the mice lost on average 3 grams of bodyweight (see Fig. 3.1.1) and food intake has steadily decreased (see Fig. 3.1.2). For this reason it was decided to stop with these treatments and to return to the high fat control diet until the end of the study (this deviation of the protocol has been discussed with the study monitor by E-mail correspondence of 5, 6 and 18 July 2006). A further deviation concerns the plasma ketone bodies determination. These compounds can only be detected in plasma in a proper way after an overnight fasting period (16 h). For this reason it could not be measured in the same plasma samples of weeks 0 and 8 (taken after 4h fasting) and therefore, it was decided to perform this determination in week -1 and week 6 of treatment (see E-mail correspondence 13 and 30 June 2006). At weeks 0, 1, 4 and 8 plasma cholesterol and triglyceride levels after 4h-fasting have been measured. In weeks 1 (7 days of treatment), 4 and 8, 15l-plasma samples have been sent to the 3M company for the determination of the test compound concentrations. 4 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 As originally planned, at weeks 0, 4 and 8, lipoprotein profiles were analyzed (at week 0, the average values of three subpools were determined, at week 4 and 8, one pooled sample per group was analyzed). Since the week 4-lipoprotein patterns, especially of groups 5 and 7, were different from expected, extra lipoprotein profiles have been analyzed in treatment-week 6 (see E-mail correspondence with study monitor of 24 July 2006). The latter plasma samples were obtained after overnight fasting (necessary for ketone body determination), which in our experience does not give different results compared to lipoprotein profiles of 4h-fasting samples. In weeks 0, 4, 6 (extra measure points; see E-mail correspondence of 24 July 2006) and 8, plasma ALAT values were determined in pooled samples per group. In week 5-6 feces has been collected from 3 subgroups per treatment-group during 3 periods of 2-3 days, and fecal bile acids, fatty acids and sterol quantities and composition were determined. Initiated by the E-mail correspondence of 25 July 2006, feces has been collected in week 4 as well. However, up to now, we see no reason yet to investigate it further. In week 7, the metabolic activity of 4 animals per group from groups 1, 3, 4 and 6 have been analyzed in metabolic cages. The following parameters have been measured: food intake, drink intake, feeding bouts, total physical activity, O2-consumption and CO2-production. From these parameters the respiratory exchange ratio (RER) and O2-consumption-related body energy expenditure have been calculated. Because the food can only be supplied in powdered form in the metabolic cages, during this period (week 7 - 7.5), the other animals in this study also received powdered food. Hereafter all animals received food in chunks again. In week 8-9 intestinal cholesterol uptake has been determined. In week 10 the mice were sacrificed, plasma was collected by means of heart punction, the liver and perigonadal fat were taken out and weighed, two small pieces of the liver were quickly frozen in liquid nitrogen for determination of lipids and eventual RNA analysis, and microsomes and mitochondria have been isolated from the remaining part of the liver. HMG-CoA reductase and cholesterol-7-hydroxylase activity have been determined in the microsomal preparations, sterol-27-hydroxylase activity has been measured in the mitochondrial fractions. 5 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 3. RESULTS OF IN VIVO EXPERIMENT (APOE3LEIDEN MICE) The resulting individual primary data and subsequent calculated data are provided in a separate appendix to this report. The mean values per group are presented in the figures and tables below. Significance of differences was calculated using the non-parametric Mann-Whitney U test and the resulting p-values are given in the appendix. In the figures, a colored asterisk at a separate measure point indicate a statistically significant difference (p<0.05) between the value of the indicated group (of the same color) and that of the control group. In the tables, this difference has been indicated by a yellow background. 3.1. Body weight and food intake measurements As shown in Figs. 3.1.1 and 3.1.2, the treatment with the high concentrations of PFHS and PFOS (groups 5 and 7) caused already from the start a remarkable decrease in body weight and food intake. After the change back to the control diet in week 2, these animals return to normal food intake, comparable to all other groups. Subsequently, group 5 animals gained weight again up to the control values, however, the body weight of the group 7 mice, although increasing again, stayed behind during the whole treatment period. The PFBS treated mice showed the same body weight increase as the controls. The Groups 4 and 6 (low concentrations of PFHS and PFOS) lost weight significantly in the last weeks of the study when compared to control animals. This is in line with the loss of fat tissue as shown below. Figure 3.1.1 Body weight Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS 35 Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 30 Bodyweight (g) 25 * * *** ** * * * ** ** ** * * ** 20 15 -2 0 2 4 6 8 10 12 change diet groups 5 and 7 to high fat control diet time (weeks) Except for the deviations in food intake behavior of groups 5 and 7 (see above), all other animals ate about the same throughout the treatment period (Fig 3.1.2). 6 of 35 TNO-report of study 3M-02 Figure 3.1.2 Food intake Values are means S.D. PFAS in ApoE3Leiden mice Final report 04-12-2006 Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS 4 Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS Food intake (g/mouse/day) 3 2 ** ** 1 0 week 0-1 week 1-2 week 2-3 week 3-4 week 4-5 week 5-6 week 6-7 change diet groups 5 and 7 to high fat control diet week 7-8 week 8-9 3.2. Plasma measurements The following parameters have been determined in plasma in the course of the study: cholesterol (Fig. 3.2.1), triglycerides (Fig. 3.2.2), lipoprotein profiles (Fig 3.2.3 - 3.2.5), ALAT (Fig. 3.2.6) and ketone bodies (Fig. 3.2.7). Figure 3.2.1 Plasma cholesterol Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 10 Plasma cholesterol (mmol/l) 8 * 6 4 2 *** ** * ** * 0 0 2 4 6 8 10 change diet groups 5 and 7 to high fat control diet Time (weeks) As seen in other animal studies, PFHS and PFOS strongly decreased plasma cholesterol. The reversal to control diet in the groups 5 and 7 resulted in a return to control values for 7 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 the PFOS group (7), however this return was delayed until week 8 for the PFHS group (5). The PFBS treatment did not decrease significantly the cholesterol levels. More or less the same results have been obtained for the plasma triglyceride (TG) levels. Figure 3.2.2 Plasma triglyceride Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS 2.5 Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS Plasma TG (mmol/l) 2.0 1.5 * 1.0 0.5 0.0 *** *** *** 0 2 4 6 8 10 change diet groups 5 and 7 to high fat control diet Time (weeks) PFHS and PFOS decreased TG levels, but also the highest concentration of PFBS (group 3) after 8 weeks of treatment; the lowest concentration of PFBS, group 2, was not different from control. TG levels rose again in the group 7 after return to the control diet, however it remained lower than the control TG level in week 8. Group 5 TG-levels turn back to control levels after the shift to control diet. The differentiation into lipoproteins showed more differences than observed with the lipid levels alone. The patterns have been determined by measuring the cholesterol contents of the different column fractions, which indicate all lipoprotein fractions (Figs 3.2.3a, 3.2.4a, 3.2.5a), and by measuring the phospholipid content, which is major a constituent of the HDL fractions (Figs 3.2.3b, 3.2.4b, 3.2.5b). The PFBS treatment resulted in only minor changes in the lipoprotein fractions in time. VLDL seems to be increased a little, but IDL/LDL and HDL are somewhat lower than these fractions in the control patterns. The low concentration of PFHS (group 4) showed the appearance of a lipoprotein fraction which is located before the HDL faction and half overlapping with the IDL/LDL fraction. Since phospholipids are a major constituent, this fraction has been designated as large, less-dense HDL. It might arise from an impaired clearance of smaller HDL by the liver. This fraction appeared to increase upon longer treatment with PFHS. Concomitantly, VLDL and IDL are strongly decreased, which might indicate a decrease in production/secretion of apoB-containing lipoproteins by the liver. The 4-week pattern of the group 5 animals (high conc. PFHS; after 2 weeks on control diet again) resembled very much the group 4 pattern with a major increase in the large HDL. However, after 6 and 8 weeks, the lipoprotein profiles went back to the control patters with again more VLDL, IDL and small HDL. 8 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 cholesterol (mmol/l) cholesterol (mmol/l) Figure 3.2.3a Week 0 1.2 VLDL 1.0 Pooled plasma lipoprotein (cholesterol) profiles ; PFBS treatment Week 4 Group 1: HF control Group 1: HF control Group 2: HF + 0.01 % PFBS Group 3: HF + 0.03 % PFBS IDL/LDL HDL 1.2 VLDL IDL/LDL HDL 1.0 cholesterol (mmol/l) 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0246 Week 6 Group 1: HF control 8 10 12 14 16 18 20 22 24 fraction Akta Group 2: HF + 0.01 % PFBS Group 3: HF + 0.03 % PFBS 1.2 VLDL IDL/LDL HDL 1.0 0.0 0246 Week 8 Group 1: HF control 8 10 12 14 16 18 20 22 24 fraction Akta Group 2: HF + 0.01 % PFBS Group 3: HF + 0.03 % PFBS 1.2 VLDL IDL/LDL HDL 1.0 cholesterol (mmol/l) 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 fraction Akta 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 fraction Akta Figure 3.2.3b Pooled plasma lipoprotein (phospholipid) profiles ; PFBS treatment Week 0 0.5 0.4 VLDL Group 1: HF control IDL/LDL HDL Week 4 Group 1: HF control 0.5 VLDL Group 2: HF + 0.01 % PFBS IDL/LDL 0.4 Group 3: HF + 0.03 % PFBS HDL phospholipids (mmol/l) Phospholipids (mmol/l) 0.3 0.3 0.2 0.2 0.1 0.1 0.0 0246 Week 6 Group 1: HF control 8 10 12 14 16 18 20 22 24 fraction Akta Group 2: HF + 0.01 % PFBS Group 3: HF + 0.03 % PFBS 0.5 VLDL IDL/LDL HDL 0.4 0.0 0246 Week 8 Group 1: HF control 8 10 12 14 16 18 20 22 24 fraction Akta Group 2: HF + 0.01 % PFBS Group 3: HF + 0.03 % PFBS 0.5 VLDL IDL/LDL HDL 0.4 phospholipids (mmol/l) phospholipids (mmol/l) 0.3 0.3 0.2 0.2 0.1 0.1 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 fraction Akta 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 fraction Akta 9 of 35 cholesterol (mmol/l) TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 Figure 3.2.4a Week 0 1.2 VLDL 1.0 Pooled plasma lipoprotein (cholesterol) profiles ; PFHS treatment Week 4 Group 1: HF control Group 1: HF control Group 4: HF + 0.006 % PFHS Group 5: HF + 0.02 % PFHS IDL/LDL HDL 1.2 VLDL IDL/LDL HDL 1.0 cholesterol (mmol/l) 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0246 Week 6 Group 1: HF control 8 10 12 14 16 18 20 22 24 fraction Akta Group 4: HF + 0.006 % PFHS Group 5: HF + 0.02 % PFHS 1.2 VLDL IDL/LDL HDL 1.0 0.0 0246 Week 8 Group 1: HF control 8 10 12 14 16 18 20 22 24 fraction Akta Group 4: HF + 0.006 % PFHS Group 5: HF + 0.02 % PFHS 1.2 VLDL IDL/LDL HDL 1.0 cholesterol (mmol/l) cholesterol (mmol/l) 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 fraction Akta 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 fraction Akta Phospholipids (mmol/l) Figure 3.2.4b Pooled plasma lipoprotein (phospholipid) profiles ; PFHS treatment Week 0 Group 1: HF control Week 4 Group 1: HF control Group 4: HF + 0.006 % PFHS Group 5: HF + 0.02 % PFHS 0.5 VLDL IDL/LDL HDL 0.5 VLDL IDL/LDL HDL 0.4 0.4 phospholipids (mmol/l) 0.3 0.3 0.2 0.2 0.1 0.1 0.0 0246 Week 6 Group 1: HF control 8 10 12 14 16 18 20 22 24 fraction Akta Group 4: HF + 0.006 % PFHS Group 5: HF + 0.02 % PFHS 0.5 VLDL IDL/LDL HDL 0.4 0.0 0246 Week 8 Group 1: HF control 8 10 12 14 16 18 20 22 24 fraction Akta Group 4: HF + 0.006 % PFHS Group 5: HF + 0.02 % PFHS 0.5 VLDL IDL/LDL HDL 0.4 phospholipids (mmol/l) 0.3 0.3 0.2 0.2 0.1 0.1 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 fraction Akta 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 fraction Akta 10 of 35 phospholipids (mmol/l) TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 Figure 3.2.5a Pooled plasma lipoprotein (cholesterol) profiles ; PFOS treatment Week 0 1.2 1.0 VLDL Group 1: HF control IDL/LDL HDL Week 4 Group 1: HF control Group 6: HF + 0.003 % PFOS 1.2 VLDL IDL/LDL 1.0 Group 7: HF + 0.01 % PFOS HDL cholesterol (mmol/l) cholesterol (mmol/l) 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0246 Week 6 Group 1: HF control 8 10 12 14 16 18 20 22 24 fraction Akta Group 6: HF + 0.003 % PFOS Group 7: HF + 0.01 % PFOS 1.2 VLDL IDL/LDL HDL 1.0 0.0 0246 Week 8 Group 1: HF control 8 10 12 14 16 18 20 22 24 fraction Akta Group 6: HF + 0.003 % PFOS Group 7: HF + 0.01 % PFOS 1.2 VLDL IDL/LDL HDL 1.0 cholesterol (mmol/l) cholesterol (mmol/l) 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 fraction Akta 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 fraction Akta Figure 3.2.5b Pooled plasma lipoprotein (phospholipid) profiles ; PFOS treatment Week 0 Week 4 Group 1: HF control Group 1: HF control Group 6: HF + 0.003 % PFOS Group 7: HF + 0.01 % PFOS 0.5 VLDL IDL/LDL HDL 0.5 VLDL IDL/LDL HDL 0.4 0.4 phospholipids (mmol/l) Phospholipids (mmol/l) 0.3 0.3 0.2 0.2 0.1 0.1 0.0 0246 Week 6 Group 1: HF control 8 10 12 14 16 18 20 22 24 fraction Akta Group 6: HF + 0.003 % PFOS Group 7: HF + 0.01 % PFOS 0.5 VLDL IDL/LDL HDL 0.4 0.0 0246 Week 8 Group 1: HF control 8 10 12 14 16 18 20 22 24 fraction Akta Group 6: HF + 0.003 % PFOS Group 7: HF + 0.01 % PFOS 0.5 VLDL IDL/LDL HDL 0.4 phospholipids (mmol/l) phospholipids (mmol/l) 0.3 0.3 0.2 0.2 0.1 0.1 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 fraction Akta 0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 fraction Akta 11 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 The lipoprotein profiles of the PFOS treated animals are remarkable. The low concentration (group 6) showed major the large-HDL fraction which is steadily increasing during the treatment. ApoB-containing lipoproteins have been decreased. Similar as with PFHS, VLDL and IDL fractions have been decreased. Group 7, which showed a recovery of plasma TC/TG levels after 2 weeks control diet, has VLDL levels similar to the control group again, but showed low IDL and a very large "large"-HDL peak, which remained the same over the rest of the treatment period. A possible explanation could be that liver VLDL production has recovered (or is enhanced), but lipoprotein clearance is still impaired. It all might be the result of the long half-life of PFOS in the body. This explains also the gradually increase of the "large"-HDL peak in time in the low concentration PFOS group 6, which in week 8 has become as large as this peak in the group 7 plasma. Since the liver seems to be strongly involved in the mechanism of action of PFHS and PFOS, the liver enzyme alanine aminotransferase (ALAT) has been determined a few times more than originally planned (Fig. 3.2.6). Figure 3.2.6 Pooled plasma ALAT activity Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 500 Pooled plasma ALAT (U/l) 400 300 200 100 0 week 0 week 4 week 6 (o/n fast) week 8 At all measure points ALAT has been increased in the PFOS-treated groups (groups 6 and 7), although from week 2 group 7 received the control diet again. PFHS increased ALAT as well (group 4), but here the return to control diet (group 5) caused a (partial) recovery of this activity. Although plasma ALAT activity is often used as a marker for liver damage, the increase in the plasma can be the result as well of an induction of the expression of the liver enzyme by the treatment, but can also result from the enlargement of the liver as seen in this study (see Fig. 3.7.1). PFBS had no or only a minor effect on plasma ALAT levels. 12 of 35 Plasma ketonbodies (mmol/l) TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 Since it was expected that the test compounds might have an effect on fatty acid metabolism, plasma ketone bodies have been determined before and after 6 weeks of treatment. As shown in Fig. 3.2.7, PFHS and, even more so, PFOS increased significantly the plasma ketone body levels. The two weeks of treatment with the highest dose, followed by 4 weeks of control diet (groups 5 and 7) still resulted in increased levels in week 6. PFBS did not significantly increase the ketone bodies. Figure 3.2.7 Plasma ketone bodies Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 5 ** 4 3 * * 2 1 0 week -1 week 6 So, PFHS and PFOS induced fatty acid oxidation, which is supported by the results of the analysis in metabolic cages performed in week 7 of treatment. 3.3. Measurements in metabolic cages After 7 weeks of treatment, 4 mice of the different treatment groups, the HF control group (group 1), the 0.03% PFBS group (group 3), the 0.006% PFHS group (group 4) and the 0.003% PFOS group (group 6) have been analyzed individually for 3 days in metabolic cages (after 1 day of acclimatization in the same cage). The following parameters have been measured: food and drink intake (Figs. 3.3.1.1 and 3.3.1.3), feeding size bouts (Fig. 3.3.1.2), moving activity (X activity; Fig. 3.3.2), O2-consumption and CO2-production (Figs. 3.3.3.1 and 3.3.3.2). From the latter parameters the respiratory exchange ratio (RER; Fig. 3.3.3.3), which gives information on the substrate that is used for energy generation, and additionally, the O2-consumption-related energy expenditure (Fig. 3.3.4) have been calculated. 3.3.1 Food and drink intake No significant differences have been observed between the groups in food intake, feeding bout size and drink intake, although the PFHS treated mice might have eaten somewhat more. 13 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 Figure 3.3.1.1 Food intake Group 1: HF control Group 4: HF + 0.006 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 12 Accumulated food intake (g) 10 8 6 4 2 0 0 12 24 36 48 60 72 Figure 3.3.1.2 Feeding bout size Time (h) Group 1: HF control Group 4: HF + 0.006 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 0.08 0.06 * Feeding bouts (g) 0.04 * 0.02 * * * ** 0.00 * Figure 3.3.1.3 0 12 24 Drink intake Group 1: HF control 36 Time (h) 48 60 72 Group 3: HF + 0.03 % PFBS Group 4: HF + 0.006 % PFHS Group 6: HF + 0.003 % PFOS 12 Accumulated drink intake (ml) 10 8 6 4 2 0 0 12 24 36 48 60 72 14 of 35 Time (h) TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 3.3.2 Moving (X) activity No significant differences have been observed between the groups in moving activity. Figure 3.3.2 X Activity Group 1: HF control Group 4: HF + 0.006 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS Average X activity per interval 1000 800 600 400 200 0 0 12 24 36 48 60 72 Time (h) 3.3.3 O2-consumption, CO2-production and respiratory exchange ratio (RER) It is remarkable to notice that differences in these parameters have been observed between the groups, while using rather small groups. O2-consumption of the PFHS and PFOS treated animals was significantly increased, while no difference between control and PFBStreated mice have been found. VO2 (ml/h/kg mouse) Figure 3.3.3.1 O2-consumption Group 1: HF control Group 4: HF + 0.006 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 5000 4000 3000 * * *** ** ** * ***** ** ** ** * * * ** * ** * **** ** ** 2000 1000 0 0 12 24 36 48 60 72 Time (h) 15 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 CO2-production of the PFHS and PFOS groups was found to be increased as well, although in much less time points compared to the increase in O2-consumption. For this reason the RER for these groups are lower (significantly for PFOS, but not yet statistically significantly for PFHS) than the RER of the control and PFBS-treated group. This is an indication that fat is preferentially used for energy generation over the use of carbohydrates. This is in agreement with an increase in fatty acid oxidation as reflected by an increased plasma ketone body level in PFHS- and PFOS-treated mice and also for the decrease in fat tissue as found later in this study. Figure 3.3.3.2 CO2-production Group 1: HF control Group 4: HF + 0.006 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 5000 VCO2 (ml/h/kg mouse) 4000 3000 * ** 2000 1000 0 0 Figure 3.3.3.3 1.2 12 24 36 48 60 72 Time (h) Respiratory exchange ratio (RER) Group 1: HF control Group 4: HF + 0.006 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS Respiratory exchange ratio 1.0 0.8 * **** 0.6 0.4 0 12 24 36 48 60 72 Time (h) 16 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 3.3.4 Energy expenditure Congruent with these observations it has been calculated that the O2-consumptiondependent energy expenditure has been increased as well by the PFHS and PFOS treatment. Figure 3.3.4 Energy expenditure Group 1: HF control Group 4: HF + 0.006 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 25 20 * *** ** ** * * *** *** * * ***** ** * 15 * * 10 Energy expenditure (kcal/hr/kg mouse) 5 0 0 12 24 36 48 60 72 Time (h) The differences observed in the metabolic cage analysis might be more pronounced by using larger groups in eventual future studies (preferable n = 8). 17 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 3.4. Fecal bile acid, sterol and fatty acid measurements In week 6 of treatment, feces have been collected and its lipid and bile acid contents determined. Fecal bile acids have been decreased by PFHS (0.006%; group 4) and even more by PFOS (0.003%). Although groups 5 and 7 were back on control diet for 4 weeks, both these compounds have still kept the bile acid excretion below the control values. PFBS-treatment seems to have increased the fecal bile acid excretion, which was even significantly different from the control group for group 2. Figure 3.4.1 Bile acid excretion Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 6 * 5 4 * * * 3 * 2 Total bile acid excretion (mol/100 gram mouse/day) 1 0 Table 3.4.1 Fecal bile acid composition group -muricholate deoxycholate cholate Group 1: HF control 6.1 2.8 14.7 5.7 7.9 2.7 Group 2: HF + 0.01 % PFBS 4.8 1.2 12.4 2.1 8.4 1.2 Group 3: HF + 0.03 % PFBS 4.2 1.3 12.8 1.2 7.6 2.3 Group 4: HF + 0.006 % PFHS 7.1 2.3 11.3 1.8 10.2 4.2 Group 5: HF + 0.02 % PFHS 5.3 0.8 13.9 2.4 8.5 3.7 Group 6: HF + 0.003 % PFOS 6.8 2.0 15.9 5.5 14.3 4.0 Group 7: HF + 0.01 % PFOS 10.8 2.7 22.1 5.5 16.4 4.9 yellow indicates a significant difference from control (p<0.05). % of bile acids lithocholate 9.6 4.1 10.7 2.4 13.1 4.5 10.9 3.3 16.0 5.2 11.3 3.2 6.1 2.4 -muricholate 20.4 5.0 19.6 7.2 19.9 4.1 17.2 4.5 16.8 5.7 12.5 3.5 16.8 4.0 -muricholate 39.8 8.8 42.2 4.4 40.9 3.4 40.9 3.8 36.6 7.0 37.0 5.7 24.2 4.0 hyodeo/urso 1.5 0.8 1.9 1.0 1.4 0.9 2.3 2.2 2.8 1.7 2.3 1.4 3.5 3.5 The latter increase with PFBS did not change the fecal bile acid composition. Concomitantly with the decrease by PFHS and PFOS, a change in composition of excreted bile acids has been observed as well, which can result from a change in bile acid synthesis pathways in the liver, but also from a change in intestinal bile acid conversion. The latter might also happen in the liver after reabsorption and return to the liver. 18 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 Only in group 4 (PFHS) was fecal neutral sterol excretion significantly increased. Fecal neutral sterols, which are primarily composed of cholesterol, are determined by food intake, intestinal absorption and liver production. Since the diet used contained 0.25% cholesterol, food intake plays a major role. This is more or less the same in all groups. Since intestinal cholesterol absorption is also the same or increased (see Fig.3.5.1), most probably the significantly increased fecal sterol excretion in group 4 (PFHS) is the result of an increased hepatic cholesterol synthesis. This is in agreement with the increased HMG-CoA reductase activity observed (see Fig. 3.8.2). Although no differences have been seen in the quantity of excreted plant sterols, which originate from the diet, the composition was slightly changed in the PFHS and PFOS groups. It might be the result of a change in intestinal absorption/resecretion. Figure 3.4.2 Sterol excretion Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 100 5 Total sterol excretion (mol/100 gram mouse/day) 80 * 4 60 3 40 2 20 1 0 0 neutral sterols phytosterols Table 3.4.2a Fecal neutral sterol composition group % of neutral sterols coprostanol coprostanon cholesterol cholestanol Group 1: HF control 0.0 0.0 0.0 0.0 94.4 0.4 2.7 0.2 Group 2: HF + 0.01 % PFBS 0.0 0.0 0.0 0.0 95.0 0.6 2.3 0.8 Group 3: HF + 0.03 % PFBS 0.0 0.0 0.0 0.0 94.6 0.7 2.7 0.4 Group 4: HF + 0.006 % PFHS 0.0 0.0 0.0 0.0 95.5 0.3 2.4 0.3 Group 5: HF + 0.02 % PFHS 0.0 0.0 0.0 0.0 94.4 1.4 3.1 1.5 Group 6: HF + 0.003 % PFOS 0.0 0.0 0.0 0.0 94.3 0.7 3.4 0.6 Group 7: HF + 0.01 % PFOS 0.0 0.0 0.0 0.0 93.7 0.7 4.6 0.6 lathosterol 2.9 0.3 2.7 0.3 2.7 0.6 2.1 0.1 2.5 0.2 2.4 0.2 1.7 0.1 19 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 Table 3.4.2b Fecal phytosterol composition group Group 1: HF control Group 2: HF + 0.01 % PFBS Group 3: HF + 0.03 % PFBS Group 4: HF + 0.006 % PFHS Group 5: HF + 0.02 % PFHS Group 6: HF + 0.003 % PFOS Group 7: HF + 0.01 % PFOS campesterol 26.2 0.8 26.2 1.0 26.5 0.7 27.9 1.1 26.6 1.1 27.6 0.8 27.9 1.0 % of phytosterols stigmasterol -sitosterol 8.2 1.1 65.6 1.1 8.1 1.7 65.7 1.5 7.7 1.9 65.9 1.7 4.3 2.0 67.7 1.3 6.5 2.7 66.9 2.1 3.6 0.5 68.8 0.8 5.0 3.5 67.1 2.9 -sitostanol 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Figure 3.4.3 Fatty acid excretion Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS 1000 Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS Total fatty acid excretion (mol/100 gram mouse/day) 800 600 * 400 * 200 0 Table 3.4.3 Fecal fatty acid composition group Group 1: HF control Group 2: HF + 0.01 % PFBS Group 3: HF + 0.03 % PFBS Group 4: HF + 0.006 % PFHS Group 5: HF + 0.02 % PFHS Group 6: HF + 0.003 % PFOS Group 7: HF + 0.01 % PFOS C14:0 2.0 0.4 1.9 0.5 1.8 0.3 1.7 0.3 2.0 0.5 1.8 0.4 2.2 0.4 C16:0 46.0 3.9 44.1 2.7 45.1 2.3 43.6 1.9 45.5 3.2 45.6 2.9 48.2 2.4 C16:1 0.38 0.04 0.35 0.13 0.39 0.04 0.35 0.02 0.39 0.04 0.34 0.03 0.35 0.03 % of fatty acids C18:0 46.0 3.7 47.5 2.9 47.0 2.4 48.6 1.9 46.2 3.2 46.6 3.0 43.5 2.7 C18:1 4.8 0.5 5.2 0.6 4.8 0.3 4.8 0.4 5.1 0.7 4.8 0.3 4.9 0.5 C18:2 0.30 0.04 0.30 0.05 0.31 0.02 0.31 0.03 0.29 0.03 0.29 0.03 0.38 0.00 C18:3 0.55 0.21 0.62 0.10 0.58 0.08 0.60 0.07 0.57 0.10 0.56 0.09 0.46 0.06 Fecal fatty acids are derived from the fat from food intake (TG) and lipids produced by the liver (PL). Most of the groups (some have large SD) showed no significant difference from control. Group 4 (PFHS) had significantly increased fecal fatty acids, whereas group 7 (2 weeks PFOS and 4 weeks control) showed a strong decrease, with a concomitant small change in fatty acid composition. 20 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 From the difference between the total fatty acid (TG) intake per subgroup and total fatty acid excreted in the feces, the fatty acid balance has been calculated. The values are expressed as the difference between total fatty acid input and output in mol fatty acid per 100 gram mouse per day. As shown in Fig. 3.4.5, all groups have more or less the same fat intake, except that in group 7 this is significantly higher. This is of interest in the light of the loss of fat tissue in the same group (see Fig. 3.6.1). Figure 3.4.5 Fatty acid balance Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS 8000 Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 6000 * Fatty acid balance (mol/100 gram mouse/day) 4000 2000 0 21 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 3.5. Intestinal cholesterol absorption The intestinal cholesterol absorption was assessed by the fecal dual-isotope method. In week 8 all mice were individually housed and received an oral dose of 200 l of olive oil, containing 1 Ci [14C]-cholesterol and 1 Ci [3H]-sitostanol, per mouse. Bodyweight and food intake were monitored, and feces were collected for the following 4 days. Feces were homogenized, pooled per mouse over the 4-day period, and dissolved in ethanolic potassium (3mol/l, 60% ethanol). Radioactivity was determined in the fecal samples to assess the ratio of [14C]-labeled cholesterol and [3H]-labeled sitostanol. Sitostanol was used as the reference compound since it is known to be poorly absorbed (< 3 %) in mice. The formula used to calculate cholesterol absorption was: % cholesterol absorption = ([14C]/[3H]-dosing mixture - [14C]/[3H]-feces) / [14C]/[3H]-dosing mixture * 100. Figure 3.5.1 Intestinal cholesterol absorption Values are means S.D. Intestinal cholesterol absorption efficiency (%) Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS 60 50 Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS * * 40 30 20 10 0 week 8-9 Both PFOS-treated groups 6 and 7 have increased intestinal cholesterol absorption. Since all groups have about the same cholesterol intake via the food and the same fecal cholesterol output (except group 4), this increase must be caused by an increased hepatic cholesterol synthesis. Indeed, the increased HMG-CoA reductase activity in these groups (see Fig. 3.8.2b) suggests an increased hepatic cholesterol synthesis. Although not significantly, the intestinal cholesterol absorption of group 4 (PFHS) seems to be slightly increased as well, which, beside additionally an increased fecal excretion, might be explained as well by an increased hepatic cholesterol synthesis. 22 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 3.6. Perigonadal fat tissue weight Figure 3.6.1 Perigonadal fat weight Values are means S.D. 04-12-2006 Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS 1.2 Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS Perigonadal fat weight (g) 1.0 0.8 0.6 0.4 * * * 0.2 0.0 week 10 PFHS (group 4) and PFOS (groups 6 and 7) strong reduced perigonadal fat tissue. Although not measured, this might reflect a decrease in total body fat tissue. This notion is supported by the decrease in body weight. Plasma leptin levels might be measured, which should be found to be decreased as a reflection of decreased fat tissue. The reduction in perigonadal fat tissue was not seen in the PFBS-treated groups and the PFHS group 5. 3.7. Liver weight and hepatic lipid contents Livers are strongly increased by PFOS (groups 6 and 7) and PFHS (group 4). The return to control diet of the group 5 PFHS-treated animals caused a return to only slightly but still significantly enlarged livers. When expressed as % of total body weight (Fig. 3.7.1b), the PFBS-treated group 3 also had slightly but significantly enlarged livers when compared to controls. As shown in Fig. 3.7.2, the lipids, triglycerides and cholesterylesters, have been increased in the livers of PFOS-treated groups 6 and 7. In the other livers these contents were comparable with that of the control mice. Free cholesterol contents have not been changed by any of the compounds. 23 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 Figure 3.7.1a Liver weight Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS 6 Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS Liver weight (g) 5 * 4 ** 3 * 2 1 0 week 10 Liver weight as % of bodyweight (%) Figure 3.7.1b Liver weight as % of body weight Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 20 * 15 * * 10 * * 5 0 week 10 24 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report Figure 3.7.2 Lipid contents liver Values are means S.D. 04-12-2006 Lipid contents liver (g/mg protein) Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 300 250 * 200 * 150 100 ** 50 0 Free cholesterol Triglycerides Cholesterolester 3.8. Liver enzyme activities: HMG-CoA reductase, cholesterol 7-hydroxylase and sterol 27-hydroxylase. Immediately after sacrifice, livers have been removed and microsomes and mitochondria have been isolated from the liver tissue. Thereafter, HMG-CoA reductase and cholesterol-7hydroxylase activity have been determined in the microsomal preparations and sterol-27hydroxylase activity in the mitochondria. The yield of microsomes per g liver is about the same in all groups (Fig. 3.8.1) Fig. 3.8.1 Microsomal preparations Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 20 isolated microsomes per gram liver (mg/g) 15 10 5 0 25 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 Although HMG-CoA reductase activity was rather low in all preparations and therefore SD's are rather large, PFHS and PFOS seem to have increased this activity. This is significantly different from control for PFHS group 4. Since the yield of the microsomes per g liver is more or less the same and large differences between the liver weights have been observed, the activity has been recalculated and expressed as total activity per liver (Fig. 3.8.2b). This might indicate that large differences in total hepatic cholesterol synthesis between the groups exist. Total HMGCoA reductase activity per liver has been increased significantly in groups 4 (PFHS), 6 (PFOS) and 7 (PFOS). Figure 3.8.2a HMG-CoA reductase activity in liver microsomes Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS 20 * 15 Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS HMG CoA reductase activity (pmol/min/mg protein) 10 5 0 Figure 3.8.2b Total HMG-CoA reductase activity per liver Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS 1000 Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 800 * Total HMG CoA reductase activity per liver (pmol/min) 600 * 400 * 200 0 26 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 Figure 3.8.3a Cholesterol 7-hydroxylase activity in liver microsomes Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 8 6 7 alpha hydroxylase activity (nmol/h/mg protein) 4 * 2 * * 0 Figure 3.8.3b Total cholesterol 7-hydroxylase activity per liver Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 120 Total 7 alpha hydroxylase activity per liver (nmol/h) 100 80 * 60 * * * 40 20 0 One of the rate limiting enzymes of bile acid synthesis, the microsomal enzyme cholesterol 7hydroxylase, has found to be decreased by PFHS (group 4) and by PFOS (groups 6 and 7). The activity in the other groups did not differ significantly from the control value (although that of group 2 seems to be somewhat enhanced). If the activity is expressed as total activity per liver a significant decrease by PFHS (group 4) and PFOS (group 6) remained, suggesting that total hepatic bile acid production might be reduced as well. 27 of 35 isolated mitochondria per gram liver (mg/g) TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 The other regulatory enzyme of bile acid synthesis, sterol 27- hydroxylase was measured in the mitochondrial preparations. It was remarkable that the yield of mitochondria (mg/g liver; Fig. 3.8.4) seemed to be increased in the PFHS-treated and PFOS-treated groups, suggesting that these compounds increased the number of mitochondria. This might be related with the observed increase of fatty acid oxidation. Figure 3.8.4 Liver mitochondrial preparations Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 40 * 30 20 * * 10 0 As expresses per mg of mitochondrial protein sterol 27-hydroxylase was found to be decreased by PFHS (group 4) and PFOS (groups 6 and 7) as well. Figure 3.8.5a sterol 27-hydroxylase activity in liver mitochondria Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS 20 27 hydroxylase activity (nmol/h/mg protein) 15 10 * * * 5 0 28 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 However, taking into account of the differences in mitochondrial yields and in liver weights and expressing the activity per liver (Fig. 3.8.5b), the total sterol 27-hydroxylase activity was found to be increased in the same groups. Figure 3.8.5b Total sterol 27-hydroxylase activity per liver Values are means S.D. Group 1: HF control Group 4: HF + 0.006 % PFHS Group 7: HF + 0.01 % PFOS Group 2: HF + 0.01 % PFBS Group 5: HF + 0.02 % PFHS 800 * 600 Group 3: HF + 0.03 % PFBS Group 6: HF + 0.003 % PFOS * * 400 200 0 Total 27 hydroxylase activity per liver (nmol/h) 29 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 4. RESULTS IN VITRO EXPERIMENT (RAT HEPATOCYTES) From the first in vitro experiments performed in project 3M-1, it was suggested that the PFAS compounds might influence the internal acetate pool, which is the substrate for hepatic cholesterol synthesis. Since fatty acid catabolism is a source for acetate, the effect of the test compounds on the incorporation of different sized radiolabelled fatty acids into cholesterol has been investigated in the present project. As a positive control, a strong and specific inhibitor of cholesterol synthesis, rosuvastatin has been taken along in each separate experiment. Identical experiments have been performed 3 - 4 times (as mentioned at the figures) Figure 4.1 Incorporation of [14C]-acetate into cholesterol in rat hepatocytes in vitro Values are means S.D. (n = 4). [14C]-Acetate incorporation 200 150 Rosuvastatin PFBS PFHS PFOS 100 Cholesterol synthesis (% of control) 50 0 0.001 0.01 0.1 1 10 100 compound concentration (mol/l) 1000 10000 Figure 4.2 Incorporation of [14C]-butyrate into cholesterol in rat hepatocytes in vitro Values are means S.D. (n = 4). [14C]-Butyrate incorporation Cholesterol synthesis (% of control) 250 200 150 Rosuvastatin PFBS PFHS PFOS 100 50 0 0.001 0.01 0.1 1 10 100 compound concentration (mol/l) 1000 10000 30 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 Rosuvastatin strongly inhibited the incorporation of the short chain fatty acids acetate and butyrate into cholesterol (Figs. 4.1 and 4.2). Although there is a large variation in the data obtained for the test compounds in the individual experiments, on average there seems to be hardly an effect of these compounds on cholesterol synthesis from acetate and butyrate. The incorporation of the longer fatty acid [14C]-laurate into cholesterol is also comparable to control values in the presence of the PFAS's (Fig. 4.3); however, there might be a slight concentration dependent decreasing effect of PFHS and PFOS. Rosuvastatin is again a much stronger inhibitor if the cholesterol synthesis from [14C]-laurate and also from [14C]palmitate (Fig. 4.4). Figure 4.3 Incorporation of [14C]-laurate into cholesterol in rat hepatocytes in vitro Values are means S.D. (n = 3). [14C]-Laurate incorporation 200 150 Rosuvastatin PFBS PFHS PFOS 100 Cholesterol synthesis (% of control) 50 0 0.001 0.01 0.1 1 10 100 compound concentration (mol/l) 1000 10000 Figure 4.4 Incorporation of [14C]-palmitate into cholesterol in rat hepatocytes in vitro Values are means S.D. (n = 3). [14C]-Palmitate incorporation 150 100 Rosuvastatin PFBS PFHS PFOS Cholesterol synthesis (% of control) 50 0 0.001 0.01 0.1 1 10 100 compound concentration (mol/l) 1000 10000 31 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 The incorporation of palmitate, the longest fatty acid used in this study, into cholesterol was influenced in a somewhat remarkable way. In the presence of the different concentrations of the test compounds on average all values are lower than control values. But there is no obvious concentration dependency in this decrease. So, it is difficult to understand that all three compounds are influencing palmitate uptake/transport/degradation in the same concentration independent way. Since the solubility of this longer fatty acid in the medium is depending on several factors and the radiolabelled compound was present in tracer amounts, it might be that the test compounds have influenced its solubility and hence its availability for uptake in the cells. So, all together the test compounds have either no or only slight effects on the incorporation of short and longer radiolabelled fatty acids into cholesterol in rat hepatocytes under the conditions used. 5. SUMMARY AND CONCLUSIONS Although the test compounds have similar chemical structures, they show differences in their biological behavior. For that reason they are discussed here separately. 5.1. Effects of PFBS in ApoE3Leiden mice Both PFBS-treated groups 2 (0.01%) and 3 (0.03%) showed no or small differences with the control group 1. No differences were found in body weight, food intake, plasma ALAT and ketone bodies, metabolic parameters as measured in metabolic cages, fecal sterol and fatty acid excretion, intestinal cholesterol absorption, fat tissue weight and liver weight, lipid contents and enzyme activities. Small differences have been seen in plasma lipid levels: a slight decrease in cholesterol and triglycerides, which was reflected in minor changes of the lipoprotein patterns; VLDL seemed to be increased a little, but IDL/LDL and HDL were somewhat lower than these fractions in the control patterns. Interesting is the small but significant increase in fecal bile acid excretion in the group 2 mice. These mice showed also an increased hepatic cholesterol 7-hydroxylase activity which might support the preceding notion, but because of the large standard deviation in this value it was not statistically significant. The suggested small increase in bile acid synthesis might be related to the slightly decreased plasma lipid levels. 5.2. Effects of PFHS in ApoE3Leiden mice The group 4 animals have been treated with 0.006% of PFHS throughout the 10 weeks of the study period. The higher dosed group 5 (0.02% of PFHS) showed already from the start deviations in food intake, which led to a large loss of body weight in the 2nd week of treatment. Therefore, from week 2, these mice were put back on control diet and remained on that throughout the rest of the treatment period. This resulted in a recovery of these animals, which, after a shorter or longer period, showed similar values as measured in the control group; however, some differences remained. The PFHS-treatment in group 4 had the following effects: Although food intake did not change, the body weight steadily decreased in the last 3 weeks. This can be explained by the prominent loss in fat tissue (measured as perigonadal fat mass; it might be better to measure plasma leptin as a marker for whole body fat). The latter effect was probably been caused by increased fatty acid oxidation, what can be concluded from increased plasma ketone bodies and a preference for fat oxidation over carbohydrate oxidation, as measured in the metabolic cage analysis (tendency towards decreased RER; group size was rather small). Additionally, O2consumption and concomitant energy expenditure was enhanced in these animals. Another indication for increased oxidation is the increased quantity of mitochondria, 32 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 isolated from the significantly enlarged livers. For further support of increased fatty acid oxidation, hepatic activity and/or mRNA's of enzymes involved in fatty acid oxidation, e.g. Acyl-CoA oxidase, should be measured. Plasma cholesterol and triglycerides were strongly decreased by PFHS throughout the treatment period, which was paralleled by a change in lipoprotein profile: ApoB-containing fractions (VLDL, IDL and LDL) were decreased; whereas, a shift in HDL towards larger, less dense particles was observed. These phenomena can be explained by a decrease in hepatic VLDL production and decreased uptake/conversion of HDL by the liver. In summary, liver functions have been changed, which is further supported by an increase in liver size/weight (liver lipid contents did not change), increased plasma ALAT values, increased HMG-CoA reductase activity and decreased cholesterol 7-hydroxylase, which are rate limiting enzymes in cholesterol and bile acid syntheses, respectively. Mitochondrial sterol 27-hydroxylase activity was changed as well. Changes in hepatic cholesterol/bile acid synthesis might have influenced the fecal lipid excretion and intestinal cholesterol absorption. PFHS increased fecal cholesterol and fatty acid excretion and decreased fecal bile acid excretion. On the other hand, changes in intestinal uptake/conversion and small changes in food intake cannot be ruled out in this respect. There is a trend observed towards increased intestinal cholesterol absorption by PFHS. Several of the liver effects of PFHS are similar to those of a PPAR agonist, e.g., like fibrates. For this reason it might be of use to further investigate the expression of other genes which are regulated by PPAR. As known, such compounds lower plasma lipids and decrease fat tissue as well, but their effect on intestinal functions is not known. As mentioned above, in the group 5 animals with a 2 weeks-treatment at the highest dose of PFHS, followed by 8 weeks on control diet again, the effects by PFHS as seen in group 4 disappeared, and the parameters measured returned more or less rapidly to control values. That was not the case in all instances the case: The livers remained a little enlarged and plasma ALAT stayed somewhat higher than control. Ketone bodies remained significantly increased up to week 6, and fecal bile acid excretion was still reduced. So it might be that during this period PFHS was not yet totally eliminated from the mice. The values of plasma PFHS concentrations, which will be determined by 3Mcompany, will give inside in this matter. 5.3. Effects of PFOS in ApoE3Leiden mice The effects of PFOS (group 6; 0.003%) on the parameters measured in ApoE3Leiden mice in this study were very similar to those of PFHS, however, the effects were in several instances more pronounced. That means that group 6 mice showed: Loss of (perigonadal) fat tissue, strongly increased plasma ketone bodies and a decreased RER, which showed a preference for fat oxidation over carbohydrate oxidation, and a concomitantly increased energy expenditure; all indications for an increased fatty acid oxidation. Additionally, increased quantities of mitochondria, have been isolated from the livers of these animals. Plasma cholesterol and triglycerides were strongly decreased, with a similar change in lipoprotein profile as observed with PFHS. ApoB-containing fractions were decreased; whereas, a shift in HDL towards larger, less dense particles was observed. Interestingly, the "large"-HDL peak increased in time, suggesting that PFOS is accumulating during the treatment period in this group. 33 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report 04-12-2006 Furthermore, the liver weight increased almost 3 times with concomitantly increased plasma ALAT levels. Liver lipids, triglycerides and cholesteryl esters were significantly increased. Hepatic microsomal HMG-CoA reductase activity was increased and cholesterol 7-hydroxylase and mitochondrial sterol 27-hydroxylase activities were decreased. PFOS did not change fecal cholesterol and fatty acid excretion, and decreased fecal bile acid excretion and composition. On the other hand, the ratio of different neutral sterols in the feces has been changed, and PFOS increased significantly intestinal cholesterol absorption. The latter might be a reflection of increased hepatic cholesterol synthesis, but other changes in intestinal cholesterol metabolism cannot be ruled out. As with PFHS, the PFOS mechanism of action might be (partly) understood as working like a PPAR agonist. This notion needs further investigation. The observation in group 7 (2 weeks of a higher dose of PFOS followed by 10 weeks on control diet) are almost identical to those in group 6. And although cholesterol levels seem to recover after 4 weeks, the lipoprotein pattern continued to contain the same large "large"-HDL fraction. These all are indications that PFOS concentrations in the mice remained at a similar level throughout the whole period and that PFOS is not eliminated readily from the body. PFOS is known to have a long elimination half-life in rats, monkeys, and humans, and determination of the plasma PFOS-concentrations at the end of the treatment period, should confirm this presumption. The only remarkable difference between group 6 and 7 is found in the fecal excretion data. Much less fatty acids have been found in the feces of group 7, which also possess a slightly different composition of fatty acids. So, intestinal fatty acid absorption (and/or conversion) seemed to be increased when PFOS is withdrawn from the food. Table 5.1 gives an overview of the effects of PFBS, PFHS and PFOS treatment on the different measured parameters. 34 of 35 TNO-report of study 3M-02 PFAS in ApoE3Leiden mice Final report Table 5.1: Overview of effect of PFBS, PFHS and PFOS treatment on measured parameters Measured parameter Bodyweight Food intake Total Cholesterol Total Triglycerides Lipoprotein profile ALAT Ketone bodies Fecal bile acids Fecal neutral sterols Fecal fatty acids Fatty acid balance Metabolic cage analysis: O2-consumption CO2-production Respiratory exchange ratio Energy expenditure Intestinal cholesterol absorption Weight perigonadal fat Liver weight Liver weight as % of bodyweight Liver lipids: Free cholesterol Triglycerides Cholesterolester Microsomes: amount HMG CoA reductase per mg protein HMG CoA reductase per liver Cholesterol 7--hydroxylase per mg protein Cholesterol 7--hydroxylase per liver Mitochondria: amount Cholesterol 27-hydroxylase per mg protein Cholesterol 27-hydroxylase per liver PFBS 0.01% PFBS 0.03% PFHS 0.006% - - - slight decrease in LDL and HDL (n.s.) - slight decrease in LDL and HDL (n.s.) - - - - 1.3 x - - - - - - - - - - - - - - - - - 1.2 x - - - - - - - - - - - - - - - - - - - - - - (-36%) (-50%) decrease in HDL, shift to LDL, decrease in VLDL/IDL 2-4.5 x 1.4 x (-25%) 1.5 x 1.2 x - slight increase (n.s.) slight decrease (n.s.) (-53%) 2.1 x 2.2 x - - 2.7 x 5.9 x (-60%) (-29%) 1.7 x (-45%) 2 x PFHS 0.02% only 2 weeks on diet (-38%)* (-54%)* decrease in HDL, shift to LDL, decrease in VLDL/IDL** 1.5-2 x 1.5 x (-18%) - - 1.2 x 1.2 x - PFOS 0.003% (-31%) (-22%) decrease in HDL, shift to LDL, decrease in VLDL/IDL 3-5 x 2.5 x (-48%) - - slight increase (n.s.) 1.2 x (-61%) 2.4 x 2.7 x - 1.4 x 1.8 x - 4.5 x (-77%) (-46%) 1.2 x (-41%) 1.8 x PFOS 0.01% only 2 weeks on diet (-46%)* (-50%)* decrease in HDL, shift to LDL, decrease in VLDL/IDL** 3-5 x 2.7 x (-17%) - (-55%) 1.2 x 1.3 x (-54%) 2.3 x 2.5 x - 1.5 x 1.9 x - 3.8 x (-65%) - 1.2 x (-29%) 2.1 x *: measured after 1 week of treatment **: measured after 2 weeks after changing diet to control diet n.s.: not significant 04-12-2006 35 of 35