Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Influence of diet fat saturation on rates of cholesterol synthesis and esterification in healthy young… Mazier, Marie Jeanne Patricia 1994

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1995-983221.pdf [ 12.87MB ]
Metadata
JSON: 831-1.0088878.json
JSON-LD: 831-1.0088878-ld.json
RDF/XML (Pretty): 831-1.0088878-rdf.xml
RDF/JSON: 831-1.0088878-rdf.json
Turtle: 831-1.0088878-turtle.txt
N-Triples: 831-1.0088878-rdf-ntriples.txt
Original Record: 831-1.0088878-source.json
Full Text
831-1.0088878-fulltext.txt
Citation
831-1.0088878.ris

Full Text

INFLUENCE OF DIET FAT SATURATION ON RATES OF CHOLESTEROL SYNTHESIS AND ESTERIFICATION IN HEALTHY YOUNG MEN by MARIE JEANNE PATRICIA MAZIER BSc Acadia University, 1978 BSc University of Guelph, 1987 MSc Universite Laval, 1989  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY  in THE FACULTY OF GRADUATE STUDIES DIVISION OF HUMAN NUTRITION  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA October 1994 (c) Marie Jeanne Patricia Mazier, 1994  In presenting this thesis in partial FULFILLMENT of the requirements for an advanced degree at The University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  Division of Human Nutrition The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date:  28 October 1994  ABSTRACT To examine the effect of diet fat type on rates of cholesterol synthesis and esterification during feeding and fasting, nine healthy male subjects were fed solid-food diets of 40% fat as predominantly either olive oil (MONO), safflower-oil margarine (POLY), or butter (SAT).  At the end of each two-week diet trial, subjects were given  deuterium (D) oxide orally and de novo synthesis was measured from D incorporation into cholesterol and interpreted as rates of fractional synthesis (FSR) (pools/day) into the rapidly exchangeable free cholesterol (FC) pool.  Absolute synthesis rates (ASR) were calculated  as the product of FSR and the FC pool.  Pool size for each subject was  obtained from analysis of the specific activity decay curve of an intravenous injection of 4-14C-cholesterol over nine months.  Synthesis  was measured over two consecutive 12-h fed periods followed by two consecutive 12-h fasted periods.  Serum samples were also assayed for  lathosterol concentration, an index of cholesterol synthesis.  Serum  cholesterol and non-HDL cholesterol concentrations were highest on the SAT diet, lowest (P<0.001) on the POLY diet and intermediate on the MONO diet, triglyceride levels were greater (P<0.03) on the SAT diet than on the POLY diet, and HDL levels were lowest (P<0.05) on the SAT diet and highest on the MONO diet. Cholesterol D enrichment and FSR during each 12-h period were greater (P<0.014) on the POLY diet than on the SAT diet; MONO enrichment and FSR were not significantly different from those on the other two diets. Similar results were obtained for rates of  ii  cholesterol esterification (P<0.001).  Deuterium enrichment data  suggested, and lathosterol data confirmed, that free cholesterol synthesis was greater during the fed period than during the fasted period (P<0.01); however, this could not be confirmed for rates of cholesterol esterification.  Results suggest that POLY fat feeding  augments de novo cholesterol synthesis without adverse effects on total serum cholesterol concentrations, and that the deleterious effects of SAT fat on serum cholesterol are not brought about by augmented de novo synthesis.  Finally, the combination of deuterium incorporation and  mathematical modelling produces estimates of daily cholesterol synthesis which are compatible with those invoked by more laborious techniques.  iii  TABLE OF CONTENTS  ABSTRACT  ii  TABLE OF CONTENTS  iv  LIST OF ABBREVIATIONS  x  LIST OF TABLES  xiv  LIST OF FIGURES  xviii  LIST OF APPENDICES  xxii  ACKNOWLEDGEMENTS  xxiv  DEDICATION  1  xxv  INTRODUCTION 1.1  1  An overview of studies linking changes in serum cholesterol concentration to the composition of diet fat intake  1.2  1  Potential mechanisms of diet fat's effect on circulating cholesterol levels  6  1.2.1 Faecal sterol excretion  6  1.2.2 Cholesterol absorption  9  1.2.3 Cholesterol synthesis  10  1.2.4 Lipoprotein composition and catabolism  13  1.2.5 Lipoprotein receptors  17  iv  1.2.6 Dietary cholesterol precursors and non-cholesterol sterols 1.3  18  Cholesterol metabolism  19  1.3.1 Production rate, turnover, and mass of exchangeable cholesterol  20  1.3.2 Origin and turnover of plasma cholesterol  23  1.3.3 Plasma cholesterol turnover relative to whole-body cholesterol turnover 1.4  1.5  1.6  27  In vivo measurement of cholesterol synthesis  30  1.4.1 Sterol balance method  30  1.4.2 Specific activity decay curves  30  1.4.3 Biochemical assays  31  1.4.4 Measurement of rate of precursor incorporation  32  1.4.4.1 Deuterium incorporation  32  1.4.4.2 Other tracers as precursors  33  1.4.5 Mass isotopomer distribution analysis  34  Model-based compartmental analysis  34  1.5.1 Computer models  35  1.5.2 SAAM/CONSAM  39  1.5.3 Cholesterol models  42  Effect of diet fat type on rates of cholesterol synthesis and esterification  46  1.6.1 Hypotheses  46  1.6.2 Rationale  46  v  1.6.3 Significance  48  1.7  Thesis objectives  50  1.8  Specific aims  50  1.8.1 Measurement of cholesterol turnover in humans  50  1.8.2 Measurement of rates of cholesterol synthesis and esterification in humans 1.8.3 Cholesterol model development  2  51 51  MATERIALS AND METHODS  52  2 .1  52  Chemicals  2.2  Equipment  2 .3  Study diets  54  2.4  Subject selection  57  2.5  Cholesterol turnover  57  2.5.1 Preparation and delivery of inj ectate  57  2.5.2 Sampling protocol  58  2.5.3 Measurement of serum specific activity  59  2.5.4 Measurement of total serum cholesterol  59  Cholesterol synthesis  60  2.6.1 Experimental protocol  60  2.6.2 Measurement of serum lipids  61  2.6.3 Measurement of serum cholesterol deuterium uptake  61  2.6  53  2.6.3.1 Validation of serum sample size  61  2.6.3.2 Extraction of total lipid  62  vi  2.6.3.3 Separation and isolation of lipid fractions  2.7  ...62  2.6.3.4 Saponification of cholesterol ester fraction  ..63  2.6.3.5 Combustion tube transfer and water evolution  ..63  2.6.3.6 Distillation and reduction  64  2.6.3.7 Mass spectrometry  65  2.6.3.8 Measurement of body water deuterium enrichment  66  2.6.4 Measurement of serum lathosterol concentration  67  Modelling data sets  69  2.7.1 Modelling turnover data  69  2.7.2 Modelling free cholesterol synthesis data  70  2.7.2.1 Linear uptake of tracer  .70  2.7.2.2 Compartmental modelling based on monoexponential tracer uptake  72  2.7.3 Modelling cholesterol ester data 2.7.3.1 Linear uptake of tracer  74 74  2.7.3.2 Compartmental modelling based on monoexponential tracer uptake 2.8  74  Statistical analyses  74  2.8.1 Subj ects' body weight and energy intake  75  2.8.2 Diet configuration  75 .  2.8.3 Serum lipids  76  2.8.4 Specific activity decay curves  76  2.8.5 Serum lathosterol  76  2.8.6 Validation of serum sample size for mass spectrometry  vii  .77  3  2.8.7 Cholesterol deuterium incorporation  77  2.8.8 Compartmental models and model parameters  78  RESULTS  79  3.1  Subjects' body weight, percent body fat, and energy intake  3 .2  Diet configuration  79  3.3  Serum lipids  79  3.4  Serum lathosterol  81  3.5  Validation of serum sample size for mass spectrometry  81  3.6  Specific activity decay curves  96  3.7  Deuterium incorporation  97  3.7.1 Free cholesterol  3.8  ..79  97  3.7.2 Esterified cholesterol  97  3.7.3 Body water  98  Compartmental models and model parameters  113  3.8.1 Free cholesterol data  113  3.8.1.1 Linear incorporation of tracer  113  3.8.1.2 Monoexponential incorporation of tracer  114  3.8.2 Esterified cholesterol data  139  3.8.2.1 Linear incorporation of tracer  139  3.8.2.2 Monoexponential incorporation of tracer  139  3.8.3 Correlation of synthesis rates with serum lathosterol .140 3.8.3.1 Linear incorporation of tracer  141  3.8.3.2 Monoexponential incorporation of tracer  141  viii  4  DISCUSSION  164  4.1  A brief review of study goals  164  4 .2  Models  165  4.2.1 Model assumptions and ambiguities  165  4.2.2 Model comparisons  168  4.3  4.4  4.5  4.6  4.2.2.1 Linear incorporation of tracer  168  4.2.2.2 Monoexponential incorporation of tracer  170  4.2.2.3 Conclusion  173  The effect of diet fat saturation  173  4.3.1 De novo cholesterol synthesis  173  4.3.2 Cholesterol esterification  177  The effect of feeding versus fasting  179  4.4.1 De novo cholesterol synthesis  179  4.4.2 Cholesterol esterification  180  Summary  182  4.5.1 Cholesterol production  182  4.5.2 Dietary recommendations  182  Conclusions  18 3  REFERENCES  185  APPENDICES  209  ix  LIST OF ABBREVIATIONS ACAT  acyl-CoA:cholesterol acyltransferase  ANCOVA  analysis of covariance  ANOVA  analysis of variance  APE  atom percent excess  apo  apoprotein  °C  degrees Celsius  CE  cholesterol ester  CETP  cholesterol ester transfer protein  CHD  coronary heart disease  CHO  carbohydrate  Choi  cholesterol  cm  centimetre  CONSAM  conversational simulation, analysis, and modelling  CV  coefficient of variation  d  day  dL  decilitre  D  deuterium  DPM  disintegrations per minute  DPS  digitonin-precipitated sterols  EME  expected maximum enrichment  FC  free cholesterol  FCR  fractional catabolic rate  FER  fractional esterification rate  FSR  fractional synthesis rate  g  unit of acceleration  g  gram  GISP  Greenland Ice Sheet Precipitation  glc  gas liquid chromatograph  h  hour  HDL  high density lipoprotein  HMG CoA  3-hydroxy-3-methylglutaryl Coenzyme A  2 H20  Deuterated water  ht  height  IDL  intermediate density lipoproteins  k  fractional turnover rate  kg  kilogram  kJ  kilojoule  KOH  potassium hydroxide  1  rate of transfer between pools  L  litre  Lath  lathosterol  LCAT  lecithin:cholesterol acyltransferase  LDL  low density lipoprotein  M  mass  m  metre  mg  milligram  min  minute xi  mL  millilitre  mm  millimetre  mmol  millimole  MNL  mononuclear leukocytes  MONO  monounsaturated fatty acid  n  number  NADPH  nicotinamide adenine dinucleotide phosphate (reduced)  nm  nanometer  °/oo  parts per thousand  P  statistical probability of differences not existing  PL  phospholipid  POLY  polyunsaturated fatty acid  PR  production rate  P:S  polyunsaturated to saturated ratio  R  rate of mass transfer  REE  resting energy requirements  SAAM  simulation, analysis, and modelling  SAT  saturated fatty acid  SD  standard deviation  SE  standard error  SLAP  Standard Light Antarctic Precipitation  SMOW  Standard Mean Ocean Water  sub  subject  t  time  TG  triglyceride  TME  theoretical maximum enrichment  /iCi  microcurie  fj.1  microlitre  /im  micrometre  AtM  micromolar  u  daily cholesterol production in the central pool  v  volume  VLDL  very low density lipoprotein  wt  weight  X  mean  Zn  zinc  Xlll  LIST OF TABLES Table 3.1  Subject body weight at beginning and end of two-week diet  periods  82  Table 3.2  Daily energy intake (kJ) on each of the three diets  83  Table 3.3  Calculated meal composition of each diet  84  Table 3.4  Percent individual fatty acid composition of fat in the three  diets  85  Table 3.5  Percent fatty acid composition of fat in the three diets ..86  Table 3.6  Mean serum lipid values induced by each diet  87  Table 3.7  Mean serum lipid values induced at each time point  88  Table 3.8  Serum cholesterol lathosterol values in subjects at 12 h  (fed) and 36 h (fasted) time points during each 48 h test period . 89 Table 3.9  Deuterium enrichment (o/oo) relative to SMOW validation trial  conducted to determine serum sample size needed for mass spectrometry  Table 3.10  90  Comparison of the closeness of fit obtained by analysis of  the specific activity decay data with a two-term versus a three-term equation  102  xiv  Table 3.11  Parameter values obtained by fitting each subject's specific  activity decay data to a three-term exponential equation Table 3.12(a)  Parameter values obtained by fitting each subject's  specific activity decay data to a three-compartment model Table 3.12(b)  (1973) and the values obtained in this  study  107 Serum free cholesterol deuterium enrichment (o/oo) relative  to SMOW at five time points on each of the three diets Table 3.15  106  Three-pool cholesterol model comparison of the values  obtained by Goodman et al.  Table 3.14  105  Parameter values obtained by fitting each subject's  specific activity decay data to a three-compartment model Table 3.13  103  108  Ratios of serum free cholesterol deuterium enrichment to  body water deuterium enrichment at four time points on each of the three diets Table 3.16  109  Serum esterified cholesterol deuterium enrichment (o/oo)  relative to SMOW at five time points on each of the three diets .110 Table 3.17  Ratios of serum esterified cholesterol deuterium enrichment  to body water deuterium enrichment at four time points on each of the three diets  ill  xv  Table 3.18  Body water deuterium enrichment (o/oo) relative to SMOW at  three time points on each of the three diets  Table 3.19  112  Free cholesterol fractional synthesis rates, calculated for  each 12-h period, using the first linear method  Table 3.20  Calculated de novo  free cholesterol synthesis during each  12-h period, using the first linear method  Table 3.21  Calculated de novo  Calculated de novo  132  Free cholesterol fractional synthesis rates and mass  synthesized derived using SAAM/CONSAM for each 24-h period  Table 3.26  129  free cholesterol synthesis during each  12-h period deriving parameter values with SAAM/CONSAM  Table 3.25  126  Free cholesterol fractional synthesis rates, calculated for  each 12-h period deriving parameter values with SAAM/CONSAM  Table 3.24  123  free cholesterol synthesis during each  12-h period, using the second linear method  Table 3.23  120  Free cholesterol fractional synthesis rates, calculated for  each 12-h period using the second linear method  Table 3.22  117  135  Fractional esterification rate, calculated for each 12-h  period, using the first linear method  xv i  142  Table 3.27  Calculated total de novo esterification during each 12-h  period, using the first linear method Table 3.28  Fractional esterification rate, calculated for each 12-h  period, using the second linear method Table 3.29  151  Cholesterol fractional esterification rates, calculated for  each 12-h period deriving parameter values with SAAM/CONSAM Table 3.31  148  Calculated total de novo esterification during each 12-h  period, using the second linear method Table 3.30  145  154  Calculated de novo esterified cholesterol evolved during  each 12-h period, deriving parameter values with SAAM/CONSAM ....157  Table 3.32  Cholesterol fractional esterification rates and mass of  esterified cholesterol generated, calculated for each 24-h period, deriving parameter values with SAAM/CONSAM  xvii  160  LIST OF FIGURES Figure 3.1 Serum total cholesterol concentration at 12 h intervals during the 48 h sampling period  91  Figure 3.2 Serum triglyceride concentration at 12 h intervals during the 48 h sampling period  92  Figure 3.3 Serum HDL concentration at 12 h intervals during the 48 h sampling period  93  Figure 3.4 Serum non-HDL cholesterol or apo B-containing lipoproteins (VLDL and LDL) concentration at 12 h intervals during the 48 h sampling period  94  Figure 3.5 Serum lathosterol concentration at 12 h and 36 h during the 48 h sampling period  95  Figure 3.6 Specific activity decay curves for each subject (A to I): turnover of serum cholesterol  99-101  Figure 3.7 Three-pool model of cholesterol turnover in man, as redrawn by author (Goodman et al.  1973)  104  Figure 3.8 Proposed model showing de novo cholesterol synthesis as measured by deuterium incorporation  XVI11  116  Figure 3.9 Calculated de novo fractional synthesis rates for free cholesterol during each 12-h period, using the first linear method . 119 Figure 3.10 Calculated de novo free cholesterol synthesis during each 12-h period, using the first linear method  122  Figure 3.11 Calculated de novo fractional synthesis rates for free cholesterol during each 12-h period, using the second linear method 125 Figure 3.12 Calculated de novo free cholesterol synthesis during each 12-h period, using the second linear method  128  Figure 3.13 Free cholesterol fractional synthesis rates, calculated for each 12-h period, using a modified version of the linear method, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau  131  Figure 3.14 Calculated de novo free cholesterol synthesis during each 12-h period, using a modified version of the linear method, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau  134  Figure 3.15 Free cholesterol fractional synthesis rates, calculated for each 24-h period, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau  xix  137  Figure 3.16 Calculated de novo free cholesterol synthesis during each 24-h period, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau  138  Figure 3.17 Fractional esterification rates calculated for each 12-h period, using the first linear method  144  Figure 3.18 Calculated de novo appearance of esterified cholesterol during each 12-h period, using the first linear method  147  Figure 3.19 Fractional esterification rates calculated for each 12-h period, using the second linear method  150  Figure 3.20 Calculated de novo appearance of esterified cholesterol during each 12-h period, using the second linear method  153  Figure 3.21 Cholesterol fractional esterification rates, calculated for each 12-h period, using a modified version of the linear method, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau  156  Figure 3.22 Calculated de novo esterified cholesterol evolved during each 12-h period, using a modified version of the linear method, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau  xx  159  Figure 3.23 Cholesterol fractional esterification rates, calculated for each 24-h period, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau  Figure 3.24 Calculated de novo  162  esterified cholesterol evolved during  each 24-h period, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau  xx i  163  LIST OF APPENDICES  Appendix l  Sample meal plan: meals for Subject H during each diet  period  Appendix 2  209  Body water deuterium enrichment (°/oo) relative to SMOW at  five time points on each of the three diets  Appendix 3  Subject body weight at beginning and end of two-week diet  periods  213  Appendix 4  Daily energy intake (kJ) on each of the three diets  Appendix 5  Individual serum lipid concentrations at 5 time points  during two-day test periods on each of the diets  Appendix 6  212  214  215  Serum lathosterol levels and ratios of lathosterol to  cholesterol at 5 time points during two-day test periods on each diets  Appendix 7  219  Deuterium enrichment (°/oo) relative to SMOW: validation  trial conducted to determine serum sample size needed for mass spectrometry  Appendix 8  221  Data used to generate C 14 cholesterol specific activity  decay curves for each subject  222  xxii  Appendix 9  Sample SAAM/CONSAM deck for estimation of parameters of  multiexponential equation best describing the specific activitydecay curve of one subject Appendix 10  229  Sample SAAM/CONSAM deck for calculation of pool sizes and  rates of exchange between pools, as well as the rate of cholesterol de novo synthesis, based on specific activity decay curves Appendix 11  Serum free cholesterol deuterium enrichment (°/oo) relative  to SMOW at five time points on each of the three diets Appendix 12  231  233  Serum esterified cholesterol deuterium enrichment (°/oo)  relative to SMOW at five time points on each of the three diets .235 Appendix 13  Sample SAAM/CONSAM deck for calculation of rates of de novo  cholesterol synthesis pool sizes and rates of exchange between pools, based on deuterium incorporation data  XXI11  237  ACKNOWLEDGEMENTS I would like to recognize the members of my supervisory committee, each of whom contributed to this study. Dr. Jiri Frohlich arranged for lipid analysis of serum samples at the Lipid Research Centre,- Dr. Sheila Innis donated chemicals, equipment, time, and expertise for the serum lathosterol assay; Dr. Linda McCargar allowed me to use her laboratory equipment to monitor subjects' weight fluctuations during diet periods,and Dr. Urs Steinbrecher injected the radioactively-labelled cholesterol into each subject. I am pleased to acknowledge the financial support I received during my training. These include two research traineeships from the Heart and Stroke Foundation of BC & Yukon, two graduate scholarships from UBC, a teaching assistantship from the School of Family and Nutritional Sciences, and a research assistantship from Dr. Peter Jones. The subjects I worked with were the best, and I am indebted to these fine individuals for their stamina in enduring the rigors of a long clinical investigation. I would like to thank the staff, Division of Laboratory Medicine, for numerous blood draws on my subjects. I am appreciative of the assistance received from Dr. Hugh Barrett, Research Facility for Kinetic Analysis, University of Washington, Seattle. He patiently taught me the nuts 'n bolts of SAAM/CONSAM and guided me about the peculiarities of my data set. Special thanks to Nur Shaw, my "peon". She cheerfully completed countless onerous tasks, and kept the subjects content with her flawless cooking and melodious entertainment. Likewise, I am indebted to Dr. Catherine Leitch for her help with the mass spec, her patient answers to myriad questions, and her willingness to sacrifice her limited free time to assist me, particularly in the balmy days of summer. Franco Ciolli taught me how to use the GC and the scintillation counter. Dr. Jennifer Hamilton coached me through the lathosterol assay, and Blair Main endured my endless jabs at his hapless veins in the name of perfecting my phlebotomy skills. I thank them for their help, their tolerance, and their humour. Thank you to all my friends for their support, and un gros merci to my best friend, my spouse, Barry Taylor.  xxiv  DEDICATION  Pour mes parents, HelfSne et Andre Mazier.  xxv  1  1.1  AN OVERVIEW  OF STUDIES  CONCENTRATION  INTRODUCTION  LINKING  CHANGES IN SERUM  TO THE COMPOSITION  OF DIET  FAT  CHOLESTEROL  INTAKE  The influence of diet fat on serum cholesterol concentration has engendered debate for over thirty years. Kinsell and Michaels (1955) showed that consumption of formula or mixed diets containing large amounts of soybean oil, a source of POLY, consistently produced a significant fall in total serum cholesterol, compared with that obtained from diets based on coconut oil, which contains shorter chain SAT. Ahrens et al.  (1957), presenting similar results using a variety of  diets which differed only in the type of fat employed, concluded that total serum cholesterol varied inversely with the iodine number of diet fats and oils.  Keys et al.  (1957), after conducting work in which both  quantity and quality of diet fat were varied, derived a predictive regression equation associating total serum cholesterol with POLY and SAT diet fat content.  Seemingly MONO had no significant effect on serum  cholesterol, although a later study (Hegsted et al.  1965) showed that  MONO-rich olive oil was as effective as POLY in lowering serum cholesterol concentration in some subjects.  Hegsted et al.  (1965)  proposed a predictive multiple regression equation, involving only diet intake of myristic acid (C14:0). , palmitic acid (C16:0), POLY, and diet cholesterol, which explained 91% of the total variance in total serum cholesterol.  This indicated that perhaps certain fatty acids, and not  1  just structural classes of fats, could alter total serum cholesterol. These authors suggested that the proportion of fatty acids in the fat consumed, rather than the percentage of energy intake  supplied as fat,  was most effective for prediction of total serum cholesterol. Recently, much effort has been directed toward clarifying diet fat saturation's effect on serum lipid profiles. Vega et al.  (1982) showed  that a 40% POLY diet lowered the cholesterol content of all lipoproteins measured (VLDL, LDL, and HDL) ,• compared with a regime in which fat was supplied primarily by SAT. Hegsted et al.  (1993) published regression  analysis of literature data which showed that while POLY actively lower serum cholesterol levels, the most powerful predictor of serum cholesterol concentrations was SAT. Mattson and Grundy (1985) compared the effects of liquid diets, in which 40% of the energy was supplied by either SAT, POLY, or MONO, on cholesterol levels.  Low density  lipoproteins and total cholesterol were reduced in patients receiving either the MONO or POLY diets, compared with those receiving the SAT diet.  The Lipid Research Clinics Coronary Primary Prevention Trial  (1986) found that decreases in SAT intake were consistently associated with reductions in total plasma cholesterol levels and LDL levels. This was confirmed by Barr et al.  (1992), who showed that a reduction in diet  fat intake from 37% to 30% of diet energy intake did not lower plasma total and LDL cholesterol concentrations unless the reduction in total fat was achieved by decreasing SAT intake.  It is now accepted that  either decreasing SAT intake or increasing POLY intake will depress  2  total serum cholesterol concentration.  Reiser et al.  (1985) found that  beef tallow, a source of SAT, had no effect on plasma cholesterol levels or LDL levels in normal men.  The tallow used in this study contained  48.5% oleic acid (C18:lw9); it is possible that mechanisms regulating serum cholesterol levels were influenced as much by the beef fat's MONO content as by its SAT content.  Beef fat is usually a rich source of  palmitic acid (C16:0), but that used in this study contained an unusually high proportion of stearic acid (C18:0), which has been shown to be as effective as oleic acid in lowering plasma cholesterol levels when it replaces palmitic acid in the diet (Bonanome and Grundy 1988; Denke and Grundy 1991; Kris-Etherton et al.  1993).  O'Dea et al.  (1990)  have demonstrated that the addition of beef fat to the diet results in increased levels of serum cholesterol.  Ng et al.  (1992), however,  compared the effects of diet palmitic acid on serum lipids to those of oleic acid, and found no significant differences in total cholesterol, HDL, LDL, or TG concentrations.  Palmitate in this study was derived  from palm oil; some other factor, perhaps non-cholesterol sterols or cholesterol precursors (Subbiah 1971; Gylling and Miettinen 1988; 1992) in the oil source, may alter the extent of palmitic acid's effect on serum lipids. Studies of the role of MONO in lowering serum cholesterol levels have yielded variable results.  Seppanen-Laakso et al.  (1993)  demonstrated that substitution of either low-erucic rapeseed or olive oil for a 25% SAT margarine in a habitual diet resulted in significant  3  decreases in both total and LDL serum cholesterol concentrations, but not in HDL levels. The magnitude of the change was similar in both oilsupplemented diets. Sirtori et al.  (1986) showed that a diet rich in  olive oil was as likely to lead to favourable plasma lipid profiles as was a diet rich in corn oil. Serum total cholesterol and LDL cholesterol levels were similarly reduced by consumption of either diet; HDL levels were decreased on the corn oil diet, but were either unchanged or increased by the olive oil diet.  Similar results were  obtained by Mensink and Katan (198 9), Wardlaw and Snook (1990), Chan et al. (1991), and Mata et al.  (1992).  McDonald et al.  al. (1991), Wardlaw et al. (1991), and Valsta et al.  (1989), Berry et (1992).  Yet  another clinical study found no difference in total HDL levels between subjects consuming a high-MONO diet and those on a high-POLY diet, although the HDL2 concentrations were 50% higher, and the HDL3 concentrations were 7% lower on the latter (Dreon et al. 1990) . Two reviews urge the increased consumption of MONO rather than POLY, arguing that MONO lower serum cholesterol levels without changing HDL levels, while consumption of POLY leads to a decline in all lipoprotein classes (Baggio et al.  1988; Grundy et al. 1989) . Chang and  Huang (1990) , however, concluded that large amounts of diet MONO raised serum LDL and VLDL levels, while total serum cholesterol and HDL were not significantly altered by consumption of such a diet. al.  Ginsberg et  (1990) compared serum lipid levels obtained on an American Step 1  diet (3 0% of energy consumed as fat, SAT:POLY:MONO 10:10:10) with a  4  MONO-enriched Step 1 diet (38% of energy consumed as fat, SAT:POLY:MONO 10:10:18).  Both were effective in reducing plasma total cholesterol and  LDL concentrations, when compared with levels seen in subjects consuming an average North American diet. Wahrburg et al.  (1992) compared low-fat  MONO and POLY diets; both lowered total and LDL cholesterol concentrations but also lowered HDL cholesterol levels. Wood et al. (1993) compared the effect on serum lipids of diets containing butter enriched with either 50% MONO, or POLY,- neither significantly altered serum total cholesterol, LDL, or HDL concentrations from baseline values.  Lichtenstein et al.  (1993) measured serum lipids in subjects  who had consumed diets enriched with canola, corn, or olive oil. Total serum cholesterol dropped on all diets, but the decline was smaller on the olive oil diet; the magnitude in decline in LDL levels was similar, and significant, on all three diets, and a significant decline in HDL concentrations was seen on the canola and corn oil diets. These differences were seen both in the fasting state and postprandially. Kris-Etherton et al.  (1993) reported that oleic acid was less  hypocholesterolemic than linoleic acid, but that their effect on serum HDL levels was similar.  Finally, Nydahl et al. (1994) found that MONO  and POLY fats were equally effective in favourably altering lipids in hyperlipidemics.  The effect of MONO fats on serum lipid levels may be  susceptible to factors such as proportion of diet as fat, length of the feeding trial, or subject characteristics: age, gender, body mass index, and pre-existing lipid levels (Clifton and Nestel 1992) .  5  Consequently, consumption of either MONO or POLY is considered prudent because of their hypocholesterolemic effects, but consumption of SAT is acknowledged to contribute to elevated serum cholesterol levels (Jackson et al.  1978; Mensink and Katan 1992; Expert Panel on Detection,  Evaluation, and Treatment of High Blood Cholesterol in Adults 1993) . This is an important dietary consideration since serum total cholesterol concentration is closely linked with the risk of CHD.  In normal  individuals, every 1 mg/dL rise in plasma total cholesterol increases CHD risk by 1-2% (Kannel et al.  1.2  POTENTIAL  MECHANISMS  CHOLESTEROL  1979).  OF DIET  FAT'S  EFFECT  ON  CIRCULATING  LEVELS  Although the response of serum sterol levels to diet fat intake has been defined, the mechanisms responsible have not been satisfactorily characterized.  Whole-body responses to diet fat have  been studied and several factors isolated and explored which may be responsible for diet fat's effect on total serum cholesterol.  1.2.1 Faecal sterol excretion In the 1960s, a number of studies were conducted to determine whether the cholesterol-lowering effect of a diet high in POLY was caused by an increase in faecal excretion of either neutral sterols or bile acids while on this diet. Avigan and Steinberg (1965) fed highfat, liquid formula diets containing either coconut oil (as a source of  6  SAT) or corn or safflower oils (as sources of POLY) to hypercholesterolemic subjects. They measured the faecal excretion of neutral sterols and bile acids, as well as serum cholesterol levels. All subjects showed lower serum cholesterol levels when on the unsaturated fat diet, but there was no obvious relationship between the magnitude of changes in serum cholesterol level and faecal sterol excretion.  Spritz et al.  (1965), working with both normal and  hypercholesterolemic men, found that the induced shifts in plasma cholesterol were not necessarily accompanied by changes in faecal excretion of cholesterol. Moore et al.  (1968), however, who conducted a  sterol balance study in young men fed either a high POLY (safflower oil) or a high SAT (butter) diet, reported that faecal excretion of neutral sterols and bile acids was greater on the POLY diet.  Connor et al.  (1969), using cocoa butter and corn oil as the sources of SAT and POLY respectively, reported an increase in total faecal sterols, but this was principally manifested as an increase in bile acids.  These early  studies, therefore, did not unequivocally link changes in diet fat saturation to changes in faecal sterol excretion. Later studies have produced similarly conflicting results. Grundy and Ahrens (1970), using a variety of combinations of fats in liquid formula diets, were unable to show a correlation between diet fat intake, faecal sterol excretion, and serum cholesterol levels in hypercholesterolemic men.  Grundy (1975) later reported, however, that  hypertriglyceridemic men showed an increase in faecal sterol excretion  7  following consumption of a POLY-enriched safflower oil diet, compared with levels found after consumption of a SAT-enriched lard diet. Conversely, Glatz and Katan (1993), working with healthy individuals, found that the faecal excretion of neutral steroids was significantly higher on a low P:S diet than on a high P:S diet.  Shepherd et al.  (1980) also reported that a 40% fat diet rich in safflower oil did not cause a consistent change in faecal sterol excretion in normal men, compared with that found in the same men receiving a high SAT diet containing butter and processed cheese. Nestel et al.  (1973) fed diets  containing 45-50% fat to healthy adults and found that the lowering of plasma cholesterol levels induced by consumption of POLY was associated with an increase in faecal sterol excretion.  When these studies were  repeated for longer periods, a new steady state of faecal excretion was established, but no significant difference was found between the two diets (Nestel et al.  1974).  Similar results were found in men when both  the level of diet cholesterol as well as the type of diet fat ingested were altered (Nestel et al.  1975).  These later studies were unable to  clearly show diet fat saturation-mediated differences in sterol faecal excretion. There are several plausible explanations for the variable results reported here, methodological considerations aside. Normo- and hypercholesterolemic, even hypertriglyceridemic, individuals may not necessarily respond to diet fats in a similar manner, and are thus difficult to compare amongst studies.  8  The rate of cholesterol  absorption may change with age.  Faecal sterol absorption in response to  a high POLY diet may be a transient response, as suggested by Nestel et al.  (1974).  Connor et al.  (1969) have suggested that POLY may cause  cholesterol loss from tissues. Other diet constituents which affect cholesterol absorption (Connor 1990; Swain et al.  1990) may not have  been well controlled in the investigations, particularly where food was self-selected.  Finally, response to diet fat may vary substantially  among individuals (Beynen et al.  1987; Katan et al.  1988a; 1988b).  1.2.2 Cholesterol absorption Wood et al. (1966), who examined the effect of a six-month transition from predominantly SAT to POLY in the diets of normal patients, suggested that POLY diets result in decreased absorption of exogenous cholesterol.  Recent studies are not in agreement; Nestel et  al. (1974), who performed cholesterol balance studies on normal men fed high-fat diets of either SAT or POLY for 36-40 days, reported that the mean cholesterol absorption did not differ between diets. Similar studies on hyperlipidemic and hypertriglyceridemic subjects fed either SAT or POLY diets for 30 days also failed to demonstrate a consistent change in cholesterol balance from one diet to another (Grundy 1975; Grundy and Ahrens 1970).  Finally, McNamara et al. (1987), who conducted  cholesterol balance studies on 50 normal men, established that the quality of diet fat does not alter diet cholesterol absorption, and that diet cholesterol has virtually no effect on hypercholesterolemia.  9  1.2.3 Cholesterol synthesis Cholesterol is an essential component of the tissues of all animals.  It is not surprising, therefore, that all metabolizing tissues  in the animal body make cholesterol. Approximately 700-900 mg of cholesterol are synthesized per person daily (Turley and Dietschy, 1982) . Synthesis rates vary considerably, both among different tissues and within a particular tissue under differing conditions (Myant 1981) . Liver and intestine are considered the most important sites for cholesterol synthesis, although there is disagreement about what percent of whole-body cholesterol synthesis is hepatic (Dietschy et al.  1993).  Cholesterol synthesis can be easily and accurately measured in vitro.  McNamara et al. (1987) measured the rate of sterol synthesis in  isolated MNL by analysis of [2- C]acetate incorporation into sterols; this method follows the rate of evolution of de novo synthesized cholesterol after addition of a radioactively-labelled precursor to the system.  No difference was noted in the rate of sterol synthesis in  individuals fed either a SAT or a POLY diet.  Dietschy (1984) measured  the activity of the rate-limiting enzyme in the cholesterol biosynthetic pathway as a response to diet fat alterations in many different species. There is a dearth of information about the mechanisms involved in in vivo  changes in cholesterol synthesis in humans in response to diet.  Spritz et al.  (1965) used sterol balance studies to demonstrate that  plasma cholesterol concentrations can change, seemingly in response to the type of diet fat consumed, independent of changes in cholesterol  10  excretion.  They were unable to resolve whether this was caused by  changes in the rate of cholesterol synthesis or shifts in cholesterol balance.  Grundy and Ahrens (1970) combined the sterol balance method  with plasma cholesterol turnover studies and were not able to show that cholesterol synthesis was altered by diet fat saturation in ten of eleven hyperlipidemic patients studied.  Recent work by Glatz and Katan  (1993), who measured whole-body cholesterol synthesis both by the sterol balance method and by serum lathosterol variation, suggests that SAT stimulate cholesterol synthesis. To date, studies comparing in vivo  cholesterol synthesis in humans  consuming either SAT, POLY, or MONO, suggest that diet fat modification alters rates of de novo cholesterol synthesis.  Jones et al.  (1991;  1994b) measured de novo cholesterol synthesis in hypercholesterolemic subjects using D incorporation.  They concluded that synthesis rates  were greater in subjects fed a corn oil-based diet compared with either a baseline diet or an olive oil-based diet, but were not significantly different from rates of synthesis in subjects fed a canola oil-based diet.  Similar findings were also seen in another study using D  incorporation to examine rates of cholesterol synthesis in elderly, mildly hypercholesterolemic subjects (Jones et al.  1994a): rates of  synthesis were greater in subjects fed diets containing corn oil than in those fed diets containing beef tallow. Fernandez et al.  (1990), working with guinea pigs, examined the  effect of diet fat quality on FSR by determining the incorporation of  11  [3H] water into DPS and measuring sterol balance. Sterol balance measurements indicated that whole-body cholesterol synthesis rates were not affected by diet fat quality, but 3H incorporation into DPS indicated that FSR in animals fed a corn oil diet was significantly lower than that in animals fed either an olive oil-based diet or a lardbased diet.  The authors suggested that diet fat-mediated effects on the  metabolic sources of NADPH could explain this discrepancy, but admitted that further studies are needed to define the effects of POLY on the generation of [3H] NADPH from [3H] water. Additionally, these authors found rates of hepatic cholesterogenesis were decreased in pigs fed olive oil diets, compared with those on the other two diets.  In a later  study, again using guinea pigs as subjects, Fernandez and McNamara (1994) confirmed that in guinea pigs fed olive oil-enriched diets HMGCoA reductase was less active than in those fed either beef tallow- or corn oil-based diets, suggesting a diminished rate of synthesis in animals fed olive oil diets. Hepatic ACAT activity, however, a measure of rates of cholesterol esterification, was independent of the type of fat consumed. Others working with animals have found either (i) no difference in hepatic cholesterol synthesis rates following consumption of SAT, POLY, or MONO (Triscari et al.  1978); (ii) increased rates following ingestion  of safflower oil when compared with either olive oil or coconut oil (Spady and Dietschy 1988) ; or (iii) decreased rates following consumption of safflower oil or beef tallow compared with lard (Mercer  12  and Holub, 1981).  Species differences, quantity of fat consumed, and  fatty acid composition of the particular fats fed are thought to contribute to these strikingly variable results.  1.2.4 Lipoprotein composition and catabolism The lowering of plasma cholesterol by POLY diets may be a result of alterations in plasma lipoprotein composition.  Several studies have  been conducted in which normal and hyperlipidemic subjects were fed either a high POLY diet or a high SAT diet.  Plasma lipoproteins were  isolated, and the fatty acid composition of the lipoprotein fractions, including TG, CE, and PL, was quantified (Glueck et al. 1986; Grundy et al.  1988; Grundy and Vega 1988; Mattson and Grundy 1985; Shepherd et al.  1980; Vega et al.  1982) . Cholesterol-to-protein, PL-to-protein, and TG-  to-protein ratios have also been measured in similar studies (Kuksis et al.  1982; Morrisett et al.  1977; Shepherd et al.  1978; Spritz and  Mishkel 1969) . Generally, when compared with a SAT diet, a POLY diet results in an increase in lipoprotein linoleic acid content, a decrease in stearic acid and palmitic acid contents, and either a decrease, an increase, or no change in oleic and palmitoleic acid contents.  Spritz  and Mishkel (1969) proposed that since unsaturated PL occupy more space than saturated PL, the greater LDL linoleic and linolenic acid contents could result in fewer molecules per particle. Morrisett et al.  (1977)  showed that lipoproteins from patients fed the POLY diet had thermotropic transitions at lower temperatures than did those from  13  patients fed the SAT diet, resulting in greater lipoprotein fluidity. Jackson and Gotto (1974) submit that altered membrane fluidity could be responsible for changes in uptake by LDL receptors, by providing less than ideal conditions for the initiation of endocytosis.  POLY diets  could, therefore, lead to significant alterations in the composition of individual lipoproteins.  Whether these changes are significant to  maintenance of the overall serum cholesterol balance is uncertain. Chait et al.  (1974) suggested that POLY may decrease the rate of  secretion of VLDL into the plasma, and Turner et al.  (1976) reported an  increase in LDL FCR, the rate at which a fraction of the pool is catabolized, in some patients fed POLY diets compared with patients receiving SAT diets. Shepherd et al.  (1978) noted an increase in LDL  FCR, as well as a decrease in apo LDL, in patients fed diets enriched with POLY, but did not remark upon changes in LDL synthesis. Vega et al.  (1982) observed that since POLY have a hypocholesterolemic effect on  all lipoproteins, the simultaneous decline in apo constituents does not result in a decreased lipid-to-protein ratio. Mattson and Grundy (1985) confirmed that the relative proportions of total protein and the various lipid components in lipoprotein fractions, including PL, are not altered by consumption of a POLY diet instead of a SAT diet, as did Brousseau et al. (1993) working with monkeys. Kuksis et al.  (1982) concluded that the decrease in plasma  cholesterol observed in individuals consuming POLY diets, compared with SAT diets, results from a decrease in the number of plasma lipoproteins,  14  rather than from a change in size or composition.  Demacker et  al.  (1991), who investigated the effect of diet fat saturation on rates of removal of TG-rich lipoproteins, concluded that the clearance of chylomicrons and their remnants is accelerated on a diet rich in POLY, compared with one of SAT. Woollett et al.  (1992a), working with  hamsters, demonstrated that substitution of POLY for SAT in the diet decreased LDL-cholesterol production rate from nearly 200% to 155%. The reduction of LDL production was much less dramatic when oleic acid was consumed (Daumerie et al.  1992).  In another study where hamsters were  again used as an experimental model, they determined that short and medium chain-length SAT, as well as C18:0, are biologically neutral with respect to regulation of plasma LDL concentration, but that C12:0, C14:0 and C16:0 SAT increase the rate of LDL formation, and hence markedly elevate plasma LDL concentrations (Woollett et al. 1992b). The lowering of LDL and VLDL concentrations, which are the main contributors to decreased serum cholesterol levels following diet supplementation with w3 POLY, may result from a reduction in the rate of synthesis of apo B (Brousseau et al.  (1993); Illingworth et al.  1984).  This reduction is not seen, however, in either hypercholesterolemic (Demke et al.  1988; Wilt et al.  1989) or hypertriglyceridemic (Deck and  Radack, 1989) subjects consuming diets supplemented with modest doses of w3 fatty acids.  In the latter group, total apo B levels increased,  whereas in the former group both total and LDL levels increased. Similar studies with baboons (Kushwaha et al.  15  1991) and monkeys (Sorci-  Thomas et al.  1989) showed no increase in either mRNA levels for hepatic  apo B or in apo B secretion from the liver. Diet fatty acid composition can also influence HDL concentrations. Diets high in SAT lead to high levels of both LDL and HDL (Knuiman et al.  1987),- when MONO-rich diets are fed, usually a reduction in LDL, but  not HDL, occurs (Baggio et al.  1988; Grundy et al.  1989).  In contrast,  high levels of POLY will cause a reduction in HDL as well as LDL levels (Sirtori et al. 1986; Vega et al. occurs is not yet understood.  1982), and the mechanism whereby this  Shepherd et al.  (1978) demonstrated a  reduction in the synthesis rate of apo Al in subjects consuming diets high in linoleic acid, compared with those fed a diet high in SAT. Vega et al.  (1982) found that total plasma apo Al levels were significantly  lower in individuals on a POLY diet, compared with those found in individuals on a SAT diet, but relative proportions of apo Al, apo All, and apo D did not change since POLY diets significantly reduced protein levels (21%) in HDL particles. Nicolosi et al.  (1990) showed that apo  Al levels were 112% higher in monkeys fed a coconut oil diet, compared with those fed a corn oil diet.  They ascribed these results to a higher  apo Al FCR in corn oil-fed monkeys. These decreases in apo Al levels, whether brought about by an increase in FCR or a decrease in synthesis in subjects fed POLY-rich diets, may provide a mechanism for the reported reduction in HDL.  16  1.2.5 Lipoprotein receptors Lipoprotein receptors in the liver and extrahepatic tissues mediate the uptake and degradation of cholesterol-carrying lipoproteins (Brown et al. 1981); the number of receptors determines the efficiency of internalization of cholesterol into cells (Brown and Goldstein 1983) . Those with familial hypercholesterolemia demonstrate abnormalities in, or an absence of, LDL receptors; resulting in elevated serum cholesterol levels and subsequent atherogenesis (Mahley and Innerarity, 1983) . It has been proposed that a diet fat may exert its effect on circulating cholesterol levels by altering the number of hepatic LDL receptors, leading to either greater or lesser clearance of cholesterol from circulation.  Fernandez and McNamara (1989) demonstrated that the  increase in specific binding of LDL in hepatic membranes of guinea pigs fed diets containing corn oil was caused by an increased number of LDL receptors in the pigs, compared with animals fed either olive oil- or lard-containing diets.  Spady and Dietschy (1985; 1988) showed a  decrease in LDL receptor-mediated cholesterol transport in hamsters fed SAT diets, compared with animals fed mixed-fat diets. Woollett et al. (1992a), who also worked with hamsters, demonstrated that substitution of POLY for SAT in the diet caused LDL receptor activity to increase from 25% to 8 0% of control values; oleic acid was shown to have a similar effect (Daumerie et al.  1992).  In another study where hamsters  were again used as an experimental model, they determined that the C12:0, C14:0 and C16:0 SAT lower hepatic receptor activity (Woollett et  17  al. 1992b).  Fox efc al.  (1986) have shown that SAT-rich diets reduce  hepatic LDL receptor messenger-RNA (mRNA) levels in the baboon, whereas POLY-rich diets increase levels. Sorci-Thomas et al. (1989), however, working with African green monkeys fed diets containing either 4 0% POLY or SAT, were unable to demonstrate any changes in hepatic LDL receptor mRNA due to diet fat. The bulk of the experimental evidence, therefore, suggests that diet fat may impact serum cholesterol by changing cholesterol uptake rates; this is discussed in a recent review article by Spady et al.  (1993).  1.2.6 Dietary cholesterol precursors and non-cholesterol sterols It has been suggested that either non-cholesterol sterols (Subbiah 1971; Gylling and Miettinen 1988; 1992), which are present in some fats such as rapeseed oil, and dietary cholesterol precursors such as squalene, which can be found in olive oil, may exert an effect on whole body cholesterol metabolism which is independent of the degree of saturation of the fat. Strandberg et al. (1990) fed 900 mg/d squalene to humans for 7-3 0 days, and noted no changes in serum cholesterol concentrations, although there was a marked increase in cholesterol faecal excretion. A more recent study examining the effect of consumption of l g squalene/d for 6 weeks on serum cholesterol concentration concluded that this quantity contributed to a 12% increase in both total and LDL serum cholesterol, but that consumption of 0.5g/d had virtually no effect (Miettinen and Vanhanen 1994) . Olive oil,  18  considered rich in this cholesterol precursor, contains 2-7 mg/g oil (Liu et al.  1976); even if squalene consumption did alter cholesterol  metabolism it is unlikely that an individual would consume enough to have an appreciable effect. Phytosterols are thought to be hypocholesterolemic in most species of animals (Andriamiarina et al. Gerson et al.  1964; Uchida et al.  1989; Bhattacharyya and Eggen 1984; 1984).  A recent study with rat liver  cells suggested that if the ratio of dietary phytosterols to cholesterol is 1:1 and dietary cholesterol is consumed in excess, a decrease is seen in both liver enzyme activity and cholesterol content (Laraki et 1993).  al.  It is not clear whether this ratio is valid in humans and  whether such an effect would significantly alter whole-body cholesterol metabolism.  As with squalene, it is unlikely that such large quantities  of plant sterols would habitually be consumed.  1.3  CHOLESTEROL  METABOLISM  Whole-body cholesterol is an amalgamation of synthesis and catabolism, as well as transfer from compartment to compartment, or pool to pool, within the body.  The study of whole-body cholesterol  metabolism can yield information concerning normal and abnormal pool sizes, how these changes with varying dietary and physiological conditions, and whether such changes are relevant to the maintenance of human health.  19  1.3.1 Production rate, turnover, and mass of exchangeable cholesterol Chobanian and Hollander (1962) gave tracer doses of 4-  C-  cholesterol to nine morbid patients at intervals ranging from 1 to 226 days prior to death, and investigated the equilibration between serum and tissue cholesterol.  They observed the rise in radioactivity of  cholesterol in various tissues after an intravenous injection of labelled cholesterol, and found that plasma cholesterol exchanges relatively rapidly with cholesterol in liver, erythrocytes, spleen, lungs and intestine, more slowly with that in skeletal muscle, adipose tissue, skin, and kidney, and much more slowly with that in arterial wall (Chobanian et al.  1962).  Goodman and Noble (1968) conducted a similar study with healthy patients.  They divided body cholesterol into Pool A, which turns over  rapidly, and Pool B, which turns over slowly.  Pool A included  cholesterol in plasma, erythrocytes, liver, bile, small intestine, and some of the cholesterol in other viscera.  Pool B included that in  skeletal muscle, adipose tissue, skin, and arterial walls.  The mean  mass of Pool A (MA) was 25 g, and the mean production rate of Pool A (PRA) was 1.35 g/day in normal individuals.  Pool B mass could not be  calculated, but size limits were estimated based on the assumption that no cholesterol is synthesized in one of the two pools, which is unlikely.  Nestel et al. (1969) found that the mean M A was 17.9 g, the  mean MB was 42.4 g, and PRA for an adult of normal weight was 1.1 g/day.  20  In a later study, Goodman et al.  (1973) concluded that total body  cholesterol was partitioned into three, rather than two, pools. The first compartment (Pool 1) consists of cholesterol which equilibrates fairly rapidly with plasma cholesterol, and includes that in plasma, erythrocytes, liver, intestine, and much of the cholesterol in other viscera such as lung, pancreas, spleen, and kidney.  The second  compartment (Pool 2) cholesterol equilibrates at an intermediate rate with plasma cholesterol, and includes that in some visceral and peripheral tissues. The third compartment (Pool 3) cholesterol equilibrates more slowly, and includes sterol in skeletal muscle and arterial walls. Additionally, it is thought that there is a pool of exceedingly slowly exchangeable cholesterol, which includes that in the nervous system, bone, and fibrous connective tissue.  The mean mass of  Pool 1 (M]_) was 23.4 g,- upper and lower limits for the other two pools were estimated.  Pool 2 ranged from 11.3 to 20.2 g and Pool 3 ranged  from 35.7 to 72.1 g per person.  Total cholesterol production rate was  estimated to be 1.1 g/day, and is equivalent to the total body turnover rate, assuming that all cholesterol catabolism and excretion occurs by the tissues which compose the first pool (Grundy and Ahrens, 1969). After examining cholesterol turnover in 24 patients classified as either normal, hypertriglyceridemia or hypercholesterolemic, Smith et al.  (1976) demonstrated that the M-^ pool size was strongly correlated  with excess body weight and serum cholesterol and TG concentrations. Goodman et al.  (1980) showed that body weight and serum cholesterol  21  levels, but not serum TG levels, were directly related to the total pool size as well as the M^ pool size. Biss et al.  (1980), however, who  examined the relationships between serum cholesterol levels and the kinetics of cholesterol metabolism in 15 healthy men, found no correlation with any of the parameters, including Ml pool size, except for a weak negative correlation with the turnover rate.  Smith et al.  (1976) and Goodman et al. (1980) also noticed that all estimates of the size of the M3 pool were correlated with excess adiposity, as did Miettinen (1976), who suggested a causal role for adipose tissue in cholesterol synthesis in slowly exchangeable cholesterol pools. Schwartz et al. (1978) used specific activity decay curves and tracer kinetics to develop a multicompartmental model describing hepatic cholesterol compartments associated with the metabolism of bile acids and biliary cholesterol.  In a similar study, this same group later  developed a model describing the kinetics of cholesterol precursor sources, turnover of HDL CE, steady state fluxes, and pool sizes (Schwartz et al. 1982). Dell et al.  (1985), who studied cholesterol turnover in baboons,  estimated that central pool cholesterol synthesis varied from 61 to 89% of the total cholesterol production rate.  If cholesterol kinetics in  humans are similar to those in baboons this suggests that 25% of total cholesterol production occurs in Pool 3 peripheral tissues.  Studies  using a variety of animals have also demonstrated the presence of sidepool cholesterol production, varying among species and with total diet  22  cholesterol (Jeske and Dietschy 1980; Spady and Dietschy 1983; Turley et al.  1981; Dietschy 1984; Spady et al.  1993).  The liver and intestine  remain the most important organs for cholesterol synthesis.  1.3.2 Origin and turnover of plasma cholesterol Cholesterol is carried in lipoproteins in the plasma, where it undergoes a complex series of exchanges among the lipoprotein fractions, including esterification of FC by LCAT and redistribution of the CE so formed.  The recommended plasma cholesterol level is less than 5.2  mmol/L in adults, 67% esterified.  The CE to total cholesterol ratio is  highest in HDL (79%), lowest in VLDL (60%), and intermediate in LDL (71%) (Myant 1981) .  LDL contains the most cholesterol (65%), VLDL the  least (9%), with HDL halfway between the two (26%) . In humans, plasma CE obtained during fasting contain a high proportion of POLY, predominantly linoleic acid; the fatty acid profile is similar in all three major lipoprotein fractions.  During fasting,  the bulk of CE is thought to be formed in blood by the LCAT reaction. The evidence for this may be summarized as follows (Myant 1981) : (a) The fatty acid of CE in the three major lipoprotein fractions is normally linoleate, found at the C2 position on the glycerol backbone of plasma lecithin and transferred to cholesterol by LCAT.  The fatty  acid would be oleate if esterification occurred primarily by ACAT. (b) In  vitro,  the rate of cholesterol esterification by LCAT is  sufficient to account for the rate of turnover of plasma CE in  23  vivo.  (c) Relative initial rates of cholesterol esterification decrease in the order HDL>VLDL>LDL; CE are formed primarily by the action of LCAT on HDL FC, then transported to VLDL and LDL (Ferezou et al.  1982) .  (d) In LCAT-deficient patients, CE is almost completely absent. Most of the cholesterol esterification in plasma takes place in HDL (Dobiasova et al.  1991).  HDL 3 acts as a substrate for LCAT  esterification while HDL 2 inhibits this process (Barter et al.  1984;  1985), and the relative ratio of the two HDL subclasses controls the rate of cholesterol esterification in plasma (Dobiasova et al. Shepherd et al.  1992).  (1978) have demonstrated that the ratio of HDL 2 to HDL 3  falls by 28% in normal men consuming a POLY diet, compared with that seen in men on a SAT diet.  Also, reduced HDL 2 levels have been observed  in patients with hypertriglyceridemia (Nikkila et al. al. 1987).  1987; Patsch et  Not only is this a factor in reverse cholesterol transport,  but since it has been shown that HDL 2 may be inhibitory to LCAT (Barter et al.  1984), the increased FER seen in dyslipidemic patients may be a  direct result of reduced inhibition by HDL 2 (Dobiasova et al. Melchert et al.  1991).  (1987) demonstrated that the fatty acid profile of  CE differed between vegetarians and non-vegetarians, with those of vegetarians being higher in POLY and those of non-vegetarians being higher in SAT.  De Backer et al.  (1989) suggested that since diet  variables account for a significant proportion of the variance in the type of serum CE, perhaps measurements of CE could be used to identify people at risk for atherosclerosis with respect to fat consumption.  24  Dullaart et al.  (1989) demonstrated that a diet fat load can induce a  shift in lipoprotein CE distribution from HDL and LDL to TG-rich lipoproteins in normal subjects but not in hypercholesterolemic subjects, although levels of total plasma CE are not increased.  This  may result from accelerated chylomicron clearance in hypercholesterolemic individuals (Weintraub et al.  1987).  In non-fasting individuals, CE formed by ACAT in the intestinal wall enter the circulation with VLDL and chylomicrons.  Cholesterol  esters can be exchanged among the lipoprotein fractions, although more slowly than free plasma CE.  Plasma CE turn over at a constant rate  under steady state conditions, with the main route of removal from plasma by LDL catabolism in extrahepatic tissues. Plasma CE turnover may be expressed either as the fraction of the total pool replaced in time (fractional rate of turnover) or as the total mass replaced in time (absolute rate of turnover).  Additionally,  there are fractional and absolute rates of turnover of CE within either particular lipoprotein fractions or fatty acid classes (Myant 1981). Plasma cholesterol esterification can be estimated in vivo by following the rate of incorporation of either labelled cholesterol or a cholesterol precursor into plasma CE. The injection of radioactive acetate or mevalonate into a normal subject is followed by a rise in specific activity of total plasma cholesterol (Zilversmit, 1960) . As the radioactive FC becomes esterified, the specific activity of FC declines, and that of CE rises. Assuming that plasma CE is synthesized  25  from a pool of FC which is in rapid equilibrium with plasma cholesterol, the FER can be calculated from the two specific activity curves (Zilversmit 1960; Nestel and Monger 1967) . The FER rate and the total amount of plasma CE may then be used to calculate the absolute rate of turnover.  Two to three percent of the whole plasma pool of CE is  removed and resynthesized each hour (Goodman 1965) . Furthermore, the rate of turnover of plasma CE is correlated with the production rate of total cholesterol in vivo  (Myant et al.  1973).  Similarly, the fractional and absolute rates of turnover of the CE in the lipoprotein fractions can be measured.  Rates of turnover of CE  in different lipoprotein fractions separated by ultracentrifugation may also be calculated by measuring the radioactivity of CE separated by thin layer chromatography.  The fractional turnover rate is highest in  HDL, intermediate in VLDL, and lowest in LDL, as seen by observing the initial appearance of radioactivity in CE (Sodhi, 1974), and is thought to reflect turnover of the lipoprotein itself. Miller et al.  (1976) have found that VLDL cholesterol  concentration is positively correlated with plasma cholesterol esterifying activity.  The mechanism behind this is not known, but data  indicate that LCAT may be important for VLDL catabolism.  Conversely,  the concentration of LDL cholesterol has been shown to be negatively correlated with plasma esterifying activity (Sodhi 1974), and Myant et al. (1973) have shown that the fractional turnover rate of LDL CE is  26  inversely related to its plasma concentration, suggesting a correlation between concentrations of plasma FC and LCAT (Miller et al. 1976) .  1.3.3 Plasma cholesterol turnover relative to whole-body cholesterol turnover Plasma FC exchanges readily among different lipoproteins (Ferezou et al.  1982), between lipoproteins and erythrocytes, and between  lipoproteins and the membranes of cells in contact with plasma (Myant 1981).  These exchanges may be observed by following the changes in  specific activity of FC during the first few hours after an intravenous injection of a radioactive cholesterol precursor (Nestel and Monger 1967).  Following the injection, the specific activity of FC rises most  rapidly in the liver.  It begins to decline, and that in plasma rises,  following the exchange of newly synthesized FC from the liver with plasma cholesterol.  Eventually, radioactive cholesterol appears in the  erythrocytes as the result of cholesterol exchange between lipoproteins and erythrocytes (Sodhi 1974) . Plasma CE also exchanges readily between lipoproteins, as discussed earlier, albeit at a slower rate than does FC. al.  Pattnaik et  (1978) isolated a protein from human plasma which was reported to  promote the exchange of CE between HDL and LDL, but with no apparent net exchange of CE.  Marcel and Vezina (1973) and Chajek and Fielding (1978)  both described a transfer protein that facilitates the bulk transfer of CE from HDL to VLDL and LDL.  Cholesterol esters are also transferred in  27  bulk from LDL to VLDL.  It is now known that CETP allows the CE of HDL  to be transported to the liver by chylomicron remnants and IDL, or by hepatic uptake of LDL.  CETP thus removes the product that inhibits LCAT  activity in HDL (Mayes 1988a). Miller et al.  (1976) investigated the relationship between plasma  lipoprotein cholesterol concentration and cholesterol turnover in normal subjects.  Subjects consuming diets high in SAT showed positive  correlations of VLDL total cholesterol with cholesterol turnover. Diets high in SAT tended to raise serum cholesterol levels (Mattson and Grundy 1985), and newly synthesized cholesterol entered plasma predominantly within VLDL (Mayes 1988a).  Subjects consuming POLY had lower. VLDL  concentrations relative to cholesterol turnover. found the slowly exchanging  Blum et al.  (1985)  cholesterol pool to be negatively  correlated with both HDL cholesterol and major HDL apo, but these relationships disappeared when allowance was made for the effect of body weight.  Additionally, neither HDL nor LDL subfractions were important  determinants of the parameters of long-term cholesterol turnover in humans (Palmer et al.  1986).  Miller (1987) did find that HDL  cholesterol level was negatively correlated with the size of the slowly exchanging cholesterol pool, but not with total plasma cholesterol. Clearly, then, much remains to be done in establishing the predictability of the cholesterol turnover response to changes in these variables.  28  It has been suggested that HDL promotes reverse cholesterol transport (Pritchard and Frohlich 1994).  Excess amounts of unesterified  cholesterol are accepted by nascent HDL.  Esterification by LCAT  generates a concentration gradient that results in an efflux of cholesterol from cell surfaces.  HDL then passes on the newly esterified  cholesterol to LDL or VLDL via CETP; CE is exchanged for TG.  The  cholesterol fraction is thus eventually delivered to the liver, converted to bile acids, and excreted.  It has also been proposed that  the net movement of CE into or out of HDL is dependent on HDL composition (Sparks et al. 1991).  In normolipidemic conditions the rate  of CE influx into HDL is greater than the efflux, resulting in a net mass movement into HDL, but in hyperlipidemia, altered HDL composition results in a net movement of CE into LDL. Different type of HDL may play different roles in reverse cholesterol transport (Pritchard and Frohlich 1994).  Patients with a  greater proportion of smaller, denser HDL3, are at greater risk for CHD than are those with larger and lighter HDL2. As noted earlier, reduced HDL2 levels have been observed in patients with hypertriglyceridemia (Nikkila et al.  1987; Patsch et al.  1987).  The greater FER combined  with possible accumulation of CE in LDL may be a significant risk factorin these individuals. HDL particles also differ by their apo complement. Most contain apo Al, and some contain both apo Al and apo All.  Differential  Immunoelectrophoresis has been used to measure the amount of particles  29  containing these different apoproteins in plasma; data suggest that particles containing apo AI are more antiatherogenic than those containing both AI and All (Pritchard and Frohlich 1994).  Similarly,  the charge of the HDL particle may affect its involvement in reverse cholesterol transport (Kwano et al.  1.4  IN  VIVO  MEASUREMENT  1993; Miida et al.  OF CHOLESTEROL  1990).  SYNTHESIS  Several different methods, have been used to measure cholesterol synthesis in  vivo.  1.4.1 Sterol balance method The sterol balance method assesses the rate of whole-body cholesterol synthesis in humans by measuring cholesterol balance: faecal excretion of cholesterol plus its acidic and neutral metabolites, minus cholesterol intake (Grundy and Ahrens 1969) . Although well established, it requires several days of steady-state conditions and thus cannot detect short-term changes in the rate of cholesterol synthesis.  1.4.2 Specific activity decay curves Specific activity decay curves, generated by measuring the rate of disappearance of radioactively-labelled cholesterol from plasma, have been subjected to multicompartmental analysis for estimates of daily production rates (Goodman and Noble 1968; Goodman et al.  1973; Schwartz  et al. 1978; 1982) . These decay curves have also been used for input-  30  output analysis, where flux is estimated by dividing the dose of the tracer by the area under the washout curve of enrichment over time (Perl and Samuel 1969; Samuel and Perl 1970; Samuel et al. 1972).  The results  obtained using these two methods agree closely with those obtained from sterol balance studies (Grundy and Ahrens 1969), but both are limiting in that they require long-term sampling and require that cholesterol turnover be in a steady state. Additionally, different results are obtained from, short-term (10-12 weeks) and long-term (50-60 weeks) studies (Goodman and Noble 1968; Goodman et al.  1973; Samuel and  Lieberman 1973; 1974,- Samuel and Perl 1969; Samuel et al.  1978).  1.4.3 Biochemical assays Several assays have been developed which correlate plasma concentrations of cholesterol precursors, with the rate of cholesterol synthesis. al.  These include measurement of plasma mevalonate (Parker et  1984), squalene (Miettinen 1969), and lathosterol (Kempen et al.  1988).  Each of these has been verified with sterol balance and HMG CoA  reductase activity  (Bjorkhem et al. 1987; Strandberg et al.  1989).  Simple albeit laborious, these assays can be used for indirect measurement of cholesterol synthesis.  31  1.4.4 Measurement of rate of precursor incorporation Other methods used to measure the rate of synthesis include those which involve measuring the rate of incorporation of a labelled precursor into cholesterol.  1.4.4.1  Deuterium  incorporation  Deuterium oxide incorporation was first used to measure lipogenesis in mice (Rittenberg and Schoenheimer 1937a; 1937b); the method was later modified for measurement of cholesterol synthesis in humans (Taylor et al.  1966a; 1966b).  Improvements in isotope ratio mass  spectrometry sensitivity since then have led to further refinements. It is now possible to administer small amounts of deuterated water over short periods and obtain reliable information about rates of cholesterol synthesis. Deuterated water enters cells readily and equilibrates quickly with intracellular water.  Little unlabelled water is generated  intracellularly, allowing the cell precursor pool enrichment to equal that of plasma.  The calculated rate of cholesterol synthesis depends  upon the rate of label incorporation per molecule of cholesterol. During the synthesis of cholesterol, hydrogen atoms from water are incorporated into the sterol molecule in three different ways.  Seven  atoms of hydrogen are incorporated directly from water, 15 atoms from NADPH, and eventually, hydrogen atoms from water are incorporated into the acetyl CoA pool, which can be used as a cholesterol precursor.  32  If  the assumptions are made that over a 48-h period there is complete equilibration of deuterated water with plasma water and with NADPH, but that the acetyl CoA pool is not yet labelled, then 81% of the hydrogen atoms per cholesterol carbon atoms, the H/C ratio, will be incorporated 2 into the molecule from  H 2 0 (Dietschy and Spady, 1984) . Other  assumptions which must be incorporated if this method is to be used have recently been summarized by Jones et al.  (1993).  These include (1) the  amount of recycling of label into other pools, such as acetate, during the time of interest; (2) the form of the mathematical equation, usually accepted as monoexponential over short periods of time, of D incorporation over time; and (3) the theoretical and actual maximum plasma cholesterol enrichment. Deuterium incorporation has been used to estimate human cholesterol synthesis rates (Jones et al. Schoeller 1990; Wong et al.  1991).  1988; 1993a; Jones and  Synthesis rates calculated from D  incorporation data have been correlated with plasma mevalonate levels (Jones et al.  1992).  1.4.4.2  tracers  Other  as  precursors  Tritiated water has been used to measure cholesterol synthesis in several animal models (Dietschy 1984; Dietschy and Spady 1984; Spady and Dietschy 1983), but the relatively large dose required render its use impractical for human studies.  Other tracers, such as 13C-acetate and  13 C-mevalonate, have also been used for incorporation studies (Liv et  33  al.  1975), but drawbacks inherent to their use include difficulties in  assessing intracellular precursor pool enrichment and label dilution by unlabelled compounds (Dietschy and Spady 1984) .  1.4.5 Mass isotopomer distribution analysis Mass isotopomer distribution analysis (MIDA) has recently been developed as a technique for measuring biosynthesis (Hellerstein and Neese 1992); it has been applied to the measurement of endogenous synthesis of plasma cholesterol in humans (Neese et al.  1993).  MIDA's  primary advantage is its ability to discriminate "true" precursor enrichment in systems where a labelled tracer has been introduced but where the rate of either entry into cells or label dilution is unknown. As such it promises to overcome the difficulties cited above which occur with the use of such tracers as  13 13 C-acetate and C-mevalonate.  1.5  ANALYSIS  MODEL-BASED  COMPARTMENTAL  The interpretation of complex biological systems often invites simplification to the point that much meaningful information is obscured or lost (Skinner et al.  1959).  For this reason, multicompartmental  models are often developed to fit experimental data obtained from biological systems.  These models mathematically amalgamate accumulated  experimental data and consolidate it with previous knowledge of the system to give a reasonable approximation of the entity being studied (Garfinkel and Fegley 1984).  34  The expanded availability of software specifically designed to deal with biochemical, nutritional, and physiological systems and the popularity of powerful computers have enabled many investigators to consider the application of tracer kinetic studies to their area of interest (Coburn 1992).  These studies, analyzed using mathematical  models, can yield quantitative and predictive information about system dynamics.  1.5.1 Computer models A mathematical model describes the relationship between variables and/or parameters of a system, and can be used for simulation and prediction (Foster and Boston 1983; Green 1992) . A system is a finite set of compartments (Foster and Boston 1983).  A compartment or pool is  an amount of material which acts kinetically in a homogeneously distinct way and is characterized by its physicochemical properties, its environment, or both (Green and Green 1990) . The number of pools in the model is determined by the physiological knowledge base and the number and identity of directly measurable pools. There are two general categories of multicompartmental models (Garfinkel and Fegley 1984) . A "black box" model is one in which the investigator knows what goes into or out of a given system, but not the steps involved.  In the second type of model, each separate component is  well understood, but how each one fits into the overall system, or whether its behaviour remains the same in the system as in isolation, is  35  not. vivo  An example of this is the application of in vitro  results to an in  system. Models can be further classified as models of data or models of  systems (DiStefano and Landaw 1984) . Models of data require few structural hypotheses about the system from which the data are obtained, and demand only that the data be based on random, normally distributed samples.  These models consist of sets of mathematical functions which  do not necessarily have any physical basis. By themselves, they cannot be used with any degree of confidence for extrapolation or prediction. Models of systems are usually based on principles and hypotheses about how a system is structured and how it functions, such as productprecursor relationships, and usually take the form of a set of differential equations. Finally, models can be described based on the number of compartments and their relationship to each other. Mammillary models have one central compartment which is connected to peripheral pools, none of which are connected to each other.  Catenary models consist of a  number of pools connected in series. A general model has any number of pools among which information is exchanged freely in all directions (DiStefano and Landaw 1984) . The tracer is the substance introduced into the system, for example a stable or radioactive isotope, for which the behaviour once in the system generates data used for model construction.  The tracee is  the compound of interest which is being followed by the tracer.  36  Tracer  kinetics refers to the inter-relationships between the tracer and the system (Green and Green 1990).  Flux rate is the rate of appearance of  tracee; turnover rate the rate of disappearance of tracer (Wolfe 1992) . The fractional turnover rate {k) is, in the steady state, the fraction of a pool lost per unit time, and in a single pool system is equal to the FSR, which is the fraction produced per unit time (Wolfe 1992).  In  a multi-pool system, the turnover rate also incorporates loss or gain from other pools.  In a multicompartmental model, k refers to the rate  parameters, or rates of exchange between pools per unit time (Wolfe 1992).  The turnover time is the time required for the total mass or  pool of tracer to be replaced by new substrate (Wolfe 1992) . A tracer, is introduced into the system under investigation, and its rate of appearance and disappearance in accessible sites (eg. plasma, erythrocytes, urine, faeces) is recorded (Green and Green 1990). These data, along with information about initial conditions and experimental design, as well as any assumptions, can be used to set up a file using modelling software.  Since most modelling programmes are  dependent on initial estimates, it is necessary to generate functions describing tracer/tracee data, for example specific activity decay data or tracer incorporation data. Modelling software programmes will block completion of the analysis if the model simulation does not fit the observed data adequately (Green and Green 1990) . Pool sizes, flux rates, and other parameters of interest are usually determined based on initial estimates provided by the modeller.  37  The programme will generate  differential equations describing the tracer and tracee behaviour and attempt to fit this to estimates of a conceptualized compartmental model's parameters. A substance can be followed from ingestion to incorporation into various bodily tissues to eventual excretion, and the information garnered used to characterize steady state whole-body flux and turnover. Knowledge of general mathematics and kinetic theory is useful when learning and applying tracer kinetic.  Berman (1982) discusses the  theory, application, and limitations of kinetic analysis, with particular reference to lipoproteins.  Ramberg et al.  (1992) elaborate  the use of tracer techniques to collect data specifically for models determining nutrient requirements.  Cobelli et al. (1987) and Toffolo et  al. (1993) discuss theoretical aspects such as differences between stable and radioactive isotopes which can affect how tracer data are interpreted and parameters calculated, both in steady and non-steady states.  Bergman (1988) reviews the usefulness of models in nutrition as  well as the steps required to develop and validate a new compartmental model.  DiStefano and Landaw (1984) examine methodological limitations  and physiological interpretations of models and expose common misconceptions and misinterpretations.  A basic understanding of the  mathematical manipulations used by software may help in planning the experiment to collect data to maximize the software's capabilities and minimize its limitations (Garfinkel and Fegley 1984).  Several articles  delve into the mathematics used to describe tracer kinetics data,  38  determine compartment number and size, and exchange rates among pools (Berman et al. 1962; Dell et al. al.  1959).  1973; Gurpide et al.  1964; Skinner et  Some understanding of statistics is appropriate,  particularly since this information may be cryptically described by modelling software (Landaw and DiStefano 1984). Ideally, the model provides a minimal but complete framework for all available data, and reflects the detailed physiology of the system to the extent that it is known, while incorporating the speculations of the modeller.  It thus becomes a useful research tool for testing  concepts, designing experiments, and gaining new insights into the system.  A test of the validity of the model is its predictive power  beyond the original data base.  Confidence in a model increases when  results from different approaches remain consistent with its predictions.  A new model need not be developed for each new set of  data: all efforts should be directed toward integration with available models (Foster and Boston, 1983).  1.5.2 SAAM/CONSAM A number of computer programmes have been developed to aid in modelling kinetic data (Collins 1992); these vary in the number of features they contain and the extent of user assistance available. Models presented in this study were developed using CONSAM (Boston et al. 1981; Boston et al. 1982; Foster and Boston 1983), the interactive version of SAAM.  First developed over 3 0 years ago, SAAM is a powerful  39  computer modelling programme which was specifically developed to allow biologists, who may not be at ease with mathematical techniques, to describe irregular patterns in biological systems using mathematical terms.  It will analyze data within the context of a proposed model,  allowing the modeller to describe a system in terms of physiological pools and flows. Taking initial estimates, it constructs and solves differential equations describing the data, and displays the model and curves, along with parameter values and confidence statistics. Data confidence limits can be integrated into the model. Built-in convergence routines provide the best fit of data to model parameters (Collins 1992). SAAM's disadvantage is that it runs only in batch mode, having been developed prior to the advent of user-interactive computers. CONSAM, a user-interactive version of SAAM, was developed to overcome this limitation (Boston et al.  1981).  CONSAM is a large FORTRAN  programme which includes in excess of 350 subroutines and 40,000 lines of code.  Its maximum size is slightly over 400,000 bytes.  It was  developed for use on a Digital Equipment Corporation VAX 11/78 0 computer.  Different versions of CONSAM have been introduced since its  inception 15 years ago; the version in use now is CONSAM 30.1, which will run on IBM-compatible computers (Boston et al. 1981).  CONSAM  incorporates a text editor, a command processor, a model processor, a solution-saving facility, a system-saving facility, and a graphics module (Foster and Boston 1983).  It allows for interrogation, display,  40  modification of the model and the solutions, and setting up and executing various modelling tasks (Berman 1982).  CONSAM gives the  modeller the power of SAAM with flexibility in solving and developing models directly, and is probably the most useful programme available for modelling in vivo kinetics. To analyze a model using CONSAM, a text data file or deck must be created using the text editor.  This includes model parameters,  observations, constraints, and statistical information.  Once a model  solution has been obtained, values of the solution and parameters can be saved, and the modeller can retrieve and compare various solutions for specific parameter sets.  CONSAM has a flexible graphics facility which  displays results in graphic or character mode.  Up to five solutions can  be presented simultaneously, facilitating comparison of simulations (Boston et al.  1981).  SAAM/CONSAM is distributed, as well as continually improved and revised, by the Resource Facility for Kinetic Analysis, University of Washington, Seattle, and the Laboratory of Mathematical Biology, National Cancer Institute, Bethesda.  Manuals are available and  distributed by these centres as well; these range from introductory to advanced and include a series of tutorials (Berman et al.  1983; Berman  and Weiss 1978; Foster et al. 1989; Kawaguchi 1990). A considerable effort is required to become conversant with SAAM/CONSAM, as data entered in decks must follow a specific order or flow.  SAAM/CONSAM is not forgiving of decks containing errors, and its  41  error messages tend to be cryptic. A entirely new edition of SAAM, SAAM II, has recently been developed at RFKA, which allows models to be designed and constructed  much more easily than previous SAAM versions.  1.5.3 Cholesterol models Much is known about the physiology and biochemistry of cholesterol, and in vitro  studies have yielded details about cholesterol  synthesis: precursors, rate-limiting reactions and enzymes, and mechanisms of cellular absorption. coming in the area of in vivo  Information has been slower in  cholesterol metabolism in humans. One  valuable approach to the study of in vivo  cholesterol metabolism is  kinetic analysis of specific activity-time curves following the administration of tracer isotopes (Schwartz 1982); this will define the number and location of cholesterol pools and quantify rates of turnover. Early investigators (Gurpide et al.  1964; Chobanian and Hollander  1962), who attempted to estimate the total activity of exchangeable cholesterol from the plasma specific activity-time curve assumed that the curve reflected mixing and turnover in a single pool of cholesterol. This resulted in overestimation of both the rate of turnover and the total exchangeable mass of cholesterol.  Goodman and Noble (1968) showed -  that a two-pool model was more satisfactory.  After following the  specific activity-time curve for ten weeks, they proposed that cholesterol transfers between two body pools, and were able to estimate pool sizes, flux rates, and daily production rates. A similar approach  42  was followed by Nestel et al. (1969), with comparable results being obtained.  Samuel et al. (1968) demonstrated that the specific activity-  time curves of some subjects exhibited a third component with a lower slope, beginning 20-30 weeks following the tracer injection.  This  suggested the presence of a third pool of cholesterol which equilibrates very slowly. Goodman et al.  In an effort to define the parameters of this model, (1973) studied the long-term turnover of plasma  cholesterol in men injected intravenously with 4-14C-cholesterol.  They  were able to confirm that a three-pool model was indeed more convincing than a two-pool model when defining total body cholesterol kinetics. Rate constants were calculated for transfers between pools (kj_j , in units of days" , for transfer from Pool j to Pool i), for the cholesterol production rate (PR, in units of g/day) which equals the rate of mass transfer out of the system from Pool 1 (RQI)/ and for the mass of the rapidly turning over pool (M1# in g). Grundy and Ahrens (1969) showed that estimates of the daily turnover of cholesterol using the two-pool model obtained by compartmental analysis were 15% higher than those obtained from sterol balance.  Samuel and Lieberman (1973) verified that many decay curves  must be fitted by three rather than two exponentials; the curves can be divided into three distinct regions, each corresponding to a particular pool of cholesterol.  Furthermore, they demonstrated that long-term (66  weeks) studies of the curves yielded estimates of cholesterol turnover that were 14% less and of total exchangeable mass 26% greater than  43  previous estimates based on short-term (10-12 weeks) studies. The close agreement between estimates of cholesterol turnover, based on the sterol balance method, and those obtained from long-term analysis of the plasma specific activity-time curves, implies that estimates of total exchangeable mass based on long-term curves are nearer the true value than are those based on the two-pool model. An obvious problem with the time-curve method is that it takes a full year to complete, and requires that subjects' cholesterol metabolism remain in a steady state.  Samuel et al.  (1978) proposed that  a combination of sterol balance and short-term (10-12 weeks) analysis of specific activity-time curves be used to estimate the total exchangeable mass of cholesterol.  This method is not, however, as accurate as long-  term analysis if a steady state can be achieved.  To circumvent problems  associated with long-term sampling in humans, Dell et al.  (1985)  proposed a sampling schedule involving six large samples, each analyzed in sextuplicate.  Although this does eliminate the weekly sampling  otherwise obligatory, considerable confidence in the three-pool model is required, as the timing of the samples is critical, and optimal times will vary amongst subjects.  Ostlund (1993), using SAAM/CONSAM to  analyze plasma specific activity decay curves of labelled cholesterol, has developed a "minimal model" for human whole-body cholesterol metabolism.  The advantage of this model is that data need be collected  for only 70 d, and its simplicity allows parameter values, specifically transfer rates between central and peripheral pools, to be determined  44  with greater statistical certainty.  Its disadvantage lies in the small  number of parameters which can be determined, although the establishment of precise parameter values will certainly facilitate future work. After the development of the three-pool cholesterol model, investigators remained keenly interested in the use of tracer kinetic studies to define human cholesterol metabolism.  Schwartz et al.  (1978)  used SAAM/CONSAM to develop a multicompartmental model describing hepatic cholesterol compartments associated with the metabolism of bile acids and biliary cholesterol.  Schwartz et al.  (1982) used SAAM/CONSAM  to develop a model describing the kinetics of cholesterol precursor sources, turnover of HDL CE, steady state fluxes, and pool sizes. The latest effort from this group is a comprehensive model which identifies and quantifies cholesterol pools and transport pathways in blood and liver (Schwartz et al.  1993).  Patients with bile fistulas 3  were administered two of the following: HDL, LDL, or VLDL containing  H-mevalonate,  14 C-cholesterol,  H-FC or CE. Specific activity data from  blood and bile samples collected for the first 1800 min were used to develop the model. This model defines FC and CE pools in each of plasma, liver, bile ducts, and extrahepatic tissues, as well as turnover rates, rates of flux between pools, and precursor incorporation rates.  45  1. 6  EFFECT  OF DIET  FAT  TYPE ON RATES  OF CHOLESTEROL  SYNTHESIS  AND  ESTERIFICATION  1.6.1 Hypotheses The hypotheses of this thesis are the following: 1.  The rate of cholesterol synthesis in humans is influenced by  the degree of saturation of diet fats. 2.  The rate of cholesterol esterification in humans is influenced  by the degree of saturation of diet fats. 3.  The rates of cholesterol synthesis and esterification are  greater postprandially than during the fasted state.  1.6.2 Rationale The degree of diet fat saturation has been conclusively linked with predictable variation in serum cholesterol concentration (Section l.l).  Several metabolic adaptations engendered by the composition of  diet fat are thought to each be partially responsible for this effect (Section 1.2). Of these, changes in cholesterol synthesis rates are possibly the least well understood, weakening our ability to predict serum cholesterol levels following a particular diet, as it is not known what fraction of the deviations in serum cholesterol concentrations seen following diet fat alterations are caused by alterations in cholesterol synthesis rates. Animal models differ widely in their metabolic  46  response to diet factors, and are thus poor predictors of human performance (Dietschy 1984; Spady and Dietschy 1988). The interaction of both diet fat type and feeding state on rates of cholesterol esterification remains unclear.  Numerous studies have  demonstrated that when diet TG are saturated, the rate of esterification decreases (Hegsted et al.  1965; Mattson and Grundy 1985; McMurry et  al.  1991; Mensink and Katan 1992); esterifying enzymes have a marked preference for POLY and MONO fatty acids (Glomset 1968; 1970; Linscheer and Vergroesen 1988). unclear.  The role of the feeding state, however, remains  Two studies carried out with baboons reported both increased  (Mott et al.  1987) and decreased (Fielding et al.  1989) rates during  fasted periods, compared with fed periods. Others working with humans have demonstrated that LCAT activity closely follows serum TG concentration (Marcel and Vezina 1973; Rose and Juliano 1977; Castro and Fielding 1985), which would suggest greater FER during the postprandial state.  Whether this is independent of diet fat type remains to be seen. Studies monitoring the effect of diet fat saturation on serum  lipids have generally been carried out in the fasting state. North American humans are, however, in a predominantly postprandial state throughout the day.  Several investigators have characterized serum  lipids in this state. Total plasma cholesterol levels after a single fat-rich meal are not significantly different from fasting values; HDL values are significantly lower than the corresponding fasting values; and LDL levels are significantly higher (Cohn et al. 1988a,- 1988b; Rifai  47  et al.  1990; De Bruin et al. 1991; Groot et al.  1991).  To date,  however, the question of whether changes in either the rates of cholesterol synthesis and esterification or plasma lipid profiles attributable to feeding state are consistent across the diet fat saturation profile has not been examined.  It is hoped that any strong  trends will help influence susceptible individuals to alter possibly harmful eating patterns, with the long term effect being an achievable reduction in the incidence of CHD. Although CHD is a major cause of mortality in both women and men, this study was carried out in men since they develop CHD much earlier in life than do women (Barr et al. al.  1989; Dullaart et al.  1992; Berry et al.  1991; De Backer et  1989; Expert Panel on Detection, Evaluation,  and Treatment of High Blood Cholesterol in Adults 1993; Ginsberg et al. 1990; Groot et al. 1991; Kris-Etherton et al. 1993).  1.6.3 Significance Atherosclerosis was recently described as "a multifactorial disorder in which nutritional factors are closely related to the manifestation of the disease" (Nestel 1989) . One factor closely linked with the risk of CHD is plasma total cholesterol concentration.  In  normal individuals, every 1 mg/dL rise in plasma total cholesterol increases CHD risk by 1-2% (Kannel et al. 1979).  Furthermore, familial  hypercholesterolemia, a common genetic disorder in which plasma cholesterol levels are greatly elevated, is present in a large  48  proportion of those who suffer myocardial infarctions before the age of 60 (Reynolds 1989).  It is generally accepted, therefore, that elevated  serum cholesterol levels are implicated in the etiology of CHD. One of the factors closely linked to increases in total plasma cholesterol is diet fat intake, particularly SAT intake.  What is not  well understood is the exact mechanisms whereby diet fat affects plasma cholesterol levels.  It has been speculated that diet fat affects  cholesterol synthesis itself, but until recently no safe, precise procedures were in place for monitoring short-term changes in cholesterol synthesis in humans following consumption of meals differing in fat quality.  The D incorporation method (Jones et al.  1988; 1993a)  offers a tool to investigate this aspect of cholesterol metabolism and clarify the exact role that diet fat plays in augmenting plasma cholesterol levels in humans.  By combining a study of cholesterol  synthesis with one of cholesterol turnover, as measured by the specific activity of radioactively-labelled serum cholesterol, a clearer understanding will be gained of the role of diet fat in altering the parameters of cholesterol turnover. These include the rates of transfer of cholesterol among the three pools previously described, the size of the pools, and the speed of cholesterol turnover. The development of a model of cholesterol synthesis and turnover in humans, using the SAAM/CONSAM programme and the data generated by this study, will allow changes in cholesterol metabolism to be more readily predicted when diet factors are altered.  49  The information thus  garnered, and the predictions which it will then be possible to make, will enable us to decide, with greater precision than is presently possible, whether diet guidelines are necessary for the general public and, if so, what those guidelines should be.  1.7  THESIS OBJECTIVES  The objectives of this thesis are the following: 1.  To determine whether the degree of diet fat saturation  influences the rates of absolute and fractional cholesterol synthesis in normal individuals. 2.  To determine whether the degree of diet fat saturation  influences the rate of cholesterol esterification in normal individuals. 3.  To determine whether rates of cholesterol synthesis and  esterification in normal individuals are greater when the subject is postprandial or fasted. 4.  To develop a comprehensive mathematical model of cholesterol  turnover in normal humans, integrating study results with present knowledge.  1.8  SPECIFIC AIMS  1.8.1 Measurement of cholesterol turnover in humans Parameters of cholesterol turnover in nine male subjects were established as described by Goodman and Noble (1968) . A model of  50  cholesterol turnover was fit to the values obtained, and production •rate, exchange rates among pools, and pool sizes' were calculated.  1.8.2 Measurement of rates of cholesterol synthesis and esterification in humans Nine male subjects were randomly assigned to three groups which received SAT, MONO, and POLY diets for two-week periods in a crossover design.  Rates of cholesterol synthesis and esterification were  determined by using deuterium incorporation methodology.  Serum samples  were also analyzed for measurement of serum lathosterol concentration, and the results obtained by the two methods compared for correlation.  1.8.3 Cholesterol model development A mathematical model of cholesterol flux in humans was developed, which combined data from the cholesterol turnover study and the cholesterol synthesis rate determinations.  51  2  2.1  MATERIALS AND METHODS  CHEMICALS  All chemicals, unless otherwise noted, were purchased from either Sigma Chemical Co., St. Louis, MO., or Canlab Ltd., Burnaby, BC. Gases*s were supplied by Medigas Inc., Burnaby, BC.  Scintillation fluid  was obtained from Fisher Scientific Ltd., Edmonton, AB.  4- 14C-labelled  cholesterol was procured from New England Nuclear, a division of Du Pont Ltd., Markham, Ont. Deuterium oxide, 99.8 APE was obtained from ICN Biomedicals, Montreal, PQ.  Cupric oxide was purchased from BDH  Chemicals, Toronto, Ont., and Zn metal was procured from Biogeochemical Laboratories, Indiana University, Bloomington IN.  The silylating  reagents used in the lathosterol assay were purchased from Chromatographic Specialties Inc., Brockville, Ont. Serum total cholesterol was assayed with an assay kit (Total Serum Cholesterol Assay, Diagnostic Chemicals Ltd., Charlottetown, PEI), which is based on an enzymatic method developed by Allain et al.  (1974).  Briefly, CE are hydrolyzed to FC by cholesterol esterase, and all cholesterol is oxidized by cholesterol oxidase to cholesten-3-one, producing hydrogen peroxide (H202)-  H 2 0 2 then combines with 4-  aminoantipyrine and phenol in the presence of peroxidase to yield a chromagen with maximum absorbance at 505 nm.  The intensity of the  colour produced is directly proportional to the amount of cholesterol in the sample.  Serum FC was assayed with a kit (Boehringer Mannheim Canada  52  Ltd., Dorval, PQ) which is similar to that described above, but does not include the cholesterol esterase.  2.2  EQUIPMENT  Vacutainer tubes (Canlab Ltd., Burnaby, BC) containing no anticoagulants were used for all blood draws. Acrodisc syringe filters, pore size 0.2/im (Gelman Sciences, Montreal, PQ), were used to coldsterilize labelled serum samples prior to injection into subjects. Thin layer chromatography plates were ordered from Baxter Corp., Burnaby, BC. Pyrex tubes used for combustion and reduction were obtained from Corning Glass Works, Corning, NY. Microcapillary tubes (2 p.1) used for measurement of serum, water D enrichment were obtained from Fisher Scientific Ltd., Edmonton, AB.  O-rings and fittings for the  distillation line were received from Columbia Valve and Fitting, North Vancouver, BC. Microvials used in the lathosterol assay were purchased from Hewlett-Packard. The scintillation counter used to evaluate serum specific activity was an ISOCAP/3 00, Nuclear-Chicago, equipped with an external standard and a series of certified samples of known concentration. Deuterium enrichment of the hydrogen gas obtained from the reduced serum cholesterol samples was determined by isotope ratio mass spectrometry (VG Isomass, 903D, Cheshire, England). The GLC used to verify the diet fatty acid profile and measure serum lathosterol levels was a Hewlett Packard  53  (HP) 5890 Series II  equipped with flame ionization detection.  Fatty acid separation was  performed on a HP-5, 25 m capillary column, 0.20 mm diameter, 0.33 /xm thickness.  Sterol separation was achieved on a Restek RTx-1, 30 m  capillary column, 0.25 mm internal diameter, 0.25 /xm thickness from Restek Corp., Bellefonte, PA. Other equipment used included two centrifuges (Silencer model H103FRS, Western Scientific Services, Burnaby, BC; International portable refrigerated centrifuge model PR-2, International Equipment Co., Needham Heights, Mass USA), a freeze-dryer (Labconco Freeze Dryer 5, Fisher Scientific Ltd., Edmonton, AB), a bioelectrical impedance analyzer (BIA) (Model 101, RJL Systems, Detroit, MI), a muffle oven (Fisher Isotemp muffle furnace model 184A, Fisher Scientific Ltd., Edmonton, AB), a spectrophotometer (Coleman Hitachi 101, model 111-050, Maywood, Illinois USA), a water bath equipped with a shaker (Precision Scientific Co., Chicago), a dri-bath modular heater (model H2105-1, Canlab Ltd., Burnaby, BC), weighing scales (Mettler models PE2000 and H54AR, Fisher Scientific Ltd., Edmonton, AB), a -80°C freezer (Series 100, Kelvinator Commercial Products Inc., Manitowoc, Wis USA), an autoclave (Market Forge Sterilimatic model 120/208-240, Market Forge Co., Everett, Mass USA), and a wrist action shaker (Burrell Corp., Pittsburg, PA USA).  2.3  STUDY  DIETS  Three study diets were designed for use in the cholesterol synthesis investigation: SAT, POLY, and MONO.  54  Research diets were  planned using Nutricom  (Smart Engineering, Vancouver, BC), a diet  analysis computer programme based on Canadian Nutrient File data (Health and Welfare Canada, 1988).  Data were inspected by a registered  dietician from St. Paul's Hospital, Vancouver, BC.  Each diet provided  40% of energy as fat, 45% as CHO, and 15% as protein.  Diets differed  only in the major fat provided: in the SAT diet it was butter, in the POLY diet it was a soft safflower oil-based margarine (Fleischmann's), and in the MONO diet it was olive oil. The estimated P:S ratio of, the SAT diet was 0.3 5; that of the POLY diet was 1.4. Meals were prepared using fresh, canned, and frozen ingredients purchased at a local supermarket. All ingredients were weighed to the nearest 0.5 g.  A 2-day diet cycle was used.  A detailed meal plan for  one of the subjects is provided as an example in Appendix 1. Fat composition was also manipulated by varying the lunchtime yogurt: lowfat was served in the POLY diet, high-fat was served in the SAT and MONO diets, and ice milk was served in the POLY pasta supper, whereas ice cream was served in the SAT and MONO pasta supper. beverages were served.  No caffeinated  Subjects were provided with a serving of diet  carbonated beverage after supper, for consumption as an evening snack. Meal times were 8:00-9:00 AM, 12:30-1:30 PM, and 5:30-6:30 PM. During each diet period, all meals were consumed under supervision in the Metabolic Research Unit, School of Family and Nutritional Sciences, University of British Columbia (UBC).  55  Subjects were instructed to eat  all the food provided, and were asked not to eat or drink anything, except water, which was not part of the prepared meals. Daily REE were estimated for each subject using the Mifflin predictive equation (Mifflin et al.  1990).  This equation is:  REE = 10 x wt (kg) + 6.25 x ht (cm) - 5 x age + 5  Equation  1  REE values were multiplied by an activity factor,of 1.7 to yield total daily energy requirements (Bell et al.  1985) .  Sample meals from each diet were blended and freeze-dried, and energy content verified by isothermal bomb calorimetry (LECO AC300, LECO Corp, St. Joseph, MI), using benzoic acid as a combustion standard. Fatty acid composition of the meals was determined by gas liquid chromatography after lipid extraction (Folch et al. trifluoride methylation (Bannon et al. 1982).  1957) and boron  Certified fatty acid  sample mixtures were used to identify peaks (Supelco Inc., Bellefonte, PA).  GLC running conditions were: initial temperature 18 0°C, ramping  l°C/min to 210°C, holding for 3 0 min. Total time of run was 60 min. The injector temperature was 300°C and the detector temperature was 320°C. mL/min.  Column flow rate was 1 mL/min and purge vent flow rate was 5 Split ratio was 100:1.  Carrier gas was helium, with nitrogen  gas used as a makeup gas at the detector.  56  2.4  SUBJECT  SELECTION  Posters were placed in public buildings about the university campus.  Screening criteria were: male, age 20-45 years, fasting serum  cholesterol concentration 3.4-5.7 mmol/L, body fat content 10-15%, low to moderate daily physical activity, no use of medication, and no chronic medical condition. Forty-five potential subjects contacted the investigator and expressed interest in participating. carefully explained to the men.  The experimental procedure was  Primary reasons for dropping out at  this point were: (1) the length of the experiment and (2) fear of receiving an intravenous injection of radioactively-labelled cholesterol.  The remaining candidates' fasting serum cholesterol  concentrations were determined, and approximately 33% of the men were rejected for having levels either too high or (in one case) too low. Nine men (25.6 +1.0 years of age) remained who met the criteria and wished to participate; these became the subjects for both the cholesterol turnover and the cholesterol synthesis rate determinations. The experimental protocol was approved by the Clinical Screening Committee for Human and Other Studies Involving Human Subjects, UBC Office of Research Services. Certificate of Approval: #C88-320. 14 The use of 4-  C-cholesterol in humans in this study was approved  by the Atomic Energy Control Board of Canada, and the procedures involved with its use were approved by the UBC Radiation Safety Office.  57  2.5  CHOLESTEROL  TURNOVER  2.5.1 Preparation and delivery of injectate The day before the experiment started, 30 mL blood samples were drawn from fasted subjects. After resting at 4°C for 15 min, samples were spun at 3000g for 20 min at 4°C, and serum was extracted.  Serum  was then inoculated with a radioactive tracer using a modified version of the method described by Goodman and Noble (1968) . Twenty-five /iCi of 4-  C-cholesterol dissolved in ethanol was used per subject.  Preliminary investigations revealed that a ratio no greater than 1:10 ethanol:serum can be used if serum precipitation is to be avoided.  For  each subject, 625 /il of labelled cholesterol in ethanol was evaporated under nitrogen gas to reduce the volume by 50%, and the resulting concentrate was injected slowly beneath the surface of 3.5 mL serum. One mL serum was used to rinse the sides of the tube in which cholesterol solution was concentrated; this was added to the original 3.5 mL.  Labelled serum was shaken gently at 37°C for two h, and at room  temperature for an additional 15 h.  Serum was then cold-sterilized by  passage through an Acrodisc syringe filter, pore size 0.2 /xm, and injected into an antecubital vein within one h of sterilization. Sterile technique was maintained throughout the procedure: all glassware was sterilized prior to use, and  needles, syringes, and plastic vials  were purchased pre-sterilized and discarded after a single use. A 10 /il injectate aliquot was reserved and counted; the resulting count was  58  multiplied by the total volume injected into the subject to determine total /xCi per load.  2.5.2 Sampling protocol Subjects were requested to report to the laboratory after an overnight fast, and a 10 mL blood sample was drawn from the antecubital vein.  Samples were subsequently drawn over a 9 month period: three  times in the first week, once per week for the remainder of the first three months, fortnightly for the following three months, and once every three weeks for the final three months. Blood was always drawn at the same time of day to minimize any effects of diurnal variation. restrictions were imposed during this study.  No diet  Total serum cholesterol  was measured in each sample, as well as serum specific activity.  2.5.3 Measurement of serum specific activity Duplicate serum sample aliquots of 200 /xl were counted by a liquid scintillation system with and without an external standard, to allow correction for quenching caused by materials in the samples such as proteins, pigments, or hemoglobin.  Certified standards were also  counted and used to construct a standard curve. for 10 min.  Samples were counted  Serum samples drawn from each subject prior to the tracer  injection were also counted and subtracted from each sample as background radiation.  Counts per 10 min were converted to DPM.  59  2.5.4 Measurement of total serum cholesterol Total serum cholesterol was measured by enzymatic assay (Allain et a.1.  1974) using 10-20 \x\ aliquots of sample.  duplicate.  Each sample was run in  A certified standard provided with the kit was used to  construct a standard curve. intraassay CV was 0.9%.  The interassay CV was 2.2% and the  The interassay CV measures variability between  test runs, for example when using the same assay or the same equipment on different days.  The intraassay CV is an estimate of the variability  in test response in the same sample measured repeatedly.  2.6  CHOLESTEROL  SYNTHESIS  2.6.1 Experimental protocol Subjects were randomly assigned to three groups which received the three diets, SAT, POLY, and MONO, in a crossover design. fed for 13 days, followed by one day of fasting.  Each diet was  Bolus doses of  2  H20  were prepared for the subjects: 0.7 g 2H^O/kg estimated total body water.  Total body water was estimated as 73% of the fat-free mass (Pike  and Brown 1975) , which was assessed using BIA (Kushner and Schoeller 1986).  On day 13, fasting baseline blood samples were drawn, followed  by administration of the oral dose of deuterated water; this was followed by breakfast. intervals for 48 h.  Blood samples were subsequently drawn at 12 h  This period corresponded to a normal "fed" day  followed by a fasted day, allowing comparison of cholesterol synthesis  60  rates and lipid levels on the three different diets, both postprandially and while the subjects were fasted.  Body water D levels were maintained  by consumption of lightly labelled water which was prepared for the subjects to carry with them while they were away from the Metabolic Unit; it contained 1.4 g  H20/kg H20.  Eight weeks elapsed between the  end of one diet and the beginning of the next.  2.6.2 Measurement of serum lipids All samples were measured in duplicate, with certified standards being used to construct standard curves. Total serum cholesterol was measured (Section 2.5.4) . Serum FC was measured by enzymatic assay (Stahler et al. 0.5%.  1977; Trinder 1969).  The intraassay CV was less than  Total serum TG was assayed enzymatically (Bucolo and David 1973) .  HDL was similarly measured after precipitation of apo B-containing lipoproteins (Warnick et al.  1985).  Rather than estimating LDL, HDL was  subtracted from total cholesterol yielding an estimate of apo Bcontaining lipoproteins, the concentration of which could then be followed during both fed and fasted periods. All of the lipid analyses except for the FC determinations were performed at the Lipid Research Centre, University Hospital-Shaughnessy Site, UBC.  61  2.6.3 Measurement of serum cholesterol deuterium uptake  2.6.3.1  Validation  of serum sample  size  A validation trial of serum sample size needed to obtain reproducible enrichment when measuring D enrichment was conducted after FC sample enrichment had been determined but before CE samples were processed.  Two samples were prepared: one consisted of serum from a  recently deuterated individual, the second was a FC standard dissolved in chloroform.  Triplicate aliquots of either one, two, three, or four  mL of each sample were processed as described below (2.6.3.2 to 2.6.3.7).  Total cholesterol concentration in serum samples was 185 +  3.3 mg/dL, and FC standards were prepared containing similar cholesterol concentrations.  2.6.3.2  Extraction  of total  lipid  To each tube, which contained 2-4 mL serum, 8 mL methanol was added.  Tubes were heated at 55°C under nitrogen for 15 min, after which  24 mL hexane-chloroform 4:1 (v/v) was added.  Samples were shaken for 15  min, 2 mL H 2 0 was added, and samples were shaken for an additional 15 min, followed by centrifugation at 1500g- for 15 min.  The upper  (organic) phase was removed and the process was repeated. the combined organic phases was dried down under nitrogen.  62  Solvent in  2.6.3.3  Separation  and isolation  of lipid  fractions  Extracts were dissolved in 2 00 /il chloroform and spotted onto thin layer silica plates, which had been recently activated by heating at 100°C for 30 min.  Plates were developed using petroleum ether-diethyl  ether-acetic acid 135/15/1.5 (v/v/v) for 1 h (Jones et al.  1988; 1993a).  Plates were air dried, and identification of cholesterol fraction was made by visualization in iodine vapor against a FC standard. Bands of FC and CE were scraped into separate tubes. Tubes containing CE were stored at -80°C until they could be saponified.  Six  mL of hexane-chloroform-diethyl ether 5/2/1 (v/v/v) was added to the tubes containing the FC fraction; these were shaken for 15 min and centrifuged at 1500gr for 15 min.  Solvent was carefully removed, and  extraction from silica was repeated twice. This was dried down under nitrogen, and transferred to combustion tubes.  2.6.3.4  Saponification  of cholesterol  ester  fraction  Silica containing the CE fraction was combined with 5 mL methanolic KOH, capped tightly> and boiled (100°C) for 30-40 min. After cooling, 3 mL H20 and 13 mL petroleum ether was added per tube, which was shaken for 5 min the centrifuged at 1500g for 5 min.  The ether  phase, the top layer, was removed and retained, and the final extraction was repeated.  The combined ether phases were dried down under nitrogen,  and were then plated on activated thin layer silica plates, as described above, to separate the FC from the free fatty acids.  63  2.6.3.5  Combustion  tube  transfer  and water  evolution  Each silica sample to which cholesterol was bound was dissolved in a small amount (< 500 ixl) of chloroform and pipetted into pre-annealed 18 cm by 6 mm Pyrex tubes containing 500 mg cupric oxide and a 2 cm piece of silver wire. Tube containing sample was then placed in liquid nitrogen and frozen (30 sec). Care was taken to avoid prolonged exposure to atmosphere once frozen: this is a potential source of water contamination.  Solvent was removed by gradual evacuation for 7 min to a  final pressure of less than 50 microns, the tube was cut and sealed using an acetylene torch.  Each tube was shaken to distribute cupric  oxide along the entire length of the tube and combusted in a 520°C oven for 4 hours, then allowed to cool to room temperature.  Moisture was  visible in each tube.  2.6.3.6  Distillation  and  reduction  Combustion water was vacuum-distilled into pre-annealed 18 cm by 6 mm Pyrex tubes containing 60+5 mg Zn metal. To accomplish this, the tube containing the sample was inserted in a cracker assembly and placed on one end of an assembly resembling an inverted W, with a U-joint at centre.  The Zn-containing tube was placed on the other end and the  system evacuated for 10 min.  The sample tube was frozen in liquid  nitrogen for 3 min, allowing sample water and carbon dioxide to freeze, and the U-joint was immersed in a methanol-liquid nitrogen mixture, (-80 C to -100°C).  The system was closed to vacuum, liquid nitrogen was  64  removed from around the sample tube, and the tube was broken in the cracker assembly, causing the pressure to rise to 200-500 microns. Carbon dioxide was evacuated by slowly opening the line to vacuum; too rapid an evacuation would cause sample water to be evacuated as well. After 30-45 sec, the system was again closed to vacuum, and the cracker tube assembly heated for 1 min.  During this stage, sample water is  being trapped in the frozen U-tube joint. After heating, the system was allowed to rest undisturbed for 5 min and the pressure returned to baseline, indicating that all the water was frozen and the carbon dioxide evacuated.  Liquid nitrogen was added around the Zn-containing  tube, just freezing the tip, the methanol-liquid nitrogen slurry was removed from around the U-joint, and the U-joint was heated for approximately 30 sec, causing the pressure to rise initially, then return to baseline.  During this stage water is being transferred from  the U-joint to the Zn metal. After resting for 5 min, the tube was cut and sealed using an acetylene torch, then heated at 520°C for 30 min, causing the reduction of sample water to hydrogen gas.  Tubes were  allowed to cool to room temperature.  2.6.3.7  Mass  spectrometry  Deuterium enrichment of hydrogen evolved from cholesterol samples was measured by isotope ratio mass spectrometry, with an internal analytical error of 0.17 per mil (°/oo).  The mass spectrometer was  calibrated daily using three standards: SMOW, SLAP, and GISP, which were  65  used to construct a standard curve. All samples for each subject were analyzed on the same day.  The traditional calculation of isotope  enrichment is some variant of the difference between the observed ratio of the sample to that of a reference gas, and is expressed as delta or del: in this study °/oo values represent parts per thousand changes in enrichment of the sample relative to the D abundance in SMOW (Cobelli et al.  1987).  The equation used is:  [(R  sample ~ RSMOw)/RSMOwl  x 100  ° °/°°  where R is the ratio of heavy to light isotopic species.  Equation  2  The per mil  designation is used because of the relatively small enrichments encountered (Jones et al. duplicate.  1988; 1993a).  Samples were analyzed in  Interassay and intraassay CV could not be determined as  samples are destroyed during analysis. Average standard values obtained were (X + SE): SMOW: 0.43 + 0.82, SLAP: -336.81 + 3.31, GISP: -140.602 + 1.89.  The overall precision level of mass spectrometer analysis in this  study was determined by averaging replicate standard deviations.  The  average standard deviations of D enrichment for body water, serum FC, and serum CE were 1.7, 8.0, and 6.2 parts per thousand (°/oo) respectively, relative to SMOW.  66  2.6.3.8  Measurement  of body water  deuterium  enrichment  To measure the D enrichment of body water, serum samples from 24and 48-h time points were diluted six fold with Vancouver tap water, reducing D enrichment to within the normal range of the mass spectrometer.  Baseline samples were not diluted.  measured in duplicate or better.  All samples were  Preparation of serum water samples for  mass spectrometry was as follows: a pre-annealed 10 cm by 6 mm Pyrex tube was placed in a holder an attached to on side of the U-joint assembly, with a Zn-containing tube on the other side; the system was evacuated for 10 min. gas introduced.  The system was then closed to vacuum and nitrogen  A 2 /xl microcapillary tube containing serum was  introduced into the empty tube, nitrogen gas was shut off, and the serum sample was frozen by partial immersion in liquid nitrogen for three minutes.  The system was evacuated until the pressure returned to  baseline, then closed to vacuum and the U-joint immersed in a methanolliquid nitrogen mixture, (-80°C to -100°C).  Liquid nitrogen was removed  from around the inlet tube containing the serum sample, and it was heated for 1 min then allowed to rest for 2 min.  The tip of the zinc-  containing tube was frozen with liquid nitrogen, the methanol-liquid nitrogen slurry was removed from the U-joint, and it was heated until the system pressure returned to baseline, indicating that the transfer of water from the serum sample to the zinc-containing tube was complete. The system was evacuated and the tube cut. the tube was cut and sealed using an acetylene torch, then heated at 520 C for 30 min, causing the  67  reduction of sample water to hydrogen gas. Tubes were allowed to cool to room temperature before analysis by mass spectrometry.  2.6.4 Measurement of serum lathosterol concentration Two time points were selected during each of the three 48-h sampling period, one each during fed and fasted periods and exactly 24 h apart to avoid any confounding diurnal effects.  Serum obtained at 12  and 36 h was assayed for lathosterol content, using the procedure described by Hamilton et al.  (1992).  An internal standard was prepared  by dissolving 5 mg 5-alpha cholestane in 5 mL chloroform, placing 50 itl in a vial and dried; once it was needed, 5 mL chloroform was added to the vial to give the desired concentration of 1 mg/100 mL.  For each  sample, 50 pil of 5-alpha cholestane (Sigma C8003), an internal standard, was added to 200 izl serum.  This was saponified with 3 mL  freshly made  methanolic KOH (94 mL methanol and 6 mL 50% KOH) for 16 h at 80°C. Samples were allowed to cool prior to the addition of 2.65 mL H 2 0. Three mL petroleum ether was added, the samples shaken, the layers separated by spinning in a centrifuge for 5 min at 1500gr, and the top layer retained.  This extraction was repeated twice, the concentrate was  transferred to a microvial, and dried under nitrogen at 45-50°C. Vials were topped with septum lined caps and stored at 4 C until needed. Prior to injection into the gas liquid chromatograph, samples were silylated by the addition of 20 itl hexamethyldisilane, 2 fil trimethylchlorosilane, and 5 til dimethylformamide into each vial.  68  This  was mixed thoroughly and allowed to rest for 5 min before a 2 /jl aliquot was injected into the GLC.  This 5 min period was adhered to strictly by  the investigator. Gas liquid chromatograph running conditions were: initial temperature 80°C, hold one min,- ramp to 12 0°C at 2 0°C/min, hold for 7 min; ramp to 249°C at 20°C/min, hold for 15 min; ramp to 269°C at 20°C/min, hold for 25 min. At the end of each run the temperature was ramped to 320°C and held for a minimum of 5 min, ridding the column of contaminants.  The injector temperature was 300°C and the detector  temperature was 320°C.  Column flow rate was 1 mL/min, split vent flow  rate was 4.5 mL/min, and purge vent flow rate was 5 mL/min.  Carrier gas  was helium, with nitrogen gas used to adjust the flow rate. Lathosterol peak identification was confirmed using a certified standard (7,(5-alpha)-cholesten-3 S-ol, Sigma C7400).  A series of  standards was prepared exactly as were the actual samples: each contained 50 /xl 5-alpha-cholestane, the internal standard, plus either 0, 10, 25, 50, 75, or 100 jxl lathosterol.  This series produced a linear  response of lathosterol to the internal standard, and was later used to determine sample lathosterol values. A series of standards was injected every second day to verify that the response was not changing. Similarly, a serum standard was used every second day and used to calculate assay variability.  The interassay CV was 11.2%.  Intraassay  CV could not be determined because samples deteriorate following silylation; repeating the injection using the same prepared sample after  69  over an hour would have yielded different results. This was confirmed by the investigator.  Lathosterol was reported as /iM and as the ratio of  lathosterol to cholesterol.  2.7  MODELLING  DATA  SETS  2.7.1 Modelling turnover data For each sample taken, two numbers were obtained: one for total serum cholesterol concentration and one for DPM per mL serum.  These  were used to generate serum cholesterol specific activity (DPM/mg cholesterol) which was adjusted to reflect theoretical C  cholesterol  enrichment if 25 /xCi had been administered to each subject instead of the variable amounts which were received.  This step was taken to allow  comparison of specific activity decay curves, which were generated by plotting DPM/mg serum cholesterol versus time in days, among the nine subjects.  One data set was obtained per subject.  Each data set was  analyzed by SAAM/CONSAM, which used the fourth order variable step-size Runge Kutta method, and the equation best describing each curve was generated (Foster and Boston 1983).  This same data was then subjected  to compartmental analysis by SAAM/CONSAM using the Chu and Berman method, with the number of compartments equal to the number of terms in the multiexponential equation best describing it (Foster and Boston, 1983).  This exercise yielded values for several parameters which are  70  unique to each subject: rates of exchange among compartments, total cholesterol turnover, and pool sizes.  2.7.2 Modelling free cholesterol synthesis data Several methods were used to develop models to calculate cholesterol synthesis; these are described below.  2.7.2.1  Linear uptake of  tracer  The first method is based on a linear rate of uptake of label into cholesterol, and is the method established by Jones et al. (Section 1.4.4.1).  (1988; 1993a)  De novo cholesterol FSR is calculated for each  period in question; in this study there were four 12-h periods per diet trial.  Use of this method assumes that de novo cholesterol originates  in one pool, the serum pool.  [del C n 0 i estero ]- (°/0°) X 2 ] /  FSR (per day) = [del body  The equation used is:  water  Equation  3  (°/oo) X 0.81 H/C X 27C/46H]  where delcj10]_estero]_ and del^^y  wa ter  are  differences in D enrichment  of each tissue expressed as parts per mil versus SMOW.  The amending  factor in the equation's denominator corrects for the absolute ratio of carbon to hydrogen atoms within the cholesterol molecule (Section 1.4.4.1); the number 2 in the numerator converts the 12 h FSR to a 24 h FSR for each period.  Absolute cholesterol synthesis was calculated by  71  multiplying the FSR value by the size of the central cholesterol pool, which was determined previously using SAAM/CONSAM to analyze the specific activity decay curves. The pool size was first divided by three to allow for calculation of de novo synthesis of FC only; use of the original pool size value would have greatly exaggerated the amount of cholesterol synthesized. The second linear method is that established more recently by Jones et al. (1994a; 1994b), where rates are calculated as the change in product enrichment over time divided by the maximum possible enrichment.  FSR (per day) = [del body  Equation  ^^cholesterol (°/oo) ] /  water  4  (°/oo) X 0.81 H/C X 27C/46H]  where delcj10]_estero]_ is the difference in serum cholesterol from zero to each time point, and the numerator is the maximum cholesterol enrichment possible.  This calculation was carried out at each time point.  Its  advantage is that it doesn't generate any negative FSR values, and its disadvantage is that it ignores isotope turnover in the later periods. This method uses the same assumptions as does the linear method outlined above.  2.7.2.2  Compartmental  modelling  based  on monoexponential  tracer  uptake  SAAM/CONSAM was used to derive FSR and de novo cholesterol values, assuming that de novo cholesterol arises from one pool only.  72  This  method was based on two assumptions: the D uptake rate is best described by a monoexponential equation, and cholesterol enrichment reaches a plateau at the TME, which is 0.478 times the plateau body water D enrichment (Jones et al.  1993a).  Use of this method assumes that all  cholesterol entering the central pool is newly synthesized, which is incorrect: diet cholesterol as well as cholesterol incoming from side pools is also incorporated.  These represent sources of label dilution,  which prevent the TME from being attained.  Consequently, the maximum  enrichment plateau was adjusted by assimilating the assumption that approximately 200 mg of cholesterol is absorbed from the diet per day (Nestel et al.  1969) with knowledge of the total exchange between pools  in each subject derived from the specific activity decay data (Section 2.7.1).  This allowed prediction of the degree in which the final  cholesterol deuterium enrichment plateau would deviate from the theoretical plateau, i.e. the EME.  For example, if compartmental  modelling predicts that 20% of the cholesterol in the central pool comes from unlabelled sources, the TME was lowered by this fraction, yielding the EME. As mentioned above, The pool size was multiplied by 0.33 to allow for the fraction of the pool which is FC. Calculations were done for each 12-h period and each 24-h period; each represents a particular physiological state, i.e., either fed or fasted and either daytime or nighttime.  The rate of cholesterol  synthesis is apparently subject to diurnal variation (Jones and Schoeller 1990) .  73  One problem with this method, however, is that SAAM will not calculate FSR values if the enrichment drops. Negative slopes are invariably seen overnight in FC D enrichment.  Negative uptake rates  reflect an influx of unlabelled cholesterol which dilutes the apparent enrichment, as well as a possible decline in de novo synthesis, apparently brought about by fasting.  To allow estimation of synthesis  during these periods, the angle of D incorporation was compared to a previous angle in the same data set where the, FSR was generated by SAAM/CONSAM, and the FSR adjusted proportionately.  Use of this method  assumes that diet fat type has no discernable effect on rates of exchange of cholesterol between pools; this will be discussed later, along with other shortcomings of this method. Subjects were fed for the first 24-h period then fasted for the second 24-h period, which represents a disruption of steady state conditions.  For this reason it was not possible to use SAAM/CONSAM to  model each 48-h period in its entirety, since steady state conditions are required for results obtained from SAAM/CONSAM to be valid.  2.7.3 Modelling cholesterol ester data  2.7.3.1  Linear  uptake  of  tracer  Rates of esterification were calculated as described in Section 2.7.2.1, using enrichment data from esterified cholesterol and body  74  water.  The pool size was multiplied by 0.67 to allow for calculation of  cholesterol esterification.  2.7.3.2  Compartmental  modelling  based  on monoexponential  tracer  uptake  Esterification rates were determined by modelling as described in Section 2.7.2.2, using enrichment data from esterified cholesterol and body water.  The pool size was multiplied by 0.67 to allow for  calculation of cholesterol, esterification.  2.8  STATISTICAL  ANALYSES  All results were expressed as X ± SE except for those generated using SAAM/CONSAM.  This software package only calculates SD, and it is  not clear either how many points are used in the calculation or whether points of the curve are weighed equally when determining statistical parameters. manually. 1974).  This makes it difficult to confidently calculate SE values  Statistical significance for all tests was set at P<0.05 (Zar  Whether or not data sets were distributed normally was confirmed  with the Kolmogorov-Smirnov D statistic; non-normal data were normalized by reciprocal or logarithmic transformation. sets were analyzed using ANOVA.  Normally distributed data  Data sets which could not be normalized  were ranked and analyzed nonparametrically.  Unless noted otherwise,  multiple comparison testing, when required, was performed with Tukey's test. N.C.).  Analyses were done using SAS version 6.04 (SAS Institute, Cary, All data sets which looked at the effect of diet were tested for  75  an effect of treatment sequence.  In no case was a significant result  found; these statistics are therefore not listed in the results.  2.8.1 Subjects' body weight and energy intake ANOVA was used to compare subjects' body weight at the start of each diet period, as well as total energy intake. ANCOVA was used to compare final body weights with initial body weight as the covariate.  2.8.2 Diet configuration Data sets were interpreted with ANOVA, to see whether the diets differed in protein, fat, CHO, cholesterol, total energy, and fatty acid composition.  2.8.3 Serum lipids Data were interpreted with ANOVA for a factorial experiment, looking for an effect of diet or time, as well as interactions.  Data  examined included: total serum cholesterol, HDL, apo B-containing (total cholesterol minus HDL), and TG concentrations, as well as the ratio of HDL to total cholesterol.  Free cholesterol, which was only measured at  one time point (t=0), was interpreted with ANOVA, examining the effect of diet fat type on free cholesterol content of total cholesterol.  76  2.8.4 Specific activity decay curves The Gauss-Markov F test was used to find the number of terms in a multiexponential equation which would best describe each specific activity decay curve (Goodman et al.  1973).  The percent improvement in  residual error when using a three-term instead of a two-term equation, a measure of closeness of fit to the equation, was also calculated for each, data set. When generating equations with SAAM/CONSAM, the best fit was generated by the programme using the Marquardt nonlinear least squares fitting technique (Marquardt 1963) .  2.8.5 Serum lathosterol Data were interpreted with ANOVA for a factorial experiment, looking for an effect of diet or time.  In addition, Spearman's  coefficient of rank correlation was calculated between both FSR and de novo synthesized cholesterol values, which were generated by modelling D uptake data, and serum lathosterol concentrations.  2.8.6 Validation of serum sample size for mass spectrometry Each data set, i.e., the deuterated plasma samples and the FC/chloroform samples, was interpreted with ANOVA.  2.8.7 Cholesterol deuterium incorporation Data were interpreted with ANOVA for a factorial experiment, looking for an effect of diet or time, as well as interactions, on each  77  parameter.  Data analyzed were: FC, CE, and body water D enrichments.  In addition, ratios of both FC enrichment and CD enrichment to body water enrichment were calculated, then analyzed in a similar fashion. Subjects did not attain identical body water D concentrations, as seen in Appendix 2, and the availability of D from body water may be an important factor influencing cholesterol D incorporation rates. Before analysis, baseline values were subtracted from each data set, setting the baseline value to zero in each case.  This emphasized actual changes  in D enrichment over time and facilitated comparisons among subjects on the three diets. Baseline D levels varied considerably from diet to diet for each subject, since the time which elapsed between the end of one test period and the beginning of the next was not long enough to allow a return to baseline values.  2.8.8 Compartmental models and model parameters When generating compartmental models to best describe data sets, the best fit was determined by SAAM/CONSAM using both linear and nonlinear least-squares fits to the data (Berman 1982; Foster and Boston 1983).  All model parameters were interpreted using ANOVA with a  factorial experiment, looking for an effect of diet or time, as well as interactions.  78  3  3.1  SUBJECTS'  BODY WEIGHT,  PERCENT  RESULTS  BODY FAT,  AND ENERGY  INTAKE  Mean body weights at the beginning and end of diet periods are in Table 3.1; individual values in Appendix 3.  Neither initial (P<0.990)  nor final weight (P<0.995) varied among the three diets, nor did they differ from the beginning to the end of each period (P<0.873). Energy intake did not vary either (P<0.950) (Table 3.2 and Appendix 4) .  3.2  DIET  CONFIGURATION  Meals did not differ in protein (P<0.971), fat (P<0.243), CHO (P<0.979), or cholesterol content (P<0.722) (Table 3.3).  Diets differed  in content of the following fatty acids (Table 3.4): C8:0 (P<0.017), C10:0 (P<0.001), C14:0 (P<0.042), C18:0 (P<0.024), C18:1 (P<0.001), and C18:2 (P<0.001), but did not differ in C16:0 (P<0.064).  Percentage of  each class of fat in each diet is shown in Table 3.5: meals within each diet did not differ significantly in their fatty acid composition with the exception of C18:1 content (P<0.038); its percentage in lunches was slightly greater than that of the spaghetti supper.  Fats not mentioned  in Table 3.4 were not seen during GLC analysis.  3.3  SERUM  LIPIDS  Serum lipids values measured during each 48-h sampling period of the diet trials are in Tables 3.6 and 3.7; individual values in Appendix  79  5.  Total cholesterol concentrations (Figure 3.1) did not change with  time of sampling (P<0.865); there was, however, an effect of diet (P<0.001), with levels on the SAT diet 12.3% greater than those seen on the MONO diet, which were 10.1% greater than those observed on the POLY diet.  The ratio of serum FC to CE was not affected by diet (P<0.256). Serum TG concentrations (Figure 3.2) showed a significant effect  of time (P<0.001): TG levels were higher at the 12 h time point, 1.5 h after supper.  Levels returned to baseline values over the next 12 h and  continued to decline, with the 48-h value lower than those at 0, 12, or 24 h.  Serum TG concentrations were also influenced by diet (P<0.026),  those on the SAT diet 22.1% greater than seen on the POLY diet. A diet effect was noted on serum HDL concentrations (Figure 3.3) (P<0.046), with levels in subjects on the MONO diet 8.2% higher than while on the SAT diet.  HDL levels varied slightly but not significantly  with time of sampling during the fed-fasted cycle (P<0.095).  The ratios  of HDL to total serum cholesterol were lower (19.7%) in subjects fed the SAT diet than with either of the other diets (P<0.001).  Ratios were not  affected by sampling time (P<0.350). Levels of non-HDL, or apo B-containing cholesterol, which is displayed in Figure 3.4, did not vary significantly with time of sampling (P<0.954), but differences were seen among levels on the three diets (P<0.001).  Concentrations observed on the SAT diet were 19.1%  higher than those reported on the MONO diet, which were in turn 12.4% greater than corresponding levels on the POLY diet.  80  3.4  SERUM  LATHOSTEROL  Serum lathosterol concentrations and lathosterol:cholesterol ratios seen during each diet period are in Table 3.8; individual numbers obtained by each subject are in Appendix 6.  Lathosterol concentrations,  displayed in Figure 3.5, were higher at the 12 h time point compared with the 36 h time point (P<0.010). found (P<0.250).  No significant diet effect was  Similarly, the ratio of lathosterol:cholesterol was  higher on all diets at the 12 h time point (P<0.008), but no significant diet effect was noted (P<0.172).  3.5  VALIDATION  OF SERUM SAMPLE  SIZE  FOR MASS  SPECTROMETRY  Mean D enrichment levels in the sample size validation trial are shown in Table 3.9; actual values are in Appendix 7.  The enrichment  results from 1 mL deuterated plasma samples were lower than those from the 2, 3, or 4 mL samples (P<0.016).  Similar results were obtained from  the FC/chloroform samples (P<0.001).  Consequently, no samples smaller  than 2 mL were used in the cholesterol synthesis and esterification rate determinations.  The SE values for this data set are smaller than the SE  values seen in the FC D 2 0 enrichment data.  This is thought to be  related to differences in equipment performance and chemical availability between the two periods when the two sets of samples were processed.  81  Table 3.1 Subject body weight at beginning and end of two-week diet periods  Diet  Init wt (kg)  F i n wt (kg)  Change (%)  MONO  69.1 + 2.2  68.7 + 2 .1  -0.66 + 0.52  POLY  68.8 + 2.0  68.5 + 2 .0  -0.42 + 0.44  SAT  68.8 + 2.1  68.7 + 2 .0  0.03 + 0.32  n=9, X + SE. Init wt: initial weight; fin wt: final weight; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Change is calculated as: ((final weight minus initial weight) divided by (initial weight)), multiplied by 100.  82  Table 3.2  Daily energy intake (kJ) on each of the three diets  Sub  X ± SE  A  12595 + 169  B  13478 + 160  C  14078 + 167  D  14530 +  E  13658 + 181  F  12989 ± 163  G  12663 + 175  H  12912 +  0  I  12716 +  2  *  4  Subjects.  83  Table 3.3  Calculated meal composition of each diet  SAT  MONO  POLY  Protein (g/1000 kJ)  9.4+0.3  9.3+0.3  9.4+0.4  Fat (g/1000 kJ)  10.8+0.2  11.0+0.1  10.7+0.1  Carbohydrate (g/1000 kJ)  27.5+0.7  27.3+0.5  27.3+0.7  Cholesterol mg/1000 kJ)  46.3+11.2  35.3+11.8  30.2+11.3  *  n=6 (2-day meal plan), X + SE.  84  Table 3.4 Percent individual fatty acid composition of fat in the three diets  Fatty  SAT  MONO  POLY  acid b  8:0  1.5 + 0 .5 a  0.0 + 0.,0 b  0.0 + 0..o  10:0  3.1 + 0 .4 a  0.0 + 0.,o  b  3.5 + 0 .4 a  14:0  7.2 + 1 .8 a  3.2 + 0.. 6 a b  2.5 + 0 .2 b  16:0  24.1 + 4 .2 a  18.6 + 1..8 a  13.3 + 0 .6 a  18:0  10.0 + 1 .2 a  6.4 + 0.,6 b  9.3 + 0..8 a :  18:1  32.5 + 4.,4 b  61.4 + 1.,9 a  24.2 + 4 .6 . b  18:2  21.7 + 5.,6 b  10.4 + 1.,3 b  47.1 + 2.,8 a  n=6 (2-day meal plan), X ± SE. MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Means across a row not sharing a common superscript are significantly different (P<0.05).  85  Table 3.5  Percent fatty acid saturation of fat * in the three diets  Fat type  SAT  C:0  45.8 + 7.8 a  28.2 + 3.0 b  28.6 + 1.9b  C:l  21.7 + 5.6b  61.4 + 4.l a  24.2 + 4 . 6 b  C:2  32.5 + 4.4 b  10.4 + 1.3b  47.1 + 2.8 a  *  MONO  POLY  n=6 (2-day meal plan), X + SE. MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Means across a row not sharing a common superscript are significantly different (P<0.05).  86  Table 3.6  Mean serum lipid values induced by each diet  Serum Lipid (mM)  SAT  Total Choi  4.54 + 0.07  4.04 a +0.05  3.67 ab +0.05  FC C  1.31 + 0.14  1.16 + 0.13  1.22 + 0.14  TG  0.94 + 0.08  0.87 + 0.08  0.77 + 0.07  HDL  0.97 + 0.02  1.05a + 0.02  1.02 + 0.02  Total -HDL  3.56 + 0.07  2.99 a + 0.05  2.66 ab + 0.05  Choi to HDL ratio  4.75 + 0.11  3.94 a + 0.10  3.69 +0.08  a b c  MONO  POLY  n=9, X ± SE. Total Choi: total cholesterol; Total-HDL: Non-HDL cholesterol; chol to HDL ratio: ratio of total cholesterol to HDL; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Significantly different from row SAT values (P<0.05). Significantly different from row MONO values (P<0.05). Measurements only taken at t=0 during each 48-h test period.  87  Table 3.7  Serum Lipid (mM)  8:00AM Day 13 (base)  Mean serum lipid values induced at each time point  8:00PM Day 13  (post)  8:00AM Day 14 (post)  8:00PM Day 14 (fast)  8:00AM Day 15 (fast)  Total Choi  3.97 + 0.12  4.00 + 0.13  4.05 ±0.10  4.16 ±0.11  4.13 ±0.11  TG  0.74 + 0.04  1.56a ±0.13  0.83 ±0.05  0.64 ±0.06  0.54 a b ±0.04  HDL  0.98 +0.03  0.97 + 0.03  0.99 ±0.03  1.04 ±0.05  1.06 ±0.03  Total -HDL  2.99 + 0.13  3.02 +0.13  3.05 ±0.10  3.08 ±0.16  2.97 ±0.12  Choi to HDL ratio  4.19 + 0.20  4.18 ±0.18  4.13 ±0.14  4.00 ±0.20  3.88 + 0.15  a b  n=9, X ± SE. All variables except cholesterol to HDL ratio are expressed as mmol/L. Total Choi: total cholesterol; TG: triglycerides; HDL: high density lipoproteins; total-HDL: Non-HDL cholesterol; chol to HDL ratio: ratio of total cholesterol to HDL; base: baseline fasted value; post: postprandial; fast: fasted. Significantly different from row baseline value (P<0.05). Significantly different from row postprandial values (P<0.05).  88  Table 3.8 Serum cholesterol lathosterol values in subjects at 12 h (fed) and 36 h (fasted) time * points during each 48-h test period Diet  Time (h)  Lath (MM)  MONO  12  23.60 +4.38  4.01 ±0.13  5.97 ±1.14  MONO  36  6.64a ±1.12  4.14 ±0.15  1.61a ±0.28  POLY  12  14.73 +5.25  3.72 ±0.13  3.89 ±1.39  POLY  36  6.54a + 1.61  3.68 ±0.11  1.75a ±0.41  SAT  12  12.41 +2.10  4.58 ±0.18  2.70 ±0.43  SAT  36  7.74 a + 1.51  4.59 ±0.23  1.64a ±0.27  Choi (mM)  Lath/Choi (/iM/mM)  n=9, X ± SE. Lath: lathosterol; Choi: cholesterol; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Significantly different from 12 h (fed) serum values for that diet.  89  Table 3.9 Deuterium enrichment (°/oo) relative to SMOW: validation trial conducted to determine serum sample size needed for mass spectrometry  Sample Volume  (mL)  FC  -150.66a +22.40  -219.89b +4.89  -233.52 b +3.13  -233.37 b +3.46  Plasma  -142.36a +13.84  -175.86b ±2.86  -169.30 b +1.92  -170.67 b +3.23  n=3, X + SE. Plasma: plasma sample from individual who had consumed deuterium tracer; FC: free cholesterol. Means across a row not sharing a common superscript are significantly different (P<0.05).  90  8:00AM  8:00PM  • •• SAT diet  8:00AM  8:00PM  iioOAM  Time of sampling - B - MONO diet - A - POLY diet  Figure 3.1 Serum total cholesterol concentration at 12 h intervals during the 48-h sampling period. n=9, X ± SE. Arrows indicate mealtimes.  91  2.2 2 ~  1.8  o  E F 1.6 w  1.4  i-  a> o 1.2 D) i_  +•»  E  3 t0) CO  1 0.8 0.6 0.4  8:00AM 8:00PM 8:00AM 8:00PM A A A Time of sampling SAT diet  - a - M O N O diet  8:00AM  POLY diet  Figure 3.2 Serum triglyceride concentration at 12 h intervals during the 48-h sampling period. n=9, X + SE. Arrows indicate mealtimes.  92  1.2 1.15 1.1 o 1.05 F E Q X  1  E 0.95  3 CO  0.90.85 0.8  8:00AM  8:00PM  I M  SAT diet  8:00AM  8:00PM  8:00AM  Time of sampling - B - MONO diet - A - POLY diet  Figure 3.3 Serum HDL concentration at 12 h intervals during the 48-h sampling period. n=9, X + SE. Arrows indicate mealtimes.  93  8:00AM  8:00PM  • •• SAT diet  8:00AM  8:00PM  8:00AM  Time of sampling - s - M O N O diet  POLY diet  Figure 3.4 Serum non-HDL cholesterol or apo B-containing lipoproteins (VLDL and LDL) concentration at 12 h intervals during the 48-h sampling period. n=3, X + SB. Arrows indicate mealtimes.  94  FASTED  FED  ^ 1 MONO L J POLY H O SAT  Figure 3.5 Serum lathosterol concentration at 12 h and 36 h during the 48-h sampling period. n=9, X + SE. Concentrations were significantly lower at the 36 h time point than at the 12 h time point, as indicated by the column superscripts. No significant diet effect was seen.  95  3.6  SPECIFIC  ACTIVITY  DECAY  CURVES  Data used to generate specific activity decay curves are shown in Appendix 8. to the  Two-term and three-term multiexponential equations were fit  C-cholesterol specific activity decay data using SAAM/CONSAM.  In all cases except one (Subject F), a three-term equation gave the best fit (P<0.05) (Figure 3.6). Calculated F values, textbook F Q5, and calculated % reduction in residual error when a three-term rather than a two-term equation is used to describe the data, are in Table 3.10. The equation fitted to each data set was:  Y=  Equation  {K(80) x exp(-P(l) x T)}  5  + {k(8D X exp(-P(2) x T) } + {K(82) x exp(-P(3) X T)}  where K(80), K(81), and K(82) represent slope intercepts of portions of each curve, and P(l), P(2), and P(3) depict slopes of portions of each curve.  These parameters are listed in Table 3.11.  A sample deck of the  type used to generate the best fit is shown in Appendix 9. The three-pool model (Goodman et al. 1973) (Figure 3.7) was fitted to the data, and estimates of de novo cholesterol synthesis, exchange rates between pools, and masses three pools were generated by compartmental analysis using SAAM/CONSAM (Tables 3.12(a) and 3.12(b)). A sample deck is provided in Appendix 10. Comparison of the values obtained by Goodman et al.  (1973), who did not use SAAM/CONSAM to  96  develop their model, and the values obtained in this study are provided in Table 3.13.  Statistical evaluation was not attempted for two  reasons: (1) all subjects in this study were normal, but only half of those who participated in the Goodman study were so; (2) while SAAM/CONSAM will estimate standard errors for parameter values, the Goodman paper lists only minimum and maximum values.  It is therefore  not possible to say with any statistical certainty that the values obtained by the two methods do not differ significantly.  3.7  DEUTERIUM  INCORPORATION  The original FC D enrichment levels are in Appendix 11 with corrected data in Table 3.14.  Ratios of FC enrichment to body water  enrichment are in Table 3.15.  3.7.1 Free cholesterol Enrichment levels were lower in subjects on the SAT diet than in those on the POLY diet (P<0.043).  Enrichment levels at 0 and 12 h lower  than those at 24, 36, and 48 h ((P<0.001).  When the ratios of FC  enrichment to body water enrichment were analyzed, these trends were repeated, for both diet (P<0.014) and time (P<0.001).  3.7.2 Esterified cholesterol The original CE D enrichment values for each subject are in Appendix 12; corrected data are in Table 3.16, and ratios of CE  97  enrichment to body water enrichment are in Table 3.17.  For CE D  enrichment, a significant interaction between diet and time was found (P<0.010); consequently each diet type and time of sampling was examined separately.  Time effects were seen within each diet: on the MONO diet,  levels at 0 and 12 h were lower than all others, while levels at 24 h were lower than those at 36 and 48 h (P<0.001).  On the POLY diet,  levels at 0 and 12 h were lower those at 24, 36, or 48 h (P<0.001).  On  the SAT diet, levels at 12, 24, 36, and 48 h were higher than baseline, and levels at 48 h were higher than at 12 h (P<0.001). were seen at either 12 h (P<0.620) or 24 h (P<0.131).  No diet effects At 36 h, levels  on the SAT diet were lower than those seen on either the MONO or POLY diets (P<0.001).  At 48 h, this pattern was repeated (P<0.004).  In  contrast to the above results, analysis of the ratios of CE enrichment to body water enrichment did not show a significant interaction effect (P<0.063).  A diet effect was seen, with levels on the SAT diet being  lower than those obtained with either the MONO or POLY diet (P<0.001). A time effect was also found, with levels at 12 h lower than all other values, and levels at 24 h lower than at either 36 or 48 h (P<0.001).  3.7.3 Body water The original body water D enrichment values for each subject are given in Appendix 2; corrected data are shown in Table 3.18.  Levels at  24 and 48 h were significantly higher than baseline values (P<0.001). significant diet effect was not found (P<0.075).  98  A  10000  o i_  a> v> o> o  +••  1000  sz u E 3 i-  a> w  E  100:  Q. Q  150 200 Time (days) Subject A  350  Subject B —§§— Subject C  Figure 3.6a Specific activity decay curves for three subjects (A to C) : turnover of serum cholesterol. Each graph depicts serum specific activity decay curves using a semilogarithmic scale: DPM per mg serum cholesterol versus time in days.  99  10000  o Q>  w o sz u E 3 a> at  E  1000  100:  CL Q  150 200 Time (days) •  Subject D -^-  Subject E  350 Subject F  Figure 3.6b Specific activity decay curves for three subjects (D to F ) : turnover of serum cholesterol. Each graph depicts serum specific activity decay curves using a semilogarithmic scale: DPM per mg serum cholesterol versus time in days.  100  10000  o a> +•»  0) ffl  o sz u  1000  E  3 »_ a> V)  E  100  a. Q  100  •  150 200 Time (days)  Subject G - * — Subject H  250  300  350  Subject I  Figure 3.6c Specific activity decay curves for three subjects (G to I): turnover of serum cholesterol. Each graph depicts serum specific activity decay curves using a semilogarithmic scale: DPM per mg serum cholesterol versus time in days.  101  Table 3.10 Comparison of the closeness of fit obtained by analysis of the specific activity decay data with a two-term versus a three-term equation  Sub  Calculated F  F  % reduction in residual error  .05  A  46.10  3.55  83.7  B  115.16  3.49  92.0  C  26.66  3.49  72.7  D  28.83  3.47  73.3  E  115.37  3.47  91.7  F  2.85  3.59  NSD  G  20.27  3.52  68.1  H  11.04  3.44  50.1  I  23.37  3.52  71.1  NSD: no significant difference,- sub: subject. The residual error, a measure of the closeness of fit, was calculated for the best fit obtained with each model. The percent reduction in residual error obtained with the three-term model was calculated as: ((residual error 2-term - residual error 3-term)/residual error 2-term)*100).  102  Table 3.11 Parameter values obtained by fitting specific activity decay data to a three-term exponential equation Sub  K(80)  K(81)  K(82)  P(D  P(2)  P(3)  4360 ±630  1431 ±176  181.9 ±87.8  0.206 ±0.042  0.027 ±0.005  0.005 ±0.002  869.1 ±259  3604 ±546  503.7 ±12 0  0.036 ±0.013  0.201 ±0.050  0.011 ±0.001  3099 ±522  1016 ±174  272.4 ±51.1  0.284 ±0.075  0.038 ±0.007  0.006 ±0.001  4501 ±8387  993..9 ±8 71  49.1 ±636  0.297 ±0.043  0.020 ±0.001  b  2522 ±460  712.,1 ±166  358.6 ±66.5  0.287 ±0.076  0.042 ±0.011  0.011 ±0.001  2834 ±215  366..0 ±48..2  0.060 ±0.005  0.008 ±0.001  3569 ±453  967..2 ±2 92  0.128 ±0.027  0.028 ±0.009  2755 ±2208  632..6 ±1331  41.83 ±32.5  0.094 ±0.0134  0.023 ±0.003  259.0 +76.4  0.191 ±0.041  0.034 ±0.006  5046 ±645  b c  1473 ±283  257.5 ±108  0.005 ±0.002  c  0.005 +0.001  X ± SD, estimated using SAAM/CONSAM. Sub: subject; K(X), in units of DPM/mg cholesterol, represent y-intercepts of each portion of a curve, should it be extended to the y-axis; P(l), P(2), and P(3), in units of days" , represent slopes of each portion of a curve. The overall equation to which the data were fitted is: Equation 5. Subject F's specific activity decay curve was best represented by a two-term, not a three-term equation,- this translates into a twopool model rather than a three-pool model. 2.53e-06 ± 7.02 e-07 is the value estimated by SAAM/CONSAM. 1.54e-08 ±3.02 e-09 is the value estimated by SAAM/CONSAM.  103  1(3,4)  1(4,3)  Figure 3.7 Three-pool model of cholesterol turnover in man, as redrawn by author (Goodman et al. 1973). n=9, X + SE. Boxes represent compartments: #3 consists of cholesterol in rapid equilibrium with plasma cholesterol, and includes cholesterol in plasma, red blood cell, and liver, together with much of the cholesterol in intestines, pancreas, spleen, kidney, and lung; #2 consists of cholesterol which equilibrates at an intermediate rate with plasma cholesterol, and includes some of the cholesterol in viscera and peripheral tissues; #4 is cholesterol which equilibrates slowly with plasma cholesterol, and includes much of the cholesterol in peripheral tissues, particularly skeletal muscle. Arrows indicate flow pathways, l(i,j) represent rates of exchange between pools, u(i) shows entry of cholesterol into pool.  104  Table 3.12(a) Parameter values obtained by fitting each subject's specific activity decay data to a three-compartment model  de novo Sub  M(2) (mg)  M(3) (mg)  M(4) (mg)  A  19814  16550  39355  881.7  293 .9  B  19082  19590  24391  1089.6  363..2  C  23489  20974  70550  1074.0  358 .0  D  12491  29551  1.565e+06  49.8  16..6  E  21664  28291  41965  1689.3  563..1  29890  40425  1025.7  341 .9  900.6  300..2  1292.5  430..8  807.0  269..0  Fa G  15425  21208  53699  H  17004  20576  117170  I  13287  14322  46488  u(3) (mg/d)  synthesis (mg/d)  Estimates obtained using SAAM/CONSAM. Legends refer to three-pool model, seen in Figure 3.7. Sub: subject; M(3) is the cholesterol pool which equilibrates most rapidly with plasma cholesterol, M(2) equilibrates at an intermediate rate, and M(4) equilibrates at a slow rate. u(3) refers to daily total cholesterol production in the central pool; FC de novo synthesis has been estimated to be approximately 33% the daily production rate. Subject F's specific activity decay curve was best represented by a two-term, not a three-term equation; this translates into a twopool model rather than a three-pool model.  105  Table 3.12(b) Parameter values obtained by fitting each subject's specific activity decay data to a three-compartment model  Sub  1(3,2)  1(2,3)  1(4,3)  1(3,4)  1(0,3)  A  0.0723 +0.0168  0.0865 +0.0238  0.0161 +0.0031  0.0068 +0.0030  0.0533 +0.0061  B  0.0750 ±0.0264  0.0731 +0.0238  0.0224 +0.0115  0.0180 +0.0046  0.0556 +0.0060  C  0.1070 +0.0295  0.1200 +0.0437  0.0380 +0.0076  0.0113 +0.0023  0.0512 +0.0066  D  0.0699 +0.0078  0.1660 +0.0274  0.0862 +0.2810  0.0007 +0.0027  0.0040 +0.2850  E  0.1200 +0.0424  0.0917 +0.0247  0.0496 +0.0177  0.0335 +0.0044  0.0597 +0.0046  0.0197 +0.0024  0.0146 +0.0017  0.0343 +0.0020  Fa  G  0.0085 +0.0033  0.0214 +0.0057  0.0381 +0.0117  0.0524 +0.0188  0.0425 +0.0043  H  0.1872 +0.0569  0.1547 +0.1255  0.0318 +0.0070  0.0056 +0.0009  0.0630 +0.0141  I  0.0716 +0.0184  0.0665 +0.0203  0.0268 +0.0046  0.0083 +0.0023  0.0564 +0.0062  X + SD, estimated using SAAM/CONSAM. Legends refer to three-pool model, seen in Figure 3.7. Sub: subject; l(i,j), in units of days" , represent rates of exchange between pools, from pool j to pool i. Pool 0 refers to a space not in any pool. Two-pool model rather than a three-pool model.  106  Table 3.13 Three-pool cholesterol model: comparison of the values obtained by Goodman et a.1. (1973) and the values obtained in this study  Parameter  Goodman a  SAAM/CONSAMb  M(2) (g)  11.3+ 1.5  17.8+ 1.5  2 0.2+ 3.0 M(3) (g)  23.4+ 1.1  22.3+ 2.0  M(4)  54.3+ 10.8 C  1(2,3) (d _1 )  35.7+ 1.4 72.1+ 4.6 5.6e-02+ 7.0e-03  1(3,2) (d 1 )  (g)  2.6e-01+  1.7e-01  1.2e-01+ 2.3e-02  1.3e-01+  5.0e-02  1(3,4) (d X)  2.6e-02+ 1.0e-03  1.2e-02+  3.3e-03  1(4,3) (d 1 )  4.0e-02+ 3.0e-03  3.9e-02+  8.9e-03  u(3) (g)  1.13+ 0.09  0.985+ 0.16  a  b c  n=6, X + SE. M(2), M(3), and M(4) are cholesterol pool sizes; for side pools 2 and 4, minimum and maximum values as estimated by Goodman et al. (1973) are shown. l(i,j) is the rate of exchange into i from j, and u(3) refers to cholesterol daily production rate from the central pool. n=9, X + SD, estimated using SAAM/CONSAM. Does not include estimated value for Subject D.  107  Table 3.14 Serum free cholesterol deuterium enrichment (°/oo) relative to SMOW at five time points on each of the three diets  Time (h)  MONO  POLY  SAT  0.00+ 0 . 0 0 a +  0.00+ 0 . 0 0 a +  12  33.25+ 5 . 9 8 b + *  42 . 09 + 12 . 6 2 b +  24  79.96+ 8 . 8 3 C + *  90.52+14.97 C +  75.85+ 8 . 3 2 c *  36  70.69+ 9 . 3 4 c + *  94.41 + 1 3 . 6 2 c +  48.27+  48  86.35+ 9 . 8 1 C + *  86 . 32 + 1 6 . 6 6 C +  0  **  0.00+ 0 . 0 0 a + 29.59+ 5.69 b *  70.40+  7.26 C * 8.23 C *  n=9, X + SE. MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Means down a column not sharing a common letter superscript are significantly different (P<0.05). Means across a row not sharing a common symbol superscript (+*) are significantly different (P<0.05). Baseline value was either added or subtracted to each value of a set to allow comparison of relative enrichment among subjects and diets. Actual values are in Appendix 11.  108  Table 3.15 Ratios of serum free cholesterol deuterium enrichment to body water deuterium enrichment at four time points on each of the ieic  three diets Time (h) 12 24  MONO  0..0088 a+ .0.0016 h+*  0 . 0220 D + 0..0022  POLY  0., 0 1 3 5 a + +0..0035  SAT  0.. 0 0 8 4 a * + 0..0014 b*  0.. 0 2 4 9 b + +0..0035  ±o..0023  0..0216"  36  0.. 0197" + o..0024  b+*  0.. 0 2 9 8 b + +0..0043  0. .0145° ±0..0021  48  0.. 0 2 3 9 D +  b+*  0.. 0 2 6 8 b + +0..0039  0,. 0202 b * + 0..0020  o.0026  b*  22=9, X + SE. MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Means down a column not sharing a common letter superscript are significantly different (P<0.05). Means across a row not sharing a common symbol superscript (+*) are significantly different (P<0.05). Ratios were obtained by using enrichment values from which the baseline value had been subtracted, allowing comparison of relative enrichment among subjects and diets.  109  Table 3.16 Serum esterified cholesterol deuterium enrichment (°/oo) relative to SMOW at * five time points on each of the three diets  Time (h)  MONO  POLY  SAT  0  0.00a +0.00  0.00a ±0.00  0 .00a ±0..00  12  7.85a +5.53  14.68 a ±5.72  10 .15 a b ±6 .21  24  25.97 b ±2.33  43.31 b ±8.51  22 .61 b c ±4..69  36  55.15 C ±5.42  55.53 b ±6.34  23.. 3 3 b c + ±4..20  48  55.46 C +4.20  65.02 b +9.02  36..47 C + +5.02  n=9, X ± SE. MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Means down a column not sharing a common letter superscript are significantly different (P<0.05). Means across a row not sharing a common symbol superscript (+) are significantly different (P<0.05). Baseline value was either added or subtracted to each value of a diet set to allow comparison of relative enrichment among subjects and diets. Actual values obtained are in Appendix 12.  110  Table 3.17 Ratios of serum esterified cholesterol deuterium enrichment to body water deuterium enrichment at four time points on each * of the three diets  Time (h)  MONO  POLY  SAT  12  0.0022c +0.0015  0.0048c +0.0017  0.0030a+ +0.0014  24  0.0073 +0.0006  0.0128 +0.0027  0.0066b+ +0.0013  36  0.0152' +0.0011  0.0175' +0.0021  0.0067c+ +0.0010  48  0.0157' +0.0012  0.0203 +0.0028  0.0103 c + +0.0015  n=9, X + SE. MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Means down a column not sharing a common letter superscript are significantly different (P<0.05). Means across a row not sharing a common symbol superscript (+) are significantly different (P<0.05). Ratios were obtained by using enrichment values from which the baseline value had been subtracted, allowing comparison of relative enrichment among subjects and diets.  Ill  Table 3.18 Body water deuterium enrichment (°/oo) relative to SMOW at three time points on each of the three diets  Time (h)  MONO  POLY  SAT  0.00c +0.00  0.00c + 0.00  0.00c +0.00  24  3 5 6 4 . 25* +127.82  3407. o r +184.93  3556.45* +37.78  48  3534.41 +95.09  3363.74* +192.95'  3456.46* +59.46  11=9, X + SE. MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Means down a column not sharing a common letter superscript are significantly different (P<0.05). Baseline value was either added or subtracted to each value of a set to allow comparison of relative enrichment among subjects and diets. Actual values obtained are in Appendix 2.  112  3.8  COMPARTMENTAL MODELS AND MODEL PARAMETERS  3.8.1 Free cholesterol data Several different methods were used to calculate FSR and de novo synthesis. 3.8.1.1  The model to which the data were fitted is in Figure 3.8.  Linear  incorporation  of  tracer  The first method used FC D enrichment data to calculate cholesterol FSR over each 12-h period (Jones et al.  1988; 1993a).  Values are in Table 3.19; means are in Figure 3.9.  No diet effects were  seen (P<0.815).  Rates varied with time (P<0.001), with numbers in the  first two 12-h periods higher than those in the second two.  These  differed from each other, with rates in the third period lower than those in the fourth period.  Daily de novo synthesis totals were  calculated as the product of FSR and 33% of the mass of the M(3), pool (Table 3.20, Figure 3.10) . An interaction between diet and time was observed (P<0.029); each treatment was thus examined separately.  On the  MONO diet, synthesis during the second 12-h period was greater than in the third and fourth period, whereas that in the first 12-h period differed from that in the third (P<0.001).  On the POLY diet, synthesis  in the first 12-h period was greater than that in the last two, and that in the second 12-h period was greater than that in the fourth (P<0.001). On the SAT diet, synthesis during the third 12-h period was less than  113  that in all other periods (P<0.001).  No diet effect was seen in the  first (P<0.526), second (P<0.805), or fourth (P<0.121) 12-h period. Differences were seen in the third 12-h period, where synthesis was lower on the SAT diet than on the other two diets (P<0.012). The second linear method used FC D enrichment data to compute FSR (Jones et al. 3.11.  1994a; 1994b).  Values are in Table 3.21; means in Figure  Differences were seen with diet (P<0.014) and time (P<0.001).  Rates were greater on the POLY diet than the SAT diet; MONO values did not differ from the other two.  Rates were lower in the first 12-h  period than in the other three periods.  Similar time trends (P<0.015)  were seen in mass synthesized, but no diet effect (P<0.265) was noted (Table 3.22, Figure 3.12).  3.8.1.2  Monoexponential  incorporation  of  tracer  This method used SAAM/CONSAM to fit enrichment data over each 12-h period to a monoexponential reaching a plateau at the TME, which is 0.478 x body water enrichment; data were then fit to the EME. A sample deck is in Appendix 13.  Fractional synthesis rates calculated using TME  and EME are presented in Table 3.23; EME means are depicted in Figure 3.13.  For data calculated with the TME, no diet effect was seen  (P<0.788), but FSR were greater in the first two 12-h periods than in the last two, and rates in the third period were lower than in the fourth (P<0.001).  For EME, patterns were similar to those observed when  the TME was used for modelling.  No diet effect was seen (P<0.699), but  114  rates were higher in the first two 12-h periods than in the last two periods, and rates in the third period were lower than in the fourth (P<0.001).  Mass synthesized values calculated using TME and EME are  presented in Table 3.24; EME means are shown in Figure 3.14.  For TME  data, no diet effect was found (P<0.894), but more cholesterol was synthesized in the first two 12-h periods than in the last two, and mass synthesized in the fourth 12-h period was greater than that in the third 12-h period (P<0.001).  For EME data, a diet effect was not seen  (P<0.873), but more cholesterol was synthesized in the first two 12-h periods than in the third, and the mass synthesized in the fourth 12-h period was greater than that synthesized in the third 12-h period (P<0.001). SAAM/CONSAM was used to calculate parameter values for each 24-h period using the TME and EME.  Fractional synthesis rates are shown in  Table 3.25; EME means are depicted in Figure 3.15.  No effect of diet  was observed in the FSR calculated using either the TME (P<0.981) or the EME (P<0.988).  Rates were higher during the fed state when compared  with the fasted state, for the FSR calculated with both TME (P<0.001) and EME (P<0.001).  Values obtained for mass of cholesterol synthesized  are displayed in Table 3.25; EME means are presented in Figure 3.16. Similar results to those obtained for FSR were seen: no significant effect of diet was seen in mass calculated using the TME (P<0.960) or the EME (P<0.9995) . Mass synthesized was greater during the fed period than during the fasted period for TME (P<0.001) and EME (P<0.001).  115  P< fD T3  s; rr W  13 >fl  H  fD  o  rr 3"  rr  fD  n fD  rr  Co  •a  Pttl  3  F-  d 3  0J  £ co  O  co  << 3  F-  fD  Cu rr  n to  fD  V  n  fD  > rrrt rt CD  3  rt  fD  CO I-i  0  fD  3 O rt C P. rr i-i 0) 3 n M  -C  o o 3  3  F- C  P>  fD  n  P,  fD  n  d  H- CO r t  fD CO pCO  F- cr r t CD 0 3"  i-i a o rr ^ s 3* en n> 0) s  > H  M  0  P- CO  O  * ^-, OJ H i fu — H 3 O a c3"o £  . M FO 13 3 O > 0 H C  P> 0) CD r t  3 rr rr Hi  n CD  rr fD  rt  O  #H  3  a . # OJ  N  fD CO CO rt fD I-i  Co  P- en h t ) • PJ fD H- O O) 3 fD P, rr H ttl  O  pCO  rt  p,  rr rr o 3" 3 ff fD fD 0 ra 3 M  ^3  rr o n P. o tJ o M 3 O fD ra •a ttl £D <, 0  F- rr  co  fD  0)  rr  m fD H P- 0 O 3 C O M rr H H i  LJ.  3  H O  Hi  P- rr p - H Cfl fD CO rr C CD 3T rr h ?r fD 3 •a o P> O m co p> S DJ M F- C fD p P- N tti CO 3 3 CO H i 3 fD fD fD LQ Hi P 3 a a rr ft) O F- & a P, 3 o o H i CD  CO r t  1  3 0) rr S O H P^-* P- rr rr fD a O O CQ C - i-j 3" fDH i-i P *o c_l. I-i o . OCO CD ^ s; — 0 p) 3 IfDQ fD P- H i ^ co fD 3 n a w 3 *v p«< s p j CD p> 3 o a & H- rr CO fD o o ^ H n o co a 0) M rr co c rr fD fD rr fD fD ^ fD IQ p j fD CO P, 3 h  30  CO  o  m  I  -<  CD O D  w o  Table 3.19 Free cholesterol fractional synthesis rates, calculated for each 12-h period, using the first linear method  Diet  0-12  12-24  24-36  36-48  MONO  0.069  0.051  -0.012  POLY  0.000  0.068  -0.014  SAT  0.079  0.059  -0.011  0.005  MONO  0.023  0.057  0.020  -0.015  POLY  0.017  0.073  -0.017  0.002  -0.057  0.012  SAT  0.017  MONO  0.048  0.055  0.009  0.016  POLY  0.018  0.055  0.014  -0.019  SAT  0.023  0.003  -0.011  -0.005  MONO  0.030  0.053  -0.012  0.069  POLY  0.035  0.077  -0.028  -0.007  SAT  0.026  0.060  -0.011  0.011  MONO  0.025  0.046  -0.005  0.008  POLY  0.096  0.001  -0.003  SAT  0.037  0.004  -0.002  0.049  MONO  0.088  0.086  -0.035  -0.014  POLY  0.103  SAT  0.044  0.094  -0.041  -0.009  MONO  0.066  0.076  -0.018  0.015  POLY  0.106  0.074  -0.021  0.022  -0.058  H  I  SAT  0.035  0.098  -0.056  0.015  MONO  0.000  0.017  -0.023  0.013  POLY  0.168  0.043  0.000  0.006  SAT  0.063  0.036  -0.063  0.084  MONO  -0.004  0.050  -0.006  -0.014  POLY  0.010  0.013  0.017  -0.013  SAT  0.000  0.093  -0.074  Sub: subject; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Fractional synthesis rates expressed as pools per day.  118  0.09  0-12  M  12-24 24-36 Time periods MONO \_\  36-48  POLY £ ~ J SAT  Figure 3.9 Calculated de novo fractional synthesis rates for free cholesterol during each 12-h period, using the first linear method. n=9, X ± SE. Periods not sharing a common superscript are significantly different. No significant diet effect was seen.  119  Table 3.20 Calculated de novo free cholesterol synthesis during each 12-h period, using the first linear method  Diet  0-12 (mg/d)  12-24 (mg/d)  MONO  380.7  POLY  24-36  (mg/d)  36-48 (mg/d)  281.4  -66.2  93.8  0.0  375.1  -77.2  SAT  435.8  325.5  -60.7  27.6  MONO  150.2  372.2  130.6  -98.0  POLY  111.0  476.7  -111.0  13.1  -372.2  78.4  SAT MONO  335.6  384.5  62.9  111.9  POLY  125.8  384.5  97.9  -132.8  SAT  160.8  21.0  -76.9  -35.0  MONO  124.9  220.7  -50.0  287.3  POLY  145.7  320.6  -116.6  -29.1  SAT  108.3  249.8  -45.8  45.8  MONO  235.8  433.8  -47.2  75.4  POLY  905.3  9.4  -28 .3  SAT  348.9  37.7  -18.9  462.1  MONO  876.8  856.8  -348.7  -139.5  POLY  1026.2  -577.9  SAT  438.4  936.6  -408 .5  -89.7  MONO  466.6  537.3  -127.2  106.0  POLY  749.3  523.1  -148.5  155.5  120  SAT-  247.4  692.8  -395.9  106.0  MONO  0.0  116.6  -157.7  89.2  POLY  1152.3  294.9  0.0  41.2  SAT  432.1  246.9  -432.1  576.1  MONO  -19.1  238.7  -28.6  -66.8  POLY  47.7  62.1  81.2  -62.1  0.0  444.0  -353.3  SAT  Sub: subject; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Synthesis values were calculated by multiplying the FSR rate by the size o the central pool and dividing by three to eliminate esterified cholesterol.  121  700  <o ;o  E  400  OT "<0 <D  c>. 0)  o » 0)  2 -200  0-12  12-24 24-36 Time periods  36-48  ^ ^ MONO L _ ] POLY [ 7 1 SAT  Figure 3.10 Calculated de novo free cholesterol synthesis during each 12-h period, using the first linear method. n=9, X + SE. Synthesis values were calculated by multiplying the FSR rate by the size of the central pool and dividing by three to eliminate esterified cholesterol. A significant interaction between diet and time was found; therefore each treatment was examined separately. Letters indicate differences due to time on each diet: on each diet, columns not sharing a common superscript are significantly different. Symbols indicate differences due to diet during each time period; only the third time period had such differences.  122  Table 3.21 Free cholesterol fractional synthesis rates, calculated for each 12-h period, using the second linear method  Diet  0-12  12-24  24-36  36-4!  MONO  0.033  0.047  0.054  0.062  POLY  0.000  0.032  0.025  0.000  SAT  0.039  0.069  0.063  0.066  MONO  0.011  0.040  0.050  0.037  POLY  0.008  0.045  0.037  0.037  0.056  0.027  0.033  SAT MONO  0.024  0.051  0.055  0.063  POLY  0.009  0.036  0.043  0.034  SAT  0.011  0.013  0.007  0.005  MONO  0.015  0.041  0.035  0.070  POLY  0.017  0.056  0.042  0.039  SAT  0.013  0.042  0.037  0.042  MONO  0.013  0.035  0.033  0.037  POLY  0.048  0.048  0.047  SAT  0.018  0.021  0.020  0.044  MONO  0.044  0.087  0.069  0.062  POLY  0.051  0.103  0.074  SAT  0.022  0.069  0.048  0.044  MONO  0.033  0.071  0.062  0.069  POLY  0.053  0.089  0.079  0.090  123  *  SAT  0.017  0.066  0.038  0.046  MONO  0.013  0.021  0.010  0.017  POLY  0.084  0.105  0.104  0.108  SAT  0.031  0.049  0.000  0.059  MONO  0.000  0.023  0.019  0.013  POLY  0.005  0.012  0.020  0.014  SAT  0.000  0.046  0.009  Sub: subject; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Fractional synthesis rates expressed as pools per day.  124  0.08  0.06 (0  § 0.04 oc to LL  0.02  0-12 **  M  12-24 24-36 Time periods  36-48  MONO ~ _ J POLY EZ3 SAT  Figure 3.11 Calculated de novo fractional synthesis rates for free cholesterol during each 12-h period, using the second linear method. n=9, X ± SE. Columns within each period not sharing a common superscript are significantly different. Periods not sharing the same number of asterisks are significantly different.  125  Table 3.22 Calculated de novo free cholesterol synthesis during each 12-h period, using the second linear method  Diet  0-12 (mg/d)  12-24 (mg/d)  24-36 (mg/d)  36-48 (mg/d  MONO  183.4  262.0  297.5  343.9  POLY  0.0  175.8  137.8  0.0  SAT  217.8  378.9  350.0  364.3  MONO  73.4  259.2  323.4  239.4  POLY  55.2  293.4  238.9  244.8  362.5  176.7  216.0  SAT MONO  166.2  355.7  386.4  441.5  POLY  62.1  253.6  301.3  236.3  SAT  78.6  87.4  50.9  32.6  MONO  62.9  171.9  147.5  290.6  POLY  72.8  232.1  175.0  161.1  SAT  53.0  176.6  153.5  176.1  MONO  118.3  332.4  310.7  349.1  POLY  449.1  455.6  439.5  SAT  174.2  194.6  183.9  413.2  MONO  437.5  864.4  692.3  621.0  POLY  508.9  1022.7  733.3  SAT  215.4  683.2  478.0  435.2  MONO  232.7  499.3  435.9  489.0  POLY  371.3  632.2  558.5  635.4  126  SAT  122.8  465.4  269.9  322.7  MONO  37.6  62.4  29.3  48.4  POLY  243.1  305.3  303.6  314.7  SAT  90.8  142.1  0.0  172.4  MONO  0.0  107.9  93.0  60.3  POLY  24.3  55.7  96.2  64.7  0.0  219.4  43.8  SAT  Sub: subject; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Synthesis values were calculated by multiplying the FSR rate by the size of the central pool and dividing by three to eliminate esterified cholesterol.  127  500  0-12 **  12-24 24-36 Time periods  WZ& MONO r _ j  POLY  CZJ  36-48 SAT  Figure 3.12 Calculated de novo free cholesterol synthesis during each 12-h period, using the second linear method. n=9, X ± SE. Synthesis values were calculated by multiplying the FSR rate by the size of the central pool and dividing by three to eliminate esterified cholesterol. Periods not sharing the same number of asterisks are significantly different. No significant diet effect was found.  128  Table 3.23 Free cholesterol fractional synthesis rates, calculated for each 12-h period deriving parameter values with SAAM/CONSAM  Diet  0-12  12-24  24-36  36-48  MONO  0.085 (0.071)  0.033 (0.027)  0.007 (0.003)  0.016 (0.011)  POLY  0.000* (0.000)  0.077 (0.064)  0.000* (0.000)  SAT  0.079 (0.066)  0.060 (0.050)  0.000* (0.000)  0.018 (0.013)  MONO  0.054 (0.045)  0.069 (0.058)  0.020 (0.016)  0.000* (0.000)  POLY  0.017 (0.013)  0.086 (0.073)  0.000* (0.000)  0.018* (0.014)  SAT  0.068 (0.058)  0.068 (0.058)  0.000* (0.000)  0.080 (0.069)  MONO  0.063 (0.049)  0.069 (0.054)  0.012* (0.009)  0.016 (0.010)  POLY  0.043 (0.033)  0.050 (0.039)  0.012 (0.007)  0.000* (0.000)  SAT  0.028 (0.020)  0.003* (0.002)  0.000* (0.000)  0.006 (0.004)  MONO  0.041 (0.029)  0.070 (0.051)  0.000* (0.000)  0.101 (0.074)  POLY  0.045 (0.033)  0.097 (0.071)  0.000* (0.000)  0.027* (0.020)  SAT  0.037 (0.026)  0.084 (0.062)  0.000* (0.000)  0.012 (0.005)  MONO  0.034 (0.026)  0.061 (0.048)  0.000* (0.000)  0.020* (0.015)  129  **  POLY  0.124 (0.098)  0.000* (0.000)  0.000* (0.000)  SAT  0.049 (0.039)  0.008* (0.006)  0.000* (0.000)  0.060 (0.048)  MONO  0.096 (0.091)  0.090 (0.086)  0.000* (0.000)  0.045* (0.043)  POLY  0.130 (0.123)  0.133 (0.126)  0.133 (0.126)  0.000* (0.000)  SAT  0.047 (0.045)  0.100 (0.095)  0.000* (0.000)  0.034* (0.033)  MONO  0.073 (0.066)  0.080 (0.072)  0.000* (0.000)  0.014 (0.011)  POLY  0.125 (0.112)  0.083 (0.075)  0.000* (0.000)  0.022 (0.019)  SAT  0.040 (0.036)  0.111 (0.100)  0.000* (0.000)  0.011 (0.008)  MONO  0.035 (0.025)  0.020 (0.012)  0.000* (0.000)  0.013 (0.006)  POLY  0.240 (0.178)  0.056 (0.042)  0.109* (0.081)  0.018* (0.013)  SAT  0.089 (0.067)  0.048 (0.035)  0.000* (0.000)  0.119 (0.089)  MONO  0.000* (0.000)  0.067 (0.056)  0.000* (0.000)  0.000* (0.000)  POLY  0.005 (0.001)  0.010 (0.006)  0.016 (0.012)  0.000* (0.000)  SAT  0.000* (0.000)  0.108 (0.092)  Sub: subject; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Values with asterisks (*) were calculated from the angle of deuterium incorporation. Fractional synthesis expressed as pools/day. Values calculated with EME; those in brackets were calculated with TME.  130  0.12  0.1 ^  0.08  i2  § 0.06 Q. DC </)  "• 0.04 0.02  0-12  12-24 24-36 Time periods MONO  36-48  POLY T 3 1 SAT  Figure 3.13 Free cholesterol fractional synthesis rates, calculated for each 12-h period, using a modified version of the linear method, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau. n=9, X + SE. Periods not sharing a common superscript are significantly different. No significant diet effect was seen.  131  Table 3.24 Calculated de novo free cholesterol synthesis during each 12-h period deriving parameter values with SAAM/CONSAM  0-12  12-:24  (mg/d)  (mg/d)  24-36 (mg/d)  36-48 (mg/d)  468 .1 (394 .3)  183 .2 (148 .3)  40.7 (18.4)  86.0 (60.8)  * 0 .0 (0 .0)  425 .5 (355 .7)  433 .7 (364 .7)  332..3 (277 .1)  * 0.0 (0.0) * 0.0 (0.0)  (71.4)  MONO  350 .4 (295 .6)  448..3 (378 .1)  132.2 (102.0)  * 0.0 (0.0)  POLY  110 .7 (82..8)  563..0 (477..5)  * 0.0 (0.0)  119.3* (89.2)  SAT  447..1 (377 .6)  447..1 (377..6)  0.0 (0.0)  523.9 (446.4)  MONO  440..1 (345..5)  481..8 (378..7)  80.9 (63.5)  114.5 (72.4)  POLY  304..1 (234..0)  348,.9 (271..3)  85.3 (45.6)  0.0 (0.0)  SAT  195..8 (142..3)  * 19,.5 (14..2)  0.0 (0.0)  35.2 (19.5)  MONO  171.,5 (122..5)  289.,6 (213..0)  0.0 (0.0)  419.0 (307.9)  POLY  188..6 (135,.6)  403,.9 (297..2)  0.0 (0.0)  114.3 (82.2)  SAT  153..1 (108,,5)  349..1 (257,.6)  * 0.0 (0.0)  48.7 (19.5)  MONO  316..9 (241..8)  571..5 (453..3)  * 0.0 (0.0)  182.8* (139.5)  Diet  MONO  POLY  SAT  132  POLY  1168.0 (927.5)  * 0.0 (0.0)  * 0.0 (0.0)  SAT  464.9 (364.5)  * 71.6 (56.1)  * 0.0 (0.0)  569.5 (450.6)  MONO  960.3 (911.4)  897.5 (852.3)  * 0.0 (0.0)  * 610.7 (430.7)  POLY  1292.4 (1230.0)  1321.7 (1255.20  SAT  470.2 (455.8)  997.3 (948.3)  * 0.0 (0.0)  344.9* (334.3)  MONO  515.8 (465.1)  566.9 (510.3)  * 0.0 (0.0)  97.7 (78.9)  POLY  881.8 (791.9)  587.3 (528.3)  * 0.0 (0.0)  153.6 (131.3)  SAT  283.5 (252.5)  787.9 (708.0)  * 0.0 (0.0)  77.8 (59.6)  MONO  308.0 (171.5)  172.9 (82.3)  * 0.0 (0.0)  112.7 (41.2)  POLY  2113.5 (1220.8)  495.9 (288.1)  961.7* (555.5)  154.4* (89.2)  786.5 (459.5)  420.6 (240.1)  * 0.0 (0.0)  1047.3 (610.4)  318.7 (268.9)  * 0.0 (0.0)  SAT MONO POLY SAT  * 0.0 (0.0) 23.9 (6.9) * 0.0 (0.0)  1321.7 (1255.2)  47.4 (28.8)  77.5 (57.2)  * 0.0 (0.O)  * 0.0 (0.0) * 0.0 (0.0)  518.1 (438.2)  Sub: subject,- MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Synthesis values were calculated by SAAM/CONSAM. Values with asterisks (*] were extrapolated from the angle of deuterium incorporation. Values calculated using the EME as a plateau; those in brackets were calculated using the TME as a plateau.  133  H1 UJ it*  3  CO fD CD  CD  s ID  n rt  s;  O  rr 3"  fD  a  *\ o O  f> §  j£J  CD CO  3  M  0)  <! 01  CD rj  rr  fD  3  CO CO  a  rt  3  CD  3  n 3-  CD <T3 3 0) H H H - 0)  t-i H-  a CD  *•  ro < PH 3 o M LQ  O  H  >3<  CO  a  CD H HO  W  u  1  to  r->  n tr  fD 0)  LQ  3  H H-  a 3  fD CO HCO  3 rr  0) H  CD  o a 3"  3 3  3  H-  X  0)  CD  rr  h \-> H- O  s'  CD  CD CO  rr  fD  Hi H CD CD  •$ a  3 O  a CD  o M  M 21  3 rr 3  fD  H  Hl Hi fD  CD  n rr  CD  CD  •  *>  a H>  CD  el-  3  rr  M  0) rr OJ  Hi  M fD CO  ^ tr ^ a H - CO  H  3 rr  CD  O 0) M O  CD H CD  0) H CD  o 0)  o  <j CD H CO H-  HLQ  n 3H 3  Q) M  O *fl  rr 3 3* n fD O 3T  LQ  3  CO H-  3  01 h-' 3  0 CD  HH] H-  Hl PfD  O  3  0)  LQ  < a 3  CD CO Htfl  3  3  SI  U X "O  rt  ^3  CO  ta  CO  1+  H-  CO HLQ  \Q  31  S  t^  CO  * CO ^ §1 a  rt  h.  CO  3  fcl HII O a  IT)  3  CD 01  t3  n  fD l-i  en 3 Xi  3 3 O 3  O O  2)  fD Hi Hi fD  3  LQ  fD  rr  a H-  £1J  coJ 3  rt  3 O  o a CO  H-  s  P) r t H rr 3- 3  M  t i *a  fD l-i H-  l-i H-  rt  3  HHi HO 0J  3  CO HLQ  a 0  z  H  I  o  o  O  I  o  o o  o o  o o  o o  o o  o o  o o  o o  o o  o o  o o  - * I O C 0 . P t U 1 0 > > l 0 0 < 0 O - J -  Free cholesterol synthesis (mg/day)  Table 3.25 Free cholesterol fractional synthesis rates and mass synthesized derived using SAAM/CONSAM for each 24-h period  Diet  FED FSR  FASTED FSR  FED CS  FASTED CS  MONO  0.062 (0.052)  0.012 (0.009)  341.7 (284.8)  63.4 (39.6)  POLY  0.039 (0.032)  0.000 (0.000)  213.2 (174.3)  0.0 (0.0)  SAT  0.077 (0.064)  0.009 (0.007)  424.8 (355.4)  49.4 (35.7)  MONO  0.058 (0.049)  0.010 (0.008)  381.3 (322.0)  66.1 (51.0)  POLY  0.018 (0.018)  0.009 (0.007)  117.5 (114.4)  59.7 (44.6)  SAT  0.069 (0.058)  0.040 (0.035)  447.4 (377.6)  262.0 (223.2)  MONO  0.065 (0.051)  0.014 (0.010)  454.2 (356.2)  97.7 (68.0)  POLY  0.046 (0.036)  0.006 (0.004)  321.5 (248.3)  42.7 (22.8)  SAT  0.015 (0.009)  0.003 (0.002)  103.8 (62.9)  17.6 (9.8)  MONO  0.046 (0.034)  0.051 (0.028)  193.3 (140.0)  209.5 (115.9)  POLY  0.053 (0.039)  0.014 (0.010)  221.4 (161.3)  57.2 (41.1)  SAT  0.043 (0.031)  0.006 (0.003)  181.0 (130.8)  24.4 (9.8)  MONO  0.039 (0.030)  0.010 (0.008)  364.6 (282.5)  91.4 (69.8)  135  POLY  0.075 (0.059)  0.000 (0.000)  707.4 (559.4)  0.0 (0.0)  SAT  0.031 (0.023)  0.027 (0.020)  292.5 (219.6)  256.7 (189.5)  MONO  0.096 (0.091)  0.023 (0.022)  958.0 (909.2)  305.4 (215.4)  POLY  0.066 (0.062)  0.067 (0.063)  654.7 (620.8)  660.9 (627.6)  SAT  0.055 (0.053)  0.017 (0.017)  552.1 (523.4)  172.5 (167.2)  MONO  0.075 (0.068)  0.007 (0.006)  530.9 (478.0)  48.9 (39.5)  POLY  0.115 (0.103)  0.011 (0.010)  813.7 (728.5)  76.8 (65.7)  SAT  0.049 (0.043)  0.006 (0.004)  343.2 (306.8)  38.9 (29.8)  MONO  0.031 (0.021)  0.007 (0.003)  270.0 (186.9)  56.4 (20.6)  POLY  0.179 (0.131)  0.064 (0.047)  1574.5 (1155.3)  558.1 (322.4)  SAT  0.077 (0.057)  0.060 (0.045)  679.0 (504.9)  523.7 (305.2)  MONO  0.032 (0.025)  0.000 (0.000)  150.5 (121.5)  0.0 (0.0)  POLY  0.006 (0.002)  0.008 (0.006)  29.0 (11.9)  38.8 (28.6)  SAT  0.055 (0.047)  265.9 (222.4)  Sub: subject; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Fasted values were extrapolated from the angle of deuterium incorporation. Fractional synthesis expressed as pools per day. CS: cholesterol synthesis, mg/d. Values calculated using the EME as a plateau; those in brackets were calculated using the TME as a plateau.  136  FED M  FASTED POLY  MONO  SAT  Figure 3.15 Free cholesterol fractional synthesis rates, calculated for each 24-h period, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau. n=9, X ± SE. Periods not sharing a common superscript are significantly different. No significant diet effect was seen.  137  800 >. re ;a o>  £  600  TJ  <D N  "</>  0)  W  400  o 1_  0) (0 0)  •*-•  o o  200  0)  a>  FED ^  FASTED  MONO [_^J POLY [TTJ SAT  Figure 3.16 Calculated de novo free cholesterol synthesis during each 24-h period, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau. n=9, X ± SE. Synthesis values were calculated by SAAM/CONSAM. Periods not sharing a common superscript are significantly different. No significant diet effect was seen.  138  3.8.2 Esterified cholesterol data As noted above for FC data, several different methods were used to calculate the rate of appearance of CE during the diet trials.  3.8.2.1  Linear  incorporation  of  tracer  The first method used CE D enrichment data to calculate FER (Jones et al.  1988; 1993a).  Values are in Table 3.26; means in Figure 3.17. A  diet effect was not seen (P<0.254), but rates in the second 12-h, period were greater than in the fourth period (P<0.034).  Daily appearance was  calculated as the product of FER and 67% of the mass of the M(3) pool. Estimated daily totals of new CE are in Table 3.27; means are in Figure 3.18.  Neither diet (P<0.309) nor time (P<0.123) affected total mass. The second linear method used CE D enrichment data to calculate  FER (Jones et al. 1994a; 1994b). Figure 3.19.  Values are in Table 3.28; means in  Rates were greater on the POLY diet than on the SAT diet,  with MONO not different from either of the others (P<0.001); rates were lower during the first 12-h period than during all others, and were lower during the second period than during the last two (P<0.001) . Estimated daily totals of new CE are in Table 3.29; means are in Figure 3.20.  3.8.2.2  Similar trends were observed as were demonstrated with FER.  Mono exponential  incorporation  of  tracer  The second method used SAAM/CONSAM to fit enrichment data from 12h periods to a monoexponential.  Rates of esterification are in Table  139  3.30; EME means are in Figure 3.21. (P<0.250) affected FER.  Neither diet (P<0.200) nor time  Values obtained by SAAM/CONSAM for the mass of  CE evolved using the TME and the EME are in Table 3.31; EME means are presented in Figure 3.22. Again, TME-evolved mass was not affected by either diet (P<0.156) or time (P<0.360), and EME-evolved mass was not affected by either diet (P<0.139) or time (P<0.358). SAAM/CONSAM was then used to calculate parameter values for each 24-h period, using in turn both the TME and EME.  Fractional  esterification rates are in Table 3.32; means are in Figure 3.23. Feeding state was not a factor affecting FER calculated using either the TME (P<0.653) or the EME (P<0.733).  A diet effect was not seen when  evaluating the FER calculated using the TME (P<0.063), but it was seen with the FER calculated with the EME: values on the POLY diet were greater than those on the SAT diet (P<0.040).  Mass of CE evolved are  displayed in Table 3.32; means are presented in Figure 3.24. As with FER, a feeding state effect was not seen for mass calculated with either the TME (P<0.795) or the EME (P<0.783).  A diet effect was not seen when  evaluating the former (P<0.089), but it was seen with EME values on the POLY diet, which were greater than on the SAT diet (P<0.005).  3.8.3 Correlation of synthesis rates with serum lathosterol Spearman correlations were used to determine whether changes in serum lathosterol concentration could be correlated with FSR and total  140  FC synthesized, based on either linear tracer or monoexponential tracer incorporat ion.  3.8.3.1  Linear  incorporation  of  tracer  For the first linear method, a positive relationship was seen between serum lathosterol concentration and FSR (Spearman Correlation Coefficient {SCC}=0.286, P<0.042). For the second linear method, a positive relationship was not seen between serum lathosterol concentration and FSR (SCC=-0.211,  3.8.3.2  Monoexponential  incorporation  of  P<0.133).  tracer  Positive relationships were not seen between serum lathosterol concentration and sets of FSR values calculated over 12-h periods calculated using TME (SCC=0.236, P<0.095) or EME (SCC=0.249, P<0.079). Positive relationships were seen between serum lathosterol concentration and both sets of FSR calculated over each 24-h period using either the TME (SCC=0.299, P<0.033) or the EME (SCC=0.356, P<0.010) .  141  Table 3.26 Fractional esterification rate, calculated for each 12-h period, using the first linear method  Diet  0-12  24-36  12-24  MONO POLY  0.023  36-48  0.015  -0.006  0.036  0.024  SAT  0.054  0.002  0.010  0.021  MONO  0.017  0.018  0.071  -0.025  POLY  -0.011  0.014  0.016  0.008  0.013  -0.008  SAT MONO  0.025  -0.002  0.042  -0.012  POLY  0.012  0.005  0.013  0.009  SAT  0.048  0.006  -0.014  -0.002  MONO  0.003  0.025  0.019  0.008  POLY  0.021  0.015  0.023  0.013  SAT  -0.021  0.022  0.013  0.023  MONO  0.001  0.041  0.002  0.006  POLY  -0.007  0.053  0.041  SAT  -0.021  0.021  0.023  0.020  MONO  -0.021  0.035  0.059  -0.015  POLY  0.051  SAT  0.010  0.030  -0.015  0.037  MONO  -0.019  0.049  0.021  0.026  POLY  0.060  0.001  0.032  -0.039  0.023  142  H  SAT  -0.003  0.027  -0.021  0.029  MONO  -0.048  0.014  0.032  0.025  POLY  0.044  0.044  0.031  0.045  SAT  0.048  0.0  MONO  0.003  0.017  0.037  -0.016  POLY  0.007  0.064  -0.005  0.003  -0.032  0.006  0.026  -0.011  SAT  *  Sub: subject; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat dietFractional esterification rates expressed as pools per day.  143  0.06  0.04 >* i2  § 3  0.02  DC UJ Li.  -0.02  0-12  ^ j  12-24 24-36 Time periods MONO \_\  36-48  POLY O J SAT  Figure 3.17 Fractional esterification rates calculated for each 12-h period, using the first linear method. n=9, X ± SE. Periods not sharing a common superscript are significantly different. No significant diet effect was seen.  144  Table 3.27 Calculated total d e novo esterification during each 12-h period, using the first linear method  Diet  0-12 (mg/d)  12-24  24-36  (mg/d)  (mg/d)  253.8  MONO  36-48 (mg/d)  165.5  POLY  -66.2  397.2  264.8  SAT  595.8  22.1  110.3  231.7  MONO  222.0  235.1  927.3  326.5  POLY  -143.7  182.8  209.0  104.5  169.8  -104.5  SAT MONO  349.6  -28.0  587.3  -167.8  POLY  167.8  69.9  181.8  125.8  SAT  671.2  83.9  -195.8  -28.0  MONO  25.0  208.2  158.2  66.6  POLY  174.9  124.9  191.5  108.3  -174.9  183.2  108.3  191.5  MONO  18.9  773.3  37.7  113.2  POLY  -132.0  999.6  773.3  SAT  -396.1  396.1  433.8  377.2  MONO  -418.5  697.4  1175.7  -298.9  POLY  1016.3  SAT  458.3  199.3  597.8  -298.9  737.3  MONO  -268.6  692.8  296.9  367.6  POLY  848.3  452.4  -551.4  SAT  14 .1  145  SAT  -42.4  381.8  -296.9  410.0  MONO  -843.9  246.1  562.6  439.5  POLY  773.5  773.5  545.0  791.1  SAT  843.9  0.0  MONO  28.6  162.3  353.3  -152.8  POLY  66.8  611.1  -47.7  28.6  -305.5  57.3  248.2  -105.0  SAT  *  Sub: subject; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Synthesis values were calculated by multiplying the FER by the size of the central pool and multiplying by 0.67.  146  700  -300  0-12  M  12-24 24-36 Time periods MONO •  36-48  POLY H H SAT  Figure 3.18 Calculated de novo appearance of esterified cholesterol during each 12-h period, using the first linear method. n=9, X + SE. Mass evolved values were calculated by SAAM/CONSAM. No significant effect of either diet or time was seen.  147  Table 3.28 Fractional esterification rate, calculated for each 12-h period, using the second linear method  Diet  12-24  24-36  36-48  MONO  0.005  0.017  0.024  POLY  0.015  0.027  0-12  SAT  0.027  0.028  0.033  0.043  MONO  0.008  0.017  0.053  0.041  POLY  0.000  0.002  0.009  0.014  0.008  0.015  0.011  SAT MONO  0.013  0.012  0.053  0.027  POLY  0.006  0.008  0.015  0.019  SAT  0.024  0.027  0.020  0.019  MONO  0.001  0.014  0.023  0.027  POLY  0.010  0.018  0.029  0.035  SAT  0.000  0.000  0.006  0.019  MONO  0.000  0.021  0.022  0.025  POLY  0.000  0.023  0.043  SAT  0.000  0.000  0.011  0.021  MONO  0.000  0.007  0.037  0.029  POLY  0.026  0.058  0.069  SAT  0.005  0.020  0.012  0.031  MONO  0.000  0.015  0.026  0.039  POLY  0.030  0.030  0.046  0.026  148  SAT  0.000  0.012  0.001  0.016  MONO  0.000  0.000  0.000  0.011  POLY  0.022  0.044  0.059  0.082  SAT  0.024  0.024  MONO  0.001  0.010  0.028  0.020  POLY  0.003  0.035  0.033  0.034  Sub: subject; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Fractional esterification rates expressed as pools per day.  149  0.05  0.04-  £ 0.03 i2 o o Q.  ffi 0.02 UJ  0-12 *  12-24 24-36 **Time period***  36-48 ***  W% MONO L _ J POLY [ 3 1 SAT  Figure 3.19 Fractional esterification rates calculated for each 12-h period, using the second linear method. n=3, X ± SE. Columns within a period not sharing a common superscript are significantly different. Periods not sharing the same number of asterisks are significantly different.  150  Table 3.2 9 Calculated total de novo esterification during each 12-h period, using the second linear method  Diet  12-24 (mg/d)  24-36  (mg/d)  36-48 (mg/d)  MONO  60.4  188.0  270.6  POLY  165.1  297.0  0-12 (mg/d)  SAT  296.8  306.4  363.0  480.0  MONO  109.9  228.9  694.6  533.8  POLY  0.0  22.6  124.3  178.7  108.9  192.4  138.1  SAT MONO  177.4  162.9  744.8  374.1  POLY  81.8  117.8  207.3  269.5  SAT  332.7  376.0  276.9  263.6  MONO  10.7  113.1  192.9  225.2  POLY  85.7  147.2  243 .0  296.9  SAT  0.0  1.7  54.1  159.0  MONO  0.0  395.1  414.2  468.2  POLY  0.0  431.0  818.3  SAT  0.0  0.0  207.7  395.3  MONO  0.0  147.2  732.7  582.4  POLY  511.8  1163.5  1389.4  SAT  96.4  398.1  248.2  616.7  MONO  0.0  214.3  363.0  548.6  151  POLY  420.2  425.3  649.8  372.0  SAT  0.0  169.0  20.4  227.2  MONO  0.0  0.0  0.0  66.1  POLY  126.6  254.6  344.6  476.2  SAT  138.9  138.1  MONO  12.0  91.3  266.8  192.0  POLY  31.5  335.0  313.4  328.5  Sub: subject; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Synthesis values were calculated by multiplying the FER by the size of the central pool and multiplying by 0.67.  152  700  0-12 *  1%£4  36-48  2j£6  ***  Time period MONO  POLY  r  I SAT  Figure 3.2 0 Calculated de novo appearance of esterified cholesterol during each 12-h period, using the second linear method. n=9, X + SE. Columns within a period not sharing a common superscript are significantly different. Periods not sharing the same number of asterisks are significantly different.  153  Table 3.30 Cholesterol fractional esterification rates, calculated for each 12-h period deriving parameter values with SAAM/CONSAM  Diet  0-12  12-24  24-36  36-48  0 .013 (0 . 0 0 8 )  MONO  0.010 (0.008)  0.010 (0.008)  0 .025 ( 0 .. 0 2 0 )  POLY  0.000 (0.000)  0.013 (0.009)  0. . 0 2 7 ( 0 .. 0 2 1 )  SAT  0.057 (0.049)  0.002* (0.002)  * 0. . 0 1 3 ( 0 ,. 0 1 1 )  0, . 0 1 9 ( 0 ,. 0 1 4 )  MONO  0.039 (0.033)  0.018 (0.014)  0. . 0 8 5 (0,. 0 7 2 )  * 0. . 0 0 0 ( 0 ,. 0 0 0 )  POLY  0.000 (0.000)  0.007* (0.005)  0. . 0 1 4 ( 0 .. 0 1 0 )  0, . 0 0 7 * ( 0 .. 0 0 5 )  SAT  0.018* (0.010)  0.018* (0.010)  0. . 0 0 9 ( 0 .. 0 0 5 )  0, . 0 0 0 * (0,. 0 0 0 )  MONO  0.032 10.024)  0.000* (0.000)  0. . 0 3 5 ( 0 .. 0 0 2 7 )  * 0. . 0 0 0 (0,. 0 0 0 )  POLY  0.009 (0.004)  0.016* (0.007)  0. . 0 1 0 ( 0 .. 0 0 5 )  * 0, . 0 0 5 ( 0 .. 0 0 2 )  SAT  0.063 (0.049)  0.012* (0.009)  * 0. . 0 0 0 (0,,000)  0. . 0 0 0 * (0,. 0 0 0 )  MONO  0.004 (0.003)  0.030 (0.021)  0 ., 0 2 3 ( 0 .. 0 1 5 )  * 0. . 0 0 0 (0.,000)  POLY  0.024 (0.016)  0.014 (0.007)  0 ., 0 2 7 ( 0 .. 0 1 8 )  0, , 0 1 1 ( 0 .. 0 0 4 )  SAT  0.000 (0.000)  0.003* (0.002)  * 0. . 0 6 5 (0..045)  0. . 0 3 9 ( 0 .. 0 2 7 )  MONO  0.001* (0.001)  0.055 (0.043)  * 0. , 0 0 3 ( 0 ., 0 0 2 )  * 0 .. 0 0 9 ( 0 .. 0 0 7 )  154  H  POLY  0.000 (0.000)  0.056 (0,044)  0.052 (0.041)  SAT  0.000 (0.000)  0.000 (0.000)  0.026 (0.019)  0.020 (0.014)  MONO  0.000 (0.000)  0.046* (0.044)  0.062 (0.059)  0.000* (0.000)  POLY  0.066 (0.063)  0.046* (0.044)  0.083 (0.079)  0.029 (0.025)  SAT  0.019* (0.018)  0.029 (0.028)  0.000* (0.000)  0.033 (0.031)  MONO  0.000 (0.000)  0.031 (0.028)  0.020 (0.017)  0.031 (0.027)  POLY  0.069 (0.062)  0.000* (0.000)  0.034 (0.030)  0.000* (0.000)  SAT  0.000 (0.000)  0.024 (0.021)  0.000* (0.000)  0.031 (0.027)  MONO  0.000 (0.000)  0.000* (0.000)  0.000* (0.000)  0.029 (0.020)  POLY  0.060 (0.044)  0.058 (0.043)  0.040 (0.029)  0.085 (0.044)  SAT  0.065 (0.050)  0.001* (0.001)  MONO  0.004* (0.003)  0 . 020 (0.015)  0.046 (0.039)  0.000* (0.000)  POLY  0.019* (0.016)  0.079 (0.067)  0.000* (0.000)  0.000* (0.000)  SAT  0.000 (0.000)  0.000 (0.000)  0.000 (0.000)  0.000 (0.000)  Sub: subject; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet. Values with asterisks (*) are those which were extrapolated from the angle of deuterium incorporation. Fractional esterification expressed as pools per day. Values calculated using the EME as a plateau; those in brackets were calculated using the TME as a plateau.  155  0.05  S 0.02 UL  0-12  12-24 24-36 Time periods  ^  MONO _ J POLY IZJ  36-48  SAT  Figure 3.21 Cholesterol fractional esterification rates, calculated for each 12-h period, using a modified version of the linear method, deriving parameter values with SAAM/C0NSAM using the expected maximum cholesterol enrichment as a plateau. n=9, X ± SE. No significant effect of either diet or time was seen.  156  Table 3.31 Calculated de novo esterified cholesterol evolved during each 12-h period deriving parameter values with SAAM/CONSAM  Diet  0-12 0-12 (mg/d)  112-24 2-24 (mg/d)  MONO  112.9* 112.9"  112.9 112.9"  (89.1)  (89.1)  0.0 0.0 (0.0)  143.7 143.7  POLY  142.4* (121.4)  209.5 (157.0)  1120.2 (948.5)  0.0* (0.0)  (62.6)  179.0 (125.1)  89.6* (62.6)  2230.4" 30.4* (131.8)  115.2 (65.9)  0.0* (0.0)  * 00.0" .0 (0.0)  498.0 (376.7)  0.0" (0.0)  147.2 (70.9)  58.5* (28.2)  (127.0)  0.0* (0.0)  0.0* (0.0)  (24.5)  251.4 251.4 (171.7)  195.9 (123.9)  0.0* (0.0)  204.9 204.9 (132.2)  1115.5 15.5 (55.4)  225.9 (149.2)  93.1 (35.3)  * 224.1" 4.1 (17.0)  541.2* (382.5)  324.7 (229.5)  48.0* (37.9)  168.2* (132.8)  515.3 515.3 (429.7)  POLY  0.0 0.0  235.4 235.4 (177.5) *  89.6" 89.6  (0.0)  SAT  2230.4" 30.4* (131.8)  MONO  449.7 449.7 (336.9)  POLY  13 9.7 (92.5)  24.6" 24.6* (21.0)  MONO  MONO  282.3 (222.7)  *  294.0 (232.2)  6634.7 34.7 (541.0)  SAT  36-48 (mg/d)  (96.4)  SAT  POLY  24-36 (mg/d)  1130.4 30.4 (56.3)  228.1" 228.1*  8881.2 81.2 (691.5)  161.8" 161.8  *  35.9" 35.9  SAT  00.0 .0 (0.0)  MONO  24.1" 24.1  *  (19.0)  (98.5) *  1033.4 1033.4 (815.7)  157  POLY  0.0 (0.0)  1063.1 (840.3)  978.8 (768.0)  SAT  0.0 (0.0)  0.0 (0.0)  484.7 (353.3)  MONO  0.0 (0.0)  932.7* (884.0)  1250.7 (1185.4)  POLY  1324.4 (1259.8)  925.0* (879.9)  1657.7 (1574.6)  389.2* (366.8)  605.4 (570.5)  MONO  0.0 (0.0)  443.7 (391.5)  POLY  984.1 (885.0)  SAT  SAT  0.0 (0.0)  MONO  0.0 (0.0)  POLY  1052.9 (780.7)  SAT  1154.3 (845.8)  289.0 (245.4)  * 0.0 (0.0)  487.2 (429.9)  345.6 (299.9) * 0.0 (0.0) 1019.4 (752.7)  * 0.0 (0.0) 543.1 (509.0) 664.0 (626.2) 437.3 (385.3) * 0.0 (0.0)  * 0.0 (0.0)  438.6 (384.7)  * 0.0 (0.0)  510.7 (341.2)  704.7 (507.8)  1058.7 (782.9)  * 17.7 (13.0)  MONO  37.8* (28.5)  189.2 (142.5)  POLY  182.0* (153.8)  762.3 (644.2)  0.0 (0.0)  0.0 (0.0)  SAT  * 0.0 (0.0)  387.6 (269.5)  445.2 (372.2)  * 0.0 (0.0)  * 0.0 (0.0)  * 0.0 (0.0)  0.0 (0.0)  0.0 (0.0)  Sub: subject; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat dietMass evolved values were calculated by SAAM/CONSAM. Values with asterisks ( are those which were extrapolated from the angle of deuterium incorporation. Values calculated using the EME as a plateau; those in brackets were calculated using the TME as a plateau.  158  800  0-12  12-24 24-36 Time periods  11^3 MONO r _ J POLY L |  36-48  SAT  Figure 3.22 Calculated de novo esterified cholesterol evolved during each 12-h period, using a modified version of the linear method, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau. n=9, X ± SE. Mass evolved values were calculated by SAAM/CONSAM. No significant effect of either diet or time was seen.  159  Table 3.32 Cholesterol fractional esterification rates and mass of esterified cholesterol generated, calculated for each 24-h period, deriving parameter values with SAAM/CONSAM  Diet  FED FER  FASTED FER  FED CE  FASTED CE  MONO  0.010 (0.008)  0.022 (0.017)  112.9 (89.1)  241.2 (189.3)  POLY  0.014 (0.009)  0.027 (0.021)  150.2 (96.4)  294.0 (232.2)  SAT  0.035 (0.028)  0.016 (0.013)  385.9 (314.0)  176.0 (139.2)  MONO  0.030 (0.025)  0.032 (0.020)  395.9 (321.5)  419.7 (266.5)  POLY  0.004 (0.003)  0.010 (0.006)  44.8 (31.3)  137.1 (80.8)  SAT  0.018 (0.010)  0.005 (0.003)  230.4 (131.8)  57.6 (33.0)  MONO  0.014 (0.008)  0.020 (0.014)  190.4 (110.0)  282.7 (191.2)  POLY  0.013 (0.006)  0.008 (0.003)  178.3 (77.4)  116.7 (41.9)  SAT  0.041 (0.032)  0.000 (0.000)  574.9 (443.6)  0.0 (0.0)  MONO  0.014 (0.007)  0.018 (0.010)  117.1 (57.1)  146.7 (81.3)  POLY  0.022 (0.013)  0.023 (0.014)  181.7 (110.3)  191.3 (117.5)  SAT  0.002 (0.001)  0.014 (0.006)  12.1 (8.5)  115.2 (53.7)  MONO  0.026 (0.019)  0.006 (0.005)  501.5 (367.0)  108.1 (85.4)  160  POLY  0.027 (0.020)  0.052 (0.041)  513.3 (380.2)  978.3 (768.0)  SAT  0.000 (0.000)  0.024 (0.018)  0.0 (0.0)  461.4 (334.0)  MONO  0.023 (0.022)  0.027 (0.025)  466.4 (442.0)  532.8 (500.0)  POLY  0.066 (0.063)  0.027 (0.025)  1327.4 (1259.8)  536.4 (509.0)  SAT  0.005 (0.004)  0.017 (0.016)  101.1 (77.5)  332.0 (313.1)  MONO  0.012 (0.009)  0.023 (0.020)  167.8 (132.9)  322.1 (277.1)  POLY  0.041 (0.036)  0.017 (0.015)  582.9 (516.5)  243.6 (215.0)  SAT  0.012 (0.011)  0.016 (0.014)  172.8 (150.0)  219.3 (192.4)  MONO  0.000 (0.000)  0.015 (0.010)  0.0 (0.0)  255.4 (170.6)  POLY  0.059 (0.044)  0.044 (0.032)  1051.0 (776.7)  780.4 (565.9)  SAT  0.039 (0.028)  MONO  0.012 (0.009)  0.023 (0.020)  113.5 (85.5)  222.6 (186.1)  POLY  0.044 (0.037)  0.000 (0.000)  472.2 (350.5)  0.0 (0.0)  SAT  0.000 (0.000)  0.000 (0.000)  0.0 (0.0)  0.0 (0.0)  691.5 (493.3)  Sub: subject; MONO: monounsaturated fat diet; POLY: polyunsaturated fat diet; SAT: saturated fat diet; CE: cholesterol esterified, mg/d. Fractional esterification rates expressed as pools per day. Values calculated using the EME as a plateau; those in brackets were calculated using the TME as a plateau.  161  0.05  FASTED  FED  ^  MONO ^ J POLY CZ3 SAT  Figure 3.23 Cholesterol fractional esterification rates, calculated for each 24-h period, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau. n=9, X ± SE. Columns within the same time period not sharing a common superscript are significantly different. No significant time effect was seen.  162  800  FASTED  FED  M  MONO [_}  POLY £ZJ SAT  Figure 3.24 Calculated de novo esterified cholesterol evolved during each 24-h period, deriving parameter values with SAAM/CONSAM using the expected maximum cholesterol enrichment as a plateau. n=9, X ± SE. Mass evolved values were calculated by SAAM/CONSAM. Columns within the same time period not sharing a common superscript are significantly different. No significant time effect was seen.  163  4  4.1  A BRIEF  REVIEW  OF STUDY  DISCUSSION  GOALS  Hypercholesterolemia is associated with CHD and atherosclerosis, and diet fat saturation is strongly implicated in its development.  Many  studies have documented the correlation between the degree of saturation of diet fatty acids and serum lipid concentrations, and several mechanisms have been proposed to explain how these changes are effected. To date, no mechanisms appear predominantly responsible for the rapid changes in serum lipid concentrations brought about by alterations in diet fat saturation. It is not clear whether, in humans, the degree of diet fat saturation influences the rates of cholesterol synthesis and esterification.  This was the primary goal of this investigation.  A  second goal was to examine whether postprandial rates of synthesis and esterification differed from those seen during fasting. The effect of diet fat saturation on rates of cholesterol synthesis and esterification were examined by measuring the rate of D incorporation into newly synthesized FC and CE isolated from serum obtained from healthy subjects at regular intervals following D ingestion.  The data obtained were incorporated into a mathematical  model, which was used to predict rates of cholesterol formation on each of three diets differing only by their fatty acid composition.  164  Serum  samples were also assayed for lathosterol concentration, to compare results from this assay with those obtained from D incorporation.  4.2  MODELS  Cholesterol metabolism is an amalgamation of its synthesis, incorporation, and destruction in the serum and 16-21 major organs, each likely possessing several compartments (Anderson and Dietschy 1977; Ostlund 1993; Wilson 1970).  Use of a model is an attempt to reduce this  complexity to understandable terms and produce relevant clinical information.  The model is the basis for any conclusions drawn about the  effect of diet fat composition on individual serum cholesterol concentrations; it is thus imperative to validate it, outlining any assumptions or shortcomings, before accepting its estimates.  4.2.1 Model assumptions and ambiguities There are several areas of uncertainty when deriving these models. First, estimates of total body cholesterol pool size are based on the three-pool model, which states that the central and rapidly turning over pool, into which arises newly synthesized FC, contains approximately 25 g of cholesterol (Goodman et al. 1973).  This was provided as data in  the modelling decks, and SAAM/CONSAM calculates the quantity of cholesterol synthesized based partly on an assessment of central pool size, which in these subjects ranged from 14.3 to 29.9 g.  Fortunately,  similar results have been reported by others using different methods of  165  estimation.  Samuel and Perl (1970) estimated a central pool size of  25.3 g, Smith et al. one of 21.4 g.  (1976) one of 25.9 g, and Schwartz et al.  (1993)  Faix et al. (1993), who measured rates of cholesterol  and fat synthesis in humans using MIDA, defined a central FC pool size of 9 g, which is similar to the above quotes since the central pool is thought to contain 33% FC and 67% CE. Variability in pool size would affect estimates of the total quantity of cholesterol synthesized de novo;  i.e., a larger pool size would translate into higher estimates of  absolute cholesterol synthesis, although obviously it would have no influence on fractional rates. This method measures synthesis only in the central pool and cannot be taken as a measure of whole-body cholesterol synthesis. Longer studies would have to account for synthesis in two or three pools,- this was recently accomplished by Zhang et al. (1994), who used D incorporation to measure rates of synthesis in piglets.  The piglets  were, however, sacrificed to allow precise determination of the rate of cholesterol synthesis in each tissue; this was not feasible in the present study because there are ethical limitations when dealing with humans. There is considerable debate as to the proportion of whole-body cholesterol synthesis occurring in the central pool.  Hepatic synthesis  is thought to be a major component of central pool synthesis, yet Spady and Dietschy (1983), who measured  3 H20 incorporation into DPS in tissues  from several animal species, reported that in monkeys hepatic synthesis  166  accounted for only 40% of whole-body newly synthesized cholesterol. Dell et al. (1985), who quantified cholesterol synthesis in baboons, found that 70% of total whole-body cholesterol synthesis occurred in the tissues which comprise the central pool.  If a similar situation exists  in humans, the data gathered here would not represent a complete look at the effect of diet fat saturation on the rate of whole-body cholesterol synthesis in humans.  Since sterol balance studies were not included in  this experiment, the quantity of cholesterol which the subjects synthesize per day, as estimated by the models, cannot be equated with daily total cholesterol production. Equilibration across the central pool is rapid: over 60% of hepatic pool free cholesterol exchanges with total plasma cholesterol per h and exchange of tracer among lipoprotein fractions occurs even more rapidly,- for example, 2.2 pools/h for VLDL/LDL into HDL (Schwartz et al. 1993).  Even red blood cells, which do lag behind plasma, do so  only by a few hours (Jones et al.  1993a), which is well within the  shortest test period used in this study.  This eliminates worries about  tracer concentrating in subcompartments, and suggests that enrichment changes in plasma reflect those in the entire pool. To date, no method has satisfactorily quantified short-term wholebody cholesterol synthesis because of uncertainties in defining both pool size and rates of exchange between pools.  In the present study,  rates of exchange were determined over a nine month period when subjects consumed a mixed diet; these rates are almost certainly altered by diet  167  fat quality.  To obtain exact exchange rates during consumption of each  diet, subjects would have to consume specific diets for a nine month period.  Unfortunately, this option is too costly to be practical.  In this study it was assumed that the fraction of deuterium incorporated into each cholesterol molecule was unchanged by diet. This may not be true; Jones et al.  (1993a) suggested that conditions causing  shifts in NADPH equilibrium may alter the commonly accepted ratio of 0.81 labelled protons per carbon atom.  For example, upregulation of the  pentose phosphate pathway, which does not exchange protons with body water and thus does not contribute to sterol enrichment during synthesis, would result in a greater proportion of the NADPH used as sources of protons during cholesterol synthesis being unlabelled, resulting in the appearance of lower synthesis rates. Similarly, lowered activity of the malic enzyme system during fasting (Mayes 1988b) could result in less availability of labelled NADPH from this source. It remains to be determined whether diet significantly alters this source of labelled protons.  4.2.2 Model comparisons  4.2.2.1  Linear  incorporation  of  tracer  Linear tracer incorporation looks at change in serum cholesterol D enrichment over each 12-h period relative to that of the precursor body water pool, adjusting for the fraction of hydrogens of cholesterol  168  derived from labelled substrate (Dietschy and Spady 1984) . This calculation does not require a complex model and allows estimation of cholesterol synthesis rates over short periods of time. The original method has been criticized, both because there is no physiological base for a model based on linear regression and because it calculates negative FSR (Foster et al., 1993, Parhofer et al., 1991). It is true that it is presently not possible to interpret negative results as calculated using the original linear method.  A negative FSR  is an indication that rates are lower than in other periods being compared, but we cannot partition out the effects of substrate recycling or incoming cholesterol from other pools. The second linear method, which calculates FSR only in comparison with the original product enrichment and not with that in the preceding period, effectively removes this problem.  Additionally, while it has been suggested that  the decrease in tracee incorporation into serum cholesterol during the fasted period when compared with the fed period simply represents a decrease in the rate of enrichment with time (Foster et al.  1993), serum  lathosterol concentrations, commonly used as an index of both whole-body (Kempen et al.  1988) and hepatic (Bjorkhem et al.  1987) cholesterol  synthesis, declined sharply in the fasted period when compared with the fed period.  This is an indicator that the decrease in the rate of  tracer enrichment during the fed period actually reflects a decline in de novo cholesterol synthesis.  169  While there may not be a physiological basis for a model based on linear regression, the initial, short-term D incorporation rate is linear.  Furthermore, this linear uptake rate is unaffected by flux  rates of other, unlabelled, material into the system, and can be taken to represents a direct measure of synthesis independent of the total whole-body production rate (Jones et al.  1994a,- 1994b).  Consequently,  the conclusion from this study is that in a simple and short-term onepool model, where entry of one tracer into one pool is examined and data points are spaced at 12-h intervals, the second linear model can yield reasonable estimates of rates of cholesterol synthesis.  4.2.2.2  Monoexponential  incorporation  of  tracer  This method uses SAAM/CONSAM to fit tracer incorporation to monoexponential equations over specific time periods, using either theoretical or expected maximum cholesterol D enrichment as a function of body water enrichment.  It then calculates rates of de  novo  cholesterol synthesis. Plasma cholesterol long-term enrichment data in a subject given continual deuterium have shown a tendency to plateau below the TME (Jones et al.  1993a; London and Schwarz 1953).  It is not likely  attainable since the plasma cholesterol pool is an open system,- i.e. cholesterol in plasma is continually being exchanged with other body pools.  The total exchange with other compartments for each subject was  estimated by SAAM/CONSAM using the specific activity decay curves, and  170  used to calculate the EME for each subject. Although these parameters are possibly affected in an unknown fashion by specific diet regimes, their incorporation into SAAM/CONSAM's calculations is more likely than not to yield reasonable estimates of synthesis and turnover rates; regardless of the diet consumed, some exchange of cholesterol between compartments will occur. Fractional synthesis rate calculations are dependent upon the time point used to calculate the slope of tracer incorporation into the tracee (Foster et al. 1993).  In this study 12-h FSR values are based on  a maximum of two points, which does not permit SAAM/CONSAM to reliably calculate error probabilities or generate monoexponential tracer incorporation curves.  Given the limitations imposed by the study  design, where only five data points were gathered during each diet period and used to estimate rates of cholesterol synthesis, it is prudent to calculate FSR using a maximum number of data points. Moreover, the flux parameters for exchange between compartments, which were generated from specific activity decay curves and incorporated into SAAM/CONSAM decks, are rates of flux per 24 h.  Undoubtedly these vary  diurnally, but this cannot be determined in this study, and use of these rates when calculating FSR over 12-h periods risks exaggerating or minimizing parameter values. As a result, FSR calculated over 12-h periods were not used to support any study conclusions, but are provided for comparison only.  171  Modelling is used to characterize steady state whole-body flux and turnover; the presence of steady state conditions is a prerequisite for modelling data (Foster et al.  1993).  This augers well for the use of  the 24-h period to calculate FSR, since subjects were fed for the first 24 h and fasted for the second 24 h.  Calculations of changes in FSR  caused by feeding state are somewhat uncertain because it is not possible to know exactly when the labelled cholesterol appearing in the serum was synthesized (Faix et al. is not attained instantly.  1993); obviously a new steady state  Previous investigations where blood samples  were taken frequently after D dosing showed labelled cholesterol appearing in serum within 4 hours of dose ingestion (Jones et al. 1992; 1993a; Jones and Schoeller 1990).  1988;  In this study samples were not  taken until at least 12 h into each time period; accordingly the data collected at 12 and 24 h are taken as representative of synthesis during each state. A similar approach has been recently used by others examining diurnal variation in lipid synthesis (Faix et al. 1993).  In  addition, changes in serum lathosterol were positively correlated with 24-h FSR values and mass of cholesterol synthesized.  Unfortunately, the  same argument as was used earlier against the use of 12-h periods can also be used here: 24-h FSR values are based on three points only.  This  yields estimated with huge standard errors, and renders it difficult to statistically distinguish curves unless the differences between them are of enormous magnitude, which was generally not the case in this study. It increases the risk of making Type II errors when analyzing the data.  172  As above, much more confidence can be placed in monoexponential curves which are generated with a large number of points, rather than with three points. The points garnered in favour of using the 24-h periods for monoexponential modelling conversely argue against using a single 48-h period to generate FSR values. While serum cholesterol concentrations do not change over the 48-h periods, food intake patterns are considerably altered and most likely perturb steady state conditions pertaining to lipid and cholesterol body equilibrium.  As stated  earlier, steady state conditions must exist during the period being modelled.  4.2.2.3  Conclusion  The above rationale supports the use of FSR and other parameter values generated by the second linear method.  These will now be used to  support any conclusions reached concerning the effect of the degree of diet fat saturation on rates of cholesterol synthesis and esterification in these nine normal men.  4.3  THE EFFECT  OF DIET  FAT  SATURATION  4.3.1 De novo cholesterol synthesis Serum lipid concentrations varied according to the type of diet fat consumed, as predicted by the literature.  173  Synthesis parameters for  FC which were calculated using the second linear method suggested that synthesis rates were greater when POLY were consumed and lesser when SAT were consumed.  This suggests that another mechanism apart from rates of  de novo synthesis is responsible for the strikingly different serum cholesterol concentrations seen in the subjects consuming each of the three diets. Animal studies suggest that differences in the rate of de novo cholesterol synthesis are indeed induced by varying the saturation of diet fat.  Hamsters consuming high POLY diets have been shown to produce  greater amounts of de .novo cholesterol than those fed either SAT or MONO diets (Spady and Dietschy 1988), although the opposite effect was seen in gerbils (Mercer and Holub 1981) and rats (Triscari et al.  1978).  Fernandez and McNamara (1994) recently confirmed that in guinea pigs fed olive oil-enriched diets, HMG-CoA reductase was less active than in those fed either beef tallow- or corn oil-based diets., suggesting a diminished rate of synthesis in animals fed olive oil diets.  Guinea  pigs and hamsters are thought to be better models of human cholesterol metabolism than are rats (Dietschy 1984) . Previous research on humans suggests, as does the present study, that diet fat saturation can alter rates of de novo cholesterol synthesis.  Jones et al.  (1991; 1994b), who measured de novo cholesterol  synthesis in hypercholesterolemic subjects using D incorporation, concluded that rates of synthesis were greater in subjects fed a corn oil-based diet compared with either a baseline diet or an olive oil-  174  based diet, but were not significantly different from rates of synthesis in subjects fed a canola oil-based diet. These findings were repeated in another experiment where D incorporation was used to measure rates of cholesterol synthesis in elderly, mildly hypercholesterolemic subjects (Jones et a.1. 1994a): rates of synthesis were greater in subjects fed diets containing corn oil than in those fed diets containing beef tallow.  Results from the abovementioned studies are strongly supportive  of the present findings.  Glatz and Katan (1993), however, who measured  cholesterol synthesis both by the concentration of lathosterol in serum of normal humans fed either POLY or SAT diets, concluded that SAT fatty acids stimulate whole-body cholesterol synthesis.  Serum lathosterol  results from the present study suggest that rates of synthesis are greater on MONO and POLY diets than on SAT diets during the fed period, although no differences were seen during the fasted period.  Results  suggest strongly that these two methods may not be measuring synthesis in the same pools but that, as suggested previously, D enrichment looks at synthesis in the central pool whereas serum lathosterol may be more indicative of either whole-body synthesis or synthesis outside the central pool. Shepherd et a.1. (1980) reported that although there was an enhanced FCR of LDL in men fed POLY- versus SAT-enriched diets, this was not accompanied by significant alterations in the rate of cholesterol synthesis.  This enhanced LDL FCR is though most likely for the low  serum cholesterol concentrations seen in the subjects consuming POLY  175  diets, even if the cholesterol FSR rates were high.  Other mechanisms  may be implicated as responsible for the low levels of serum cholesterol seen while on these diets.  Several studies have shown enhanced faecal  sterol excretion on POLY diets compared with SAT diets (Connor et 1969; Grundy 1975; Moore et al. 1985).  As a result, Jones et al.  al.  1968; Nestel et al. 1975; Oh and Monaco (1994a) has suggested that POLY fats  may affect central pool cholesterol concentrations by enhancing hepatic cholesterol elimination, up-regulating removal of circulating sterol and thus invoking higher rates of synthesis.  The results of the present  study support this hypothesis. Not all studies support these findings.  Early cholesterol balance  studies on hyperlipidemics receiving liquid diets containing either butter, safflower oil or sunflower oil were unable to demonstrate appreciable differences in rates of cholesterol synthesis amongst the three diets (Grundy and Ahrens 1970); a more recent balance study reports similar findings (McNamara et al.  1987).  As reported earlier,  not all studies observed changes in faecal steroid excretion with POLY diets (Avigan and Steinberg 1965; Grundy and Ahrens 1970; Shepherd et al.  1980; Spritz et al.  1965).  The question of how POLY fats affect  cholesterol metabolism is not completely resolved, although the present study's results are clearly compatible with those published elsewhere, suggesting an indirect role of POLY fats on serum cholesterol concentrations.  176  4.3.2 Cholesterol esterification Fractional esterification rates were greater on the POLY diet than on the SAT diet, with MONO values not differing from those on either of the other two diets. Less cholesterol esterification occurs during periods of high SAT intake because esterifying enzymes have a marked preference for POLY and MONO fatty acids (Glomset 1968; 1970; Linscheer and Vergroesen 1988).  This results in an expanded hepatic pool of FC  when SAT fat is consumed.  Although this was not reflected in the .ratio  of serum FC to CE of the subjects, they did have elevated serum cholesterol levels when compared with those on the other two diets. In a recent review of hepatic metabolism pertaining to LDL homeostasis, Dietschy et al.  (1993) contend that numerous studies,  conducted both in animals (Carr et al. 1992; Fernandez and McNamara 1991; Fernandez et al. 1992; Hennessy et al.  1992) and humans (Hegsted  et al. 1965; Mattson and Grundy 1985; McMurry et al. 1991; Mensink and Katan 1992) support the view that when diet TG are predominantly saturated, the rate of esterification decreases and simultaneous increases in LDL-C production are observed, along with decreases in hepatic LDL receptor mRNA (Fox et al. diet fats are unsaturated.  1986).  The converse is true when  Additionally, most of the cholesterol  esterification in plasma takes place in HDL (Dobiasova et al.  1991);  HDL3 acts as a substrate for LCAT esterification while HDL2 inhibits this process (Barter et al.  1984; 1985).  Shepherd et al.  (1978) have  demonstrated that the ratio of HDL2 to HDL3 falls by 28% in normal men  177  consuming a POLY diet, compared with a SAT diet.  The relative rates of  appearance of CE during each both the SAT and POLY diet periods concur with the relevant literature. Correlations between low levels of plasma CE and ingestion of MONO diets have been shown in humans (Baggio et al.  1988; De Backer et al.  1989; Grundy 1987; Pal and Davis 1991) and in animals (Jones et al. 1990).  In this investigation, rates of appearance of, CE did not differ  significantly, while the MONO diet was being consumed, from rates on either of the other two diets.  In addition, serum HDL concentrations  were very high in subjects consuming the MONO diet.  Dreon et al. (1990)  have shown that levels of HDL2 are higher and those of HDL3 lower on safflower and corn oil diets compared with peanut and olive oil diets. On the other hand, Valsta et al.  (1992) showed an opposite effect with  diets supplemented with either sunflower oil or rapeseed oil.  Since  fractions of HDL2 and HDL3 were not measured in this study, it is cannot be concluded with certainty that rates of cholesterol esterification are low on the MONO diet because of the inhibitory effect of a high percentage of HDL2, but the effect of SAT and POLY diets on rates of cholesterol esterification are straightforward.  178  4.4  THE EFFECT  OF FEEDING  VERSUS  FASTING  4.4.1 De novo cholesterol synthesis Cholesterol synthesis parameters which were computed using either the first linear method or SAAM/CONSAM were greater during the fed state than during the fasted state. Serum lathosterol concentrations and ratios of serum lathosterol concentration to serum cholesterol concentration corroborated these results. This is in agreement with earlier reports comparing rates of cholesterol synthesis between fed and fasted periods in humans (Jones et al.  1988; 1993a) and animals (Jeske  and Dietschy 1980; Kelley and Story 1985; Kempen et al. West 1976; Weingand and Daggy 1991).  1986; Turley and  Rates of cholesterol synthesis  measured using the second linear method did not, however, differ when subjects were fed or fasted;  this may indicate, as mentioned earlier,  that this linear method is not suitable for longer-term measurement because it doesn't allow differentiation between lower D enrichment levels attributable to lower rates of synthesis or loss of tracer to other compartments. The process whereby rates of cholesterol synthesis are altered by feeding state is unknown.  It has been suggested that during periods of  food restriction cholesterol precursors such as acetyl-CoA associated with stored TG may be used as energy substrates, thus limiting their availability as substrates for HMG-CoA reductase and limiting sterol synthesis (Jones et al.  1988; 1993a).  179  It is unclear whether the D  incorporation method would detect such differences.  Over the short time  period of the study (48 h), acetyl-CoA derived from fat stores is unlikely to be labelled; any de novo cholesterol synthesis arising from this source would therefore not be detected.  It may be that during the  fasted period an influx of cholesterol from other body pools prevents cholesterol from being synthesized by feedback inhibition of HMG-CoA reductase. Additionally, this enzyme's activity is decreased by glucagon and glucocorticoids, both of which would be present in elevated concentrations during fasting (Mayes 1988a).  The rate of cholesterol  synthesis is plainly sensitive to the individual's feeding state.  4.4.2 Cholesterol esterification Parameters for CE which were determined using SAAM/CONSAM were similar whether subjects were fed or fasted.  Conversely, those measured  with the second linear method showed an increase in esterification during the fasted period, compared with the fed period.  Fielding et al.  (1989), who worked with baboons as a model of human cholesterol metabolism, found that in animals fed a high-fat diet fasting levels of LCAT activity were greatly increased compared with those seen during the fed period.  There is some question as to whether the baboon is a good  model for humans however; Mott et al.  (1987) found an increased  esterification rate following chronic SAT feeding in baboons; counter to what has been reported in humans, as well as in the present investigation.  Rose and Juliano (1977) measured LCAT activity following  180  ingestion of a high-fat meal and noted that LCAT activity closely followed serum TG concentration, peaking five h following consumption of a high-fat liquid meal and remained elevated for over seven h, with a mean increase of 37.2 % over baseline levels. A smaller, but still significant, increase was also reported by Marcel and Vezina (1973) . Castro and Fielding (1985) also demonstrated an increase in the plasma esterification rate following a meal; this increase was suppressed over 98% in the presence of an LCAT inhibitor, suggesting that ACAT plays a minor role as a source of plasma CE during fed periods.  It may be that  during the fed period, when LCAT is very active, only a small proportion of the cholesterol which is esterified by LCAT contained D, but that during the fasted period, when LCAT activity is reduced, a greater proportion of available FC is D-labelled.  This would occur, for  example, if endogenous, rather than de novo synthesized, cholesterol was esterified.  Unfortunately, in this study LCAT activity was not assayed;  it is therefore difficult to draw any conclusions concerning its activity under different conditions. A more simple explanation may be that rates of esterification cannot be compared from time period to time period in this study because the labelled precursor pool size of FC varied enormously; the percentage of labelled FC would determine how much newly esterified CE would be labelled. this study.  This was not determined in  The effect of feeding versus fasting on rates of  cholesterol esterification, as determined using D incorporation, remains ambiguous.  181  4.5  SUMMARY  4.5.1 Cholesterol production It is interesting to compare the results of this investigation to those proposed in the literature by other researchers using techniques ranging from sterol balance techniques to MIDA.  Faix et al. (1993),  proposed a daily synthesis range of 608 + 146 mg per day in women, 621 + 111 mg per day in obese men and 697 + 55 mg per day in men of normal weight.  Ferezou et al. (1982) estimated daily cholesterol input to be  1.110 + 0.10 g per day.  Glatz and Katan (1993), produced figures of  599.3 + 328.7 mg synthesis per day on a high P:S diet and 719.2 + 320.9 mg per day on a low P:S diet.  Goodman et al.  cholesterol turnover of 1.13 + 0.09 g per day.  (1973) calculated daily Neese et al. (1993)  estimated normal cholesterogenesis in women to be 568 + 55 mg per day; and Ostlund (1993) arrived at daily cholesterol production rate of 1.13 + 0.01 g per day.  None of the numbers quoted above arise from studies  dealing with D incorporation methods,- it is encouraging that this methodology plus modelling with SAAM/CONSAM produces figures for daily cholesterol production which are comparable to those from studies which employed laborious, costly or conceptually difficult techniques.  4.5.2 Dietary recommendations One of the goals of this study was to strengthen the dietary recommendations for Canadians in terms of fat intake.  182  It has been  exhaustively documented that a high SAT intake raises serum cholesterol concentrations, which can lead to detrimental effects on human arteries and increase the overall CHD risk.  It is a wistful hope of researchers  that one mechanism can some day be positively pinpointed which brings about this effect. Unfortunately, this was not the case in the present study, as we were not able to show an increased rate of synthesis when SAT was consumed: indeed, the opposite effect was observed.  However,  this study combines with many others in showing that, whatever the mechanism involved, serum cholesterol levels are much greater on SAT diets than on POLY or MONO diets. As such, it supports the present dietary recommendations.  4. 6  CONCLUSIONS  This study examined the effect of varying the degree of diet fat saturation and feeding state on rates of cholesterol synthesis and esterification in nine normal young men. The data suggest that rates of FC synthesis are greater when POLY fats are consumed than when SAT fats are consumed, with MONO rates being intermediate.  This occurred even though serum cholesterol  concentrations were clearly greater while subjects were on the SAT diet than when on either of the other two diets. These findings have been previously demonstrated in hypercholesterolemic and elderly individuals, but had not yet been demonstrated in healthy young men.  Rates of  esterification are also subject to diet fat's influence, being impeded  183  by SAT diets and accelerated by POLY diets. Together, these findings support an overall impact of diet fat saturation on total serum cholesterol concentrations. Free cholesterol synthesis is greater during the fed state than when subjects are fasted; this was suggested by the D incorporation data and confirmed by the serum lathosterol data.  Discrepancies between  these two methods are thought to reflect the pools in which each method measures cholesterol synthesis. Whether, feeding versus fasting affects rates of cholesterol esterification remains to be determined.  Longer-  term studies may be more effective for studying this problem; alternatively, better ways of measuring the proportion of the FC cholesterol pool which is labelled could be developed. Finally, the D incorporation method yields estimates of daily cholesterol synthesis which are in agreement with those obtained by many other researchers using other methods.  This augers well for its  continued use as an accurate technique for measurement of de novo lipid synthesis.  184  REFERENCES  Ahmed AA, Holub BJ. 1984. Alteration and recovery of bleeding times, platelet aggregation, and fatty acid composition of individual phospholipids in platelets of human subjects receiving a supplement of cod liver oil. Lipids 19:617-624. Ahrens EH Jr, Insull W Jr, Blomstrand R, Hirsch J, Tsaltas TT, Peterson ML. 1957. The influence of dietary fat on serum-lipid levels in man. Lancet 1:943-953. Allain CC, Poon L, Chan SG, Richmond W, Fu P. 1974. Enzymatic determination of total serum cholesterol. Clin Chem 20:470-475. Anderson JM, Dietschy JM. 1977. Regulation of sterol synthesis in 16 tissues of rat. I. Effect of diurnal light cycling, fasting, stress, manipulation of enterohepatic circulation and administration of chylomicrons and triton. J Biol Chem 252:36463651. Andriamiarina R, Laraki L, Pelletier X, Debry G. 1989. Effects of stigmasterol-supplemented diets on fecal neutral sterols and bile acid excretion in rats. Ann Nutr Metab 33:297-303. Avigan J, Steinberg D. 1965. Sterol and bile acid excretion in man and the effects of dietary fat. J Clin Invest 44:1845-1856. Baggio G, Pagnan A, Muraca M, Martini S, Opportuno A, Bonanome A, Ambrosio GB, Ferrari S, Guarini P, Piccolo D, Manzato E, Corrocher R, Crepaldi G. 1988. Olive-oil-enriched diets: effects on serum lipoprotein levels and biliary cholesterol saturation. Am J Clin Nutr 47:960-964. Bang HO, Dyersberg J, Nielsen AB. 1971. Plasma lipid and lipoprotein pattern in Greenlandic west-coast Eskimos. Lancet 1:1143-1146. Bannon CD, Craske JD, Har NT, Harper NL, O'Rourke KL. 1982. Analysis of fatty acid methyl esters with high accuracy and reliability. II. Methylation of fats and oils with boron trifluoride-methanol. J Chrom 247:63-69. Barr SL, Ramakrishnan R, Johnson C, Holleran S, Dell RB, Ginsberg HN. 1992. Reducing total dietary fat without reducing saturated fatty acids does not significantly lower total plasma cholesterol concentrations in normal males. Am J Clin Nutr 55:675-681.  185  Barter PJ, Hopkins GJ, Gorjatschko L. 1985. Lipoprotein substrates for plasma cholesterol esterification: influence of particle size and composition of the high density lipoprotein subfraction 3. Atherosclerosis 58:97-107. Barter PJ, Hopkins GJ, Gorjatschko L, Jones ME. 1984. Competitive inhibition of plasma cholesterol esterification by human high density lipoprotein subfraction 2. Biochim Biophvs Acta 793:260268. Bell L, Jones PJH, Telch J, Clandinin MT, Pencharz PB. 1985. Prediction of energy needs for clinical studies. Nutr Res 5:123-129. Bergman RN. 1988. Modeling in nutrition and metabolism. Prog Food Nutr Sci 12: 211-225. Berman M. 1982. Kinetic analysis and modeling: theory and applications to lipoproteins. In: Berman B, Grundy SM, Howard G (eds) Lipoprotein Kinetics and Modeling, Academic Press, Inc.; pp. 3-36. Berman M, Beltz WF, Greif PC, Chabay R, Boston RC. 1983. CONSAM User's Guide. Washington, D.C.: US Government Printing Office. Berman M, Shahn E, Weiss MF. 1962. The routine fitting of kinetic data to models: a mathematical formalism for digital computers. Biophvs J 2:275-287. Berman M, Weiss MJ. 1978. SAAM Manual. Washington, D.C.: US Government Printing Office. Berry EM, Eisenberg S, Haratz D, Friedlander Y, Norman Y, Kaufman NA, Stein Y. 1991. Effects of diets rich in monounsaturated fatty acids on plasma lipoproteins - the Jerusalem Nutrition Study: high MUFAs vs. high PUFAs. Am J Clin Nutr 53:899-907. Beynen AC, Katan MB, Van Zutphen LFM. 1987. Hypo- and hyperresponders: individual differences in the response of serum cholesterol concentration to changes in dietary. Adv Lipid Res 22:115-171. Bhattacharyya AK, Eggen DA. 1984. Effects of feeding cholesterol and mixed plant sterols on the fecal excretion of acidic steroids in rhesus monkeys. Atherosclerosis 53:225-232. Biss K, Ho KJ, Taylor CB. 1980. Correlation of kinetic parameters of cholesterol metabolism with serum cholesterol levels. In: Schettler G, Weizel A (eds) Atherosclerosis V, Springer-Verlag, Berlin, pp. 512-515.  186  Bjorkhem I, Miettinen T, Reihner E, Ewerth S, Angekin B, Einarsson K. 1987. Correlation between serum levels of some cholesterol precursors and activity of HMG-CoA reductase in human liver. J Lipid Res 28:1137-1143. Blum CB, Dell RB, Palmer RH, Ramakrishnan R, Seplowitz AH, Goodman DS. 1985. Relationship of the parameters of body cholesterol metabolism with plasma levels of HDL cholesterol and the major HDL apoproteins. J Lipid Res 26:1079-1088. Bonanome A, Grundy SM. 1988. Effect of dietary stearic acid on plasma cholesterol and lipoprotein levels. N Engl J Med 318:1244-1248. Boston RC, Greif PC, Berman M. 1981. Conversational SAAM--an interactive program for kinetic analysis of biological systems. Comp Prog Biomed 13: 111-119. Boston RC, Greif PC, Berman M. 1982. CONSAM (Conversational version of the SAAM modeling program) . In: Berman B, Grundy SM, Howard G (eds) Lipoprotein Kinetics and Modeling, Academic Press, Inc.; pp. 437-460. Brongeest-Schoute HC, van Gent CM, Luten JB, Ruiter A. 1981. The effect of various intakes of w-3 fatty acids on the blood lipid composition in healthy human subjects. Am J Clin Nutr 34:17521757. Brousseau ME, Stucchi AF, Vespa DB, Schaefer EJ, Nicolosi RJ. 1993. A dietary enriched in monounsaturated fats decreases low density lipoprotein concentrations in cynomolgous monkeys by a different mechanism than does a dietary enriched in polyunsaturated fats. J Nutr 123:2049-2058. Brown MS, Goldstein JL. 1983. Lipoprotein receptors on the liver: control signals for plasma cholesterol traffic. J Clin Invest 72:743-747. Brown MS, Kovanian PT, Goldstein JL. 1981. Regulation of plasma cholesterol by lipoprotein receptors. Science 221:628-635. Brox JH, Killis JE, Gunnes S, Nordoy A. 1981. The effect of cod liver oil and corn oil on platelets and vessel wall in man. Thromb Haemost 46:601-611. Brox JH, Killis JE, Osterud B, Holme S, Nordoy A. 1984. Effect of cod liver oil on platelets and coagulation in familial hypercholesterolemia (Type Ila). Acta Med Scan 213:137-144.  187  Bucolo G, David H. 1973. Quantitative determination of serum triglycerides by the use of enzymes. Clin Chem 19:476-482. Carr TP, Parks JS, Rudel LL. 1992. Hepatic ACAT activity in African green monkeys is highly correlated to plasma LDL cholesteryl ester enrichment and coronary artery atherosclerosis. Arteriosclerosis & Thrombosis 12:1274-1283. Castro GR, Fielding CJ. 1985. Effects of postprandial lipemia on plasma cholesterol metabolism. J Clin Invest 75:874-882. Chait A, Onitiri A, Nicoll A, Rabaya E, Davies J, Lewis B. 1974. Reduction of serum triglyceride levels by polyunsaturated fat. Studies on the mode of action and on very low density lipoprotein composition. Atherosclerosis 20:347-364. Chajek T, Fielding CJ. 1978. Isolation and characterization of a human serum cholesteryl ester transfer protein. Proc Natl Acad Sci (USA) 75:3445-3449. Chan JK, Bruce VM, McDonald BE. 1991. Dietary alpha-linolenic acid is as effective as oleic acid and linoleic acid in lowering blood cholesterol in normolipidemic men. Am J Clin Nutr 53:1230-1234. Chang NW, Huang PC. 1992. Effects of dietary monounsaturated fatty acids on plasma lipids in humans. J Lipid Res 31:2141-2147. Childs MT, King IB, Knopp RH. 1990. Divergent lipoprotein responses to fish oils with various ratios of eicosapentaenoic acid and docosahexaenoic acid. Am J Clin Nutr 52:632-639. Chobanian AV, Burrows BA, Hollander W. 1962. Body cholesterol metabolism in man. II. Measurement of the body cholesterol miscible pool and turnover rate. J Clin Invest 41:1738-1744. Chobanian AV, Hollander W. 1962. Body cholesterol metabolism in man. I. The equilibration of serum and tissue cholesterol. J Clin Invest 41:1732-1737. Clifton PM, Nestel PJ. 1992. Influence of gender, body mass index, and age on response of plasma lipids to dietary fat plus cholesterol. Arteriosclerosis & Thrombosis 12:955-962. Cobelli C, Toffolo G, Bier DM, Nosadini R. 1987. Models to interpret kinetic data in stable isotope tracer studies. Am J Physiol 253: E551-E564.  188  Coburn SP. 1992. Application of models to the determination of nutrient requirements: an overview. J Nutr 122: 687-689. Cohn JS, McNamara JR, Cohn SD, Ordovas JM, Schaefer EJ. 1988a. Postprandial plasma lipoprotein changes in human subjects of different ages. J Lipid Res 29:469-479. Cohn JS, McNamara JR, Schaefer EJ. 1988b. Lipoprotein cholesterol concentrations in the plasma of human subjects as measured in the fed and fasted states. Clin Chem 34:2456-2459. Collins JC. 1992. Resources for getting started in modeling. J Nutr 122: 695-700. Connor WE. 1990. Dietary fiber - nostrum or critical nutrient? N Engl J Med 322:147-152. Connor WE, Witiak DT, Stone DB, Armstrong ML. 1969. Cholesterol balance and fecal neutral steroid and bile acid excretion in normal men fed dietary fat of different fatty acid composition. J Clin Invest 47:1517-1534. Daumerie CM, Woollett LA, Dietschy JM. 1992. Fatty acids regulate hepatic low density lipoprotein receptor activity through redistribution of intracellular cholesterol pools. Proc Natl Acad Sci USA 89:10797-10801. De Backer G, De Craene I, Rosseneu M, Vercaemst R, Kornitzer M. 1989. Relationship between serum cholesteryl ester composition, dietary habits and coronary risk factors in middle-ages men. Atherosclerosis 78:237-243. De Bruin TWA, Brouwer CB, Gimpel JA, Erkelens DW. 1991. Postprandial decrease in HDL cholesterol and HDL apo A-l in normal subjects in relation to triglyceride metabolism. Am J Physiol 260:E492-E498. Deck C, Radack K. 1989. Effects of modest doses of omega-3 fatty acids on lipids and lipoproteins in hypertriglyceridemic subjects. Arch Int Med 149:1857-1862. Dell RB, Mott GE, Jackson EM, Ramakrishnan R, Carey KD, McGill HC, Goodman DS. 1985. Whole body and tissue cholesterol turnover in the baboon. J Lipid Res 26:327-337. Dell RB, Ramakrishnan R, Palmer RH, Goodman DS. 1985. A convenient sixpoint blood sampling schedule for determining whole-body cholesterol kinetics in humans. J Lipid Res 26:575-582.  189  Dell RB, Sciacca R, Lieberman K, Case DB, Cannon PJ. 1973. A weighted least-squares technique for the analysis of kinetic data and its application to the study of renal xenon washout in dogs and man. Circ Res 32: 71-84. Demacker PNM, Reijnen IGM, Katan MB, Stuyt PMJ, Stalenhoef AFH. 1991. Increased removal of remnants of triglyceride-rich lipoproteins on a dietary rich in polyunsaturated fatty acids. Bur J Clin Invest 21:197-203. Demke D, Peters G, Linet O, Metzler C, Klott K. 1988. Effects of a fish oil concentrate in patients with hypercholesterolemia. Atherosclerosis 70:73-80. Denke MA, Grundy SM. 1991. Effects of fats high in stearic acid on lipid and lipoprotein concentrations in man. Am J Clin Nutr 54:10361040. Dietschy JM. 1984. Regulation of cholesterol metabolism in man and in other species. Klin Wochen 62:338-345. Dietschy JM, Spady DK. 1984. Measurement of rates of cholesterol synthesis using tritiated water. J Lipid Res 25:1469-1476. Dietschy JM, Spady DK, Turley SD. 1993. Role of liver in the maintenance of cholesterol and low density lipoprotein homeostasis in different animal species, including humans. J Lipid Res 34:16371659. DiStefano JJ III Landaw EM. 1984. Multiexponential, multicompartmental, and noncompartmental modeling. I. Methodological limitations and physiological interpretations. Am J Physiol 246: R651-R664. Dobiasova M, Stribrna J, Sparks DL, Pritchard PH, Frohlich JJ. 1991. Cholesterol esterification rates in very low density lipoproteinand low density lipoprotein-depleted plasma. Relation to high density lipoprotein subspecies, sex, hyperlipidemia, and coronary heart disease. Arteriosclerosis & Thrombosis 11:64-70. Dobiasova M, Stribrna J, Pritchard PH, Frohlich JJ. 1992. Cholesterol esterification rate in plasma depleted of very low and low density lipoproteins is controlled by the proportion of HDL2 and HDL3 subclasses: study in hypertensive and normal middle-aged and septuagenarian men. J Lipid Res 33:1411-1418. Dreon DM, Vranizan KM, Krauss RM, Austin MA, Wood PD. 1990. The effects of polyunsaturated fat vs. monounsaturated fat on plasma lipoproteins. JAMA 263:2462-2466.  190  Dullaart RPF, Groener JEM, van Wijk H, Sluiter WJ, Erkelens DW. 1989. Alimentary lipemia-induced redistribution of cholesteryl ester between lipoproteins. Studies in normolipidemic, combined hyperlipidemic, and hypercholesteroleimc men. Arteriosclerosis 9:614-622. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. 1993. Summary of the second report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel II). JAMA 269:3015-3023. Faix D, Neese R, Kletke C, Wolden S, Cesar D, Coutlangus M, Shackleton CHL, Hellerstein MK. 1993. Quantification of menstrual and diurnal periodicities in rates of cholesterol and fat synthesis in humans. J Lipid Res 34:2063-2075. Ferezou J, Rautureau J, Coste T, Gouffier E, Chevallier F. 1982. Cholesterol turnover in human plasma lipoproteins: studies with stable and radioactive isotopes. Am J Clin Nutr 36:235-244. Fernandez ML, Lin ECK, McNamara DJ. 1992. Differential effects of saturated fatty acids on low density lipoprotein metabolism in the guinea pig. J Lipid Res 33:1833-1842. Fernandez ML, McNamara DJ. 1989. Dietary fat-mediated changes in hepatic apoprotein B/E receptor in the guinea pig: effect of polyunsaturated, monounsaturated, and saturated fat. Metabolism 38:1094-1102. Fernandez ML, McNamara DJ. 1991. Regulation of cholesterol and lipoprotein metabolism in guinea pigs mediated by dietary fat quality and quantity. J Nutr 121:934-943. Fernandez ML, McNamara DJ. 1994. Dietary fat saturation and chain length modulate guinea pig hepatic cholesterol metabolism. J Nutr 124:331-339. Fernandez ML, Yount NY, McNamara DJ. 1990. Whole body and hepatic cholesterol synthesis rates in the guinea pig: effect of dietary fat quality. Biochim Biophvs Acta 1044:340-348. Fielding PE, Jackson EM, Fielding CJ. 1989. Chronic dietary fat and cholesterol inhibit the normal postprandial stimulation of plasma cholesterol metabolism. J Lipid Res 30:1211-1217.  191  Folch J, Lees M, Sloane SGH. 1957. A simple method for the extraction and purification of total lipids from animal tissues. J Biol Chem 226:497-509. Foster DM, Barrett PHR, Toffolo G, Beltz WF, Cobelli C. 1993. Estimating the fractional synthetic rate of plasma apolipoproteins and lipids from stable isotope data. J Lipid Res 34:2193-2205. Foster DM, Boston RC. 1983. The use of compartmental analysis: the SAAM and CONSAM programs. In: Robertson J (eds) Compartmental Distribution of Radiotracers, Chemical Rubber Company, pp. 73-142. Foster DM, Boston RC, Jacquez JA, Zech LA. 1989. The SAAM tutorials: an introduction to using conversational SAAM version 30. Resource Facility for Kinetic Analysis, Seattle, Washington, USA. Fox J, McGill HC, Carey KD. 1986. In vivo regulation of hepatic LDL receptor mRNA in the baboon. Differential effects of saturated and unsaturated fat. J Biol Chem 262:7014-7020. Garfinkel D, Fegley KA. 1984. Fitting physiological models to data. Am J Physiol 246: R641-R650. Gerson T, Shorland FB, Dunckley GG. 1964. The effects of S-sitosterol on the metabolism of cholesterol and lipids in rats on a diet low in fat. Biochem J 92:385-390. Ginsberg HN, Barr SL, Gilbert A, Karmally W, Deckelbaum R, Kaplan K, Ramakrishnan R, Holleran S, Dell RB. 1990. Reduction of plasma cholesterol levels in normal men on an American Heart Association Step 1 dietary or a Step 1 dietary with added monounsaturated fat. N Engl J Med 322:574-579. Glatz JFC, Katan MB. 1993. Dietary saturated fatty acids increase cholesterol synthesis and fecal steroid excretion in healthy men and women. Eur J Clin Invest 23:648-655. Glomset JA. 1968. The plasma lecithin:cholesterol acyltransferase reaction. J Lipid Res 9:155-167. Glomset JA. 1970. Physiological role of lecithin:cholesterol acyltransferase. Am J Clin Nutr 23:1129-1136. Glueck CJ, Gordon DJ, Nelson JJ, Davis CE, Tyroler other correlates of changes in total and low cholesterol in hypercholesterolemic men: the Coronary Primary Prevention trial. Am J Clin  192  HA. 1986. Dietary and density lipoprotein Lipid Research Nutr 44:489-500.  Goodman, DS. 1965. Cholesterol ester metabolism. Physiol Rev 45:747-839. Goodman DS, Noble RP. 1968. Turnover of plasma cholesterol in man. J Clin Invest 47:231-241. Goodman DS, Noble RP, Dell RB. 1973. Three-pool model of the long-term turnover of plasma cholesterol in man. J Lipid Res 14:178-188. Goodman DS, Smith FR, Seplowitz AH, Ramakrishnan R, Dell RB. 1980. Prediction of the parameters of whole body cholesterol metabolism in humans. J Lipid Res 21:699-713. Green MH. 1992. Introduction to modeling. J Nutr 122: 690-694. Green MH, Green JB. 1990. The application of compartmental analysis to research in nutrition. Ann Rev Nutr 10: 41-61. Groot PHE, van Stiphout WAHJ, Krauss XH, Jansen H, van Tol A, van Ramshorst E, Chin-On S, Hofman A, Creswell SR, Havekes L. 1991. Postprandial lipoprotein metabolism in normolipidemic men with and without coronary artery disease. Arteriosclerosis & Thrombosis 11:653-662. Grundy SM. 1975. Effects of polyunsaturated fats on lipid metabolism in patients with hypertriglyceridemia. J Clin Invest 55:269-282. Grundy SM. 1987. Monounsaturated fatty acids, plasma cholesterol, and coronary heart disease. Am J Clin Nutr 474:1168-1175. Grundy SM. 1989. Monounsaturated fatty acids and cholesterol metabolism: implications for dietary recommendations. J Nutr 119:529-533. Grundy SM, Ahrens EH Jr. 1969. Measurements of cholesterol turnover, synthesis, and absorption in man, carried out by isotope, kinetic, and sterol balance methods. J Lipid Res 10:91-107. Grundy SM, Ahrens EH Jr. 1970. The effects of unsaturated dietary fats on absorption, excretion, synthesis, and distribution of cholesterol in man. J Clin Invest 49:1135-1152. Grundy SM, Florentin L, Nix D, Whelan MF. 1988. Comparison of monounsaturated fatty acids and carbohydrates for reducing raised levels of plasma cholesterol in man. Am J Clin Nutr 47:965-969. Grundy SM, Vega GL. 1988. Plasma cholesterol responsiveness to saturated fatty acids. Am J Clin Nutr 47:822-824.  193  Gurpide E, Mann J, Sandberg E. 1964. Determination of kinetic parameters in a two-pool system by administration of one or more tracers. Biochemistry 3:1250-1255. Gylling H, Miettinen TA. 1988. Serum noncholesterol sterols related to cholesterol metabolism in familial hypercholesterolemia. Clin Chim Acta 178:41-50. Gylling H, Miettinen TA. 1992. Non-cholesterol sterols, absorption and synthesis of cholesterol and apolipoprotein A-l kinetics in a Finnish lecithin-acyltransferase deficient family. Atherosclerosis 95:574-579. Hamilton JJ, Synnes A, Innis SM. 1992. Plasma cholesterol and lathosterol levels in term infants in the early neonatal period. Pediatr Res 31:396-400. Harris WS, Connor WE. 1980. The effects of salmon oil upon plasma lipids, lipoproteins, and triglyceride clearance. Trans Assoc Am Physiol 91:148-155. Harris WS, Connor WE, McMurry MP. 1983. The comparative reductions of the plasma lipids and lipoproteins by dietary polyunsaturated fats; salmon oil versus vegetable oils. Metabolism 32:179-184. Health and Welfare Canada. 1988. Canadian Nutrient File. Ottawa: Canadian Government Printing Centre. Hegsted DM, Ausman LM, Johnson JA, Dallal GE. 1993. Dietary fat and serum lipids: an evaluation of the experimental data. Am J Clin Nutr 57:875-883. Hegsted DM, McGandy RB, Myers ML, Stare FJ. 1965. Quantitative effects of dietary fat on serum cholesterol in man. Am J Clin Nutr 17:281295. Hellerstein MK, Neese RA. 1992. Mass isotopomer distribution analysis: a technique for measuring biosynthesis and turnover of polymers. Am J Physiol 263:E988-E1001. Hennessy LK, Osada J, Ordovas JM, Nicolosi RJ, Stucchi AF, Brousseau ME, Schaefer EJ. 1992. Effects of dietary fats and cholesterol on liver lipid content and hepatic apolipoprotein A-I, B, and E and LDL-receptor mRNA levels in cebus monkeys. J Lipid Res 33:351-360. Illingworth DR, Harris WS, Connor WE. 1984. Inhibition of low density lipoprotein synthesis by dietary n-3 fatty acids in humans. Arteriosclerosis 4:270-275.  194  Jackson RL, Gotto AM. 1974. Phospholipids in biology and medicine. N Engl J Med 290:87-93. Jackson RL, Taunton OD, Morrisett JD, Gotto AM. 1978. The role of dietary polyunsaturated fat in lowering blood cholesterol in man. Circ Res 42:447-453. Jeske DJ, Dietschy JM. 1980. Regulation of rates of cholesterol synthesis in vivo in the liver and carcass of the rat measured using [3H] water. J Lipid Res 21:364-376. Jones PJH, Leitch CA, Li ZC, Connor WE. 1993a. Human cholesterol synthesis measurement using deuterated water. Theoretical and procedural considerations. Arteriosclerosis & Thrombosis 13:247253. Jones PJH, Leitch CA, Pederson RA. 1993b. Meal-frequency effects on plasma hormone concentrations and cholesterol synthesis in humans. Am J Clin Nutr 57:868-874. Jones PJH, Lichtenstein AH, Leitch CA, Schaefer EJ. 1991. Differences in cholesterol kinetics in humans fed corn oil versus olive oil. Circ 84SII:218 (abst). Jones PJH, Lichtenstein AH, Schaefer EJ. 1994a. Interaction of dietary fat saturation and cholesterol level on cholesterol synthesis measured using deuterium incorporation. J Lipid Res 35:1093-1101. Jones PJH, Lichtenstein AH, Schaefer EJ, Namchuk GL. 1994b. Effect of dietary fat selection on plasma cholesterol synthesis in older, moderately hypercholesterolemia humans. Arteriosclerosis & Thrombosis 14:542-548. Jones PJH, Pappu AS, Illingworth DR, Leitch CA. 1992. Correspondence between plasma mevalonic acid levels and deuterium uptake in measuring human cholesterol synthesis. Eur J Clin Invest 22:609613. Jones PJH, Ridgen JE, Benson AP. 1990. Influence of dietary fatty acid composition on cholesterol synthesis and esterification in hamsters. Lipids 25:815-820. Jones PJH, Scanu AM, Schoeller DA. 1988. Plasma cholesterol synthesis using deuterated water in humans: effect of short-term food restriction. J Lab Clin Med 111:627-633. Jones PJH, Schoeller DA. 1990. Evidence for diurnal periodicity in human cholesterol synthesis. J Lipid Res 31:667-673.  195  Kannel WB, Castelli WP, Gordon T. 1979. Cholesterol in the prediction of atherosclerotic disease. New perspectives based on the Framingham study. Ann Int Med 90:85-91. Katan MB, Berns MAM, Glatz JFC, Knuiman JT, Nobels A, deVries JHM. 1988a. Congruence of individual responsiveness to dietarycholesterol and to saturated fat in humans. J Lipid Res 29:883892. Katan MB, Van Gastel AC, De Rover CM, Van Montfort MAJ, Knuiman JT. 1988b. Differences in individual responsiveness of serum cholesterol to fat-modified diets in man. Bur J Clin Invest 18:644-647. Kawaguchi E. 1990. The CONSAM's User's Manual. Resource Facility for Kinetic Analysis, Seattle, Washington, USA. Kelley MJ, Story JA. 1985. Effect of dietary and feeding state on modulation of hepatic HMG-CoA reductase in rats. Lipids 20:53-55. Kempen HJM, Glatz JFC, Gevers Leuven JA, van der Voort HA, Katan MB. 1988. Serum lathosterol concentration is an indicator of wholebody cholesterol synthesis in humans. J Lipid Res 29:1149-1155. Kempen HJM, Soeterik F, de Lange J. 1986. Secretion of lipoprotein lipid and synthesis of fatty acids, cholesterol and triacylglycerol by hepatocytes of fasted-refed rats. Biochim Biophys Acta 876:494499. Keys A, Anderson JT, Grande F. 1957. Prediction of serum cholesterol responses of man to changes in fats in the dietary. Lancet 2:959966. Kinsell LW, Michaels GD. 1955. Letter to the editor. Am J Clin Nutr 3:247-253. Knuiman JT, West CE, Katan MB, Hautvast JGAJ. 1987. Total cholesterol and high density lipoprotein cholesterol in populations differing in fat and carbohydrate intake. Arteriosclerosis 7:612-619. Kris-Etherton PM, Derr J, Mitchell DC, Mustad VA, Russell ME, McDonnell ET, Salabsky D, Pearson TA. 1993. The role of fatty acid saturation on plasma lipids, lipoproteins, and apolipoproteins: I. Effects of whole food diets high in cocoa butter, olive oil, soybean oil, dairy butter, and milk chocolate on the plasma lipids of young men. Metabolism 42:121-129.  196  Kuksis A, Myher JJ, Geher.K, Jones GJL, Shepherd J, Packard CJ, Morrisett JD, Taunton OD, Gotto AM. 1982. Effect of saturated and unsaturated fat diets on lipid profiles of plasma lipoproteins. Atherosclerosis 41:221-240. Kushner RF, Schoeller DA. 1986. Estimation of total body water by bioelectrical impedance analysis. Am J Clin Nutr 44:417-424. Kushwaha RS, McMahan CA, Mott GE, Carey KD, Reardon CA, Getz GS, McGill HC Jr. 1991. Influence of dietary lipids on hepatic mRNA levels of proteins regulating plasma lipoproteins in baboons with high and low levels of large high density lipoproteins. J Lipid Res 32:1929-1940. Kwano M, Milda T, Fielding CJ, Fielding PE. 1993. Quantitation of prebeta-HDL-dependent and non specific components of the total efflux of cellular cholesterol and phospholipid. Biochem 32:5025-5028. Landaw EM, DiStefano JJ III. 1984. Multiexponential, multicompartmental, and noncompartmental modeling. II. Data analysis and statistical considerations. Am J Physiol 246: R665-R677. Laraki L, Pelletier X, Mourot J, Debry G. 1993. Effects of dietary phytosterols on liver lipids and lipid metabolism enzymes. Ann Nutr Metab 37:129-133. Lichtenstein AH, Ausman LM, Carrasco W, Jenner JL, Gualtieri LJ, Goldin BR, Ordovas JM, Schaefer EJ. 1993. Effects of canola, corn, and olive oils on fasting and postprandial lipoproteins in humans as part of a National Cholesterol Education Program Step 2 dietary. Arteriosclerosis & Thrombosis 13:1533-1542. Linscheer WG, Vergroesen AJ. 1988. Lipids. In: Shils ME, Young VR (eds) Modern Nutrition in Health and Disease, 7th ed., Lea & Febiger, Philadelphia, PA, pp. 72-107. Lipid Research Clinic Program Manual of Laboratory Operations, Vol 1. Bethesda MD: Clinics Program. National Institutes of Health, 1974 (DHEW publications (NIH) 75-628) . Liu GC, Ahrens KEH, Schreibman PH Jr, Crouse JR. 1976. Measurement of squalene in human tissues and plasma: validation and application. J Lipid Res 17:38-45. Liv GCK, Ahrens EH Jr, Schreigman PH, Samuel P, McNamara DJ, Crouse JR. 1975. Measurement of cholesterol synthesis in man by isotope kinetics of squalene. Proc Natl Acad Sci USA 72:4612-4616.  197  London IM, Schwarz H. 1953. Erythrocyte metabolism. The metabolic behaviour of the cholesterol of human erythrocytes. J Clin Invest 32:1248-1252. Lorenz R, Spengler U, Fischer S, Drum J, Weber PC. 1983. Platelet function, thromboxane formation, and blood pressure control during supplementation of the Western dietary with cod liver oil. Circ 67:504-511. Mahley RW, Innerarity TL. 1983. Lipoprotein receptors and cholesterol homeostasis. Biochim Biophys Acta 737:197-222. Marcel YL, Vezina C. 1973. Lecithin:acyltransferase of human plasma. Role of chylomicrons, very low, and high density lipoproteins in the reaction. J Biol Chem 248:8254-8259. Marquardt DW. 1963. An algorithm for least-squares estimation of nonlinear parameters. J Soc Indust Appl Math 11:431-441. Mata P, Alvarez-Sala LA, Rubio MJ, Nuno J, De Oya M. 1992. Effects of long term monounsaturated vs. polyunsaturated-enriched diets on lipoproteins in healthy men and women. Am J Clin Nutr 55:846-850. Mattson FH, Grundy SM. 1985. Comparison of effects of dietary saturated, monounsaturated, and polyunsaturated fatty acids on plasma lipids and lipoproteins in man. J Lipid Res 26:194-202. Mayes PA. 1988a. Cholesterol synthesis, transport, and excretion. In: Murray RK, Granner DK, Mayes PA, Rodwell VW (eds) Harper's Biochemistry, 21st ed., Appleton & Lange, Norwalk, CT, pp. 226240. Mayes PA. 1988b. Regulation of cholesterol metabolism. In: Murray RK, Granner DK, Mayes PA, Rodwell VW (eds) Harper's Biochemistry, 21st ed., Appleton & Lange, Norwalk, CT, pp. 186-197. McDonald BE, Gerrard JM, Bruce VM, Corner EJ. 1989. Comparison of the effect of canola oil and sunflower oil on plasma lipids and lipoproteins and on in vivo thromboxane A2 and prostacyclin production in healthy young men. Am J Clin Nutr 50:1382-1388. McMurry MP, Cerqueira MT, Connor SL, Connor WE. 1991. Changes in lipid and lipoprotein levels and body weight in Tarahumara Indians after consumption of an affluent dietary. N Engl J Med 325:1704-1708.  198  McNamara DJ, Kolb R, Parker TS, Batwin H, Samuel P, Brown CD, Ahrens EH Jr. 1987. Heterogeneity of cholesterol homeostasis in man: responses to changes in dietary fat quality and cholesterol quantity. J Clin Invest 79:1729-1739. Melchert HU, Limsathayourat N, Mihajlovic H, Eichberg J, Thefeld W, Rottka H. 1987, Fatty acid patterns in triglycerides, diglycerides, free fatty acids, cholesteryl esters and phosphatidylcholine in serum from vegetarians and non-vegetarians. Atherosclerosis 65:159-166. Mensink RP, Katan MB. 198 9. Effect of a dietary enriched with monounsaturated or polyunsaturated fatty acids on levels of lowdensity and high-density lipoprotein cholesterol in healthy women and men. N Engl J Med 321:436-441. Mensink RP, Katan MB. 1992. Effect of dietary fatty acids on serum lipids and lipoproteins. A meta-analysis of 27 trials. Arteriosclerosis & Thrombosis 12:911-919. Mercer NJH, Holub BJ. 1981. Measurement of hepatic sterol synthesis in the Mongolian gerbil in vivo using [3H]water: diurnal variation and effect of type of dietary fat. J Lipid Res 22:792-799. Miettinen TA. 196 9. Serum squalene and methyl sterols as indicators of cholesterol synthesis in vivo. Life Sci 8:713-721. Miettinen TA. 1976. Cholesterol synthesis in slowly exchangeable cholesterol pool in man. In: Schettler G, Weizel A. (eds) Atherosclerosis V, Springer-Verlag, Berlin, pp. 305-308. Miettinen TA, Vanhanen H. 1994. Serum concentration and metabolism of cholesterol during rapeseed oil and squalene feeding. Am J Clin Nutr 59:356-363. Mifflin MD, St Jeor ST, Hill LA, Scott BJ, Daugherty SA, Koh YO. 1990. A new predictive equation for resting energy expenditure in healthy individuals. Am J Clin Nutr 51:241-247. Milda T, Fielding CJ, Fielding PE. 1990. Mechanism of transfer of LDLderived free cholesterol to HDL subfractions in human plasma. Biochem 29:10469-10474. Miller NE. 1987. On the association of body cholesterol pool size with age, HDL cholesterol and plasma total cholesterol concentration in humans. Atherosclerosis 67:163-172.  199  Miller NE, Nestel PJ, Clifton-Bligh P. 1976. Relationships between plasma lipoprotein cholesterol concentrations and the pool size and metabolism of cholesterol in man. Atherosclerosis 23:535-547. Moore RB, Anderson JT, Taylor HL, Keys A, Frantz ID Jr. 1968. Effects of dietary fat on the fecal excretion of cholesterol and its degradation products in man. J Clin Invest 47:1517-1534. Morrisett JD, Pownall HJ, Jackson RL, Segura R, Gotto AM, Taunton OD. 1977. Effects of polyunsaturated and saturated fat diets on the chemical composition and thermotropic properties of human plasma lipoproteins. In-. Holman RT and Kunau WH eds., Polyunsaturated Fatty Acids, American Oil Chemists' Society, Champaign, 111: pp.139-161. Mott GE, Jackson EM, Prihoda TJ, McMahan CA. 1987. Effects of dietary cholesterol and fat, sex, and sire on lecithin:cholesterol acyltransferase activity in baboons. Biochim Biophys Acta 919:190198. Myant NB. 1981. Cholesterol synthesis in animal tissues. Jn: The Biology of Cholesterol and Related Steroids, 1st ed., London: William Heinemann Medical Books Ltd.: pp.341-395. Myant NB. 1981. The plasma cholesterol: composition and metabolism. In-. The Biology of Cholesterol and Related Steroids, 1st ed., London: William Heinemann Medical Books Ltd.: pp. 505-567. Myant NB, Balasubramaniam S, Moutafis CD, Mancini M, Slack J. 1973. Turnover of cholesteryl esters in plasma low-density and highdensity lipoproteins in familial hyperbetalipoproteinaemia. Clin Sci Mol Med 45:551-560. Neese RA, Faix D, Kletke C, Wu K, Wand AC, Shackleton CHL, Hellerstein MA. 1993. Measurement of endogenous synthesis of plasma cholesterol in rats and humans using MIDA. Am J Physiol 264:E136E147. Nestel PJ. 1986. Fish oil attenuates the cholesterol induced rise in lipoprotein cholesterol. Am J Clin Nutr 43:752-757. Nestel PJ. 1989. Current strategies for atherosclerosis and lowering cholesterol. Clin & Exp Hyper - Ther & Prac All(5&6):915-925. Nestel PJ, Havenstein N, Homma Y, Scott TW, Cook LJ. 1975. Increased sterol excretion with polyunsaturated-fat high-cholesterol diets. Metabolism 24:189-198.  200  Nestel PJ, Havenstein N, Scott TW, Cook LJ. 1974. Polyunsaturated ruminant fats and cholesterol metabolism in man. Aust N Zea J Med 4:497-501. Nestel PJ, Havenstein N, Whyte HM, Scott TW, Cook LJ. 1973. Lowering of plasma cholesterol and enhanced sterol excretion with the consumption of polyunsaturated ruminant fats. N Engl J Med 288:379-382. Nestel PJ, Monger EA. 1967. Turnover of plasma esterified cholesterol in normocholesterolemic and hypercholesterolemic subjects and its relation to body build. J Clin Invest 46:967-974. Nestel PJ, Whyte HM, Goodman DS. 1969. Distribution and turnover of cholesterol in humans. J Clin Invest 48:982-991. Ng TKW, Hayes KC, DeWitt GF, Jegathesan M, Satgunasingam N, Ong ASH, Tan D. 1992. Dietary palmitic and oleic acids exert similar effects on serum cholesterol and lipoprotein profiles in normocholesterolemic men and women. J Am Coll Nutr 11:383-390. Nicolosi RJ, Stucchi AF, Kowala MC, Hennessy LK, Hegsted DM, Schaefer EJ. 1990. Effect of dietary fat saturation and cholesterol on LDL composition and metabolism. In vivo studies of receptor and nonreceptor-mediated catabolism of LDL in cebus monkeys. Arteriosclerosis 10:119-128. Nikkila EA, Taskinen Mr, Sane T. 1987. Plasma high-density lipoprotein concentration and subfraction distribution in relation to triglyceride metabolism. Am Heart J 113:543-548. Nydahl MC, Gustafsson I-B, Vessby B. 1994. Lipid-loering diets enriched with monounsaturated or polyunsaturated fatty acids but low in saturated fatty acids have similar effects on serum lipid concentrations in hyperlipidemic patients. Am J Clin Nutr 59:115122. O'Dea K, Traianedes K, Chisholm K, Leyden H, Sinclair AJ. 1990. Cholesterol-lowering effect of a low-fat dietary containing lean beef is reversed by the addition of beef fat. Am J Clin Nutr 52:491-494. Oh SY, Monaco PA. 1985. Effect of dietary cholesterol and degree of fat unsaturation on plasma lipid levels, lipoprotein composition, and fecal steroid excretion in normal young adult men. Am J Clin Nutr 42:399-413.  201  Ostlund RE Jr. 1993. A minimal model for human whole body cholesterol metabolism. Am J Physiol 265:E513-E520. Pal S, Davis PJ. 1991. Effects of different types of polyunsaturated fatty acids on cholesterol esterification in human fibroblasts. Biochem Int 25:281-288. Palmer RH, Nichols AV, Dell RB, Ramakrishnan R, Lindgren FT, Ging EL, Blum CB, Goodman DS. 1986. Lack of a relationship in humans of the parameters of body cholesterol metabolism with plasma levels of subfractions of HDL and LDL, or with apoE isoform phenotype. J Lipid Res 27:637-644. Parhofer KG, Barrett PHR, Bier DM, Schonfeld G. 1991. Determination of kinetic parameters of apolipoprotein B metabolism using amino acids labeled with stable isotopes. J Lipid Res 32:1311-1325. Parker TS, McNamara DJ, Brown CD, Kolb R, Ahrens EH Jr, Alberts AW, Tobert J, Chen J, De Schepper PJ. 1984. Plasma mevalonate as a measure of cholesterol synthesis in man. J Clin Invest 74:795-804. Patsch JR, Prasad S, Gotto AM, Patsch W. 1987. High density lipoprotein. 2. relationship of the plasma levels of this lipoprotein species to its composition, to the magnitude of postprandial lipemia, and to the activities of lipoprotein lipase and hepatic lipase. J Clin Invest 80:341-347. Pattnaik NM, Montes A, Hughes LB, Zilversmit DB. 1978. Cholesteryl ester exchange protein in human plasma: isolation and characterization. Biochim Biophvs Acta 530:428-438. Perl W, Samuel P. 1969. Input-output analysis for total input rate total transfer mass of body cholesterol in man. Circ Res 25:191-199. Philipson BE, Harris WS, Connor WE. 1981. Reductions of plasma lipids and lipoproteins in hyperlipidemic patients by dietary w-3 fatty acids. Clin Res 29:628A. Philipson BE, Rothrock DW, Connor WE, Harris WS, Illingworth DR. 1985. Reductions of plasma lipids, lipoproteins, and apoproteins by dietary fish oils in patients with hypertriglyceridemia. N Engl J Med 312:1210-1219. Pike RL, Brown ML. 1975. Body composition. In: Nutrition: an integrated approach, 2nd ed., New York: John Wiley and Sons Ltd.: pp.775-813. Pritchard PH, Frohlich JJ. 1994. The clinical significance of lipoprotein heterogeneity. Can J Cardiol 10 Suppl B:38-43.  202  Ramberg CF Jr, Krishnamurti CR, Peter D, Wolff JE, Boston RC. 1992. Application of models to determination of nutrient requirements: experimental techniques employing tracers. J Nutr 122: 701-705. Reiser R, Probstfield JL, Silvers A, Scott LW, Shorney ML, Wood RD, O'Brien BC, Gotto AM, Insull W Jr. 1985. Plasma lipid and lipoprotein responses of humans to beef fat, coconut oil and safflower oil. Am J Clin Nutr 42:190-197. Reynolds GA. 1989. Rational theory of familial hypercholesterolemia. Circ 79:1146-1148. Rifai N, Merrill JR, Holly RS. 1990. Postprandial effect of a high fat meal on plasma lipid, lipoprotein cholesterol and apolipoprotein measurements. Ann Clin Biochem 27:489-493. Rittenberg D, Schoenheimer R. 1937a. Deuterium as an indicator in the study of intermediary metabolism. VIII. Hydrogenation of fatty acids in the animal organism. J Biol Chem 117:485-490. Rittenberg D, Schoenheimer R. 1937b. Deuterium as an indicator in the study of intermediary metabolism. XI. Further studies on the biological uptake of deuterium into organic substances, with special reference to fat and cholesterol formation. J Biol Chem 121:235-253. Rose HG, Juliano J. 1977. Regulation of plasma lecithin:cholesterol acyltransferase. II. Activation during alimentary lipemia. J Lab Clin Med 89:524-532. Samuel P, Holtzman CM, Meilman E, Perl W. 1968. Effect of neomycin on exchangeable pools of cholesterol in the steady state. J Clin Invest 47:1806-1818. Samuel P, Lieberman S. 1973. Improved estimation of body masses and turnover of cholesterol by computerized input-output analysis. J Lipid Res 14:189-196. Samuel P, Lieberman S. 1974. Simplified method for the long-term study of turnover and body masses of cholesterol by input-output analysis. Am J Clin Nutr 27:1214-1220. Samuel P, Lieberman S, Ahrens EH Jr. 1978. Comparison of cholesterol turnover by sterol balance and input-output analysis, and a shortened way to estimate total exchangeable mass of cholesterol by the combination of the two methods. J Lipid Res 19:94-102.  203  Samuel P, Perl W. 1970. Long-term decay of serum cholesterol radioactivity: body cholesterol metabolism in normals and in patients with hyperlipoproteinemia and atherosclerosis. J Clin Invest 49:346-357. Samuel P, Perl W, Hotzman CM, Rochman ND, Lieberman S. 1972. Long-term kinetics of serum and xanthoma cholesterol radioactivity in patients with hypercholesterolemia. J Clin Invest 51:266-278. Sanders TAB, Roshanai F. 1983. The influence of different types of w-3 polyunsaturated fatty acids on blood lipids and platelet function in healthy volunteers. Clin Sci 63:6491-6499. Schwartz CC. 1982. Cholesterol kinetics and modeling: introduction. In: Berman B, Grundy SM, Howard G eds, Lipoprotein Kinetics and Modeling, Academic Press, Inc.: pp. 309-312. Schwartz CC, Berman B, Vlahcevic ZR, Halloran LG, Gregory DH, Swell L. 1978. Multicompartmental analysis of cholesterol metabolism in man. Characterization of the hepatic bile acid and biliary cholesterol precursor sites. J Clin Invest 61:408-423. Schwartz CC, Berman M, Vlahcevic ZR, Swell L. 1982. Multicompartmental analysis of cholesterol metabolism in man. Quantitative kinetic evaluation of precursor sources and turnover of high density lipoprotein cholesterol esters. J Clin Invest 70:863-876. Schwartz CC, Zech LA, VanderBroek JM, Cooper PA. 1993. Cholesterol kinetics in subjects with bile fistula. Positive relationship between size of the bile acid precursor pool and bile acid synthesis rate. J Clin Invest 91:923-938. Seppanen-Laakso T, Vanhanen H, Laakso I, Kohtamaki H, Viikair J. 1991. Replacement of margarine on bread by rapeseed and olive oils; effects on plasma fatty acid composition and serum cholesterol. Ann Nutr Metab 37:161-174. Shepherd J, Packard CJ, Grundy SM, Yeshurun D, Gotto AM Jr, Taunton OD. 1980. Effects of saturated and polyunsaturated fat diets on the chemical composition and metabolism of low density lipoproteins in man. J Lipid Res 21:91-99. Shepherd J, Packard CJ, Patsch JR, Gotto AM Jr, Taunton OD. 1978. Effects of dietary polyunsaturated and saturated fat on the properties of high density lipoproteins and the metabolism of apolipoprotein A-l. J Clin Invest 58:1582-1592.  204  Singer P, Jaeger W, Wirth M. 1983. Lipid and blood pressure lowering effect of mackerel dietary in man. Atherosclerosis 49:99-108. Sirtori CR, Tremoli E, Gatti E, Montanari G, Sirtori M, Colli S, Gianfranceschi G, Maderna P, Dentone CZ, Testolin G, Galli C. 1986. Controlled evaluation of fat intake in the Mediterranean dietary: comparative activities of olive oil and corn oil on plasma lipids and platelets in high-risk patients. Am J Clin Nutr 44:635-642. Skinner SM, Clark RE, Baker N, Shipley RA. 1959. Complete solution of the three-compartment model in steady state after single injection of radioactive tracer. Am J Physiol 196:238-244. Smith FR, Dell RB, Noble RP, Goodman DS. 1976. Parameters of the threepool model of the turnover of plasma cholesterol in normal and hyperlipidemic humans. J Clin Invest 57:137-148. Sodhi HS. 1974. Current concepts of cholesterol metabolism and their relationship to lecithin:acyltransferase. Scan J Clin Lab Invest 33(S137):161-163. Sorci-Thomas M, Wilson MD, Johnson FL, Williams DL, Rudel LL. 1989. Studies on the expression of genes encoding apolipoproteins B100 and B48 and the low density lipoprotein receptor in nonhuman primates. Comparison of dietary fat and cholesterol. J Biol Chem 264:9039-9045. Spady DK, Dietschy JM. 1983. Sterol synthesis in vivo in 18 tissues of the squirrel monkey, guinea pig, rabbit, hamster, and rat. J Lipid Res 24:303-315. Spady DK, Dietschy JM. 1985. Dietary saturated triacylglycerols suppress hepatic low density lipoprotein receptor activity in the hamster. Proc Natl Acad Sci USA 82:4526-4530. Spady DK, Dietschy JM. 1988. Interaction of dietary cholesterol and triglycerides in the regulation of hepatic low density lipoprotein transport in the hamster. J Clin Invest 81:300-309. Spady DK, Woollett LA, Dietschy JM. 1993. Regulation of plasma LDLcholesterol levels by dietary cholesterol and fatty acids. Ann Rev Nutr 13:355-382. Sparks D, Frohlich JJ, Pritchard PH. 1991. Lipid transfer proteins, hypertriglyceridemia, and reduced highOdensity lipoprotein cholesterol. Am Heart J 122:601-607.  205  Spritz N, Ahrens EH Jr, Grundy SM. 1965. Sterol balance in man as plasma cholesterol concentrations are altered by exchanges of dietary fat. J Clin Invest 44:1482-1493. Spritz N, Mishkel MA. 1969. Effects of dietary fats on plasma lipids and lipoproteins: an hypothesis for the lipid-lowering effect of unsaturated fatty acids. J Clin Invest 48:78-86. Stahler F, Gruber W, Stinshoff K, Roschlau P. 1977. A reliable enzymatic assay for cholesterol determination. Med Lab 30:29-37. Strandberg TE, Tilvis RS, Miettinen TA. 1989. Variations of hepatic cholesterol precursors during altered flows of endogenous and exogenous squalene in the rat. Biochim Biophys Acta 1001:150-156. Strandberg TE, Tilvis RS, Miettinen TA. 1990. Metabolic variables of cholesterol during squalene feeding in humans: comparison with cholestyramine treatment. J Lipid Res 31:1637-1643. Subbiah RMT. 1971. Significance of dietary plant sterols in man and experimental animal. Mavo Clin Proc 46:549-559. Swain JF, Rouse IL, Curley CB, Sacks FM. 1990. Comparison of the effects of oat bran and low-fiber wheat on serum lipoprotein levels and blood pressure. N Engl J Med 322:147-152. Taylor CB, Mikkelson B, Anderson JA, Forman DT. 1966a. Human serum cholesterol synthesis measured with the deuterium label. Arch Path 81:213-231. Taylor CB, Mikkelson B, Anderson JA, Forman DT, Franz G. 1966b. Human serum cholesterol synthesis. Minn Med 49:728-733. Toffolo G, Foster DM, Cobelli C. 1993. Estimation of protein synthetic rate from tracer data. Am J Phvsiol 264: E128-E135. Trinder P. 1969. Determination of glucose in blood using glucose oxidase with an alternative oxygen acceptor. Ann Clin Biochem 6:24-27. Triscari J, Hamilton JG, Sullivan AC. 1978. Comparative effects of saturated and unsaturated lipids on hepatic lipogenesis and cholesterogenesis in vivo in the meal-fed rat. J Nutr 108:815-825. Turley SD, Anderson JM, Dietschy JM. 1981. Rates of sterol synthesis and uptake in the major organs of the rat in vivo. J Lipid Res 22:551569.  206  Turley SD, Dietschy JM. 1982. Cholesterol metabolism and excretion. Arias I, Popper H, Schacter D, Schafritz DA, eds. The Liver, Biology and Pathology. New York; Raven Press: pp.163-172.  In:  Turley SD, West CE. 1976. Effect of cholesterol and cholestyramine on sterol synthesis in the liver, ileum, and lung of the guinea pig. Lipids 11:571-577. Turner JD, Monell R, Brown VW. 1976. Heterogenous responses to polyunsaturated fat in low density lipoprotein turnover in hyperlipoproteinemia. Circ 54:4. Uchida K, Takase H, Nomura Y, Takeda K, Takeuchi N, Ishikawa Y. 1984. Changes in biliary lipids and fecal bile acids in mice after treatments with diosgenin and Ssitosterol. J Lipid Res 25:236-245. Valsta LM, Jauhiainen M, Aro A, Katan MB, Mutanen M. 1992. Effects of a monounsaturated rapeseed oil and a polyunsaturated sunflower oil dietary on lipoprotein levels in humans. Arteriosclerosis & Thrombosis 12:50-57. Vega GL, Groszek E, Wolf E, Grundy SM. 1982. Influence of polyunsaturated fats on composition of plasma lipoproteins and apolipoproteins. J Lipid Res 23:811-822. von Lossonczy TO, Ruiter A, Brongeest-Schoute HC, van Gent CM, Hermus RJJ. 1978. The effect of a fish dietary on serum lipids in healthy human subjects. Am J Clin Nutr 31:1340-1346. Wahrburg U, Martin H, Sandkamp M, Schulte H, Assmann G. 1992. Comparative effects of a recommended lipid-lowering dietary vs. a dietary rich in monounsaturated fatty acids on serum lipid profiles in healthy young adults. Am J Clin Nutr 56:678-683. Wardlaw GM, Snook JT. 1990. Effect of diets high in butter, corn oil, or high-oleic acid sunflower oil on serum lipids and apolipoproteins in men. Am J Clin Nutr 51:815-821. Wardlaw GM, Snook JT, Lin MC, Piangco MA, Kwon JS. 1991. Serum lipid and apolipoprotein concentrations in healthy men on diets enriched in either canola oil or safflower oil. Am J Clin Nutr 54:104-110. Warnick GR, Nguyen T, Albers AA. 1985. Comparison of improved precipitation methods for quantification of high-density lipoprotein cholesterol. Clin Chem 31:217-222.  207  Weingand KW, Daggy BP. 1991. Effects of dietary cholesterol and fasting on hamster plasma lipoprotein lipids. Eur J Clin Chem Clin Biochem 29:425-428. Weintraub MS, Eisenberg S, Breslow JL. 1987. Different patterns of postprandial lipoprotein metabolism in normal, type Ila, type III and type IV hyperlipoproteinemic individuals. Effect of treatment with cholestyramine and gemfibrozil. J Clin Invest 79:217-229. Wilson JD. 1970. The measurement of the exchangeable pools of cholesterol in the baboon. J Clin Invest 49:655-665. Wilt TJ, Lofgren RP, Nichol KL, Schorer AE, Crespin L, Bownes D, Eckfeldt J. 1989. Fish oil supplementation does not lower plasma cholesterol in men with hypercholesterolemia. Results of a randomized, placebo-controlled crossover study. Ann Int Med 111:900-905. Wolfe RR. 1992. Terminology for tracer studies. In: Radioactive and Stable Isotope Tracers in Biomedicine. Principles and Practice of Kinetic Analysis, Wiley-Liss, Inc., pp. 19-21. Wood PDS, Shioda R, Kinsell LW. 1966. Dietary regulation of cholesterol metabolism. Lancet 2:604-607. Wood R, Rubena K, O'Brien B, Tseng S, Martin G. 1993. Effect of butter, mono- and polyunsaturated fatty acid-enriched butter, trans fatty acid margarine, and zero trans fatty acid margarine on serum lipids and lipoproteins in healthy men. J Lipid Res 34:1-11. Woollett LA, Spady DK, Dietschy JM. 1992a. Saturated and unsaturated fatty acids independently regulate low density lipoprotein receptor activity and production rate. J Lipid Res 33:77-88. Woollett LA, Spady DK, Dietschy JM. 1992b. Regulatory effects of the saturated fatty acids 6:0 through 18:0 on hepatic low density lipoprotein receptor activity in the hamster. J Clin Invest 89:1133-1141. Zar JH. 1974. Biostatistical Analysis. Englewood Cliffs, NJ: PrenticeHall, Inc. Zhang S, Wong WW, Hachey DL, Pond WG, Klein PD. 1994. Dietary cholesterol inhibits whole-body but not cerebrum cholesterol synthesis in young pigs. J Nutr 124:717-725. Zilversmit DB. 1960. The design and analysis of isotope experiments. Am J Med 29:832-848.  208  Appendix 1 period  Sample meal plan: meals for Subject H during each diet  MONOUNSATURATED FAT DIET Breakfast  Quantity (q)  Bread, whole wheat Jam/Syrup Cereal, bran flakes Milk (2%) Orange juice Egg Olive oil Butter  45.5 21.5 71.5 264.0 108.0 125.5 25.0 11.0  Lunch 7 6 .. 5 7 6 .. 5 4 1 . .0 2 0 .. 5 1 7 .. 5 1 4 3 . .0 2 5 3 .. 5 179.. 0  Bread, whole wheat Chicken Lettuce Olive oil Mayonnaise Orange Apple juice Yogurt (3.25%) Supper 1 Chicken Spaghetti Tomato sauce Broccoli Olive oil Butter Ice cream Chocolate syrup  74.5 248.5 14 9.0 74.5 25.0 5.0 99.5 25.0  Supper 2 Turkey Cranberry sauce Potatoes Carrots Milk (2%) Olive oil Butter Canned peaches  79.0 53.0 211.5 79.0 322.0 28.5 10.5 264.0  209  POLYUNSATURATED FAT DIET Breakfast  Quantity (a)  Bread, whole wheat Jam/Syrup Cereal, bran flakes Milk (2%) Orange juice Egg Margarine, safflower  44.5 21.5 70.0 259.5 106.5 123.5 40.5  Lunch Bread, whole wheat Chicken Lettuce Margarine, safflower Mayonnaise Orange Apple juice Butter Yogurt (2%)  74.0 74.0 3 9.5 21.5 19.5 13 7.5 244.0 3.0 196.5  Supper 1 Chicken Spaghetti Tomato sauce Broccoli Margarine, safflower Ice milk Chocolate syrup  76.5 254.5 153.0 76.5 41.0 102.0 25.5  Supper 2 Turkey Cranberry sauce Potatoes Carrots Milk (2%) Margarine, safflower Butter Canned peaches  76.5 51.0 204.5 76.5 312.0 44.0 2.0 256.0  210  SATURATED FAT DIET Breakfast  Quantity (g)  Bread, whole wheat Jam/Syrup Cereal Milk (2%) Orange juice Egg Butter Margarine, safflower  45.0 21.5 70.5 260.0 106.5 123.5 3 0.0 10.5  Lunch Bread, whole wheat Chicken Lettuce Butter Mayonnaise Orange Apple juice Yogurt (3.25%)  79.5 79.5 42.5 26.5 12.7 148.0 262.5 185.5  Supper 1 Chicken Spaghetti Tomato sauce Broccoli Margarine, safflower Butter Ice cream Chocolate syrup  74.5 248.5 14 9.0 74.5 12.0 23.0 99.5 25.0  Supper 2 Turkey Cranberry sauce Potatoes Carrots Milk (2%) Margarine, safflower Butter Canned peaches  78.0 52.0 207.5 78.0 316.5 12.5 34.0 259.5  211  Appendix 2 Body water deuterium enrichment (°/oo) relative to SMOW at five time points on each of the three diets  Sub  Time (h)  A  0 24 48  74.97+ 0.83 4105.56+3.73 4206.47+ 5.88  71.01+ 0.13 3743.89+30.43 3820.02+14.33  -71.43+ 1.39 3407.92+28.47 3341.74+21.07  B  0 24 48  73.61+ 1.27 4014.21+29.83 4107.29+1.69  31.01+ 0.54 4250.64+8.28 4064.83+10.57  -58.76+ 3.71 3410.55+14.15 3341.28+27.57  C  0 24 48  14.59+ 1.49 3917.49+2.18 3786.42+36.43  28.16+ 1.27 3882.00+5.98 3818.58+26.03  -52.78+ 0.93 3421.73+11.24 3310.79+ 0.74  D  0 24 48  -33.51+ 1.45 3747.60+16.67 3369.94+11.87  -55.43+ 0.29 3576.47+2.35 3422.05+3.79  25.32+ 0.54 3559.02+0.71 3108.25+37.63  E  0 24 48  5.80+ 0.42 3631.01+6.60 3606.60+6.13  -60.87+ 1.45 2519.60+30.75 2308.80+5.74  -64.87+ 0.98 3683.24+1.88 3859.30+12.70  F  0 24 48  107.02+ 2.66 3804.28+ 9.53 3572.33+ 0.29  -53.15+ 1.62 1624.56+ 3.21 1536.11+ 1.05  28.25+ 0.75 3317.39+ 1.56 3223.69+ 4.14  G  0 24 48  -74.05+ 0.07 3371.12+7.31 2784.05+15.51  41.98+ 0.56 3801.21+20.76 3745.68+12.78  71.06+ 2.33 3585.69+12.14 3646.41+18.09  H  0 24 48  -44.13+ 2.42 3384.17+ 0.74 3324.22+16.85  56.99+ 1.51 3866.81+10.09 3825.82+8.89  18.66+ 0.95 3790.79+ 9.16 3675.17+26.50  I  0 24 48  -66.37+ 2.46 2162.60+8.08 3110.51+19.95  51.01+ 0.87 3503.25+20.51 3836.19+14.55  75.52+ 0.52 3799.30+9.67 3650.41+2.36  *  X +  MONO  POLY  SE.  212  SAT  Appendix 3 periods  Sub  G  Diet  MONO POLY SAT MONO POLY SAT MONO POLY SAT MONO POLY SAT MONO POLY SAT MONO POLY SAT MONO POLY SAT MONO POLY SAT MONO POLY SAT  Subject body weight at beginning and end of two-week diet  Init wt (kg)  Fin wt (kg)  Change  64.7 64.1 63.2 72.2 70.9 70.5 77.9 77.4 76.6 77.9 77.0 77.9 71.5 71.0 71.5 61.5 61.6 63.3 60.0 60.6 59.2 69.1 67.4 68.0 67.4 69.0 68.6  64.8 63.3 63.5 72.6 71.9 70.6 76.0 75.5 76.1 77.0 77.2 78.2 71.6 71.5 70.1 62.2 61.1 63. 57. 61. 60. 68. 66. 68.0 67.8 68.1 68.5  + 0.15 -1.25 + 0.47 + 0.55 + 1.41 + 0.14 -2.44 -2.46 -0.65 -1.16 + 0.26 + 0.39 + 0.14 + 0.70 -1.96 + 1.14 -0.81 + 0.47 -3.50 + 0.99 + 1.52 -1.45 -1.34 0.00 + 0.59 -1.30 -0.15  (%)  init wt: initial weight; fin wt: final weight. Change is calculated as: ((final weight minus initial weight) divided by (initial weight)), multiplied by 100.  213  Appendix 4  Daily energy intake (kJ) on each of the three diets  Sub  MONO  POLY  SAT  A B C D E F G H I  13180 13757 13799 14523 13033 12707 12058 12912 12719  12309 13753 14657 14523 14000 12707 12966 12912 12711  12297 12924 13778 14544 13941 13552 12966 12912 12719  214  Appendix 5 Individual serum lipid concentrations at 5 time points during two-day test periods on each of the diets.  Sub  Diet  Time (h)  Choi (mmol/L)  TG (mmol/L)  HDL. (mmol/L)  A  MONO  0 12 24 36 48  4.09 4.20 3.81 4.15 4.32  1.15 1.92 1.18 0.99 0.72  0.94 0.94 1.02 1.00 1.10  POLY  0 12 24 36 48  4.31 3.49 3.58 3.65 3.41  0.41 2.40 0.96 0.79 0.59  1.32 0.85 0.85 0.91 0.92  SAT  0 12 24 36 48  4.49 4.49 4.51 4.49 4.52  1.13 3.37 1.35 1.05 0.82  0.78 0.74 0.88 0.94 0.90  MONO  0 12 24 36 48  4.24 4.26 4.20 3.94 3.70  0.74 1.51 0.63 0.51 0.46  1.16 1.14 1.24 1.24 1.24  POLY  0 12 24 36 48  3.68 3.89 3.88 3.62 3.40  0.66 1.13 0.74 0.44 0.43  1.13 1.25 1.21 1.19 1.24  SAT  0 12 24 36 48  4 .42  0.87  0.92  4.58 4.58 4.21  0.87 0.62 0.50  0.96 1.02 1.04  0 12 24 36 48  4.04 3.73 4.19 3.92 4.19  0.62 1.06 0.57 0.29 0.33  1.10 1.02 1.10 1.12 1.26  MONO  215  Sub  Diet  Time (h)  Choi (mmol/L)  TG (mmol/L)  HDL (mmol/L)  POLY  0 12 24 36 48  2.30 2.03 3.53 3.70 3.85  0.47 1.08 0.50 0.32 0.28  1.02 1.00 1.03 1.14 1.13  SAT  0 12 24 36 48  4.47 4.22 4.29 4.24 4.34  0.64 1.02 0.52 0.32 0.29  0.76 0.90 0.84 0.96 1.04  MONO  0 12 24 36 48  3.34 3.36 3.36 3.40 3.61  0.52 0.77 0.39 0.36 0.37  1.26 0.95 0.96 0.90 1.02  POLY  0 12 24 36 48  2.99 3.11 3.11 3.24 3.30  0.57 1.00 0.57 0.38 0.31  0.76 0.80 0.82 0.84 0.86  SAT  0 12 24 36 48  3.53 3.80 3.60 3.58 3.67  0.62 1.66 0.60 0.47 0.42  0.84 0.84 0.86 0.86 0.92  MONO  0 12 24 36 48  4 .15 3.60 3.97 4.98 3.52  0.41 0.89 0.46 0.52 0.80  1.07 1.12 1.12 1.46 0.89  POLY  0 12 24 36 48  3.64 3.74 3.61 3.84 3.65  0.51 0.72 0.56 0.30 0.31  0.88 0.88 0.88 1.00 0.98  SAT  0 12 24  4.52 4.69 4.62  0.49 0.76 0.87  1.04 1.22 1.06  216  Sub  H  Diet  Time  Choi  (h)  (mmol/L)  TG (mmol/L)  HDL (mmol/L)  36 48  4.63 4.62  0.68 0.53  1.10 1.14  MONO  0 12 24 36 48  4.04 4.16 3.95 4.27 4.39  1.01 2.28 1.36 1.12 0.90  0.68 1.06 0.94 0.88 0.96  POLY  0 12 24 36 48  3.86 3.77  1.03 2.47  0.88 1.02  3.88 3.80  0.81 0.85  0.94 0.96  SAT  0 12 24 36 48  4.48 4.58 4.51 4.64 4.68  0.93 2.10 1.20 1.18 0.85  0.98 0.88 0.96 0.96 1.00  MONO  0 12 24 36 48  4 .21 4.30 4 .14 4.42 4.49  0.73 1.91 0.74 0.61 0.67  0.85 0.85 0.90 0.99 0.93  POLY  0 12 24 36 48  2.53 4.18 4.14 4.47 4.16  0.69 1.06 0.95 0.96 0.61  0.94 0.94 1.00 1.04 0.96  SAT  0 12 24 36 48  4.90 5.34 5.10 5.68 5.10  0.72 1.55 0.82 0.83 0.55  0.88 0.98 0.94 1.04 0.98  MONO  0 12 24 36 48  4.21 4 .18 4.29 4.19 3.95  0.84 2.68 1.03 0.74 0.46  0.84 0.80 0.97 0.95 0.99  217  Sub  Diet  Time (h)  Choi (mmol/L)  TG (mmol/L)  HDL (mmol/L)  POLY  0 12 24 36 48  3.69 3.67 3.65 3.76 3.64  0.72 1.46 0.90 0.61 0.43  1.00 0.90 0.98 1.00 1.16  SAT  0 12 24 36 48  5.05 5.07 5.25 5.00 5.12  1.08 2.00 1.31 0.92 0.64  0.82 0.84 0.92  MONO  0 12 24 36 48  3.80 4.04 3.96 4.23 4.12  0.71 1.09 0.73 0.57 0.58  1.18 . 1.16 1.22 1.30 1.34  POLY  0 12 24 36 48  3.68 3.62 3.40 3.25 2.43  0.96 1.11 0.66 0.50 0.45  1.22 1.16 1.18 1.18 1.40  SAT  0 12 24 36 48  4.41 4.36 4 .47 4.48 4.52  0.84 1.35 0.78 0.68 0.45  1.11 1.11 1 .20 1.20 1.22  218  1.00  Appendix 6 Serum lathosterol levels and ratios of lathosterol to cholesterol at 5 time points during two-day test periods on each diets  Sub  Diet  Time (h)  Lath (/xmol/L)  MONO  12 36  16.7 + 4.7 +  5.25 1.10  4.0 + 1.25 1.1 ± 0.36  POLY  12 36  6.3 4.2 +  1.51  1.8 1.1  Ratio Lath/Choi (/xmol/mmol)  ± 0.41  SAT  12 36  22.4 + 10.84 7.3 + 2.02  MONO  12 36  44.4 3.6 +  POLY  12 36  28.0 + 25.80 3.2 + 0.19  7.2 0.8  MONO  12 36  20.4 + 10.03 13.2 + 4.20  5.5 + 2.69 3.4 + 1.07  POLY  12 36  SAT  2.22  5.0 + 2.42 1.6 ± 0.45 10.4 0.9 + 0.56  ± 6.63 ± 0.05  ± ±  1.57 0.09  1.8 1.0  12 36  11.4 + 9.5 ±  0.11 2.55  2.7 + 0.03 2.2 ± 0.60  MONO  12 36  17.0 5.9  ± 13.90 ± 4.79  5.0 + 4 .14 1.7 ± 1 .41  POLY  12 36  2.7 1.9  ± ±  0.14 0.02  0.9 0.6  ± 0.04 ± 0.01  SAT  12 36  9.2 3.1  ± ±  1.38 0.82  2.4 0.9  ± 0.36 ± 0.23  MONO  12 36  43.1 + 11.37 9.5 + 0.20  12.0 1.9  ± 3 .16 ± 0.04  POLY  12 36  48.6 + 15.6 +  3.38 0.06  13.0 4.1  ± 0.90 ± 0.12  SAT  12 36  11.9 + 10.1 +  2.93 2.04  6.7 3.9  219  ± 0.42 ± 0.02  2.5 ± 0.63 2.2 + 0.44  Sub  Diet  n=2,  Time (h)  Lath  MONO  12 36  28.2 + 11.87 7.6 + 0.10  6.6 + 2.78 1.8 i 0.02  POLY  12 36  13.3 + 15.6  3.6 + 0.94 4.0  SAT  12 36  7.2 3.6 +  0.25  1.6 0.8  ± 0.06  MONO  12 36  10.8 + 3.8 +  1.81 0.21  2.4 0.9  ± 0.41 ± 0.05  POLY  12 36  11.7 + 8.1 +  1.08 1.63  2.6 1.9  ± 0.24 ± 0.39  SAT  12 36  19.5 + 14.8 +  5.38 7.50  3.7 2.6  ± 1.01 ± 1.32  MONO  12 36  15.1 + 4.2 +  6.46 0.09  3.6 + 1.54 1.0 ± 0.02  POLY  12 36  11.8 + 3.4 ±  8.31 0.22  3.2 0.9  SAT  12 36  9.8 ± 9.2 +  7.19 0.99  1.9 + 1.42 1.8 ± 0.20  MONO  12 36  16.5 7.4 + 1.20  4.1 1.7  POLY  12 36  2.7 + 0.07 7.0  0.7 2.2  SAT  12 36  7.9 + 2.75 4.2 + 1.55  1.8 +_ 0.63 0.9 + 0.35  X ±  Ratio Lath/Choi (jumol/mmol)  (/xmo:L/L)  SE.  220  3.56  ± 2.26 ± 0.06  ± 0.28 ± 0.02  Appendix 7 Deuterium enrichment (°/oo) relative to SMOW: validation trial conducted to determine serum sample size needed for mass spectrometry  Sample FC  Plasma  Volume (mL)  Enrichment  1  -134.15 -130.65 -187.19  2  -212.13 -222.13 -225.41  3  -234.86 -228.58 -237.13  4  -231.29 -229.87 -238.96  1  -128.70 -164.78 -133.61  2  -179.91 -171.83 -175.83  3  -166.21 -171.30 -170.40  4  -174.63 -165.66 -171.71  221  Appendix 8 Data used to generate C decay curves for each subject.  cholesterol specific activity-  Subject A (Calculated injected dose: 18.3 /iCi C14) Day  DPM/mL serum  1 4 6 11 17 24 31 41 47 72 74 82 96 107 124 145 152 164 180 194 215 245 257 278  5827 4158 3051 2141 1404 1094 985 1303 703 723 430 336 250 221 175 137 143 128 127 232 81 87 69 60  Serum chol (mg/dl)  DPM/mg chol  DPM/mg chol (corr)  161 154 161 171 163 166 168 202 202 190 190 171 154 146 154 182 185 182 173 154 189 180 195 178  3619 2700 1895 1252 861 659 586 645 348 381 226 196 162 151 114 75 77 70 73 151 43 48 35 34  4944 3689 2589 1710 1177 900 801 881s 475s 520 309 268 222p 207P 155 103 106m 96m 100 206 59 66 48 46  Subject B (Calculated injected dose: 18.6 jiCi C ) 1 4 6 13 19 26 33 40 51 61  4807 3279 2713 1505 1132 919 718 687 681 522  152 151 164 164 160 162 168 192 180 176  3163 2172 1654 918 708 567 427 358 378 297  4251 2919 2223 1233 951 762 574 481s 509s 399  222  Day  DPM/mL serum  69 76 83 90 95 106 125 139 151 163 181 195 216 237 258 279  409 339 294 306 288 223 165 145 110 135 105 96 109 87 61 25  Serum chol (mg/dl) 166 157 163 168 176 171 169 198 181 194 151 171 199 203 186 181  DPM/mg chol  246 216 180 182 164 130 98 73 61 70 70 56 55 43 33 14  DPM/mg chol (corr) 331 290 242 245 220P 175? 131 98 82m 94m 93 75 74 58 44 19  Subject C (Calculated injected dose: 14.1 /zCi C 14 ) 1 3 6 14 21 25 28 35 41 48 55 62 69 80 90 104 125 137 153 167 181 202 224 244  3303 2828 1614 882 738 662 678 550 390 278 302 222 231 200 180 148 278 135 117 86 83 92 53 69  177 173 171 204 187 178 164 164 157 161 138 146 169 159 166 178 185 175 141 148 150 169 165 162  1866 1635 944 432 395 372 413 335 248 173 219 152 137 126 108 83 150 77 83 58 55 54 32 43  3309 2898 1674 767s 700s 659 733 595 440 306 388 270 242P 223P 192 147 266m 137m 147 103 98 97 57 76  223  tO tsj if  M -J CO  p" -J CO  f if OJ  H -J O  if if 00  o >C V,  Ul CO ui  Ul OJ  00  a\  -J U)  en  if 00 if  co  CO  -J CO 00  -J  H  H CO  to Ul  Co NJ  -J  (^  H CO CO  P1 CO Ul  H Ul  if  if  en to  00  IO H CO CO if Ul Ch OJ  P1 P1 t o - J ISO U l CO CO to -J i f ISO i f CO U l  M -J  P> -J  if if H to  M  IV) [O Ul (Jl  H tO Ul ^J  00 to if  H  if  Ul  M  H  H f  O  H-  O  1=  to o  CO CD  & 0  d  CD  3 3 3  tJ t l t )  M P ' P ' p ' P ' t O O J O J O J t O O J U l i f U l ^ J C r i ^ I V O C O t m ^ < i * s i c o c o u i w u i m i o w u j ( i ) U ) t o H ( J i t o o M ) o o i t o o 0 ^ * > H U l W l D O ( I 1 0 s l x l l B ( J l U ) > I I - J U ) U l t O ^ K ) O P O U  P P ' t O M t O P ' p ' O J t O O J i f i f t f U I O O CO i f if t o. i:,f f- U l t D s l f f l U U O U I O J ^ l D O v J O H O H O l 0 1 i f t f U l U l < £ > O J U > U l V O V D > f O C o p J < ] i f < I U l l X > - O . U j U l l £ > < I  to O O i s l U  if J ] l  H M J U l ^ ^ W O J O  P1 if  o  00  if  CD  CD  0  &  P.  CD O rt CD  3  LJ. CD O rt fD  a.  M OJ rt fD  H  n c  pj  o  a  C tr n> n rt  00  H-  P ' P ' P ' M H P ' H P ' P ' P ' P ' P ' P ' P ' P ' P ' P ' P ' P ' P ' P ' P ' M P ' P ' P ' P ' o i u i m o i c i i i i ^ u i * o i ^ u i m t o p i i i * . * . i o u j w u i u i u i u i u i u i ! ^ O J 0 1 < l i f O U 1 0 0 0 C ^ O i f O J C O C O l O t O t O k £ > P ' P ' C O C O i f U l t O U l  Ol 01 <] if O -J CO H  tO if  M M M t o o j t o t o t o t O i f o j i f i j i c r i c n c o c r i o o ~J (T\ V£> OJ O i f c o o t o c n o u i ^ i o j u i o i f i f u i - j ^ j i f H CO l£> tO OJ -J O t » - j o i * > t o p u i m u i u i o i ' j u i o i o W i l '  H  W M W M M P P H I - ' H 0 < J l * ' M O ( » 0 1 l A J P I - J « ) ( D O ) v ] O l 0 1 U l l t > U ) W W I O P I - ' v l W W H O P U l ^ l W ( O l » ) ( D H * > C O O l i J 0 1 l I I - J ( O U l l O H m * P  H-  h-1 0 i= H P) rt CD  O PJ  Cd  _i. CD 0 rt  &  C  00  to to  ixi 01  Ul if  H CO H  VD CO  kO  on OJ  OJ o  P1 -J H>  H  Ul  O-i  co cn  Pi  D  3  '—'  IQ  i S o O S Pi M ^ Pi 3  D  IQ  o s  O  ^O_^ O3-  &  O  n en ^-, 3 tr CD cQ  3  C \ 3 3  i-i  CO D CD 13  ^  D 0  Day  37 39 48 55 62 69 76 81 88 92 111 125 139 151 181 202 222 244 265 306  DPM/mL serum  536 514 410 411 294 349 271 285 258 197 158 166 161 104 107 82 85 69 34 13  Serum chol (mg/dl) 157 162 152 152 151 179 180 191 201 168 172 174 191 188 191 200 194 190 226 249  DPM/mg chol (corr)  DPM/mg chol  408p 380 323 323 233 233 180 178m 154m 140m 110 114 101 66s 67 49 52 43 18 6  341 317 270 270 195 195 151 149 128 117 92 95 84 55 56 41 44 36 15 5  Subject F (Calculated injected dose: 14.9 juCi C 1 4 ) 1 4 11 16 18 27 34 41 48 55 60 67 72 90 104 115 159 173 177 194  3893 3137 1638 1255 1015 807 633 589 428 394 268 373 316 309 209 163 109 112 113 50  200 164 150 160 160 165 158 156 176 175 177 186 161 205 185 201 176 185 179 178  1947 1913 1092 784 634 489 401 378 243 225 151 201 196 • 151 113 81 62 61 63 28  3266 3209P 1832p 1316 P 1064 821 672 633 408 378 254m 336m 329m 253 190 136s 104 102 106 47  225  Day  215 236 286  DPM/mL serum  83 72 67  Serum chol (mg/dl) 161 166 193  DPM/mg chol  DPM/mg chol (corr)  52 43 35  86 73 58  Subject G (Calculated injected dose: 13.7 ^Ci C 1 4 7 12 18 25 30 32 39 47 53 70 75 82 87 103 114 126 161 175 196 217 238 253 281  3965 3011 2031 1642 1178 1169 798 604 429 457 449 306 414 357 234 204 174 306 164 113 100 94 86 71 61  166 164 156 176 189 185 176 187 174 166 162 178 191 214 192 205 207 193 208 194 212 164 193 176 193  2389 1836 1302 933 623 632 453 323 247 275 277 172 217 167 122 100 84 159 79 58 47 57 45 40 32  )  4359 3350 2376 1702 1137m 1153m 827m 589 450 502 506 314 396s 304s 222s 182 153 289p 144 106 86 105 81 74 58  Subject H (Calculated injected dose: 19.4 jzCi C 1 4 ) 1 4 6 11 18 25 34 39  6209 3536 2801 2145 1548 1005 877 584  195 193 193 194 190 181 192 201  3184 1832 1451 1106 815 555 457 291  4103 2361 1870 1425 1050 716 589 374  226  Day  46 51 53 60 69 76 83 90 95 102 107 125 139 151 181 203 224 244 265 305  DPM/mL serum  523 227 361 339 327 215 233 155 187 192 281 116 78 128 99 52 69 78 82 55  Serum chol (mg/dl) 185 175 173 171 166 177 194 191 202 210 227 196 196 204 196 187 177 196 186 165  DPM/mg chol  DPM/mg chol (corr) 364m 167m 269 255 254 157 155 105 119s 118s 160s 76 51 8lP 65 36 50 51 57 43  283 130 209 198 197 121 120 81 93 91 124 59 40 63 51 28 39 40 44 33  Subject I (Calculated injected dose: 12.1 /iCi C 14,) 1 3 5 9 15 22 27 29 36 44 50 57 64 71 78 83 96 127 155 170  5375 3318 3274 2520 1166 892 635 687 492 429 387 344 316 266 271 225 234 100 98 103  177 175 192 188 186 189 161 161 161 162 164 165 158 175 202 195 201 165 163 175  3037 1896 1705 1340 627 472 394 427 306 265 236 208 200 152 134 115 116 61 60 59  6274 3917 3523 2769 1295 r 975 m 815m 882 631 547 488 431 413 314s 277s 238s 241 125P 124 122  227  Day  183 203 224 240 280  m P s  DPM/mL serum  119 101 101 51 75  Serum chol (mg/dl) 160 176 171 205 190  DPM/mg chol  DPM/mg chol (corr)  74 57 59 25 39  154 119 122 51 82  DPM: disintegrations per minute; chol: cholesterol; DPM/mg chol (corr): values adjusted to give theoretical C enrichment if 25 /iCi had been administered per subject; these values were used to generate specific activity decay curves. samples taken while subject was on monounsaturated fat diet, samples taken while subject was on polyunsaturated fat diet, samples taken while subject was on saturated fat diet.  228  Appendix 9 Sample SAAM/CONSAM deck for estimation of parameters of multiexponential equation best describing the specific activity decay curve of one subject.  A SAAM3 0 SUBJECT E DPM/MG CHOL C C FIT A SUM OF EXPONENTIALS TO THE DATA C H DAT C C G O O ) IS THE "INITIAL CONDITIONS" C XG(23)=K(80)*EXP(-P(1)*T)+K(81)*EXP(-P(2)*T)+ K(82)*EXP(-P(3)*T) C C 115G(23) *0.836 FSD=0.1 0 1 4 5 11 18 25 32 37 39 48 55 62 69 76 81 88 92 111 125 139 151 181 202 222 244 265 306  2385 1543 1274 0722 484 443 448 341 317 270 270 195 195 151 149 128 117 092 095 084 055 056 041 44 36 15 05  H PAR C  229  K(80) K(81) K(82) P(l) P(2) P(3)  2 .029216E+03 4 .737480E+02 4 .660952E+02 2 .807654E-01 1..369968E-02 6..345450E-02  1..OOOOOOE+00 1..OOOOOOE+00 1..000000E-02 1..000000E-02 9..999998E-04 9..999998E-04  Y  230  1..000000E+05 1 .000000E+05 1 .000000E+05 1..OOOOOOE+00 1..OOOOOOE-01 1..000000E-01  Appendix 10 Sample SAAM/CONSAM deck for calculation of pool sizes and rates of exchange between pools, as well as the rate of cholesterol de novo synthesis, based on specific activity decay curves.  A SAAM3 0 SUBJECT E COMPARTMENTAL ANALYSIS C C PURPOSE CHOLESTEROL TURNOVER AND KINETICS C DETERMINATION OF POOL SIZES AND RATES OF C EXCHANGE C H DAT C MASSES (M) AND STEADY STATE INPUT (U) M(2) M(3) M(4) U(3) C FLUXES R(3,2) R(2,3) R(3,4) R(4,3) R(0,3) 103 /4.598E+07 FSD=0.1 C UNITS ARE DPM/MG SERUM CHOLESTEROL VERSUSSTIME (DAYS) C AFTER CONVERSION (LINE 18) UNITS ARE FRACTION OF C INJECTED DOSE PER MG SERUM CHOLESTEROL VERSUS TIME C (DAYS) C 20.9UCI 14C-LABELLED CHOLESTEROL INJECTED C 0 1 4 5 11 18 25 32 37 39 48 55 62 69 76 81 88 92  2385 1543 1274 722 484 443 448 341 317 270 270 195 195 151 149 128 117  231  Ill 125 139 151 181 202 222 244 265 306  92 95 84 55 56 41 44 36 15 5  H PAR C IC(3) IS THE INITIAL FRACTION OF THE TOTAL INJECTED C AMOUNT ICO) 1 C K(3) IS THE INVERSE OF THE SPACE OF DISTRIBUTION OF M(3) C ITS UNITS ARE /ML K(3) 6.387736E-05 0.000000E+00 1.000000E+00 C RATES OF EXCHANGE IN UNITS OF /DAY L(3,2) 1.197982E-01 9.999998E-04 1.000000E+02 L(2,3) 9.173752E-02 5.000000E-02 1.000000E+02 L(4,3) 4.963755E-02 4.000000E-02 5.000000E-01 L(3,4) 3.346296E-02 9.999998E-05 5.000000E-01 L(0,3) 5.971058E-02 5.000000E-03 1.000000E+02 H STE M(3)=1.80714/K(3) C AVERAGE SERUM CHOLESTEROL CONCENTRATION OVER TIME WAS C 180.714MG/DL, OR 1.80714MG/ML U(3) 1.689259E+03 1000000  232  Appendix 11 Serum free cholesterol deuterium enrichment (°/oo) relative to SMOW at five time points on each of the three diets.  Sub  Time (h)  A  0 12 24 36 48  -63..03 + 1 .33 3..91 53..55 42 .17 + 3 .66 58 .54 + 0 .92  0 12 24 36 48  -62..19 + 5 .05 -40 .73+23 .00 13 .55+:L0 .49 32..33 17..71  -114 .43 -97 .75 + 2 .28 -25 .79 -42 .25 -40..48 + 1 .93  -240 .96  0 12 24 36 48  -75..05+24..03 -31..46 + 7..25 18..23 + 0..51 26..30 40..73 + 2..62  -102..68 + 6 .50 -86..43 + :L9 .89 -36..37 + 0 .06 -23..89 + 0..86 -40..89+20..83  -174..57+ 7..52 -156..18+19..01 -154..13+19..01 -162..66± 4,.68 -167..03+ 3..54  -209..15 + -181..52 + -121..00 + -142..68 + -147..97 +  -104..00+ 1..04 -82 .36+11..12 -31..93+ 7..46 -41 .3 7+ 4..00 -32 .14+ 3..44  G  MOINfO  PO]^Y  -70,.81 + -74,.35 + -14,.29 + -26 .50 +  2 .34 7 .25 5 .01 0 .32  SAT  -255 -198 -156 -163 -160  .22+ 6..05 .37+17 .65 .32+ 5..20 .87+15 .01 .14+15 .32  -149 .83+35 .54 -196 .54+ 3..51 -186 .66+13..68  0 12 24 36 48  -146..24 + -120..31 + -75..34 + -85..3 9 + -26..38 +  0 12 24 36 48  -133..19 + 0,.41 -110..32 + 1,.64 -68,.96 + 7..57 -73.,15 + :L8 .79 . -65,,72 + 5.,41  -189..97 -133..60 -132..82 + 6..05 -134..80  0 12 24 36 48  -68..2 8 + 3,.58 7..02 80,.4 9 + 2,.50 50,.87 + 0,.40 38..61+22..12  -241..96 + 3 .14 -194..75+:L4,.76 -147..07+32..92 -173 . .92 + 2 .53  47 .92+ 5 .38 81..43+20,.16 154 .19+ 3 .20 122 .27+ 3..51 115..61  0 12 24 36 48  -246..18+:L0.,36 -196..58+:L7..90 -139..72 + 0..51 -153.,24 -141.,94  -49..82 43.,94 109..69 + 2,.96 91..08+31,.75 110..50  -86..03+ 4..03 -56..59+26..66 25..56+ 9..67 -21..30 -8..66+ 4 .36  0,.83 1,.49 0,.12 5..23 8 .88  233  7 .00 0 .25 7 .54 4 .30 3 .04  -137..27+ -103..35+ -99..38+ -101..45+ -56..81±  2..14 0..08 1..92 1,.17 3,.01  H  0 12 24 36 48  -123.31+15.54 •102.33 -88 .45+ 1.22 •106.95+ 0.83 -96.28+ 5.84  0 12 24 36 48  -156.32 -159.20+ 8.35 -124.97+14.57 -129.31+ 2.99 -138.79  X +  •112.84 35.66+ 73.72+ 73.79+ 79.46+  1.34 0.36 1.11 4.99  -88.83+27.14 -79.95+13.02 -68.47+16.59 •53.69+ 4.94 •65.18+16.54  SE.  234  -14.72+1.75 41.85 73.88 16.90 92.74+11.92 -99.97+16.81 •100.07+ 3.47 -20.47 -84.11  Appendix 12 Serum esterified cholesterol deuterium enrichment ( /oo) relative to SMOW at five time points on each of the three diets.  Sub  Time (h)  A  0 12 24 36 48  B  MONO  40.52 51.15+ 5.18 73.59+2.62 88.12+ 6.34  SAT  -56.51+ 1.93 -61.74+ 1.80 -30.10+ 3.59 -9.00±4.30  -278.33+ 7.04 -239.79+12.16 -238.54+12.83 -231.18 -216.00+14.60  -75.27+ -85.95± -71.87+ -56.59± -48.42+  -265.50+19.24  0 12 24 36 48  -34.37+ -18.38+ -1.10+ 66.63+ 43.24+  C  0 12 24 36 48  -49.79 -26.65 -28.50+ 9.65 -2.32 -0.99+0.11  D  0 12 24 36 48  -162.15+ -159.33+ -138.95+ -122.58+ -115.96+  3.10 0.76 1.39 2.82 1.96  -274.71+13.66 -258.52+23.24 -246.90+ 0.13 -228.79+ 0.42 -218.62+ 3.33  -104.29+ -122.42+ -102.36+ -93.31+ -71.89+  E  0 12 24 36 48  -113.06+ 2.19 -112.36+ 3.71 -75.07+2.63 -73.23 -67.99+ 8.08  -257.20+ 9.57 -261.24+ 0.93 -230.29 -206.10  -97.58+ 7.07 -116.92+ 0.08 -98.08+0.70 -77.46+6.08 -59.96+ 0.21  F  0 12 24 36 48  -42.13+ 7.81 -59.65+15.31 -29.52+ 0.83 20.61 7.74+6.07  -254.08±13.07 -230.45+ 2.89 -200.36+ 1.32 -189.93+4.12  21.79+ 0.54 29.25+ 1.27 52.60+ 1.00 41.00 65.72+2.47  0 12 24 36 48  -262.54+10.49 -276.65+ 5.19 -239.81+ 5.55 -224.04 -204.35+10.08  -20.83+ 1.02 31.92+ 7.09 32.56+ 2.62 60.74+ 2.60 25.87+11.60  -58.69+17.25 -61.40+ 3.25 -38.54+ 1.27 -56.26+ 0.96 -31.59+ 5.28  G  3.12 1.16 1.79 3.36 6.45  POLY  2.72 1.91 1.02 0.51 0.15  -251.87+13.28 -241.43+ 0.47 -248.23+ 0.47  -96.29+2.16 -85.66+14.25 -80.98+ 0.66 -69.34+4.93 -61.24+7.68  -261.18 -222.48 -217.44+24.71 -228.97+4.63 -230.52+13.53  235  3.70 2.63 0.57 1.48 1.11  H  0 12 24 36 48  -189.53+22.67 -228.43+32.38 -217.33+ 5.03 -191.46+ 3.66 -171.18+ 3.59  -55.64 -17.15+ 21.75+ 49.10+ 89.09+  0 12 24 36 48  -262.62+23.41 -260.90+13.58 -249.42+ 2.16 -224.04+ 4 . 6 1 -234.86+ 8.80  •68.12+ 3.18 •62.40 -7.23 •11.16+ 5.72 -8.40+ 0.30  X +  SE.  236  1.92 4.46 6.63 5.43  -72.55 -29.48 -29.74+ 4.19  -53.69+20.67 •81.03+ 4.27 •75.81+ 2.28 •53.86+30.85 •63.41+10.18  Appendix 13 Sample SAAM/CONSAM deck for calculation of rates of de novo cholesterol synthesis pool sizes and rates of exchange between pools, based on deuterium incorporation data.  A SAAM3 0 C C C  COMPARTMENTAL ANALYSIS SUBJECT ?  c H PAR L(3,10) 9.990048E-04 L(0,3) 8.900167E-04 H DAT XFF(10)=F(10) 110QL 0 1 12 24 36 48 2400 H DAT 103 0 12 12 24 24 36 36 48 48 2400 H STE M (3)=28291/3 U(3) 8.393153E+00 Y  9.999998E-06 9.999998E-09  1.000000E+02 1.000000E+03  *0.478 0 3625 3625 3625 3600 3600 3600 PSD=0.1 0 21.7 24.03 58.88 69.60 46.77 73.34 63.65 71.30 1721  1.0E+05  237  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088878/manifest

Comment

Related Items