Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Aspects of cholesterol metabolism in suckling and adult guinea pigs and term and preterm human infants Hamilton, Jennifer Jane 1992

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

Item Metadata

Download

Media
831-ubc_1992_fall_hamilton_jennifer_jane.pdf [ 5.54MB ]
Metadata
JSON: 831-1.0086529.json
JSON-LD: 831-1.0086529-ld.json
RDF/XML (Pretty): 831-1.0086529-rdf.xml
RDF/JSON: 831-1.0086529-rdf.json
Turtle: 831-1.0086529-turtle.txt
N-Triples: 831-1.0086529-rdf-ntriples.txt
Original Record: 831-1.0086529-source.json
Full Text
831-1.0086529-fulltext.txt
Citation
831-1.0086529.ris

Full Text

ASPECTS OF CHOLESTEROL METABOLISM IN SUCKLING AND ADULT GUINEA PIGS AND TERM AND PRETERM HUMAN INFANTS by JENNIFER JANE HAMILTON B.Sc, University of Guelph, 1983 M.Sc, University of British Columbia, 1986  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Pathology)  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA September 1992 © Jennifer Jane Hamilton, 1992  In  presenting  degree freely  this  at the available  copying  of  department publication  of  in  partial  fulfilment  of  University  of  British  Columbia,  I  for  this or  thesis  reference  thesis by  this  for  his  and  scholarly  or  thesis  study. I further  for  her  purposes  gain  shall  permission.  (Signati  Department  of  Pathology  The University of British Vancouver, Canada  Date  DE-6 (2/88)  Columbia  Oct. 14 92  requirements  agree  that  agree  may  representatives.  financial  the  be  It not  that  the  by  understood be  allowed  an  advanced  Library shall make  permission  granted  is  for  for  the that  without  it  extensive  head  of  my  copying  or  my  written  ABSTRACT Early nutrition can affect responses to dietary fat and cholesterol in later life. Normal postnatal cholesterol metabolism must be understood before this phenomenon is explained. In many animal species, including humans, plasma LDL-cholesteroll levels increase soon after birth. The mechanism for this is unknown although hypotheses include the onset of feeding, hormonal changes at birth and regression of the fetal zone of the adrenal. An increased rate of cholesterol synthesis and/or decreased LDL receptor number may be involved. The objectives of this thesis were to measure 1) developmental changes in indices of the rate of cholesterol synthesis (using plasma lathosterol levels and hepatic 3-hydroxy-3-methylglutaryl coenzyme A (HMG CoA) reductase activity) and hepatic LDL receptor mass in postnatal guinea pigs 2) plasma lathosterol in humans at birth and 4 days of age and 3) plasma lathosterol in preterm infants during 10% Intralipid-induced hypercholesterolemia soon after birth. HMG CoA reductase is the rate limiting enzyme of cholesterol synthesis, while lathosterol is a sterol intermediate. Levels of plasma lathosterol, measured by gas-liquid chromatography, have been correlated to hepatic and whole body cholesterol synthesis and are therefore possibly an in vivo index of cholesterol synthesis. Hepatic LDL relative receptor mass was measured by ligand blotting using gold-labelled BVLDL as the ligand. Liver cholesterol levels were also determined to test the hypothesis that plasma cholesterol could be from hepatic stores after birth. In infants, apo AI and B were measured as indicators of the presence of HDL and LDL+VLDL, respectively. In the suckling guinea pig, changes in LDL-cholesterol levels coincided with changes in hepatic HMG CoA reductase activity. Hepatic HMG CoA reductase activity was higher at 4 than at 1 and 8 days of age and was higher still in adults. Plasma lathosterol levels peaked at 8 days of age and were decreased in adults. The LDL receptor mass was equal among the 3 suckling ages and was 2-fold higher in adults. The hepatic content of cholesterol did not change between 1 and 8 days of  ii  age. In the term infant, plasma lathosterol levels decreased between birth and 4 days of age despite increased plasma cholesterol and apo B. Among preterm infants, plasma lathosterol levels increased with the infusion of lipid. In infants given negligible lipid, hypercholesterolemia and increased plasma lathosterol levels were not found. Apo AI or B levels did not increase with lipid administration. These data from the guinea pig, indicate 1) that "postnatal hypercholesterolemia" may be related to increased cholesterol synthesis, although the quantitative importance of this is unknown, and 2) that hepatic and extrahepatic cholesterol synthesis may change independently after birth. The changes in guinea pig plasma cholesterol levels during suckling do not appear to be related to a change in guinea pig LDL receptor mass. Interestingly, however, an inverse relationship between LDL receptor mass and plasma cholesterol levels appeared to be present in late suckling versus adult ages. Contrary to results of the guinea pig studies, data from human term infants suggest that the postnatal increase in plasma cholesterol is not coincident with increased cholesterol synthesis. Finally, data from preterm infants support the theory that 10% Intralipid infusion results in the transfer of free cholesterol from cellular membranes to plasma, thus enhancing cholesterol synthesis to replace lost cholesterol. The preterm infant infused with 10% Intralipid may be analogous to laboratory animals in whom early perturbation of cholesterol metabolism has long term effects.  iii  TABLE OF CONTENTS  ABSTRACT  ii  TABLE OF CONTENTS  iv  ABBREVIATIONS  ix  LIST OF TABLES  xi  LIST OF FIGURES  xii  ACKNOWLEDGEMENTS  xiii  1 INTRODUCTION  1  1.1 Cholesterol transport and metabolism  1  1.1.1 LDL receptor  3  1.1.2 Cholesterol synthesis  8  1.2 Fetal cholesterol transport and metabolism  11  1.3 Postnatal cholesterol transport and metabolism  13  1.3.1 Changes in energy substrate  16  1.3.2 Initiation of enteral feeding  16  1.3.3 Change in hormonal milieu  19  1.3.4 Change in cholesterol homeostasis  20  1.3.5 Reduced uptake of lipoproteins by the adrenal  21  1.3.6 Physiologic cholestasis  22  1.3.7 Efflux of lipid from tissue stores  22  1.4 Preterm infant plasma cholesterol levels  23  1.5 Significance  25  1.6 Rationale  25  iv  1.7 Thesis objectives  28  1.8 Specific aims  29  1.8.1 Guinea pig experiments  29  1.8.2 Term infant study  29  1.8.3 Preterm infant studies  30  2 METHODS AND MATEMALS  31  2.1 Chemicals  31  2.2 Clinical supplies  31  2.3 Equipment  32  2.4 Guinea pig experiments  33  2.4.1 Experimental design  33  2.4.2 Guinea pig housing and care  33  2.4.3 Guinea pig sacrificing  33  2.4.4 Preparation of tissues  34  2.4.4.1 Plasma and liver  34  2.4.4.2 Microsomes  34  2.4.4.3 Plasma membranes  34  2.4.5 Plasma determinations  35  2.4.5.1 Cholesterol  35  2.4.5.2 Lathosterol  35  2.4.5.3 Lipoproteins  36  2.4.6 Liver determinations  37  2.4.6.1 Total lipid and cholesterol  37  2.4.6.2 HMG CoA reductase activity  37  v  2.4.6.3 LDL receptor relative mass  39  2.4.6.4.1 Lipoproteins  41  2.4.6.4.2 Membrane solubilization  41  2.4.6.4.3 SDS-PAGE and electroelution  42  2.4.6.4.4 Gold-lipoprotein conjugation  42  2.4.6.4.5 Ligand blotting  42  2.4.6.4.6 LDL receptor identification  43  2.4.6.4.7 Linearity  43  2.4.6.4.8 Immunoblotting  44  2.5 Human studies  44  2.5.1 Subject selection  44  2.5.1.1 Term infants  44  2.5.1.2 Preterm infants  45  2.5.1.3 Normal adults  45  2.5.1.4 Hypercholesterolemic adults  45  2.5.2 Subject care and blood sampling  46  2.5.2.1 Term infants  46  2.5.2.2 Preterm infants  46  2.5.2.3 Normal adults  47  2.5.3 Plasma analyses  47  2.6 Statistical analyses  48  2.6.1 Guinea pigs  48  2.6.2 Term infants  48  2.6.3 Preterm infants  48  vi  2.6.4 Adults  49  2.7 Ethical approval  49  2.7.1 Guinea pigs  49  2.7.2 Term infants  49  2.7.3 Preterm infants  49  3 RESULTS  50  3.1 Guinea pig experiments  50  3.1.1 Animal body and liver weight  50  3.1.2 Plasma cholesterol and lipoproteins  50  3.1.3 Hepatic lipid  50  3.1.4 Cholesterol synthesis  55  3.1.5 LDL receptor mass  56  3.2 Term infant study  61  3.2.1 Infant body weights  61  3.2.2 Plasma cholesterol and apolipoproteins  61  3.2.3 Plasma lathosterol concentrations  63  3.2.4 Plasma lathosterol:cholesterol ratios  65  3.3 Preterm infant studies  65  3.3.1 Preliminary cholesterol study  65  3.3.2 Lathosterol study  71  3.3.2.1 Cord and day 3-4 subset  71  3.3.2.2 Longitudinal subset  73  4 DISCUSSION  78  4.1 Guinea pig experiments  78  vii  4.1.1 Early developmental changes in cholesterol synthesis 78 4.1.1.2 Early developmental changes in LDL receptor relative mass  82  4.1.1.3 Early developmental changes in liver cholesterol content  84  4.1.2 Suckling versus adult cholesterol metabolism  85  4.1.2.1 Suckling versus adult cholesterol synthesis  85  4.1.2.2 Suckling versus adult LDL receptor relative mass  86  4.1.3 Guinea pig study summary  87  4.2 Term infant study  89  4.3 Plasma lathosterol as an indicator of cholesterol synthesis in the neonate  91  4.4 Speculation about increased plasma cholesterol levels after birth  92  4.5 Preterm infant study  93  4.5.1 Effect of 10% Intralipid on plasma lathosterol  93  4.5.2 Effect of 10% Intralipid on plasma apoproteins  94  4.5.3 Hepatic versus extrahepatic cholesterol synthesis during 10% Intralipid infusion  95  4.5.4 Preterm versus term plasma cholesterol and lathosterol levels  96  4.5.5 Summary  97  4.6 Future studies  97  4.7 Overall conclusions  99  5 REFERENCES  102  viii  ABBREVIATIONS  ANOVA  analysis of variance  apo  apoprotein  BM  breast milk  BPD  bronchopulmonary dysplasia  BSA  bovine serum albumin  /SVLDL  beta very low density lipoprotein  °C  degrees celsius  Ci  curie  d  day  DPM  disintegrations per minute  EBM  expressed human breast milk  EDTA  ethylenediamine tetraacetic acid  F  infant formula  g  gram  HDL  high density lipoprotein  HMD  hyaline membrane disease  HMG CoA  3-hydroxy-3-methylglutaryl coenzyme A  IDL  intermediate density lipoprotein  kDa  kilodaltons  L  litre  LDL  low density lipoprotein  mol  mole  ix  M  molar  mL  milliliter  mRNA  messenger ribonucleic acid  NADP  nicotinamide adenine dinucleotide phosphate  NADPH  nicotinamide adenine dinucleotide phosphate, reduced form  NEC  necrotizing enterocolitis  ng  nanogram  p  statistical probability of mean differences not existing  pmol  picomole  PMSF  phenylmethylsulfonyl fluoride  PDA  patent ductus arteriosus  SD  standard deviation  SDS-PAGE  sodium dodecyl sulfate-polyacrylamide gel electrophoresis  TBS  Tris buffered saline  Tris  tris(hydroxymethyl)aminomethane  VLDL  very low density lipoprotein  v  volume  wt  weight  x  LIST OF TABLES 1.1 Animals experiencing elevated postnatal plasma cholesterol levels  15  1.2 Some examples of late effects of early nutrition  26  3.1 Body and liver weights of guinea pigs in experiments 1-3  51  3.2 Plasma cholesterol, hepatic HMG CoA reductase activity and hepatic LDL receptor relative mass from suckling and adult guinea pigs  53  3.3 Plasma cholesterol, apoproteins and lathosterol levels from term infants at birth (cord) and at 4 days of age  62  3.4 Plasma cholesterol and lathosterol values in adult reference populations  64  3.5 Characteristics of preterm infant populations  67  3.6 Details of infants presented in Figures 3.8 and 3.9  68  3.7 Levels of cholesterol, apo AI and B, and lathosterol in plasma of preterm infants at birth (cord) and 3-4 days of age 3.8 Details of infants presented in Figures 3.10 and 3.11  72 74  xi  LIST OF FIGURES 1.1 An overview of lipoprotein metabolism  2  1.2 A diagramatic representation of the LDL receptor  4  1.3 An overview of cholesterol biosynthesis  9  2.1 Ligand blotting and immunoblotting techniques for detecting the LDL receptor  40  3.1 Effect of age on guinea pig plasma sterols, plasma lipoproteins and liver lipids  52  3.2 Verification of the identity of plasma lipoproteins from 1 day old and adult guinea pigs by apoprotein content on SDS-PAGE  54  3.3 Verification of the identity of guinea pig LDL receptor as detected by ligand blotting  57  3.4 Detection of guinea pig LDL receptor by immunoblotting and ligand blotting  58  3.5 Representative example of quantitation of LDL receptor relative mass from guinea pig liver membranes by ligand blotting 3.6 A representative blot of the comparison of LDL receptor relative mass in 8  59  day old and adult guinea pig liver membrane by ligand blotting  60  3.7 Organization of preterm infant studies  66  3.8 Plasma cholesterol levels and 10% Intralipid infusion rates of 4 representative preterm infants 3.9 Plasma cholesterol levels and 10% Intralipid infusion rates of 2 representative preterm infants with minimal 10% Intralipid infusion  69 70  3.10 Plasma cholesterol, lathosterol and apo AI and B levels and 10% Intralipid infusion rates of 6 representative preterm infants  75  3.11 Plasma cholesterol, lathosterol and apo AI and B levels and 10% Intralipid infusion rates of 2 representative preterm infants with minimal 10% Intralipid administration  76  4.1 Comparison of developmental changes in hepatic HMG CoA reductase activity of the rat, pig and guinea pig  79  xii  ACKNOWLEDGEMENTS  Thank you, first, to Dr. Sheila Innis, my research supervisor, who shared in the many trials and frustrations of this project. Second, I must acknowledge the constant guidance of Dr. Peter Hahn, my supervisor emeritus. My sincere thanks, also, to my thesis committee members, Dr. Juri Frohlich, Dr. Peter Jones, Dr. Gillian Lockitch and Dr. Dana Devine who endured years of committee meetings and spent considerable time reading drafts of this work. Further gratitude goes to Drs. Frohlich and Pritchard who have accepted me into their fold at the Lipid Research Group of the University of British Columbia and tolerated my lack of productivity while completing this work. I would also like to recognise my collaborators and friends Dr. Min Phang who took part in the preterm infant study, Dr. Ann Synnes who was involved in the term infant study, Dr. Nancy Auestad and Roger Dyer who shared in the development of the ligand blot assay, and Laurie Nicol who measured HMG CoA reductase in the guinea pig samples. Sincere thanks as well to Murray Mackinnon who had the difficult role of statistical consultant for this project. Thank you also, to my fellow students, colleagues and friends: Dianne Arbuckle, Wendy Cannon, Roberta Egby, Dr. Jane Gardner, Dr. Neil Haave, Dr. Nina Hrboticky, Janette King, Rebecca Ng, France Rioux, Marie Yeo, and Dr. Katharine Wall. My parents, Sally and Donald Hamilton and in-laws, Bunny and Donald Meikle were also unfailing in their support. The days, weekends and weeks of babysitting were especially appreciated. Thanks go also to the security department of the Children's Hospital Site who escorted me to my car on many late evenings. Finally, but most importantly, I would like to acknowledge my husband, Drew Meikle, without whom this work could never have been completed. Thank you also to my sons, Alexander and James, who were born in the course of this degree, and too often encountered an absent or tired mother.  xiii  Chapter 1 INTRODUCTION  In the fetus, near term, the quantity of cholesterol in plasma and its lipoprotein distribution are similar among most mammalian species. After birth there seems to be a universal increase in plasma LDL-cholesterol levels which is known as "postnatal hypercholesterolemia". This term, however, is perhaps a misnomer because it may represent a normal physiological occurance. Understanding the perinatal changes in plasma cholesterol levels may be central to determining how early nutrition affects responses to dietary fat and cholesterol in later life. In this thesis, early postnatal cholesterol metabolism was studied in guinea pigs and humans. Also, the preterm human infant was investigated as an example of how early nutrition can influence cholesterol metabolism.  1.1 Cholesterol transport and metabolism Cholesterol is a 27 carbon steroid molecule ubiquitously present as a structural component of animal cellular membranes. It is also the precursor for the production of steroid hormones, the substrate for synthesis of bile acids and a facilitator of lipid absorption and transport. Sources of cholesterol in the body include the diet and de novo synthesis. Excretion of cholesterol occurs directly, through the bile, or is accomplished by its conversion to bile acids and subsequent fecal loss.  Cholesterol is carried through the plasma by lipoproteins, which are spherical complexes of  protein (apoproteins) and lipid (free and esterified cholesterol, phospholipid and triglyceride). The lipoprotein type or class is identified by its size and density (dependent on the ratio of protein to lipid) and the apoproteins present. The complement of apoproteins dictates the role that each of the lipoproteins play. Generally, cholesterol metabolism and transport can be divided into exogenous and endogenous paths (Figure 1.1). Exogenous cholesterol is absorbed at the intestine, packaged along with triglyceride into chylomicrons and secreted into the blood via the lymph. Lipoprotein lipase,  1  EXOGENOUS PATHWAY  ENDOGENOUS PATHWAY  ,  Bile acids +  LDL DIETARY CHOLESTEROL  -*-  Cholesterol  t  1  \  \  1 , INTESTINE  EXTRAHEPATIC TISSUES  LIVER  ,  1  t CHYLOMICRONS  CHYLOMICRON REMNANTS  1  LCAT IDL  VLDL  t  *"~  HDL  t  \  CAPILLARIES  CAPILLARIES  lipoprotein lipase  lipoprotein lipase  Figure 1.1 An overview of lipoprotein metabolism. Dietary and biliary cholesterol enter the circulation via chylomicrons. Lipoprotein lipase allows the removal of triglyceride from these particles at the capillaries. The resulting chylomicron remnants are cleared by hepatic receptors. VLDL, produced by the liver, transport triglyceride to the periphery like chylomicrons. VLDL remnant particles, IDL, are taken up by the liver or metabolized to LDL. LDL are mostly cleared by hepatic receptors. HDL are synthesized by the intestine and liver and with lecithincholesterol acyltransferase (LCAT) allow transport of cholesterol from the periphery to the liver, (adapted from Goldstein et ah, 1983) 2  present at the capillary endothelium of muscle and adipose tissue, facilitates the removal of triglyceride from these particles. The resultant chylomicron remnants are cleared by remnant or LDL receptors in the liver. During fasting, and when cellular cholesterol needs are not met by exogenous supply, cholesterol is synthesized endogenously and is exported from the liver in triglyceride-rich very low density lipoproteins (VLDL). VLDL also interact with lipoprotein lipase at the capillary endothelium leading to the release of triglyceride from VLDL and resulting in intermediate density lipoproteins (IDL). IDL may be cleared by either hepatic low density lipoprotein (LDL) receptors or presently undefined scavenger receptors. Remaining IDL is converted to LDL which eventually binds to LDL receptors on either hepatic or extrahepatic cells. Also, LDL can be taken up by a thus undefined receptor-independent mechanism. Finally, high density lipoprotein (HDL) is secreted by the liver and the intestine and it, along with lecithin:cholesterol acyl transferase (LCAT) is responsible for "reverse cholesterol transport" which allows cholesterol to be transported from the periphery to the liver. LCAT catalyzes the formation of cholesteryl esters from free cholesterol and phospholipid fatty acid. There is no doubt that cholesterol is essential for life and that cholesterol synthesis is necessary for normal cell growth and function (Kandutsch and Chen, 1977). Clearly, however, an excess of cholesterol in the plasma is hazardous. High plasma LDL-cholesterol levels are associated with increased risk of coronary heart disease. Thus, it is very important that plasma cholesterol levels be tightly regulated. The LDL receptor is thought to be central to cholesterol homeostasis.  1.1.1 LDL receptor In 1974, Goldstein and Brown, studying fibroblasts, proposed the mechanism by which cholesterol metabolism might be regulated (reviewed by Brown and Goldstein, 1986 and illustrated in Figure 1.2). They hypothesized that LDL is taken up by a glycoprotein cell surface receptor present on most cells. The LDL receptor, or LDL (B/E) receptor, is synthesized in the endoplasmic  3  Figure 1.2 A diagramatic representation of the LDL receptor. Receptor is synthesized in the endoplasmic reticulum (1) and undergoes addition of sialic acid residues and glycosylation in the Golgi apparatus (2). Finally the receptor migrates to clathrin coated pits in the cell membrane where it is available for LDL binding (3). Invagination of the pits occurs at regular intervals and bound LDL (4) is transported into the cell (5). An ATPase pump on the vesicle, creates an acid environment, allowing dissociation of the LDL receptor and LDL (6). The receptor returns to the cell surface (7) while LDL is degraded in lysosomes (8). Free cholesterol, which is eventually released from the lysosome causes repression of further cholesterol synthesis and LDL receptor synthesis and stimulates acyl cholesterol acyl transferase (ACAT) and thus, cholesteryl ester storage, (adapted from Brown and Goldstein, 1985) 4  reticulum and then undergoes glycosylation and addition of sialic acid residues in the Golgi apparatus. After 45 minutes the receptor is found concentrated in clathrin coated pits on the cell surface. These regions undergo endocytosis to form vesicles every 3-5 minutes, regardless of whether or not lipoprotein has bound to the LDL receptor. Apoprotein (apo) B100, the sole apoprotein on LDL, is the ligand responsible for this interaction. Apo E, present on chylomicrons, IDL, VLDL and a subspecies of HDL, can also bind to the receptor. The subsequent dissociation of LDL and the receptor is accomplished by the creation of an acidic (pH 6.5) environment within the vesicle by a membrane bound-ATPase proton pump. The liberated receptor then shuttles back to the plasma membrane where it is again able to bind LDL. The half-life of the receptor is approximately 25 hours. Meanwhile, the vesicle, still containing LDL, fuses with a lysosome which degrades the lipoprotein apoprotein and hydrolyses the lipid constituents. Cholesterol delivered by LDL, or an oxidized derivative, exerts feedback control on cellular cholesterol metabolism by 1) decreasing cholesterol synthesis, 2) decreasing the concentration of LDL receptor mRNA and 3) activating acyl CoA:cholesterol acyltransferase activity which allows storage of cholesterol as cholesterol esters. The overall effect of the above system is that there is an inverse relationship between LDL receptor number and plasma LDL-cholesterol levels. When there is abundant LDL-cholesterol in the plasma, the receptor is down-regulated and when plasma LDL-cholesterol levels are low, increased numbers of receptors are present on the cell surface. Generally, there is also a concordance between LDL receptor number and 3-hydroxy-3-methyl glutaryl (HMG) CoA reductase (E.C.I. 1.1.1.34) activity. HMG CoA reductase is the rate limiting enzyme of cholesterol synthesis (Shapiro and Rodwell, 1971). A demand for cholesterol by the cell is usually met by increasing both the receptor number and cholesterol synthesis. When abundant cholesterol is present, both are depressed. The importance of a functional LDL receptor is demonstrated by the effect of having mutant genes which produce defective LDL receptors. In this condition, known as familial hypercholesterolemia, high levels of plasma LDL-cholesterol result, leading to premature atherosclerosis.  5  Hepatocyte and fibroblast LDL receptors are indistinguishable (Havinga et al., 1987, Semenkovich and Ostlund, 1987). In many species, most LDL receptors are found in the liver. In the hamster, for instance, >90% of the LDL removed from the plasma is via the hepatic LDL receptor (Spady et al., 1983a). It is, therefore, thought that this organ is essential for cholesterol homeostasis. This has been confirmed in humans by the success of liver transplantation as a treatment for homozygous familial hypercholesterolemia in a six year old girl. Five months after the surgery, plasma cholesterol levels had dropped from ~28.4 to ~7.8 mM (Bilheimer et al., 1984). Besides dietary cholesterol, which decreases receptor dependent hepatic uptake of cholesterol in the hamster (Spady and Dietschy, 1988), there are other factors which affect LDL receptor expression and binding. High levels of dietary saturated fat, in hamsters, lead to decreased hepatic receptor dependent cholesterol uptake (Spady and Dietschy, 1988), while feeding polyunsaturated versus saturated fat to guinea pigs causes LDL receptor levels to increase (Ibrahim and McNamara, 1988). Hormones can also influence the LDL receptor. Glucocorticoids, for example, cause decreased LDL receptor binding to cultured rat hepatocytes (Salter et al., 1987 and 1988). Conversely, injection of estrogen in rats increases the number of hepatic LDL binding sites (Kovanen et al. 1979a). Estrogen also induces the entire LDL receptor pathway in cultured human hepatoma cells (Semenkovich and Ostlund, 1987). Insulin and triiodothyronine increase LDL receptor numbers on cultured human fibroblasts (Chait et al., 1979a and b). These hormones have a similar effect on cultured rat hepatocytes, increasing binding of LDL to the receptor (Salter et al., 1987 and 1988). Because insulin exerts its effect as early as 1 hour after addition to the cultured cells, it may modulate binding by mechanisms other than changing receptor number (Salter et al., 1988). Soutar and Knight (1990) have argued, however, that there is no clear evidence of any physiological modulation of LDL receptor activity once receptors have been synthesized. Finally, Bihain et al. (1989) found that free fatty acids in the medium of cultured human fibroblasts inhibits LDL binding to the receptor. The concentration of fatty acids required for this effect, though, was above levels  6  generally observed in human plasma. LDL receptor activity can be measured in several ways. In vivo, this is commonly done by determining the rate of disappearance of [125I]LDL from the circulation. This method, however, does not differentiate between receptor dependent and independent uptake of LDL. Another criticism of this procedure is that the autologous LDL that is used for labelling may not be "metabolically active". Because autologous LDL is from the plasma, it potentially represents a subspecies of LDL that is not readily taken up by the receptor (Witztum et ah, 1985). This same criticism applies to all other methods that use autologous LDL. Cultured cells have also been used to measure uptake of lipoproteins by the LDL receptor. Generally the receptor and cholesterol synthesis are first upregulated by incubating the cells in lipoprotein depleted serum. The ability of added lipoproteins to inhibit cholesterol synthesis (usually measured by HMG CoA reductase activity) is assayed. LDL receptor activity is measured in vitro by an assay which relies upon quantitation of [125I]LDL binding to the receptor on plasma membrane or microsomal preparations. Bound and free [125I]LDL are separated by ultracentrifugation or filtration, and data is evaluated by Scatchard analysis to determine the affinity of binding (Kd) and the number of LDL receptors (Vm). This can also be done with tissue homogenate (Rudling and Peterson, 1985). A disadvantage of membrane assays is the presence of high levels of nonspecific binding, especially in the liver (Kovanen et al., 1979b). Finally, Western blotting has been recently used to quantitate receptor mass (Wade et al., 1985, Soutar et al., 1986, Gherardi et al., 1988). Membrane proteins are first separated by electrophoresis, which allows differentiation between the LDL receptor and proteins of other sizes. Various techniques including ligand blotting and immunoblotting are then employed to visualize, verify the identity of, and quantitate the LDL receptor band. Soutar et al. (1986) showed that this method allows detection of the expected inverse relationship between hepatic LDL receptor protein content and levels of plasma LDL-cholesterol in humans. It does not, however, independently assess the affinity of lipoprotein binding to the receptor.  7  1.1.2 Cholesterol synthesis Central to the LDL receptor theory is the ability of cholesterol synthesis to be regulated. Knowledge of this has been reviewed by Fielding and Fielding (1985). Cholesterol synthesis involves 30 different enzymes that are located in both the cytosol and the microsomes. First, acetyl CoA molecules are used to synthesize HMG CoA (Figure 1.3) and then this is converted to mevalonate by microsomal HMG CoA reductase. Eventually lanosterol is converted to cholesterol via desmosterol. An additional route between lanosterol and cholesterol also occurs. Lathosterol (5a-cholest-7-en-3j8-ol, 3/S-hydroxy-5a-,7-cholestene or A7 lathosterol) is an intermediate of this pathway. The mechanism(s) by which cholesterol, or an oxygenated derivative, decreases cholesterol synthesis is not fully understood. HMG CoA reductase activity, however, may be affected by 1) decreasing the amount of active enzyme, 2) decreasing transcription of the reductase gene and/or 3) increasing the rate of enzyme degradation. Several factors are known to affect the rate of cholesterol synthesis. The rate decreases in rats during fasting (Andersen and Dietschy, 1977) and in the presence of plasma cholesterol associated with LDL, HDL or chylomicrons (Nervi and Dietschy, 1975). In isolated rat hepatocytes, insulin increases and glucagon decreases the cholesterol synthesis rate (Ingebritsen, 1979). Stress and the interruption of the enterohepatic circulation also increase rat hepatic cholesterol synthesis (Andersen and Dietschy, 1977). Enterohepatic circulation experiments were done by administering the drug, cholestyramine, which binds bile acids in the intestine and prevents their reabsorption. Bile acid synthesis from cholesterol and cholesterol synthesis itself, are thus stimulated. Finally, feeding a high fat diet increases the rate of cholesterol synthesis in rat liver (Bortz, 1967). Fielding and Fielding (1985) warned that there is no method that determines the absolute rate of cholesterol synthesis. A commonly used index of this is the incorporation of label from pH]water or pH]water into cholesterol of various tissues. More commonly, in laboratory animals at least, the activity of hepatic HMG CoA reductase is used to assess the rate of cholesterol synthesis. This  8  acetyl CoA HMG CoA HMG CoA reductase mevalonate squalene lanosterol  dihydrolanosterol  —  methostenol  desmosterol CHOLESTEROL -  -  HO  LATHOSTEROL  HO  Figure 1.3 An overview of cholesterol biosynthesis.  9  enzyme is a transmembrane glycoprotein located primarily in the smooth endoplasmic reticulum. Tight control of cholesterol synthesis is accomplished by inactivation of HMG CoA reductase by reversible phosphorylation. The enzyme also exhibits a marked diurnal rhythm (Rodwell et al., 1976). A criticism of using HMG CoA reductase activity as a gauge of cholesterol synthesis is that at some times, for example, in late gestation in the rat, activity is not correlated with cholesterol synthesis as measured by [3H]water incorporation (Haave and Innis, 1991). This might be explained by the use of mevalonate for products other than cholesterol, such as dolichol and ubiquinone under certain physiological conditions (as discussed by Haave and Innis, 1991). In order to conduct in vivo studies of cholesterol metabolism conveniently, it is necessary to assay rates of cholesterol synthesis in the plasma. Established methods include the measurement of plasma mevalonate (Parker et al., 1984) and [2H]water incorporation into plasma cholesterol (Jones and Schoeller, 1990). The former involves laborious methodology while the pH]water technique requires relatively large blood volumes sampled repeatedly. Quantitation of plasma lathosterol by gas liquid chromatography is another, less well known, option (Farkkila and Miettinen, 1988). Plasma lathosterol is present both free and esterified and is carried on lipoproteins at concentrations 1000-fold less than cholesterol. It is believed that free lathosterol "leaks" out of cells at a rate proportional to the rate of cholesterol synthesis and is partially esterified in the plasma by LCAT (Kempen et al., 1988). Serum concentrations of lathosterol have been correlated with hepatic HMG CoA reductase activity in rats. Animals fed cholestyramine resin, which is known to increase the rate of cholesterol synthesis, had greatly increased serum lathosterol concentrations (Strandberg et al., 1989). The same correlation was also found in humans with and without cholestyramine resin administration (Bjorkhem et al., 1989). In addition, serum lathosterol concentration was indicative of rates of whole body cholesterol synthesis in humans (Kempen et al., 1988). Cholesterol balance was measured by subtracting cholesterol intake from fecal cholesterol excretion. Patients with familial hypercholesterolemia were also studied before and after treatment with MK-733, an inhibitor  10  of HMG CoA reductase. As predicted, serum lathosterol levels were lower after such treatment. The lathosterol:cholesterol ratio has also been used by many authors as an indicator of the rate of cholesterol synthesis (Farkkila and Miettinen, 1988, Kempen et al., 1988). Use of this ratio supposedly corrects for the number of lipoprotein (acceptor) particles in the serum (Kempen et al., 1988). The plasma lathosterol concentration has been used to study cholesterol synthesis in many circumstances such as the investigation of possible inhibitors of cholesterol synthesis (Kempen et al., 1988, Gylling et al., 1989, Reihner et al., 1990) and rates of cholesterol synthesis in various disorders including hepatic (Nikkila and Miettinen, 1988) and intestinal (Farkkila and Miettinen, 1988) diseases. One recent occasion when lathosterol was found to be an unsatisfactory indicator of cholesterol synthesis was with rabbits (Meijer et al., 1992). The sterol ratio was only reliable when comparing groups, but not individual animals.  1.2 Fetal cholesterol transport and metabolism The fetus is in an anabolic state where most energy requirements are met by carbohydrate delivered across the placenta. A large demand exists for cholesterol as a structural component for membranes and as substrate for steroid hormone synthesis. Possible sources for cholesterol include the maternal circulation, via the placenta, and de novo synthesis. Connor and Lin (1967) studied placental transport of cholesterol in the rabbit and guinea pig by feeding 4-[14C]cholesterol before and during pregnancy and estimated that 20-24% of fetal cholesterol was derived from the maternal circulation at some time during gestation. As a transport mechanism for cholesterol crossing the placenta has not been defined, the theory of placental transport of this sterol remains controversial. Regardless, it seems that the majority of fetal cholesterol is synthesized by the fetus. In humans, Carr and Simpson found relatively high rates of fetal (8-22 weeks gestation) hepatic cholesterol synthesis as indicated by both HMG CoA reductase activity (1981a) and [3H]water incorporation  11  (1982). Other human fetal tissues that synthesize cholesterol include the adrenal, testis, brain and ovary (Carr and Simpson, 1982). Relatively high rates of cholesterol synthesis have also been reported in fetal rat liver (McNamara et al., 1972, Haave and Innis, 1991). The relationship between the LDL receptor and cholesterol synthesis appears to be fully developed from early fetal life in humans. Cai et al. (1991) recently verified the presence of the LDL receptor in human fetal (20-32 week gestation) liver by immunoblotting. Goldstein et al. (1974) found that incubation of amniotic fluid cells (16 weeks gestation) in a lipoprotein-free medium stimulated HMG CoA reductase activity indicating the presence of a functional LDL receptor before birth. Other fetal and neonatal cells also seem capable of expressing functional LDL receptors. These include cultured fetal hepatocytes (10-24 weeks gestation, Carr and Simpson, 1984), newborn foreskin cells (Goldstein et al, 191A, Shakespeare and Postle, 1979), fetal lung fibroblasts strain MRC-5 (Shakespeare and Postle, 1979) and cord blood lymphocytes (Andersen and Johansen, 1980). Further evidence that the LDL receptor is likely to function the same way in utero as it does in the adult was obtained from a fetus who was diagnosed with homozygous familial hypercholesterolemia (Brown et ah, 1978). Its amniotic fluid cells had no functional LDL receptors and, as one would then expect, elevated plasma cholesterol levels were also detected. Human fetal cholesterol levels are highest before 16 weeks gestation (~2.2 mM) and fluctuate until term (Johnson et al, 1982). Johnson et al. (1982) hypothesized that the decrease in plasma cholesterol levels at 12-20 weeks gestation is because of a 10-fold increase in adrenal size, which has a large requirement for cholesterol. The rise in plasma cholesterol levels at 20-32 weeks gestation, they suggest, is because of liver growth causing heightened rates of cholesterol synthesis. Finally, the decrease in plasma cholesterol levels towards 40 weeks gestation coincided with the accelerated use of LDL-cholesterol by the fetal adrenal. LDL-cholesterol is a substrate for the synthesis of dehydroisoandrosterone sulfate which is a precursor for placental estrogens. Fetal plasma LDL-cholesterol and dehydroisoandrosterone sulfate levels are inversely correlated after 31  12  weeks gestation (Parker et al., 1983). Cai et al. (1991), however, believe that human fetal hepatic LDL receptors regulate fetal serum cholesterol levels. They found that receptor binding increased between 10-40 weeks gestation as LDL-cholesterol levels decreased. Both adrenal and hepatic LDL receptor levels may increase towards term. The exogenous pathway of lipoprotein transport is largely unused before birth except for the small amounts of lipid present in amniotic fluid (McConathy et al., 1981) which is continuously swallowed by the fetus. It is not surprising, therefore, that there is a virtual lack of chylomicrons before birth and that rat (Erickson et al., 1988), guinea pig (Bohmer et al., 1972), human (Averna et ah, 1991) and other newborn animal plasma lipoprotein profiles differ from adult. In the rat, most fetal plasma cholesterol is associated with LDL in contrast to HDL in the adult (Erickson et al., 1988). In the fetal guinea pig, cholesterol is fairly equally distributed between LDL and HDL, but HDL is present only in very small amounts in the adult (Bohmer et al., 1972). In humans, the major difference is that HDL is of greater prominence in the fetus than the adult; the LDL:HDLcholesterol ratio is 1.58 in cord blood but 2.54 in the adult (Averna et al., 1991). Also, plasma lipoproteins and apoproteins, with the exception of apo E, are at lower concentrations in cord than in adult plasma (McConathy and Lane, 1980, Blum et al., 1985, Tenenbaum et al., 1988, Averna et al., 1991). In adults, apo E is primarily associated with VLDL, but in cord plasma, large amounts of a subspecies of HDL containing apo E are present. Possibly, there is a unique metabolic role for HDL in the fetus as will be discussed in section 1.3.5. The fetal distribution of apoproteins among lipoproteins might be related to the lack of triglyceride-rich lipoproteins in the fetal circulation (Averna et al, 1991).  1.3 Postnatal cholesterol transport and metabolism After birth, energy requirements are met by free fatty acids from stored and ingested triglyceride. Plasma cholesterol and lipoprotein levels change rapidly. Data from the Bogalusa  13  Heart Study (Berenson et al, 1982) show that plasma cholesterol levels increase from —1.8 mM in cord blood to 3.5 mM at 6 months of age. However, most of this rise occurs within the first week after birth (Kaplan and Lee, 1965, Darmady et al, 1972, Potter, 1977, Van Biervliet et al., 1980, Stozicky et al., 1982, Kirstein et al., 1985). This phenomenon of postnatal increase in plasma cholesterol levels is seemingly universal among mammalian species (Table 1.1). In humans, within a month of birth, apoprotein levels have increased and the plasma lipoprotein profile approximates that of adults (Van Biervliet et al, 1980). Most of the increase in plasma cholesterol is due to LDL- and VLDL-cholesterol (Van Bierviet et al, 1980 and 1981, Kirstein et al, 1985, Tenenbaum et al, 1988). This is generally true of most species studied, exceptions being the lamb (Leat et al, 1976) and the pig (Hollanders et al, 1985) in which HDL levels increase as well. In humans, plasma cholesterol levels continue to increase in childhood, decrease slightly at adolescence (Berenson et al, 1982) and then rise throughout adult life. In contrast, among most of the species listed in Table 1.1, plasma cholesterol levels decrease again at weaning. An exception to the continual increase in plasma cholesterol levels in humans, was reported by Whyte and Yee (1958). These authors found a gradual decrease in plasma cholesterol levels in children from New Guinea after 1 year of age. This could reflect a different diet from that typically found in North America. Roberts et al (1979) suggest that the maintenance of higher cholesterol levels in humans versus other animals is the result of weaning to a diet richer in fat and cholesterol than many typical adult animal diets. These authors maintained elevated plasma cholesterol levels in suckling rabbits by weaning them to a high fat and cholesterol diet. The same was demonstrated in calves (Wiggers et al, 1971, Jacobson et al, 1973) and monkeys (Greenberg, 1970). The influence of weaning on plasma cholesterol levels was also shown in rats (McNamara et al, 1972) and pigs (Carroll et al, 1973), where early weaning accelerated and late weaning delayed the fall in plasma cholesterol levels normally found when solid food was introduced.  14  Table 1.1 Animals experiencing elevated postnatal plasma cholesterol levels Animal  Reference  cat  Hamilton and Carroll, 1977  dog  Hamilton and Carroll, 1977  sheep  Carroll et al., 1973  rabbit  Friedman and Byers, 1961 Laird and Fox, 1970 Roberts et al., 1979  Pig  Carroll <tf al., 1973 Johansson and Karisson, 1982 Hollanders et al., 1985 Jones et al., 1990  guinea pig  Drevon and Norum, 1975 Bohmer et al., 1972 Li et al., 1979a  rat  Viktora«f a/., 1960 Bizzi et al., 1963 Hahn and Koldovsky, 1966 Harris et al., 1966 Ness et al., 1979 Johansson, 1983 Haave and Innis, 1991  cow  Carroll et al., 1973 Rudling and Peterson, 1984  15  Although it seems clear that diet affects plasma cholesterol levels near the time of weaning in many species, the initial increase in plasma cholesterol levels after birth might not be dependent upon diet. Some hypotheses have been offered to explain the postnatal rise in plasma cholesterol levels, but a diversity of opinion still exists. These theories, outlined below, range from obvious changes that occur at parturition, such as the shifts in nutrient supply and hormones, to a potential maturation of cholesterol homeostasis.  1.3.1 Changes in energy substrate At birth, there is a shift from primary utilization of carbohydrate for energy to utilization of fat (Van Duyne and Havel, 1959). Csako et al. (1974) speculated that increased plasma cholesterol is an indirect result of the hypoglycemia found after birth before enteral feeding is established and when glycogen reserves have been depleted. Hypoglycemia results in adipose tissue triglyceride lipolysis and increased plasma free fatty acid concentrations. The liver takes up this free fatty acid, reesterifies it and packages it into VLDL for export into the plasma. Rates of cholesterol synthesis would, thus, probably increase to provide cholesterol for VLDL synthesis. Increased plasma LDLcholesterol levels might occur upon metabolism of this VLDL. Also, as mentioned earlier (section 1.1.1), free fatty acids may independently cause elevation of plasma LDL levels by inhibiting binding of LDL to the LDL receptor. The levels of free fatty acids used to study this phenomenon, however, were above the levels generally found in human plasma (Bilhain et al., 1989). Not considered by these authors was the potential of high plasma free fatty acid concentrations after birth affecting the LDL receptor.  1.3.2 Initiation of enteral feeding It is commonly suggested that "postnatal hypercholesterolemia" results from the onset of feeding (Carroll et al, 1973, McConathy and Lane, 1980, Strobl et al, 1983, Andersen, 1985). As  16  discussed below, dietary cholesterol, triglyceride and protein have all been linked with this phenomenon. Additionally, the compostion of dietary fat has been investigated in this context. Milk contains more cholesterol than the postnatal diets of many animals. Friedman and Byers (1967) found that dietary cholesterol was necessary for the postnatal elevation of plasma cholesterol in rabbits. Carroll et al. (1973) hypothesized that the even higher cholesterol and lipid content of colostrum compared to milk was responsible for the postnatal elevation of plasma cholesterol in several species. More recently, however, several authors have shown that in human infants, plasma cholesterol levels in the first week after birth, are equally increased above cord levels with both human milk and formula feeding (Potter, 1977, Ginsberg and Zetterstrom, 1980, Van Biervliet et al., 1981, Lane and McConathy, 1986). Human milk contains 0.25-0.50 mM cholesterol (Jensen, 1989) while formulas contain little. In the pig, also, there was no difference between plasma cholesterol levels of 5 day old pigs fed by the sow and those fed formula from birth (Jones et al., 1990). Although the many compositional differences between milk and formulas were not controlled for in these studies, the data suggest that dietary cholesterol is not necessary for the early postnatal elevation of plasma cholesterol levels. To prove this, one would need to measure plasma cholesterol levels in humans or animals fed formula with and without added cholesterol. Unfortunately, this experiment has not been done with blood sampled during ihs first week after birth. Harris et al. (1966) concluded that dietary triglyceride, and not cholesterol, was responsible for elevated cholesterol levels in suckling rats at 20 days of age. The effect of dietary triglyceride on the initial postnatal rise in plasma cholesterol levels is unknown. This would be difficult to study as triglyceride feeding is necessary to meet caloric requirements after birth and, thus, cannot be eliminated from the diet. In adult guinea pigs (Fernandez and McNamara, 1991) and rabbits (Bauer et al., 1990), however, fat feeding does increase plasma cholesterol levels. The mechanism of this effect may be down-regulation of LDL receptor number (Angelin et ah, 1983, Rudling and Peterson,  17  1985, Fox et al., 1987), or increased hepatic cholesterol synthesis (Fernandez and McNamara, 1991). The type of dietary fat may also have an effect on early postnatal plasma cholesterol levels. Human milk triglyceride contains -30% saturated fatty acids (Jensen, 1989). Infant formulas, in contrast, are relatively enriched in polyunsaturated fatty acids and contain little long-chain saturated fatty acids. In adults of many species, dietary polyunsaturated fatty acids are hypocholesterolemic when compared to saturated fatty acids (human: Hegsted et al., 1965, Shepherd et al., 1980, hamster: Spady and Dietschy, 1988, rhesus monkey: Chong et al., 1987). The effect of the degree of saturation of dietary fat may be mediated by changes in LDL receptor expression (Fox et al., 1987, Ibrahim and McNamara, 1988, Fernandez and McNamara, 1989). In older infants (3-16 months of age), too, plasma cholesterol levels can be influenced by the degree of fat saturation in the same way as adults (Nestel et al., 1979). One would predict, therefore, that young infants fed human milk would have higher plasma cholesterol levels than infants fed formula. But, as mentioned earlier, in at least the first week after birth, this is not the case. Because the composition of these diets differ in so many ways, similarities in plasma cholesterol levels between infants fed human milk and artifical formula may only be a net effect. For example, saturated fatty acids in human milk may induce an increase in plasma cholesterol levels but insulin and triiodothyronone in milk may cause up-regulation of the LDL receptor and decreased plasma cholesterol levels by increasing receptor-mediated clearance. The observation that human milk versus formula feeding results in higher plasma cholesterol levels in older infants (Darmady, 1972), suggests, though, that one week of age is not sufficient for differences to occur. Dietary protein also affects plasma cholesterol levels. Carroll and Huff (1977) found that rabbits weaned to a diet containing casein rather than soy protein had higher plasma cholesterol levels. This effect occurred only after weaning, but not in the suckling period (Roberts et al., 1979). Finally, a number of studies where sweet tea alone was fed to newborn infants for the first 3-  18  5 days after birth were done in the late 1950's (reviewed by Hahn and Koldovsky, 1966). Plasma cholesterol levels still rose in these infants, but not to the extent of those fed human milk. One might conclude from this that diet is not solely responsible for the postnatal increase in plasma cholesterol levels; however, the potential effect of energy deprivation on plasma cholesterol levels was not considered. Little and Hahn (1990) in a recent review of this field, state that the postnatal changes in plasma cholesterol levels depend, to some extent, on the degree of consumption of milk, although other factors, undoubtedly, are involved too. After 40 years of study, little more is known of the influence of diet on the phenomenon of "postnatal hypercholesterolemia".  1.3.3 Change in hormonal milieu McConathy and Lane (1980) and Strobl et al. (1983) hypothesized that a change in hormonal milieu at birth influences plasma cholesterol levels. As noted earlier (section 1.1), several hormones are regulators of cholesterol synthesis and LDL receptor binding and expression, and thus affect cholesterol metabolism in adult animals. Some specific hypotheses relating changes in hormones to plasma cholesterol levels after birth have been proposed. First, Plonne" et al. (1990) suggest that a postnatal decrease in insulin causes decreased LDL receptor binding and increased plasma LDL levels in rats. A decrease in plasma insulin levels occurs in humans and rats in the first few hours after birth (Blazquez et al., 1974). More important may be the ratio of insulin to glucagon, which also changes after birth (Blazquez et al., 1974). Another hypothesis, by Rudling and Peterson (1985), is that a postnatal decrease in estrogens, and possibly estrogen receptors, causes increased plasma cholesterol levels by down-regulating (actually removal of up-regulation) the LDL receptor number. In humans, between birth and 8 days of age, serum estradiol decreases from approximately 700 ng/dl to 1-2 ng/dl (Winter et al, 1976). Growth hormone production is increased at birth, in response to hypoglycemia. This causes mobilization of fatty acids from adipose tissue (Csako et al, 1974), which, as mentioned earlier (section 1.3.1), would result in VLDL production by the liver  19  and consequently an increase in cholesterol synthesis and plasma cholesterol levels. Finally, the hormones present in human milk (Koldovsky and Thornburg, 1987) may affect cholesterol metabolism after birth. Nonetheless, this would not account for changes in plasma cholesterol found in infants fed formula.  1.3.4 Change in cholesterol homeostasis Brown and Goldstein (1984) hypothesized that there are more LDL receptors in newborn humans (similar levels as in adult experimental animals) than adult humans. This, they suggest, may be responsible for the lower plasma LDL-cholesterol levels in cord compared to adult plasma. Supportive evidence for this theory comes from animal studies by Rudling and Peterson (1985) who found higher binding of [125I]LDL to fetal than young (<2 months of age) or adult bovine liver homogenate. They also found an inverse correlation between plasma cholesterol levels and receptor binding leading to the hypothesis that the change in cholesterol levels after birth is primarily the result of decreased numbers of hepatic LDL receptors. Increased binding of LDL to hepatic membranes was also found in young versus adult rats (Erickson et al., 1988, Auestad et al., 1991) and dogs (Mahley et al., 1981), and in fetal versus adult pigs (Mahley et al., 1981). These data from animals contradict the LDL receptor theory (Brown and Goldstein, 1986) because adult animals have lower plasma cholesterol levels than young animals (opposite to findings in humans). One would have expected relatively higher LDL receptor numbers in the adult animals. Studies of human cultured fibroblasts (Shakespeare and Postle, 1979) and lymphocytes (Andersen and Johansen, 1980) revealed no age related differences in LDL receptor binding. Results of these experiments are perhaps not surprising for a number of reasons. First, cultured fibroblasts may not be similar to hepatocytes in this respect. Also, the cells were in isolation from any possible modulating factors present in the plasma, in vivo. And finally, incubation of cells in lipoprotein-free serum was used to maximally up-regulate the receptor number and cholesterol synthesis. Differences  20  between ages might disappear under these circumstances. Increased numbers of hepatic LDL receptors in young versus adult animals may be necessary to meet the cholesterol needs of the growing liver (Mahley et al., 1981, Plonne" et al., 1990). Similarly, some tumour cells also have relatively large numbers of LDL receptors presumably because of the high cholesterol requirement for cell replication (Lombardi et al., 1989). It seems improbable that there is a sudden decrease in demand for cellular cholesterol at birth leading to a decrease in LDL receptor activity. Newborn animals remain in a period of rapid growth for some time. Instead, one might expect that a decrease in the receptor may be the effect of increased plasma LDL-cholesterol levels. Increased LDL-cholesterol levels could come about because of a reduction in demand for cholesterol by a particular organ, such as the adrenal, discussed below.  1.3.5 Reduced uptake of lipoproteins by the adrenal The fetal zone of the human adrenal has a high requirement for cholesterol for the production of steroid hormones (Carr and Simpson, 1981b). Considerable evidence suggests that this organ obtains cholesterol from plasma LDL (Carr and Simpson, 1981b) via the LDL receptor (Brown et al., 1979). Carr and Simpson (1981b) hypothesized that when the fetal zone of the adrenal regresses at birth, the demand for LDL-cholesterol suddenly decreases and as a result, plasma LDL-cholesterol levels increase. In support of this theory, it was found that anencephalic fetuses, whose adrenal glands are atrophied, have diminished numbers of adrenal LDL receptors (Carr et al., 1980) and plasma LDL-cholesterol levels 3-fold higher than normal fetuses (Parker et al., 1980). The proposed importance of LDL-cholesterol in supplying the adrenal with cholesterol is compatible with the hypothesized role for the high levels of the subspecies of HDL containing apo E found in human fetal blood. This particle can interact with the LDL (B/E) receptor with greater affinity than LDL because of the larger number of apoprotein molecules per lipoprotein (i.e. there is only one apo B per LDL particle). It was thus suggested, therefore, that HDL containing apo E could deliver  21  cholesterol to peripheral cells to compensate for the low plasma LDL levels in fetal blood (Innerarity et al, 1984, Blum et al, 1985, Van Biervliet et al, 1986). Criticism of the "adrenal" hypothesis is two-fold. First, it does not account for the increase in VLDL-cholesterol seen after birth (Andersen, 1985). VLDL production, however, could probably start independently of changes in the adrenal as hypothesized by Csako et al. (1974). Also, the adrenal theory is dependent on the regression of the fetal zone of the adrenal, which is not an anatomical feature shared by all species experiencing "postnatal hypercholesterolemia".  1.3.6 Physiologic cholestasis At birth, bile acid synthesis and secretion is immature (Lester, 1980). In the first few days postpartum, bile flow rates are slow and the bile acid pool size is small. One might expect that as the 2 modes for cholesterol excretion (bile acid production and cholesterol in bile) are slow to develop, that plasma cholesterol levels would be high after birth. Li et al. (1979a) hypothesized that in guinea pigs, immature bile acid production, as measured by 7a-hydroxylase, was responsible for "postnatal hypercholesterolemia". On the contrary, later work by this group (Yunker and Subbiah, 1985) demonstrated that the activity of this enzyme was actually higher in the neonate than the adult and decreased after birth. In the rat, though, the activity of 7a-hydroxylase is low during the suckling period (Hahn and Innis, 1984). It is possible that the combination of an influx of dietary cholesterol together with cholestasis after birth may result in elevated plasma cholesterol levels. One would expect that hepatic LDL receptors and cholesterol synthesis would consequently be down-regulated in reponse to increased hepatic cholesterol levels.  1.3.7 Efflux of lipid from tissue stores The rapid metabolic changes that occur at birth may be expected to cause redistribution of  22  cholesterol pools. In particular, the initiation of exogenous cholesterol metabolism and the enterohepatic circulation may result in changes in the liver cholesterol pool. Also, some species (i.e. humans and guinea pigs) have relatively fatty livers at birth which lose lipid after birth (Widdowson and McCance, 1955). Hepatic cholesterol may also be released from the liver into the plasma compartment. Another tissue that regresses postnatally in some species is adipose tissue. Adipose tissue may be important for cholesterol storage (Krause and Hartman, 1984) and release significant amounts of cholesterol into the plasma.  1.4 Preterm infant plasma cholesterol levels Normal changes in preterm infant plasma cholesterol levels after birth are less well defined than for infants born at term. There are many reports of hypercholesterolemia in this population during intravenous lipid administration (Higgs et al., 1974, Franklin et al., 1976, Griffin et al., 1979, Bargen-Lockner et al, 1983, Kao et al, 1984, Rovamo, 1985, Berkow et al, 1987, Cooke et al., 1987, Brans et al, 1988 and 1990, Haumont et al, 1988, Lima et al, 1988). The plasma cholesterol levels observed are generally greater than those measured either in utero (Johnson et al., 1982) or in healthy term infants (Potter, 1977). Hypercholesterolemia also occurs in term infants (Griffin et al, 1979, Dahms and Halpin, 1980, Bargen-Lockner et al, 1983, Rovamo, 1985) and adults (Thompson et al, 1975, Untracht, 1982) administered intravenous lipid. In contrast, hypercholesterolemia does not occur in enterally fed preterm infants (Genzel-Boroviczeny et al., 1988, T>o&c\etal, 1990). Intralipid is an emulsion of soybean oil triglyceride and egg yolk phospholipid commonly administered to patients unable to tolerate adequate enteral nutrition. These emulsion particles are meant to simulate chylomicrons. Once infused, they associate with C apoproteins and then undergo triglyceride hydrolysis by lipoprotein lipase (Untracht, 1982, Messing et al., 1990). The resulting phospholipid-rich vesicles, along with excess phospholipid from the emulsion, combine with more C  23  apoproteins, albumin and equimolar amounts of free cholesterol to form particles which are commonly referred to as lipoprotein X (LpX). An indicator of the presence of LpX is a high ratio of free:esterified cholesterol in the plasma. LpX has a half-life of approximately 2 days in the plasma and is thought to be taken up, unmetabolized, by the liver (Untracht, 1982). A recent report by Tashiro et al. (1991) confirms that essentially all of the hypercholesterolemia occurring in adult humans during 10% Intralipid infusion can be accounted for by increased plasma levels of LpX. Another common result of lipid infusion in preterm infants of less than 33 weeks gestation is hypertriglyceridemia. Reasons for this perhaps include immature lipoprotein lipase activity (Shennan et al., 1977, Dhanireddy et ah, 1981), low serum carnitine levels (Rovamo et al, 1988), hepatic immaturity (Filler et al., 1980) and a small adipose tissue mass (Shennan et al, 1977, Filler et al., 1980, Dhanireddy et al, 1981). Even if triglyceride is cleared, however, infants still experience hypercholesterolemia which develops more rapidly than in adult patients (Franklin et al., 1976, Papadopoulos et al., 1988). One hypothesis to explain this earlier hypercholesterolemia in preterm infants is immature LCAT activity (Jain, 1985, Amr et al, 1988, Papadopoulos et al, 1988). With low LCAT activity, phospholipid "clearance" from the plasma might be impaired. LpX is also found in the plasma with familial LCAT deficiency (Glomset et al, 1973). Perhaps exacerbating the situation in preterm infants is the possible further decrease in LCAT activity that has been associated with lipid infusion in humans (Untracht, 1982) and rats (Amr et al., 1988). Innis et al. (1985) and Messing et al. (1990), however, found no decrease in LCAT activity during total parenteral nutrition in adult humans. Further, Forte et al. (1989) reported that Intralipid infusion actually increases LCAT activity in preterm infants. In support of the theory that excess phospholipid is responsible for hypercholesterolemia in preterm infants infused with 10% Intralipid, Haumont et al. (1989) found no hypercholesterolemia in infants infused with a 20% solution of Intralipid. 20% Intralipid contains half the phospholipid (per kilojoule) of a 10% Intralipid solution.  24  According to Griffin et al. (1979), approximately 50% of the free cholesterol necessary for LpX formation originates in tissue cellular membranes. To replace membrane cholesterol lost to the plasma, cellular cholesterol synthesis rates are expected to increase. This hypothesis is supported by rat studies in which increased HMG CoA reductase activity was reported after phospholipid or Intralipid infusion (Jakoi and Quarfordt, 1974, Innis and Boyd, 1983). Also, in vitro incubation of mouse fibroblasts (Bartholow et al., 1982) and rat hepatoma cells (Yau-Young et ah, 1982) with phospholipid containing solutions caused sterol efflux.  1.5 Significance Numerous authors have reported that factors in the perinatal period affect cholesterol metabolism in later life. For example, Innis (1983) found in rats that maternal cholestyramine feeding during gestation and the first 14 days of lactation, caused increased hepatic and intestinal HMG CoA reductase activity in adult animals and increased plasma cholesterol levels when adults were challenged with a high fat and cholesterol diet at 3 months of age. Some findings of other authors, of late effects of postnatal diet, are outlined in Table 1.2. The ontogeny of cholesterol metabolism must be understood before the mechanisms of the effects of early nutrition are elucidated. The hypercholesterolemia that occurs in preterm infants administered Intralipid may be an example of a perturbation of cholesterol metabolism in the neonatal period with long term implications.  1.6 Rationale The cause of the rise in plasma LDL-cholesterol levels soon after birth has not been definitively established. Two factors likely to be involved in many of the proposals detailed in section 1.3 are decreased LDL receptor number and increased cholesterol synthesis. The hypothesis that the change in energy substrate at birth would cause increased plasma cholesterol levels relies on  25  Table 1.2 Some examples of late effects of early nutrition Species  Neonatal manipulation  Adult effect  Reference  rabbit  animals suckled on high or low cholesterol milk  hypercholesterolemia in response to dietary cholesterol if suckled on high cholesterol milk  Roberts and West, 1974  pig  fed high or low cholesterol formula  t plasma cholesterol in response to a challenge of dietary cholesterol if fed a low cholesterol formula neonatally  Reiser et al., 1979  rat  early wean to high carbohydrate diet  t plasma cholesterol  Hahn and Kirby, 1973  rat  lactating dams fed a semipurified or stock diet  higher HMG CoA reductase activity was found in pups fed stock diet, which persisted into adulthood  Reiser et al.. 1977  rat  lactating dams fed high fat and cholesterol diet  I HMG CoA reductase activity and 17ahydoxylase activity in response to a challenge of a high fat and cholesterol diet  Naseem et al., 1980  rat  early weaning  J hepatic HMG CoA reductase activity in response to a challenge of dietary cholesterol  Naseem et al., 1980  rat  pregnant and lactating dams and weanling rats fed high fat or sucrose diets  t plasma cholesterol levels in response to a challenge of high fat if fed a high fat diet neonatally  Coates et al., 1983  pregnant and lactating dams and weanling rats fed vegetable oil or animal fat diets  i plasma cholesterol and I HDL-cholesterol in response to a high fat and cholesterol challenge if fed animal fat neonatally  O'Brien et al., 1983  rat and mouse  raised in small (n=4) or large litters (n= 14) from birth  small litters had t hepatic HMG CoA reductase activity and large litters had t adipose tissue HMG CoA reductase activity  Hahn, 1984  guinea pig  cholestyramine feeding  I plasma cholesterol and t sterol excretion in response to a challenge of dietary cholesterol  Metal.,  guinea pig  premature weaning  * 7a-hydroxylase activity  Subbiah et al., 1985  baboons  suckled versus formula fed  suckled animals had I HDL-cholesterol, cholesterol synthesis and steroid excretion and t VLDL+LDL/HDL ratio  Mottetal.,  26  1980  1990  increased rates of cholesterol synthesis for export of VLDL from the liver. The potential effect of the ingestion of dietary fat may also involve decreased LDL receptor number or increased hepatic cholesterol synthesis. A direct developmental decrease in hepatic LDL receptors may also affect an increase in plasma cholesterol levels. If numbers of LDL receptors are reduced at birth, less circulating LDL would be cleared and plasma concentrations would increase. Also, if the adrenal uptake of LDL from the plasma was suddenly decreased, the subsequent increase in plasma LDL levels could cause down-regulation of hepatic LDL receptor number and cholesterol synthesis. LDL receptors and cholesterol synthesis were investigated in the guinea pig and cholesterol synthesis was studied in human infants. The guinea pig is an appropriate animal for studying the development of cholesterol metabolism for several reasons. First, like most animal species, it exhibits increased plasma LDLcholesterol levels soon after birth. It should be noted here, however, that Hamilton and Carroll, (1977) reported no "postnatal hypercholesterolemia" in the guinea pig. This may be due to sampling blood at birth and 10 days of age, thus missing the peak in plasma cholesterol levels which has been since found at 5 days of age (Li et al., 1979b). Second, guinea pigs are similar to humans in that a large percentage, 10.1%, of their body weight at birth is fat (Widdowson, 1950). Fat accounts for 16.1% of body weight in newborn humans and 1-2% in newborn rats, mice, rabbits, cats and pigs (Widdowson, 1950). Also, similar to humans but unlike rats, guinea pigs have 1) relatively fatty livers at birth (Widdowson and McCance, 1955), 2) most of their plasma cholesterol associated with LDL rather than HDL (Bohmer et al., 1972) and 3) a hepatic contribution to whole body cholesterol synthesis similar to primates (Spady and Dietschy, 1983). Finally, the newborn and young guinea pig is a convenient animal to study because of its relatively large size at birth. The ontogeny of cholesterol synthesis has not been previously investigated in humans. Increased rates of cholesterol synthesis might account for the elevated postnatal levels of cholesterol that occur independent of the cholesterol content of the diet (Potter, 1977). One might expect that if  27  the plasma cholesterol levels are similar in infants fed formula (little cholesterol) and human milk, that rates of cholesterol synthesis may be higher in those fed formula. Support for this hypothesis is from Levin et al. (1989) who speculate that the low level of HMG CoA reductase mRNA in suckling rats is due to increased dietary cholesterol. Also, Jones et al. (1990) found higher rates of cholesterol synthesis (as indicated by HMG CoA reductase activity and [3H]water incorporation into cholesterol) in 5 day old pigs fed formula rather than sow milk. The common method of assessing the rate of cholesterol synthesis, assay of hepatic HMG CoA reductase activity, is unsuitable for use in infants because of the need for liver biopsy. Plasma lathosterol levels, however, can readily be measured in very small plasma samples and offer a potential indicator of cholesterol synthesis rates in human infants. Although Intralipid (10%) associated hypercholesterolemia is thought to be partially the result of enhanced cholesterol synthesis, this has not been studied in the preterm infant. Plasma lathosterol, therefore, might also be useful as a potential indicator of the rate of cholesterol synthesis in preterm infants requiring Intralipid administration who were studied from birth until approximately 4 weeks of age.  1.7 Thesis objectives  The objectives of this thesis were the following:  1) to measure some aspects of cholesterol metabolism (LDL receptor mass, hepatic HMG CoA reductase activity, plasma lathosterol concentrations and hepatic cholesterol content) in young and adult guinea pigs to determine if changes in these coincide with developmental changes in plasma cholesterol and lipoprotein levels.  28  2) to determine whether or not the plasma lathosterol concentration increases between birth and 4 days of age in term gestation human infants. In addition, human milk and formula feeding were compared to determine whether or not infants fed formula have higher plasma lathosterol concentrations than infants fed breast milk.  3) to determine whether or not the hypercholesterolemia experienced by preterm infants administered 10% Intralipid is associated with enhanced plasma lathosterol concentrations.  1.8 Specific aims 1.8.1 Guinea pig experiments To measure the following in young (1, 4, and 8 days of age) and adult guinea pigs: 1) hepatic lipid and cholesterol content (Experiment 1) 2) plasma cholesterol and lathosterol concentration (Experiment 2) 3) plasma cholesterol concentration, hepatic HMG CoA reductase activity and hepatic LDL receptor relative mass (Experiment 3) 4) the distribution of cholesterol among plasma lipoproteins (in some samples from Experiments 1 and 2)  1.8.2 Term infant study To determine plasma lathosterol, cholesterol and apo AI and B (indicative of HDLand LDL+VLDL-cholesterol, respectively) concentrations in blood sampled from term infants at birth (cord) and 4 days of age. To determine the effect of breast milk and formula feeding on the above concentrations.  29  1.8.3 Preterm infant studies To document the association between plasma cholesterol levels and 10% Intralipid administration at this facility (Preliminary Cholesterol Study). To measure plasma lathosterol, total and free cholesterol and apo AI and B concentrations longitudinally in preterm infants (23-32 weeks gestation at birth) and to determine whether changes found coincided with parenteral administration of 10% Intralipid.  30  Chapter Two  METHODS AND MATERIALS  2.1 Chemicals Unless otherwise noted, all chemicals were purchased from Sigma Chemical Co., St. Louis, MO. Solvents were reagent grade and obtained from BDH Chemicals Canada Ltd., Vancouver, BC. Plasma total cholesterol was measured using the enzymatic kit produced by Diagnostic Chemicals Ltd., Charlottetown, PE. Free cholesterol was assayed using the kit of Boehringer Mannheim Canada Ltd., Dorval, PQ. The guinea pig anaesthetic, sodium pentathol (Somnotol) was from M.T.C. Pharmaceutical, Hamilton, ON. Guinea pig and rabbit chow were from Purina Mills, Inc., St. Louis, MI. The silylating reagents, hexamethyldisilane, trimethylchlorosilane and dimethylformamide for gas chromatography were purchased from Pierce, Rockford, IL. DL[methyl-3H]HMG CoA (10.9 Ci/mmol) and [2-14C]mevalonic acid (50.1 mCi/mmol), necessary for the HMG CoA reductase assay, were obtained from Dupont (Canada) Inc., Missisauga, ON. Aqueous scintillation fluid was from Amersham Canada Limited, Oakville, ON. Unless otherwise mentioned, ligand blotting chemicals and supplies were purchased from Bio-Rad Laboratories, Mississauga, ON. Colloidal gold (G10, Aurobeads) was also from Amersham. Sodium suramin was from Miles, Inc., FBA Pharmaceuticals, West Haven, CT. Goat anti-human apo B IgG was obtained from Medix Biotech Inc., Foster City, CA, and rabbit anti-goat IgG and goat peroxidase-anti-peroxidase were from Bio/Can Scientific, Mississauga, ON. Reagents and immunochemicals necessary for apoprotein analyses were purchased from Beckman Instruments, Inc., Palo Alto, CA.  2.2 Clinical supplies Infant formula Enfalac, Enfalac 20 and ProSobee were from Mead Johnson, Belleville, ON and Similac Special Care 68 and 81 and Pregestimil from Ross Laboratories, Columbus, OH.  31  Intravenous glucose (D5W or DlOW) was produced by Baxter Corp., Toronto, ON and amino acids (Vamin A or B) and 10% Intralipid by KabiVitrum Canada Ltd. Newmarket, ON. Cholestyramine resin, Questran"' was purchased from Bristol Laboratories, Belleville, ON.  2.3 Equipment Centrirugation at < 10,000 g was performed in a J-6B low speed centrifuge from Beckman Instruments Inc., with a swinging bucket rotor. Procedures requiring centrirugation at 10,000 g were done in a high speed International Equipment Company (Needham Heights, MA) centrifuge and rotor (#874). Beckman ultracentrifuges (L8-55 and L7-55) and rotors (50 Ti, 50.2 Ti and 50.3 Ti) were used for 100,000 g spins. Other items obtained from Beckman included Quickseal ultracentrifuge tubes for ultracentrifugation of density gradients, a beta-counter (LS9000) used for measuring 3H and an Array Protein System for measuring apoproteins by nephelometry. Gas chromatography was performed on a 3400 gas liquid chromatograph from Varian Canada Ltd., Georgetown, ON equipped with flame ionization detection, a Varian 402 data system and a RTx-1, 25 m capillary column, 0.25 mm internal diameter, 0.25 /un thickness from Restek, Corp., Bellefonte, PA. Electrophoresis and electroelution were performed on a Bio-Rad Mini-Protean II apparatus and transfer cell. The dialysis tubing (Spectra/Por, molecular weight cutoff 12,000-14,000) was from Baxter Canlab, Burnaby, BC. Densitometry was performed on a video densitometer (Model 620) also purchased from Bio-Rad. Homogenization was performed using a Polytron homogenizer from Janke and Kunkel GMBH and Co., Staufen, Germany and a tissue homogenizer (Potter Elvejhem) from Eberbach Corporation, Ann Arbor, MI. Statistical comparisons were performed with the aid of the Number Cruncher Statistical System, version 5.1 (Kaysville, UT).  32  2.4 Guinea pig experiments 2.4.1 Experimental design Data were collected over 3 years from 3 different groups of animals. This was due to the gradual development of the thesis project experimental goals and methodologies used. Hepatic lipid and cholesterol were measured in the first experiment, plasma cholesterol and lathosterol in the second and plasma cholesterol, hepatic microsomal HMG CoA reductase activity and LDL receptor relative mass (on hepatic plasma membranes) in the third. Lipoprotein analyses were performed on samples from experiments 1 and 2 when plasma volumes were sufficient (~2 mL).  2.4.2 Guinea pig housing and care Guinea pigs (Hartley) were purchased from the Animal Unit of the Department of Zoology, University of British Columbia. Animals were housed in large plastic bins lined with woodshavings and given free access to guinea pig chow (St. Louis, MO), alfalfa cubes and drinking water supplemented with ascorbic acid (50 mg/L, fresh daily). Lights were on from 0630-1830 h and off from 1830-0630 h.  2.4.3 Guinea pig sacrificing Guinea pigs were anaesthetized intraperitoneally with sodium pentathol (0.1 mg/kg body weight, between 1130 and 1230 h). Neonates of both sexes were sacrificed at 1 d (6-24 h), 4 d (+ 12 h) and 8 d (+ 12 h) of age. No more than 2 littermates were present in each age group. Adult male animals in experiments 1 and 2 were 600-1,200 g, while the animals in experiment 3 were younger (350-400 g).  33  2.4.4 Preparation of tissues 2.4.4.1 Plasma and liver Blood was collected, by puncture of the abdominal aorta, into tubes containing EDTA (1 mg/10 mL) and 5,5'-dithiobis-(2-nitrobenzoic acid) (1 mg/10 mL) and plasma was obtained following centrifugation (4°C at 1,000 g for 15 min). Plasma was stored at -70°C unless used for lipoprotein analysis. The liver was quickly excised, rinsed, blotted dry, weighed and minced (on ice). Liver was stored at -70°C unless used for microsome or plasma membrane preparation.  2.4.4.2 Microsomes All of the following procedures were performed at 4°C. The liver was homogenized in 3 volumes 225 mM sucrose with 25 mM Tris HCl (pH 7.8), 10 mM glutathione, 25 fiM leupeptin and 25 fiM aprotinin. Homogenization was with 3 strokes of a Potter-Elvejhem homogenizer. The homogenate was centrifuged for 20 min at 10,000 g and the resulting supernatant centrifuged for 1 h at 100,000 g. The microsomal pellets were resuspended in 100 mM sucrose containing 50 mM KC1, 40 mM KH2P04, 30 mM EDTA and 20 mM dithiothreitol (pH 7.2).  2.4.4 Plasma membranes Plasma membranes were prepared according to Kovanen et al. (1979b). All procedures were done at 4°C. In brief, liver was homogenized with a Polytron homogenizer at setting 10 for 10 s, twice in 10 volumes buffer A (10 mM Tris (pH 7.5), 1 mM CaCl2, 150 mM NaCl, 25 pM leupeptin and 25 /iM aprotinin). The homogenate was centrifuged for 15 min at 500 g and the resulting supernatant again for 15 min at 10,000 g. This final supernatant was then centrifuged for 1 h at 100,000 g. The pellet was washed twice by resuspending in buffer A and recentrifuging and then stored at -70°C.  34  2.4.5 Plasma determinations 2.4.5.1 Cholesterol Plasma total cholesterol was measured by enzymatic kit assay using 10-20 (il aliquots of sample. The cholesterol standard provided with the kit was used to construct a standard curve. Quality control samples (purchased aliquots of normal serum) were run with each assay. The intraassay and interassay coefficients of variation of the cholesterol assay kit were <2%.  2.4.5.2 Lathosterol Plasma lathosterol was extracted and quantitated by gas liquid chromatography as previously described (Bjorkhem et al., 1987, Farkkila and Miettinen, 1988) with minor modification. In brief, 5a-cholestane (internal standard) was added to 25-100 fi\ plasma and the mixture was saponified with 50% KOH and methanol (6:94, v/v) for 1 h at 80°C. The nonsaponifiable lipid was extracted 3 times with petroleum ether, dried and silylated with hexamethyldisilane: trimethylchlorosilane: dimethylformamide, 20:2:5, v/v/v, 5 min at room temperature. The sterols were then injected into the gas chromatograph. Samples were injected at 80°C and after remaining at this temperature for 1 min, the oven temperature was programmed to increase to 120°C (20"C/min), hold for 7 min, rise to 249°C (20°C/min), hold for 15 min, rise to 269°C (20°C/min), and hold for 20 min. The oven was then heated to 320°C (20°C/min) for 5 min prior to subsequent analyses. The injector and detector were set at 300°C and 320°C, respectively. The carrier gas (helium) flow rate was 1.25 mL/min with the inlet splitter set at 100:1 and the relays programmed to come on at 0.7 min after the run start. Lathosterol peak identification was confirmed using an authentic standard. A plasma sample was used as a quality control and run every 2 days during the gas liquid chromatography analysis. The interassay coefficient of variation was 11%. Intraassay variation could not be properly measured as samples were injected 5 minutes after silylation. Repeating this injection to determine intraassay  35  variability would have changed the assay conditions. As done previously (Farkkila and Miettinen, 1988) lathosterol was reported as /iM and as the ratio of lathosterol: cholesterol (102 x mmol lathosterol/mol cholesterol).  2.4.5.3 Lipoproteins Plasma lipoproteins were separated by continuous density gradient ultracentrifugation as described by Innis (1989) but modified as follows for smaller plasma volumes. This technique allows separation of lipoproteins, and was chosen because precise density divisions between fractions were unknown. In pigs, the densities of lipoproteins differed between fetal and adult serum (Hollanders et ah, 1985). Therefore, it could not be assumed that the densities of lipoproteins from young and adult guinea pigs were the same. A continuous density gradient of guinea pig plasma lipoproteins was created by first placing 4 mL distilled H 2 0 into a Quick-seal ultracentrifuge tube. This was sequentially underlayered with 6 mL of a NaBr solution (density = 1.02 g/mL) and finally, the plasma sample, first adjusted to a volume of 3 mL with saline and then to a density of 1.25 g/mL with solid NaBr. The tube was sealed and centrifuged in a fixed angle rotor (50 Ti) at 15°C and 100,000 g for 20 h and stopped with no brake. Fractions (0.5 mL) were removed from the gradient through a needle inserted into the base of the tube. The fraction densities were determined gravimetrically. The total cholesterol in each gradient fraction was quantitated after chloroformmethanol extraction (Folch et al., 1957) by enzymatic kit assay as for the plasma after dissolving in isopropanol. The cholesterol carried by each lipoprotein class was calculated from a plot of cholesterol concentration versus the fraction number. The resulting graph peaks were identified as triglyceriderich lipoproteins (chylomicrons + VLDL), HDL or LDL on the basis of their relative densities. The cholesterol content of the fractions composing each peak was then summed to give an estimate of cholesterol carried by each lipoprotein class. Lipoprotein identity was verified using plasma from 1  36  day old and adult guinea pigs by 4-18% SDS-PAGE (Laemmli, 1970). Lipoproteins were sequentially floated by ultracentrifugation (Mills et al., 1984) without washing to avoid loss of apoproteins. The density cuts used were determined from the above continuous density gradient ultracentrifugation and were as follows: triglyceride-rich lipoproteins= < 1.006 g/mL, LDL= 1.006 - 1.05 g/mL, HDL= 1.06 - 1.21 g/mL. Ultracentrifugation isolation times were 21 h, 20 h and 48 h for the 3 lipoproteins respectively.  2.4.6 Liver determinations 2.4.6.1 Total lipid and cholesterol Lipid was extracted from frozen, homogenized liver with chloroform-methanol (Folch et al., 1957) and was measured gravimetrically. Lipid aliquots were dissolved in 2-propanol and assayed for cholesterol with the enzymatic kit assay described above.  2.4.6.2 HMG CoA reductase activity Microsomal HMG CoA reductase activity was assayed as the rate (pmol/min) of formation of [3H]mevalonate from [3H]HMG CoA as illustrated below (Goodwin & Margolis, 1976). Addition of HC1 converts [3H]mevalonate to [3H]mevalonolactone which can be extracted. Glucose-6-phosphate dehydrogenase makes the conversion of fHJHMG CoA to [3H]mevalonate irreversible and regenerates NADPH. The pH optimum and subcellular location of the reductase activity was previously determined to be the same in guinea pigs as for rats (Li et al., 1979b). Some alterations were made to the assay because hepatic reductase activity was 10-fold lower in the adult guinea pig than in the rat (McNamara, 1984). These modifications included the use of greater quantities of tissue in the assay and less dilution of the final extract for scintillation counting.  37  from microsomes  [3H]HMG CoA +  ['HJmevalonate HMG CoA reductase  +  NADPH  NADP  NADPH  NADP  HO  ['H]mevalonolactone  glucose-6-phosphate dehydrogenase 6-phosphogl uconate  glucose-6-phosphate  Briefly, 300 /d of microsomal suspension (0.1-0.5 mg protein) was aliquoted in duplicate into 20 mL screw-cap tubes containing 50 /xl of 100 mM sucrose containing 50 mM KC1, 40 mM KH2P04 and 30 mM EDTA at pH 7.2. The mixture was preincubated for 20-30 min at 37°C in an oscillating water bath. The reaction was started with the addition of 100 /d of a substrate-cofactor mix (800 mM HMG CoA, 120 mM glucose-6-phosphate, 8 mM NADP, 7 international units of glucose-6-phosphate dehydrogenase and 1.5 fiCi/mL DL-[methyl-3H]HMG CoA) and stopped, 20 min later, by the addition of 50 pA 12 N HCl. The mixture was then incubated for a further 30 min before being removed from the water bath and addition of approximately 0.5 g sodium sulfite. Mevalonolactone was extracted from the reacted mixture with the addition of 10 mL toluene followed by 2-3 min of vigorous shaking on an Eberbach shaker. After allowing the test-tubes to stand at room temperature for 30 min, a portion of the toluene extract was aliquoted into scintillation vials. Aqueous scintillation fluid was then added and the toluene counted for 3H content. Blanks were run as above with HCl added before the substrate cofactor mix. Recovery was determined from the percentage of [14C]mevalonic acid recovered in toluene extracts as [14C]mevalonolactone.  38  2.4.6.3 LDL receptor relative mass Detection of the LDL receptor was attempted by both ligand blotting, using gold-labelled rabbit 0VLDL, and indirect immunoblotting which are illustrated in Figure 2.1. The former was found to be superior at detecting the receptor protein bands and, thus, was used here. Handley et al. (1981) and Roach et al. (1987) have also successfully used gold-lipoprotein conjugates for the detection of lipoprotein receptor on nitrocellulose. In brief, plasma membrane proteins were separated by non-reducing SDS-PAGE and transferred to nitrocellulose. The nitrocellulose blot was then incubated with gold-labelled lipoproteins. Visualization of gold-/3VLDL/LDL receptor bands was enhanced by silver staining. Dot blotting for the LDL receptor has been done by Roach et al. (1987) and Maggi and Catapano (1987) but was not used here because of the occasional detection of proteins at molecular weights other than that of the LDL receptor. /3VLDL, rather than LDL, was used as the ligand because of its increased affinity for the receptor, and hence, greater sensitivity of detection. Wade et al. (1985) have previously shown that both LDL and /3VLDL can be used to quantitate bovine adrenal membrane LDL receptor. jSVLDL (as reviewed by Kovanen, 1987) is a cholesteryl ester-rich VLDL remnant which accumulates in the plasma when the LDL receptor number decreases in response to cholesterol feeding. The immunoblotting procedure was performed as above except that the blot was incubated with unlabelled human LDL. Human and guinea pig LDL are cleared equally well from guinea pig plasma (Steinbrecher et al., 1983). A solution of antibody directed against human apo B was then added followed by other antibodies directed against the anti-apo B. Visualization was accomplished using a peroxide/chloronaphthol system. Direct immunoblotting with an antibody raised against the guinea pig LDL receptor was not done here because antibody was not available commercially. Neither, apparently, is there any mention of such an antibody prepared by other investigators in the literature. The IgG-Q monoclonal antibody (available from Amersham Canada Ltd.) raised against bovine adrenal LDL  39  A. Ligand Blotting  silver precipitate  °oo ° o o  o  a° ld © VLDLQ  o o o o o Otfft\ u  m  LDL  o  receptor  xxxxxxxxxx  XXXXXXXXXX  nitrocellulose  B. Immunoblotting  goat peroxidase anti -peroxidase  bridging rabbit anti-goat IgG  goat anti-human apo B  appB LDL bj^-^itf  *-DL receptor  XXXXXXXXXX  nitrocellulose  Figure 2.1 Ligand blotting (A) and immunoblotting (B) techniques for detecting the LDL receptor. Protein from liver plasma membrane preparations was first separated by SDS-PAGE and transferred to nitrocellulose. To detect the LDL receptor protein by ligand blotting, colloidal gold was conjugated to rabbit jSVLDL and incubated with the nitrocellulose. After washing, the visible gold signal was enhanced by silver staining. Immunoblotting was done by first incubating the nitrocellulose with human LDL. This was followed by a series of immunoglobulin incubations (goat anti-human apo B, rabbit anti-goat bridging antibody and goat peroxidase anti-peroxidase). The protein band was visualized using a peroxide/chloronaphthol system. 40  receptor does not cross react with LDL receptor from the mouse, rat, Chinese hamster, rabbit or dog (Beisiegel et al., 1981). The extent to which the IgG-C? monoclonal antibody would cross react with the guinea pig LDL receptor is apparently unknown.  2.4.6.4.1 Lipoproteins Male adult rabbits (New Zealand) from the Animal Unit of the University of British Columbia were used to produce /SVLDL. Rabbits were fed rabbit chow mixed with 100 g corn oil and 20 g cholesterol/kg for 2 wk. /SVLDL (density = < 1.006 g/mL) was isolated by ultracentrifugation (2 h at 100,000 g at 15°C) of plasma (approximately 15 mL). The protein content of the final lipoprotein was measured according to the assay of Markwell et al. (1981). Because of the high triglyceride content of the samples, they were extracted (Folch et al., 1959) before spectrophotometry. Adult human blood was obtained from female volunteers for the isolation of LDL. Plasma was prepared as above and LDL was isolated by adjusting plasma to a density of 1.02 g/mL with solid NaBr and floating VLDL by ultracentrifugation (21 h at 100,000 g at 15°C). After removing VLDL, the remaining solution was adjusted to a density of 1.05 g/mL with solid NaBr and LDL was floated (20 h at 100,000 g at 15°C). LDL was washed once by resuspending in a solution of 1.05 g/mL and repeating the ultracentrifugation.  2.4.6.4.2 Membrane solubilization All procedures were performed at 4°C. The frozen plasma membrane pellet was resuspended (400 mg/mL) in buffer B (250 mM Tris-maleate (pH 6.0), 2 mM CaCl2, 4 /iM leupeptin, 12 /iM aprotinin, 2 mM phenylmethylsulfonyl fluoride (PMSF)) using 18 and 21 gauge needles. Following this, NaCl, Triton X-100 and distilled water were added to achieve a final concentration of 125 mM Tris-maleate (pH 6.0), 1 mM CaCl2, 2 /iM leupeptin, 6 /iM aprotinin, 1 mM PMSF, 160 mM NaCl,  41  1 g/100 mL Triton X-100. Samples were mixed using a vortex over the next 10 min and centrifuged (at 100,000 g for 1 h). The soluble protein (supernatant) was stored in aliquots at -70°C and the protein content determined according to Lowry et al. (1951).  2.4.6.4.3 SDS-PAGE and electrocution Membrane protein samples were randomly allocated to lanes and were separated by SDSPAGE (Laemmli, 1970) on 1.5 mm thick, 6% acrylamide gels, under non-reducing conditions (55 min, 200 v). Proteins were electroeluted to 0.45 n nitrocellulose, without methanol (2 h, 290 mA) and the blots were stored at -70°C.  2.4.6.4.4 Gold-lipoprotein conjugation Colloidal gold was dialysed overnight against 5 mM citrate phosphate (pH 5.5) at 4°C to adjust the pH for conjugation of the lipoprotein. Gold was then forcefully added to 0VLDL (diluted to 0.3 mg/mL in a glass tube) with a glass pipette to achieve a concentration of 15 /ig /3VLDL protein/mL gold. This was mixed without using a vortex, by inversion, and bovine serum albumin (BSA) added as a stabilizer to achieve a concentration of 1 g/100 mL. The mixture was then dialysed overnight against Tris-buffered saline (TBS; 20 mM Tris (pH 8.0) and 90 mM NaCl) at 4°C and used within 12 h.  2.4.6.4.5 Ligand blotting Non-specific binding was blocked by incubating blots with a 5% BSA (wt/v) blocking solution (2 mM CaCl2, TBS) for 2 h at room temperature on a slowly shaking platform. The blocking solution was then replaced by the blotting solution (7.5 jig/mL gold-/$VLDL, 2.5% BSA (wt/v), 2 mM CaCl2, TBS) and incubated at room temperature for 1.5 h again with slow shaking. Subsequently, blots were washed as follows: twice quickly with BSA-free blocking solution, for 5  42  min with blocking solution containing 1 % BSA (w/v), for 10 min with BSA-free blocking solution and for 2 min in deionized water. Detection of the receptor bands was enhanced using a silver-based gold enhancement kit. A quality control lane was present on every gel and data were expressed as sample/control absorbance after video densitometry. Analyses were performed twice using 30 and 40 fig of sample protein. Densitometry was done in duplicate.  2.4.6.4.6 LDL receptor identification and characterization The band representing the LDL receptor was identified by performing ligand blotting on hepatic membrane protein from 4 d old and adult guinea pigs in the presence of either EDTA (20 mM) with no added CaCl2, or sodium suramin (5 mg/mL). The molecular weight of the receptor band was determined by comparison to high molecular standards run under reducing conditions (with 2-mercaptoethanol). The molecular weight and density of the hepatic receptor band was also compared to that found on adult guinea pig adrenal plasma membranes (prepared as above). The ability of ligand to saturate the receptor was tested by incubating identical blots of samples from all of the ages studied, with different concentrations of gold-|8VLDL (2, 7.5 and 15 ng/mL).  2.4.6.4.7 Linearity Linearity studies were done to ensure that for all the ages studied 1) the concentration of protein chosen for comparison of the receptor mass was within the linear range of the binding assay and 2) the slope of the data (change in absorbance (density)/increase in protein applied to the gel) was constant. These experiments were done by assaying various amounts of membrane protein (10100 ng, 2-4 samples from each age) by ligand blotting and performing regression analysis on the densitometry results.  43  2.4.6.4.8 Immunoblotting Nonspecific lipoprotein binding was blocked, as above, overnight at 4°C on a slow shaking platform. Thereafter, procedures were done at room temperature unless otherwise mentioned. Blots were then incubated with 50 /xg/mL LDL in immunoblotting solution (TBS, 2 mM CaCl2 and 5% BSA, pH 7.4) for 1.5 h. This was then rinsed with washing solution (TBS, 2 mM CaCl2, 0.5 % BSA, pH 7.4) once rapidly, twice for 20 min each and once again rapidly. Nonspecific antibody binding was blocked with 4% normal rabbit serum in immunoblotting solution for 45 min. Rabbit serum was prepared from fresh rabbit blood which was left to sit at room temperature for 30 min in a glass tube without anticoagulants and centrifiiged at 3,000 rpm for 15 min. Reaction with primary antibody was accomplished by incubating the blot in immunoblotting solution containing goat antihuman apo B (3 /tg/mL), for 2 h. Blots were again washed as above and then incubated with bridging antibody, rabbit anti-goat IgG (1:1000), in immunoblotting solution for 1 h. After washing again, blots were incubated with goat peroxidase-anti-peroxidase (1:300) in immunoblotting solution. The blot was finally washed 3 times in TBS with 2 mM CaCl2 for 5 min each. The bands were visualized by adding 4-chloro-l-napthol in ice cold methanol to 0.018% H202, incubating with the blot for 5 min and then rinsing the blots with water once again.  2.5 Human studies 2.5.1 Subject selection 2.5.1.1 Term infants Single birth term infants born by elective caesarean section at Grace Hospital, Vancouver, BC, between April and August 1990 were studied. Infants were recruited until there were 2 groups of 6 infants: those fed human milk and those fed formula. Only infants delivered by elective caesarean section were included in this study because it is known that many variables of labour and delivery can affect cord plasma cholesterol levels (Cress et al., 1977, Boulton, 1979). Possibly,  44  these might also influence cholesterol synthesis rates. In addition to labour and delivery possibly affecting cord plasma cholesterol levels, maternal hypertension or diabetes, of any origin, may also have an effect (Boulton, 1979). Thus, infants of mothers with these conditions were excluded from the study. All infants were apparently medically and metabolically normal.  2.5.1.2 Preterm infants The preliminary study, initially included all infants of 24-32 wk gestation who were admitted to the Special Care Nursery of the British Columbia Children's Hospital, Vancouver, BC, between April and August 1989. Data were used if infants were of the appropriate birth weight for gestational age and if at least 5 plasma samples were obtained, the first being within 2 wk of birth. Twenty-one infants met these criteria. Included in the second study, the lathosterol study, were appropriate for gestational age infants born at 23-32 wk gestation at Grace Hospital and admitted to the above mentioned Special Care Nursery between February and August 1990 (n=22). A requirement of a concurrent study was that infants be intubated from birth.  2.5.1.3 Normal adults Normal, healthy Caucasian male adults were recruited for this study if they met the following criteria: nonsmokers, no history of heart disease, not taking medications, not drinking more than 4 cups of coffee/d, were within 1 SD of their ideal body weight and ate a typical North American diet as verified by a 24 h food intake record. Six participants were included.  2.5.1.4 Hypercholesterolemic adults Plasma from 6 patients with familial hypercholesterolemia was obtained from the Lipid Clinic of University Hospital, Shaughnessy Site, Vancouver, BC. Blood was sampled by venepuncture  45  between 0500 h and 1000 h. Patients were undergoing treatment with oral cholestyramine resin, which increases cholesterol synthesis and plasma lathosterol concentrations (Bjorkhem et al., 1987, Gylling et al., 1989). Included were 2 women and 4 men ranging in age from 16-72 years. The dose of cholestyramine resin prescribed ranged from 4-16 g/d over a duration of 1-5 years effecting a reduction of plasma LDL-cholesterol levels of 1.5-2.6 mM.  2.5.2 Subject care and blood sampling 2.5.2.1 Term infants Choice to breast-feed or use cow's milk based infant formula was made by the mothers freely, without the advice of anyone involved in this study. The infants were fed ad lib approximately every 3 h and their body weights were recorded daily. At birth, the umbilical cord was clamped and blood (1 mL) was collected by puncture of the umbilical vein within 2 min of delivery. Blood sampled at 4 days of age (0.5 mL) was collected by venepuncture of the vein on the dorsum of the hand. This was done between 0815 h and 1545 h when the infants were between 92 and 103 h of age.  2.5.2.2 Preterm infants Medical procedures and nutritional support were at the discretion of the attending physicians with no regard to this study. Generally, infants were nourished via a peripheral vein immediately after birth with glucose, followed by the inclusion of amino acids and 10% Intralipid (0.5-4 g/kg body weight/d) over the next wk. Lipids were mixed with heparin (2 IU/mL lipid) and were infused at a constant rate on a continuous basis. Heparin causes the release of lipoprotein lipase from the capillary endothelium and facilitates triglyceride clearance from the plasma. Expressed human milk and/or formula were gradually introduced by nasogastric bolus as tolerated. Gestational age, birthweight, sex, daily weight gain and enteral and parenteral feeding data were recorded. Serum  46  bilirubin levels and any prenatal maternal treatment with dexamethasone were recorded from infants charts. Blood samples for the preliminary study were collected by heel prick into capillary tubes for hematocrit analysis as part of the medical care of the infant. These tubes were spun and the hematocrit read within 2 h of sampling. Plasma, for cholesterol assay, was collected by breaking the tube at the interface of the plasma and the red cells. Additional samples were acquired when excess plasma remained from other required laboratory tests. Umbilical cord blood was collected for the lathosterol study as described above for term infants. In addition, blood (0.5 mL) was sampled at 3-4 d of age and, when possible, two to five more times during the next 4 wk at approximately 1000 h. Blood was always taken by venepuncture of the vein on the dorsum of the hand.  2.5.2.3 Normal adults Blood was sampled by venepuncture before meals at 1700 h (± 25 min) and 0820 h (+ 20 min) the next morning. These sampling times represent the nadir and peak, respectively, of the cholesterol synthesis diurnal cycle as measured by deuterium incorporation (Jones and Schoeller, 1990).  2.5.3 Plasma analyses Plasma was prepared and stored, and sterols were measured as previously described. Plasma free cholesterol levels were determined by a specific enzymic assay kit similar to that used for total cholesterol but lacking cholesterol esterase. Plasma apo AI and B were measured in previously frozen plasma samples by immunoprecipitation. In this assay, plasma was incubated in buffers containing antibodies prepared against human apo AI or B. The antigen-antibody complexes scatter light which is detected by nephelometry. The nephelometer was calibrated using standards provided  47  with the equipment. Also provided were quality control serum samples which contained relatively high and low concentrations of apo AI and B. These were assayed with each group of samples.  2.6 Statistical analyses 2.6.1 Guinea pigs One-way ANOVA was used to assess differences due to animal age or experiment. Where significant differences were found (p<0.05), Fisher's Least Significant Difference method was used to compare individual means.  2.6.2 Term infants Group and time effects and their joint effect on the data were examined using a multivariate analytic approach (Roy's greatest root (Cole and Grizzle, 1966)) to the repeated measures design. When there was evidence of an interaction between group and time on an effect, group effects were examined separately at each time using t tests and similarly, paired / tests were used to examine the time effects for each group. Analysis of the baseline differences data was also performed on each of the parameters measured to determine the effect of diet on the change in values between birth and 4 days of age. Student's t tests were used to compare the diets.  2.6.3 Preterm infants Variations between individual infants in time of onset, dosage and duration of Intralipid feeding were unavoidable as this study operated within the confines of the best treatment for each patient. Therefore, the only statistical comparisons possible were paired t tests between samples taken at birth (cord) and at 3-4 d of age as infants' treatment was relatively similar to this point.  48  2.6.4 Adults One-sided, paired t tests were used to compare normal adult data at 1700 h and 0830 h. A series of two-sided t tests were used to compare infant to normal adult values while one-sided t tests were used to compare both infant and normal adult values to cholestyramine-treated patient data.  2.7 Ethical approval 2.7.1 Guinea pigs The procedures involving animals were approved by the University of British Columbia Animal Care Committee.  2.7.2 Term infants Study protocol was approved by the clinical screening committees for research and other studies involving human subjects of the University of British Columbia, British Columbia Children's and Grace Hospitals. Informed consent was obtained from all volunteers and parents, to allow their infant's participation.  2.7.3 Preterm infants The study protocol was approved by the clinical screening committees for research and other studies involving human subjects of the University of British Columbia, British Columbia Children's Hospital and Grace Hospital. Informed consent was obtained from all parents to allow their infant's participation in the lathosterol study. Ethical approval for use of blood samples for the preliminary study was also obtained.  49  Chapter 3 RESULTS  3.1 Guinea pig experiments 3.1.1 Animal body and liver weight Body and liver weights of guinea pigs for the 3 experiments are given in Table 3.1. Adult and 8 day old animals were lighter in experiment 3 than 1 or 2. There was no significant increase in liver or body weight during the first week after birth probably due to the large degree of variability. There was a trend, however, toward higher weights at 8 days of age than in younger animals.  3.1.2 Plasma cholesterol and lipoproteins In both experiments 2 (Figure 3.1a) and 3 (Table 3.2), plasma cholesterol levels increased by approximately 50% between 1 and 4 days of age. In experiment 2, levels had decreased again by 8 days of age and levels in adults were not different from those at 1 day of age. In experiment 3, mean concentrations of cholesterol were less at 8 than at 4 days of age but this was not a significant difference. Adult guinea pigs in experiment 3 had lower plasma cholesterol than 1 day old animals. Most of the changes in plasma cholesterol levels with age could be accounted for by changes in LDL-cholesterol levels (Figure 3.1b). Lipoprotein identity was confirmed by their apoprotein composition determined by SDS-PAGE (Figure 3.2). Levels of HDL-cholesterol did not increase between 1 and 4 days of age and were decreased at 8 days of age. Even lower levels of HDLcholesterol were found in adult plasma. These observations regarding lipoprotein levels in the early postnatal period are in agreement with the reports of others (Bohmer et al., 1972, Li et al., 1979a).  3.1.3 Hepatic lipid It has been long known that guinea pig liver lipid content decreases to adult levels within the first week of birth (Imrie and Graham, 1920). Here, too, total hepatic lipid decreased from 19.4 to  50  Table 3.1 Body weight and liver weights of guinea pigs in experiments 1-3 (g)1.  Age (days)  Adult Experiment  Liver weight  Body weight  Liver weight  Body weight  Liver weight  Body weight  Liver weight  Body weight  1  4.83 ± 0.43'  106 ± 7"  4.16 ± 1.38*  112 ± 25*  5.53 ± 1.60*  176 ± 30*  25.71 ± 9.91"  880 ± 295b  6  6  7  7  8  8  4  4  5.16 ± 0.32"  106 ± 7*  4.18 ± 0.88*  107 ± 17*  5.40 ± 1.64*  164 ± 45*  25.71 ± 9.91"  880 ± 295*  3  3  5  5  8  8  4  4  4.06 ± 0.97  103 ± 23"  4.42 ± 0.51*  106 ± 13*  5.15 ± 0.77*  133 ± 22*  17.39 ± 2.24°  412 ± 55"  4  4  7  7  11  11  8  8  n  2 n  3 n  'Values are means ± SD. One-way ANOVA followed by Fisher's Least Significant Difference analysis was used to determine differences between experiments and ages. Values in a row not sharing the same superscripts are significantly different, p< 0.01. 2  Adult and 8 day old animals were lighter in experiment 3 than in 1 or 2 (p<0.05). No other differences between experiments were found.  51  A. cholesterol lathosterol o a  4]  o E CO _D  CL  B.  c  j  HDL-cholesterol F7/lLDL-cholesterol ^ H | triglyceride—rich lipoprotein-cholesterol  h  ~o a. o CL  C. t o t a l lipid  20 o o en  en ^E o  cholesterol  15 10  "D 'Q_  o  5  "5 o  c  •  JZ  O  0 DAY 1  [1  DAY 4  DAY  7  ^ ^  ADULT  Age  Figure 3.1 Effect of age on guinea pig plasma sterols (A), plasma lipoproteins (B) and liver lipids (C). Values are means ± SD. The number of animals used in (C, experiment 1) and (A, experiment 2) are given in Table 3.1. Plasma lipoprotein determinations were performed on some animals from both experiments 1 and 2 (n= 9, 4, 8 and 6, for each age respectively). Data points sharing the same superscript are not statistically different (p > 0.05) from like determinations according to one-way ANOVA, followed by Fisher's Least Significant Difference method. 52  Table 3.2 Plasma cholesterol, hepatic HMG CoA reductase activity and hepatic LDL receptor relative mass from suckling and adult guinea pigs (experiment 3)1 Age (days) 1  4  8  Adult  n  4  7  11  8  Plasma cholesterol (mM)  2.46 ± 0.50*  3.72 ± 0.98b  3.28 ± 0.93*b  0.76 ± 0.22=  Hepatic HMG CoA reductase activity (pmol/(min x mg protein))  11.6 ± 3.1*  29.6 ± lO.^  10.5 ± 5.6«  51 ± 24.5d  Hepatic LDL (relative mass)  0.31 ± 0.18*  0.35 ± 0.06*  0.38 ± 0.15*  0.85 ± 0.11b  receptor  'Values are means ± SD. One-way ANOVA followed by Fisher's Least Significant Difference analysis was used to determine differences between ages. Values in a row not sharing the same superscripts are significantly different, p<0.05.  53  ^  m Al  -  -  •  *  »-. • =.i  .•^•i  Figure 3.2 Verification of the identity of plasma lipoproteins from a 1 day old (gel 1, lanes 1-3) and adult (gel 2, lanes 4-6) guinea pig by apoprotein content on SDS-PAGE. Lipoproteins were separated by sequential ultracentrifiigation. Density cuts were determined by continuous density gradient ultracentrifiigation and were as follows: HDL (lanes 1 and 4) = 1.06-1.21 g/ml, LDL (lanes 2 and 5) = 1.006-1.06 g/ml and triglyceride-rich lipoproteins (lanes 3 and 6) = 1.006 g/ml. Samples and molecular weight markers (not shown) werefractionatedby 4-18% SDS-PAGE and visualized with Coomassie Blue stain. The major apoprotein bands were identified according to Guo et al., 1982. HDL contained apo Al and C apoproteins. LDL had primarily apo B and the triglyceride-rich lipoprotein fraction contained apo B and C apoproteins. Apo E, which is usually present on HDL and VLDL (Guo et al., 1982), could not be identified on these gels. Apo E is commonly lost as a result of ultracentrifiigation procedures (Fainaru et al., 1977). Other protein bands on the gels are either additional apoproteins or contaminants present due to the lack of washing the fractions (ie. albumin). 54  4.6% of liver wet weight between birth and 8 days of age (Figure 3. lc). The hepatic concentration of cholesterol, however, did not change with age. Further, the amount of cholesterol (mg) per whole liver did not change in the first 8 days after birth (14.2 + 5.3, 12.4 ± 3.7 and 17.0 ± 5.4 mg (mean + SD) at 1, 4 and 8 days of age respectively).  3.1.4 Cholesterol synthesis Both plasma LDL-cholesterol (Figure 3.1) and hepatic HMG CoA reductase activity (Table 3.2) increased between 1 and 4 days of age and decreased between 4 and 8 days of age. The plasma lathosterol concentration (Figure 3.1a) did not significantly increase (p=0.07) between 1 and 4 days of age. Levels at 8 days of age were higher than at day 1, however, suggesting an increasing trend in the first 8 days after birth. Hepatic HMG CoA reductase activity was higher in the adult than suckling animals (Table 3.2) while plasma lathosterol concentrations were lower adult that 8 day old animals. Mean activities of hepatic HMG CoA reductase activity found here in adult guinea pigs were similar to those reported by Ibrahim and McNamara (1988). The plasma sterol ratio was higher in the adult than at the suckling ages studied (data not shown). High variability was associated with the results for both HMG CoA reductase activity and plasma lathosterol concentration. The variability in HMG CoA reductase activity within an age group was higher than found in the rat (Haave and Innis, 1988). Turley et al. (1976) measured cholesterol synthesis by radioactive acetate incorporation into cholesterol, and also reported much more individual variation in guinea pigs than rats. There was no evidence of periodicity in HMG CoA reductase activity at any age studied (data not shown). Turley et al. (1976) have suggested that guinea pigs do not exhibit such a phenomenon, possibly due to different meal eating habits compared to other species. The rat, for example, which does have a marked diurnal variation in hepatic HMG CoA reductase activity, has  55  only one peak of eating activity per 24 hours and eats most of its food in the the dark period. In contrast, the guinea pig has three periods of eating activity per 24 hours and eats equal amounts of food in the dark and light periods.  3.1.5 LDL receptor mass LDL receptor mass was similar at all suckling guinea pig ages studied; however, in adults the relative mass was more than 2-fold higher (Table 3.2). This is apparently the first demonstration of the guinea pig LDL receptor by ligand blotting. Its molecular weight, approximately 140 kDa, is comparable to that of other species studied (summarized by Gherardi et al., 1988). The identity of the LDL receptor was confirmed by its molecular weight, its lack of binding to gold-/3VLDL in the presence of suramin or EDTA and the abundant presence of receptor in membranes prepared from adrenal tissues (Figure 3.3). Further, immunoblotting revealed that the receptor could also bind to human LDL (Figure 3.4). The linear range (—20-70 ng membrane protein, example in Figure 3.5) and slope (0.014 ± 0.007 absorbance units//ig membrane protein, (mean + SD, n= 12)) did not differ with age. The regression coefficient (r2) was 0.956 + 0.026 (mean + SD, n= 14). A representative blot is shown in Figure 3.6. Ligand binding to the receptor protein was found to be saturable at all ages studied. Densitometry absorbance measured after incubating blots with 2 /ig/ml gold-/3VLDL was less than with 7.5 or 15 /*g/ml. There was no difference in absorbance between 7.5 and 15 j*g/ml. Besides the major band visible at approximately 140 kDa, a lighter molecular weight band was occasionally seen at approximately 115 kDa (see Figure 3.3, adrenal sample). Others have hypothesized that this represents either a precursor of the mature receptor (Fong et al., 1989) or a fragment produced as a result of the membrane preparation (Friedman et al., 1987). A very high molecular weight band was also visible on some blots, but was not EDTA or suramin sensitive. Possibly, this could be remnant receptor protein.  56  1  2  3  4  200-  11697-  fc^Hg^to;  66-  Figure 3.3 Verification of the identity of guinea pig LDL receptor as detected by ligand blotting. Solubilized membrane protein and molecular weight markers werefractionatedby SDS-PAGE and transferred to nitrocellulose. Molecular weight markers were myosin (200 kDa), /3-galactosidase (116 kDa), phosphorylase b (97 kDa) and bovine serum albumin (66 kDa). Ligand blotting was performed by incubating blots with gold-labelled /3VLDL (7.5 /ig/ml) and the bands were enhanced with silver staining as described in Chapter 2. Membrane from 4 day old guinea pig liver (200 jig) was incubated with the /SVLDL alone (lane 1) or in the presence of EDTA (20 mM, lane 2) or suramin (5 mg/ml, lane 3). Similar results were found for adult liver membrane (not shown). Lane 4 represents 20 /xg of adult guinea pig adrenal membrane incubated with 0VLDL as above. 57  1  Figure 3.4 Detection of guinea pig LDL receptor by immunoblotting and ligand blotting. Solubilized membrane protein from an adult guinea pig liver was fractionated by SDS-PAGE and transferred to nitrocellulose. Lane 1 (50 fig membrane protein) was incubated with gold-labelled /8VLDL as in Figure 3.3 and gold staining was enhanced with silver stain. Lane 2 (80 fig membrane protein) was incubated with LDL (50 /ig/ml) followed by goat anti-human apo B, rabbit anti-goat IgG and goat peroxidase-anti-peroxidase as described in Chapter 2. The band was visualized with 4-chloro-lnaphthol and H202. 58  1 2  3  4  5  1.0 o 9>  V O C  o  0.8 0.6  -Q k_  O (0  <  0.4 0.2 10  20  30  40  50  60  70  Membrane Protein ^.g)  Figure 3.5 Representative example of quantitation of LDL receptor relative mass from guinea pig liver membranes by ligand blotting. Solubilized membrane protein from a 4 day old guinea pig (18.2, 27.3, 36.5, 54.7 and 72.9 /*g protein in lanes 1 to 5 respectively) was fractionated by SDS-PAGE and transferred to nitrocellulose. Blots were incubated with gold-labelled /SVLDL (7.5 /tg/ml), the bands were enhanced with silver staining as described in Chapter 2 and scanned twice by a video densitometer to yield the absorbance areas plotted. The r2 values for the two lines are 0.964 and 0.993 and the slopes are 0.014 and 0.013 absorbance units/^g protein respectively. Similar results were found for all ages investigated. 59  Figure 3.6 A representative blot of the comparison of LDL receptor relative mass in 8 day old and adult guinea pig liver membrane by ligand blotting. Solubilized membrane protein samples (40 \i% protein) were randomly assigned to gels and lanes, fractionated by SDS-PAGE and transferred to nitrocellulose. Blots were incubated with gold-labelled /5VLDL (7.5 /tg/ml), the bands were enhanced with silver staining as described in Chapter 2. Adult membrane was run in lanes 1 and 2, membrane from 8 day old animals was in lanes 3 to 6 and a control sample was in lane 7. 60  3.2 Term infant study  3.2.1 Infant body weights Only one infant (fed formula) regained his birth weight by 4 days of age. Of the other 4 d old infants, those fed formula were 95.6 + 1 . 9 (5) % (mean + SD (n)) and those fed human milk were 92.8 ± 2.2 (6) % of their birth weights.  3.2.2 Plasma cholesterol and apolipoproteins The finding of a 40% increase in plasma cholesterol levels over the first 4 days postpartum (Table 3.3) is consistent with the results of others (Potter, 1977, Stozicky et al, 1982, Strobl et al., 1983, Lane and McConathy, 1986). Using multivariate analysis, which takes advantage of the paired samples (cord and day 4), higher plasma cholesterol levels were found both at birth and at 4 days of age for the infants fed formula rather than human milk (p=0.05). Another method of assessing the effect of diet used in this study was analysis of the baseline differential (described in Table 3.3). Using this method, no significant difference (p<0.05) in the increase in plasma cholesterol levels from birth to 4 days of age, was found between the two diet groups. This is in agreement with previous reports of no differences in plasma cholesterol levels between breast and formula feeding at 3 days (Lane and McConathy, 1986) and 5 days (Potter, 1977) of age. Most infants fed formula were oriental while the majority of infants fed human milk were Caucasian. A t test was conducted to compare data from oriental and Caucasian infants to determine whether or not race affected the cord plasma cholesterol levels. No significant difference was found (j?=0.71). There was no significant change in plasma apo AI concentrations between birth (cord) and 4 days of age (Table 3.3) but an increase of approximately 66% in apo B levels. These data are in agreement with previous findings for apo AI (Potter, 1977, Stozicky et al., 1982, Strobl et al., 1983, Lane and McConathy, 1986) and apo B (Stozicky et al., 1982, Strobl et al., 1983, Lane and  61  Table 3.3 Plasma cholesterol, apoproteins and lathosterol levels from term infants at birth (cord) and at 4 days of age1 Diet2  Cholesterol  Cord  BM  1.63 ± 0.21  2.22 ± 0.21  (mM)  F  1.78 ± 0.34  2.56 ± 0.23  Apo AI  BM  0.63 ± 0.10  0.61 ± 0.10  (g/L)  F  0.67 ± 0.10  0.73 ± 0.12  ApoB  BM  0.18 ± 0.03  0.31 ± 0.07  (g/L)  F  0.18 ± 0.03  0.30 ± 0.09  BM  5.69 ± 0.52  4.99 ± 1.40  F  7.09 ± 1.42  5.64 ± 0.91  BM  352 ± 56  226 ± 74  F  404 ± 97  220 ± 24  Lathosterol OiM) Lathosterol/cholesterol ((102 x mmol lathosterol)/mol cholesterol)  MULTIVARIATE ANALYSIS3 p values  Day 4  Group/Diet x Time  Group/ Diet  Time  0.4  0.05  0.0001  BASELINE ANALYSIS3 (Day 4 minus cord)  0.59 ± 0.28 0.76 ± 0.36  0.054  -0.04 ± 0.04 0.03 ± 0.09  0.8  0.8  0.0001  0.13 ± 0.06 0.12 ± 0.07  0.4  0.07  0.03  -0.70 ± 1.50 -1.45 ± 1.38  0.3  0.4  0.0003  -126 ± 103 -184 ± 93  'Values are means ± SD, n=6 for each determination. BM: human milk (breast fed), F: formula. 'Group and time effects and their joint effect on the data were examined using a multivariate analytic approach to the repeated measures design. 4 When there was evidence of an interaction between group and time on a measure (Apo AI), group effects were examined separately at each time using t tests. Paired t tests were used to examine the time effects for each group. No significant differences (p<0.05) were found for Apo AI when groups (diets) or times were analysed separately. There was a mild difference, however, when groups were compared at day 4 (p=0.06). 3 In addition, the baseline differential was calculated for each parameter and BM and FF groups were compared using t tests. No significant differences (p<0.05) were found for any of the parameters. Again, however, there was a mild difference for apo AI (p=0.06). 2  62  McConathy, 1986). Therefore, the rise in cholesterol levels was probably associated with VLDL and/or LDL rather than HDL. Others (Stozicky et al., 1982, Strobl et al., 1983) have reported that cholesterol is fairly equally distributed between HDL and LDL in cord blood but after birth, LDLcholesterol predominates.  3.2.3 Plasma lathosterol concentrations Normal adult plasma lathosterol concentrations were similar to those reported in 3 recent studies (Farkkila and Miettinen, 1988, Nikkila and Miettinen, 1988, Miettinen et al., 1990). In addition, cholestyramine resin treatment of hypercholesterolemic patients (Table 3.4) resulted in marked elevation in plasma lathosterol levels and the lathosterol:cholesterol ratio as previously found (Bjorkhem et al., 1987, Gylling et al., 1989), thus validating the present methodology. It is not known whether or not plasma lathosterol levels show periodicity, although this has been demonstrated for methyl sterols (Miettinen, 1982). Data here indicates that plasma lathosterol levels may be an index of this phenomenon; plasma lathosterol concentrations were greater at 0830 h than at 1700 h (p=0.04, Table 3.4), the proposed peak and nadir of cholesterol synthesis. Plasma lathosterol concentrations decreased between birth and 4 days of age and were not influenced by diet during the first 4 days after birth (Table 3.3). It is not known whether or not the newborn human has an established diurnal cycle of cholesterol synthesis by 4 days of age. In order to contrast neonatal values to adult levels, cord and neonatal blood samples were collected over a short time span and compared to the adult values at both 1700 h and 0830 h. As no differences between infants fed formula and human milk were apparent, infant data were pooled before comparison to adult values. No differences were found between neonatal (cord or day 4) and adult lathosterol concentrations at either extreme of the adult diurnal cycle.  63  Table 3.4 Plasma cholesterol and lathosterol values in adult reference populations1.  Cholestyramine-  Normal adult males  treated patients  Cholesterol (mM)  range=  Lathosterol  0*1)  range=  Lathosterol/Cholesterol ((102 x mmol lathosterol) /mol chol)  range=  1700 h  0830 h  4.85 ± 0.33  4.85 ± 0.52  6.09 ± 0.89  (4.58-5.33)  (4.24-5.66)  (4.60-7.29)  5.50 ± 1.52 a,b  6.18 ± 1.80 b  20.48 ± 13.41 e  (3.39-7.24)  (3.72-8.09)  (10.47-43.47)  112.5 ± 26.0 b,c,d  129.4 ± 42.2 b,c,d  322.9 ± 168.4  (73-140)  (78-177)  (170-595)  'Values are means ± SD, n=6 for each determination. One-tailed paired t tests were used to compare normal adult data at 1700 h and 0830 h to infant data (Table 3.3) and normal adult values to cholestyramine-treated patient data. Two-tailed t tests were used to compare infant data (Table 3.3) to adult data. 'Significant differences are indicated as follows: a = normal adults, 1700 h versus 0830 h, />=0.04, b= normal versus cholestyramine-treated adults, />=0.02, c = normal adults versus cord, p<0.0001, d= normal adults versus 4 day old infants, />=0.002-0.0002, e = infants (cord or 4 day old) versus cholestyramine-treated adults, p=0.02  64  3.2.4 Plasma lathosterol: cholesterol ratios Plasma lathosterol:cholesterol ratios were higher at birth than at 4 days of age (Table 3.3) and were also considerably higher in neonatal than adult plasma (Table 3.4). This was likely caused by the lower plasma cholesterol levels in the cord than day 4 and neonatal than adult samples, respectively, rather than higher lathosterol levels. The plasma lathosterolrcholesterol ratio found in the cholestyramine-treated patients was higher than in normal adults at both sampling times, but comparable to those found in neonatal samples. The observation that precursor levels are the same, yet product is reduced, leads one to suspect an increased demand for, or turnover of, plasma cholesterol in the rapidly growing neonate. The similar ratios in neonatal (Table 3.3) and cholestyramine-treated patient plasma (Table 3.4) are likely reflections of lower concentrations of both sterols in the neonates.  3.3 Preterm infant studies A diagram of the organization of the preterm infant studies along with relevant Figures and Tables is shown in Figure 3.7.  3.3.1 Prel iminary cholesterol study Infants were divided into 2 groups based on their history of intravenous lipid feeding. One group consisted of infants who had received > 2 g Intralipid/kg body weight/day for more than 2 days and is known as the "with Intralipid infusion" group. The other group, "with minimal Intralipid infusion", had less than this amount of intravenous lipid. Infants with Intralipid infusion (n= 12) had lower mean gestational ages and birth weights and began enteral feeds later than infants with minimal Intralipid infusion (n=9, Table 3.5). Changes in plasma cholesterol versus postnatal age and 10% Intralipid administration of representative infants (Table 3.6) are plotted in Figures 3.8 and 3.9. The clinical profiles of both groups  65  Preterm Infants  Pr^aauaary  Lftffcasteral  Choiestettd ;SMy  with Intralipid infusion (n=12) Tables 3.5, 3.6 Figure 3.8  with minimal  Intralipid infusion (n=9) Tables 3.7, 3.8 Figure 3.9  Paired cord and day 3-4 sampling (n=ll) Table 3.7  Longitudinal sampling (n=10) TabU 3.5  with Intralipid infusion (n=8) TabU 3.8 Figure 3.10  Figure 3.7 Organization of preterm infant studies. 66  with minimal Intralipid infusion (n=2) TabU 3.8 Figure 3.11  Table 3.5 Characteristics of preterm infant populations1 Preliminary Study  Lathosterol Study  With Intralipid  With Minimal Intralipid  (Longitudinal i.v. nutrition)  12  9  10  Birth weight (g)  range=  954 ± 343 (655-1800)  1227 ± 170 (955-1475)  1030 ± 172 (655-1225)  Gestational Age (weeks)  range=  26.0 ± 2.2 (24-31)  28.9 ± 1.3 (28-31)  26.6 ± 1.6 (23-29)  Day of onset of Intralipid Infusion  6.5 ± 2.6 (3-12)  -  range=  3.9 ± 1.3 (3-7)  range=  14.3 ± 8.4 (1-33)  3.1 ± 1.6 ( n = 7 4 (1-5)  6.2 ± 3.2 (3-12)  42/42/16  38/12/50  40/20/40  Sex(F/M)  5/7  3/6  4/6  Delivery (SVD/CS3)  8/4  5/4  5/5  10/0/2  8/0/1  8/2/0  Day of onset of Enteral Feeds % EBMVFormula/Mix  Race (C/O/EI*)  'Values are means ± SD. 2  EBM: expressed human breast milk  'SVD/CS: spontaneous vaginal delivery/caesarian section, C/0/EI:caucasian/oriental/eastindian 'only 8 of the 9 infants received enteral feeds during the study and one infant received enteral feeds late, at 13 days of age and is not included in the table.  67  Table 3.6 Details of infants presented in Figures 3.8 and 3.9  Infant  Enteral Feeds2  Clinical Features3  6  EBM d!6-22, Pregestimil d23-38, d46-  HMD, BPD, NEC, gastrointestinal surgery d6, 38  SVD  3  SCF 68 dl5-22, SCF 81 d22-  severe IVH  24  SVD  2  EBM d33-57, Prosoybee d57-  HMD, BPD, NEC, IVH  720  25  SVD  2  EBM d8-14, d24-63, SCF 81 d62-70, EBM d71-  HMD, BPD, PIE  1000  28  CS  2  EBM d6-36, SCF 81 d36-46, EBM/SCF 81 d46-  HMD, BPD, twin  955  28  CS  2  SCF 68 dl3-34, SCF 81 d34-  HMD, BPD, NEC  Sex  Birth weight (g)  M  1800  31  SVD  M  980  25  700  M  M  Gestational Age (weeks')  Delivery tSVD/CS1)  Onset of amino acid infusion (d)  'SVD: spontaneous vaginal delivery, CS: caesarean section 2  EBM: expressed breast milk, SCF: Special Care Formula, d: age in days  3  HMD: hyaline membrane disease, BPD: bronchopulmunary dysplasia, IVH: intraventricular hemmorhage, PIE: pulmonary interstitial emphysema, NEC: necrotizing enterocolitis  68  T3  h j "TCI  Plasma Cholesterol (mM)  o er o bo  Plasma Cholesterol (mM) O  N3 1  •f»  en  CD  o  1  O  »3  >  NJ O  i-i  5 O "  a  o  8 »  05  o  <^ 3 a1 !—  00  o  « >5 0\  ON S "  °  I I II  -^OOJ^  5" P  DD  O  1-1  o cr O •£• I-I  "8 i-t  -tOCJ-J"  Intralipid ( g / k g BW)  _ MW+ .  Intralipid ( g / k g BW)  E.  F.  o 10  o o  E CO  o  40  60  80  Age ( d )  Figure 3.9 Plasma cholesterol levels (O) and 10% Intralipid infusion rates (solid bars) of 2 representative preterm infants with minimal 10% Intralipid infusion. Neither infant received >2 g Intralipid/kg body weight/day for more than 2 days. Infants are described in Table 3.6. 70  were typical for infants of this degree of immaturity and included hyaline membrane disease, bronchopulmonary dysplasia, septicemia, patent ductus arteriosus, intraventricular hemorrhage and necrotizing enterocolitis. Plasma cholesterol levels rose coincident with lipid administration (Figure 3.8). The highest level measured for each infant ranged from 4.06-10.70 mM. In all cases, cholesterol levels decreased after cessation of lipid infusion. The highest plasma cholesterol level found among infants without 10% Intralipid infusion was 5.17 mM (Figure 3.9). Because plasma cholesterol concentrations increased with 10% Intralipid administration, this population was considered suitable for the study of cholesterol synthesis during lipid infusion.  3.3.2 Lathosterol study Infants were divided into 2, not mutually exclusive subsets: 1) those for whom paired cord and day 3-4 samples were available (n= 11) and 2) those for whom at least 3 additional samples were collected after 3-4 days of age (n= 10, Figure 3.7). Five infants fit into neither group. Paired cord and day 3-4 data were not obtained for subjects if the blood had clotted in the cord before sampling, if the infant was too sick for investigation or if the blood sample was too small for analysis. Longitudinal sampling was discontinued if extubation (according to concurrent study protocol) or death occurred. The clinical profiles of the participating infants were similar to those in the preliminary study. Serum bilirubin did not exceed 209 /iM (12.2 mg/100 mL) in any infant.  3.3.2.1 Cord and day 3-4 subset Data from these infants are shown in Table 3.7. Two infants were female and 9 were male, 2 were oriental and 9 were Caucasian, and 2 were born by caesarean section and 9 by vaginal delivery. The mean plasma cholesterol concentration increased between birth and 3-4 days of age but there was no change in plasma apo AI or B concentrations. The cord plasma cholesterol and apo  71  Table 3.7 Levels of cholesterol, apo AI and B, and lathosterol in plasma of preterm infants at birth (cord) and 3-4 days of age*  Infant  Birth weiRht (g)  Gestational Aee (weeks)  Apo AI fe/L)  Cholesterol (mmol/L)  Apo B (s/L)  Lathosterol (umol/L)  Lathosterol (102 x mmol): Cholesterol (mol)  cord  dav 3-4  cord  dav 3-4  cord  dav 3-4  cord  dav 3-4  cord  dav 3-4  SW  920  26  1.32  2.53  0.35  0.48  0.22  0.39  3.03  2.48  229  98  AB  975  28  2.12  1.76  0.52  0.44  0.20  0.20  8.90  1.21  419  69  NC  1066  27  1.34  2.74  0.22  0.33  0.28  0.34  2.48  3.28  185  120  NP  685  23  1.60  1.37  0.50  0.40  _2  0.14  7.16  2.07  446  151  4.97  3.05  300  190  JM  1195  29  1.66  1.60  0.50  0.43  0.16  _2  SK  1130  28  1.29  2.43  0.61  0.53  0.09  0.31  1.84  2.01  142  83  JRa  1480  32  2.15  2.51  0.36  0.67  0.17  0.37  4.34  1.84  202  74  JRb  1700  32  3.39  2.53  0.89  0.76  0.36  0.20  4.68  1.73  138  69  2.84  2.30  162  114  2  MR  1025  27  1.76  2.02  0.56  0.38  _2  EL  875  26  1.19  1.50  0.54  0.56  0.10  0.15  5.17  2.04  435  136  SI  1050  26  2.64  3.67  0.48  0.49  0.35  0.12  12.23  9.70  464  264  Mean  1100  27.6  1.86  2.24  0.50  0.50  0.21  0.25  5.24  2.88  284  124  SD  282  2.7  0.67  0.68  0.17  0.13  0.10  0.11  3.11  2.33  133  60  n  11  11  11  11  11  11  9  11  11  11  11  P  0.06  9 0.5  0.5  0.01  'Paired two-tailed t tests were used to compare plasma levels at birth (cord) and 3-4 days of age except for cholesterol where a one-tailed t test was used. 'Insufficient sample volume for analysis.  72  0.007  AI concentrations were similar to published levels for infants of similar gestational age (Higgs et al., 1974, Ginsberg and Zetterstrom, 1980, Lane and McConathy, 1983, Sharma et al., 1983, Parker et al., 1987, Amr et al., 1988, Genzel-Boroviczeny et al., 1988, Desci et al., 1990). There was, however, considerable variation among infants. The plasma cholesterol concentration decreased in 3/11, did not change in 1, and increased in 8/11 infants over the first 3-4 days after birth. The plasma lathosterol concentration and the ratio of lathosterolxholesterol, however, decreased. There were no differences in any of these parameters between infants whose mothers were (n=4) and were not given dexamethasone before delivery. The administration of small amounts of 10% Intralipid (< 1 g/kg body weight/d, n=2) before 3-4 days of age did not have any discernable effect.  3.3.2.2 Longitudinal subset Characteristics of the infants studied are shown in Table 3.5. Three infants were oriental and the remainder were Caucasian. Cholesterol, apo AI and B, and lathosterol concentrations and 10% Intralipid intake for representative infants (Table 3.8) are plotted in Figures 3.10 and 3.11. The rise in plasma lathosterol was coincident with the infusion of 10% Intralipid and the increase in plasma cholesterol levels. In the 3 infants for whom 10% Intralipid was stopped before the final sample was taken, plasma sterol concentrations returned to preinfusion levels (for example, Figure 3.10, Infant L). Minimal Intralipid was administered to 2 of the infants and in these, plasma lathosterol and cholesterol levels rose slightly after birth but showed no further increase during the 11 days studied. The percentage of plasma total cholesterol found in the free form (free/total cholesterol x 100) increased considerably between birth and the final sampling point with 10% Intralipid infusion. The percent plasma free cholesterol, at the first and last sampling points, for the infants shown in Figure 3.10 were as follows: G= 37 to 86%, H= 40 to 63%, 1= 52 to 78%, J= no data, K= 52 to 88% and L= 26 to 81%. The comparable changes for infants infused with minimal 10% Intralipid  73  Table 3.8 Details of infants presented in Figures 3.10 and 3.11  Infant  H  M  Sex  Birth weight (g)  Gestational Age (weeks>  Delivery (SVD/CS1)  Onset of amino acid infusion (d)  Enteral Feeds2  Clinical Features3  1130  28  SVD  EBM/SCF 68 d!2-  birth asphyxia, HMD, PIE, pneumonia, IVH, PDA  1080  28  CS  EBM/SCF68dll-12, d!9-  HMD, BPD, NEC, PDA  925  26  CS  EBM/SCF 68 d2-5, EBM d5-  HMD, septicemia, PDA  M  685  23  SVD  SCF 68 d5-8  birth asphyxia, HMD,NEC,PDA, spastic quadriplegia  M  875  26  SVD  EBM d8-14, dl8-  HMD, BPD, PDA, respiratory infection dl6  M  1140  26  SVD  EBM dl4-  septicemia, PDA, inguinal hernia surgery d8  M  920  26  SVD  EBM d5-  HMD, pneumonia, IVH, septicemia  M  1025  27  CS  EBM/SCF 68 d4-  perinatal asphyxia, HMD, BPD, PDA  M  'SVD: spontaneous vaginal delivery, CS: caesarean section 2  EBM: expressed breast milk, SCF: Special Care Formula, d: age in days  'HMD: hyaline membrane disease, BPD: bronchopulmunary dysplasia, PDA: patent ductus arteriosus, IVH: intraventricular hemmorhage, PIE: pulmonary interstitial emphysema, NEC: necrotizing enterocolitis  74  Cholesterol (mW)  Apoproteins  (g/0  40  7 6 5 4 3 2 Apoproteins  Intralipid ( g / k g BW)  15 10  0  0  1  Age (d)  25 - 20  5  0  Lathosterol (uM)  30  1  0.6 0.4 0.2 0.0 3 2  (g/0  35  Jll  10  15  Age (d)  Cholesterol (mil)  Apoproteins  (g/0  Figure 3.10 Plasma cholesterol (•), lathosterol (n) and apo AI (i >) and B (O) levels and 10% Intralipid infusion rates (solid bars) of 6 representative preterm infants. All data is plotted. Infants are described in Table 3.8. 75  Cholesterol (mM)  40 35  8 7  30 25  6 5  20 15  4 3 2 1  10 5 0  0  Apoproteins  0.6 0.4  (9/0  0.2 0.0 3 2 1  Figure 3.11 Plasma cholesterol (•), lathosterol (n) and apo AI ( • ) and B (O) levels and 10% Intralipid infusion rates (solid bars) of 2 representative preterm infants with minimal Intralipid administration. No infant received >2 g Intralipid/kg body weight/day for more than 2 days. All data is plotted. Infants are described in Table 3.8. 76  (Figure 3.11) were M= 32 to 45% and N= 59 to 66%. Apo AI and B levels did not increase during 10% Intralipid infusion. In one of the two infants with minimal Intralipid infusion (Figure 3.11), plasma apo AI increased. Cholesterol and apo B levels, however, did not change appreciatively in either infant. Others have reported a transient increase in apo AI (Genzel-Boroviczeny et al., 1988), HDL-cholesterol (Ginsberg and Zetterstrom, 1977 and 1980, Desci et al., 1990) or LDL+VLDL-cholesterol (Ginsberg and Zetterstrom, 1980, Desci et al., 1990) in preterm infants given only enteral nutrition in the first month after birth.  77  Chapter 4 DISCUSSION  4.1 Guinea pig experiments 4.1.1 Early developmental changes in cholesterol synthesis The transient increases in guinea pig hepatic HMG CoA reductase activity and plasma lathosterol concentration after birth suggest that increased cholesterol synthesis may contribute to the postnatal elevation of plasma cholesterol levels in this species. The actual effect of these changes on plasma cholesterol levels, however, is unknown. Differences in cholesterol synthesis rates at various ages in guinea pigs have not previously been measured. Li et al. (1979b) attempted this, but high variability and the small number of animals studied prevented conclusions from being made. In agreement with the present findings in guinea pigs, hepatic HMG CoA reductase activity in pigs also increased between birth and 5 days of age (Jones et al., 1990) although only one newborn animal was studied. The postnatal changes in HMG CoA reductase activity and the plasma lathosterol concentration measured as potential indices of cholesterol synthesis in the guinea pig are contrary to results of studies in the rat. Carroll (1964) gave the first indication that rates of cholesterol synthesis in the rat were depressed during suckling by studying radiolabeled acetate incorporation into cholesterol. McNamara et al. (1972) later reported that rat hepatic HMG CoA reductase activity increased 4-fold in late gestation, then decreased 4-fold within the first day after birth and another 4fold by 10 days of age (see Figure 4.1). At weaning, activity dramatically increased again and then finally decreased to adult levels 2 weeks later. These HMG CoA reductase activity results were duplicated by Ness et al. (1979) and Hahn and Walker (1979). The same trend was also found by Leoni et al. (1984) who determined [3H]water incorporation into cholesterol and by Levin et al. (1989) who measured tissue HMG CoA reductase mRNA levels. An exception in this latter study was that at weaning, no change in mRNA levels was evident. Some caution should be used with the  78  700  RAT PIG  600  GUINEA PIG O  500  CL  en  E 400 c  E 300 o E 200 Q-  100  0 I  0  .  .  i  I  10  i  i  i  I  20  i  i  i  L_  30  AGE ( d a y s )  Figure 4.1 Comparison of developmental changes in hepatic HMG CoA reductase activity in the rat (McNamara et al., 1972), pig (Jones et ah, 1990) and guinea pig (data from this thesis). Weaning occurred at different times in these species. In rats, abrupt weaning was at 20 days of age, guinea pigs may have ingested some solid food in the first 8 days after birth and the pigs were given small amounts of solid starter feed between 21 and 25 days of age. 79  hepatic HMG CoA reductase activity data because Haave and Innis (1991) have reported that in vitro reductase activity is not correlated with rates of cholesterol synthesis (measured by [3H]water incorporation) in rats at 1 day before term. It should also be noted that despite a decrease in reductase activity at birth, the activity of hepatic HMG CoA reductase is still 10-fold higher in the suckling rat than in the suckling guinea pig or pig (see Figure 4.1). Relative to those of the rat, the developmental changes in hepatic cholesterol synthesis rates in the suckling guinea pig and pig seem of little significance. The parallel increases in hepatic HMG CoA reductase activity and plasma lathosterol concentrations between 1 and 4 days after birth support the use of these techniques of measuring cholesterol synthesis in the young guinea pig. The divergence between these two measures between 4 and 8 days of age, however, may cause some doubt. Plasma lathosterol levels were highest at 8 days of age, while reductase activity was maximal at 4 days after birth and decreased at 8 days of age. There are reasons to query both methods. Measuring hepatic HMG CoA reductase activity in vitro may not be adequate because, as mentioned above, reductase activity and cholesterol synthesis may not be correlated (Haave and Innis, 1991). Also, the use of plasma lathosterol has not been previously verified as an indicator of cholesterol synthesis rates in the guinea pig or the newborn of any species. Fielding and Fielding (1985) caution that there is no absolute method for measuring the rate of cholesterol synthesis and suggest that two different methods, relying on different sets of assumptions, be used in experiments. This could be an example of where two methods of assay give valuable information. The divergence between the two measures could indicate independent developmental changes in hepatic and extrahepatic cholesterol synthesis. In adult humans, plasma lathosterol is reported to represent both whole body (Kempen et al., 1988) and hepatic (Bjorkhem et al., 1987) cholesterol synthesis rates. Both are probably true in the adult human, under the steady state conditions of relatively constant body weight, dietary intake and plasma lipoprotein levels.  80  Meijer et al. (1992) hypothesize that in animals who synthesize the majority of cholesterol at extrahepatic sites that plasma lathosterol may be more indicative of whole body rather than hepatic cholesterol synthesis. According to Turley and Dietschy (1982) hepatic cholesterol synthesis as a percentage of whole body cholesterol synthesis in rats, guinea pigs and rabbits was approximately 60, 20 and 20%, respectively. The percentage for humans was estimated to be slightly less that 20%. If plasma lathosterol also represents the rate of whole body cholesterol synthesis in the postnatal guinea pig, data here would indicate that extrahepatic cholesterol synthesis remains elevated at 8 days of age while hepatic cholesterol synthesis peaks at 4 days of age. Increased peripheral cholesterol synthesis may be needed to provide cholesterol for use as a structural component of membranes during this period of rapid growth or may be secondary to decreased expression of LDL receptors or affinity of binding to these receptors. Because changes in hepatic HMG CoA reductase activity rather than plasma lathosterol levels were better correlated with postnatal plasma LDL-cholesterol levels, this suggests that increased plasma cholesterol may be of hepatic origin. Data from the pig (Jones et al., 1990) support this finding of apparently independent changes in rates of hepatic and extrahepatic cholesterol synthesis after birth; rates of hepatic and intestinal cholesterol synthesis varied independent of one another at 5 and 10 days of age with the greater increase found in the liver. In rats, also, Levin et al. (1989) report that HMG CoA reductase mRNA levels decreased similarly in the liver and lung after birth but transiently increased in the kidney and brain at 14 days of age. The postnatal changes in HMG CoA reductase activity in the rat have been associated with the diet because the increase in reductase activity normally found at weaning can be accelerated and delayed with early and late weaning (McNamara et al., 1972). The cholesterol in milk is considered to be responsible by some authors (Chevalier, 1964, Ness et al., 1979, Leoni et al., 1984, Bruenger and Rilling, 1986, Levin et al., 1989); however, dietary fat has also been investigated. In support of the importance of dietary fat and/or cholesterol, ligation of the bile duct, which impedes fat absorption and the enterohepatic circulation of bile acids and cholesterol, caused HMG CoA reductase  81  activity to increase in suckling rats (McNamara et al., 1972). These data indicate that in the rat, dietary fat and/or cholesterol may decrease reductase activity. Finally, a protein in milk that inhibits cholesterol synthesis has been proposed to exist in rat (McNamara et al., 1972) and cow (Boguslawski and Wrobel, 1974) milk. Although the changes in plasma cholesterol levels and HMG CoA reductase activity in the suckling rat have been associated with the milk, the cause of the increase in plasma cholesterol levels in the first 1-3 days after birth has not been established. Evidently, in the rat, "postnatal hypercholesterolemia" does not result from increased hepatic cholesterol synthesis (McNamara et al., 1972). In the guinea pig (as indicated from the results of this thesis) and pig (Jones et al., 1990), however, changes in the rate of cholesterol synthesis may be more important.  4.1.1.2 Early developmental changes in LDL receptor relative mass The (B/E) LDL receptor protein was detected by ligand blotting and immunoblotting of guinea pig hepatic membranes with gold-labelled /3VLDL and LDL, respectively. The LDL receptor protein band was identified by its molecular weight, which was comparable to LDL receptors from other species (reviewed by Gherardi et al., 1988). Also, lipoprotein binding to the LDL receptor was calcium dependent and inhibited in the presence of suramin. In addition, the receptor protein was concentrated in membranes prepared from the guinea pig adrenal and was saturable. Finally, another confirmation of the receptor identity was performed in this laboratory, using identical techniques in rats (Auestad et al., 1991). 17a-ethinyl estradiol was injected into rats and caused an 8-fold increase in the mass of hepatic LDL receptor protein. 17a-ethinyl estradiol in this species is known to upregulate the hepatic LDL receptor (Kovanen et al., 1979a). The expected inverse relationship between changes in plasma LDL-cholesterol levels and hepatic LDL receptor mass found in the adult human (Soutar et al., 1986) was not present in the suckling guinea pig. In experiment 3, receptor mass was constant during the first 8 days after birth  82  while plasma cholesterol levels increased between 1 and 4 days of age and then decreased, although not significantly, by 8 days of age. The time between the first and fourth day after birth should have been sufficient for down-regulation of the LDL receptor to occur, assuming that the half-life of the guinea pig hepatic LDL receptor is similar to that of the human fibroblast LDL receptor (approximately 25 hours, Brown and Goldstein, 1986). There are several possible explanations for why down-regulation did not occur between 1 and 4 days of age in the guinea pig. First, the great demand for cholesterol by the liver in this period of rapid growth could prevent the LDL receptor from being down-regulated. Second, the expected changes in LDL receptor number may have occurred on extrahepatic rather than hepatic cells. Fox et a\. (1987) propose that this might explain discrepancies in their studies of adult baboon LDL receptors. Third, the hepatic LDL receptor may be maximally down-regulated at both 1 and 4 days of age. Finally, hematopoietic cells which make up a large proportion of cells in the liver of fetal animals (Lafeber et ah, 1984) and persist for some time after birth could have diluted the concentration of receptor on the membranes. It is not known if membranes from these cells have the same concentration of LDL receptors as the parenchymal cells which rapidly increase in number after birth. Lack of detection of differences in hepatic LDL receptor mass in early postnatal life could also result from the methodology used. The limit of sensitivity of the ligand blot assay might not allow detection of small differences among the ages. Also, it is assumed that both binding affinity and efficiency of receptor protein isolation were uniform across the ages. In some situations, binding affinity has been found to differ between treatments (i.e. Fernandez and McNamara (1991) found that the receptor binding affinity increased when a high fat diet was fed to adult guinea pigs). Examination of the liver membranes in vitro also ignores possible in vivo modulators (as described in section 1.1.1) which may be present in plasma after birth. For example, Salter et al. (1988) showed that in the presence of insulin, LDL binding to the receptor on cultured adult rat hepatocytes  83  increased in as little as 1 hour, which would probably not be enough time to allow a change in receptor number. Finally, Fong et al. (1989) found that the intracellular pool of LDL receptors, in a number of cell types, is of variable size. In cultured rat adipocytes, the intracellular and plasma membrane receptor pools were affected differently by insulin treatment (Sather et al., 1988). Ligand blotting detects both intracellular and extracellular receptor and, thus, would not allow recognition of different pool sizes at different ages.  4.1.1.3 Early developmental changes in liver cholesterol content No evidence was found to support the belief that fetal hepatic stores of lipid might contribute to the postnatal rise in plasma cholesterol levels. In agreement with this interpretation, data of Li et al. (1979a) showed no significant changes in guinea pig liver cholesterol content in the first 5 days after birth. In other species, similar results have been reported. Jones et al. (1990) found that the hepatic cholesterol content was constant in pigs fed sow's milk between birth and 5 days of age. Bizzi et al. (1963) and McNamara et al. (1972) also reported the same in rats during development. It seems probable that rather than being exported as lipoproteins, liver triglyceride is oxidized after birth. Newborn guinea pigs, particularly, are in a state of fasting or "physiologic undernutrition" (Widdowson and McCance, 1955). The enzymes necessary for fatty acid oxidation in this species are developed within 48 h of parturition (Stanley et al., 1983). Also in the guinea pig, gluconeogenesis occurs within 2 hours of birth and is probably supported by these liver lipid stores (Raghunathan and Arinze, 1977). McNamara et al. (1972) tested another hypothesis, too, by measuring liver cholesterol content in developing rats. Adult rats fed cholesterol have increased hepatic cholesterol levels and decreased hepatic HMG CoA reductase activities (Shapiro and Rodwell, 1971). It was proposed that the high levels of cholesterol in rat milk would increase hepatic cholesterol levels and explain the decreased reductase activity in suckling rat liver. As the hepatic cholesterol content did not change through  84  development, McNamara et al. (1972) concluded that dietary cholesterol does not contribute to increased plasma cholesterol levels during this period. On the contrary, Haave and Innis (1991) did find increased cholesteryl esters in the livers in the first 48 h after birth providing evidence that dietary cholesterol may depress hepatic cholesterol synthesis rates after birth.  4.1.2 Suckling versus adult cholesterol metabolism The higher mass of LDL receptor protein and increased HMG CoA reductase activity in adult compared to suckling guinea pig liver is in accordance with the theory that when plasma LDLcholesterol levels are low, LDL receptor production and cholesterol synthesis rates are up-regulated (Brown and Goldstein, 1986).  4.1.2.1 Suckling versus adult cholesterol synthesis Hepatic HMG CoA reductase activity was lower in the newborn than adult guinea pig while the opposite was true for plasma lathosterol concentrations. Using the logic described earlier (section 4.1.1), and assuming that HMG CoA reductase activity and plasma lathosterol concentrations represent cholesterol synthesis, this may indicate that the distribution of cholesterol synthesis between hepatic and extrahepatic tissues may be different in young and adult guinea pigs. In accordance with this finding, Spady et al. (1983b) found that total body sterol synthesis decreased with age in hamsters studied at 1, 2-3 and 12 months of age. Most of the decrease was extrahepatic synthesis. These authors speculate that this reflects reduced growth as the animals age and an increase in the proportion of adipose tissue and skeletal muscle, which have relatively low rates of sterol synthesis. In agreement with the higher adult hepatic HMG CoA reductase activity found here, Levin et al. (1989) report higher hepatic HMG CoA reductase mRNA levels in the adult than the neonate in a few human samples studied. Increased extrahepatic cholesterol synthesis in the suckling animal might affect plasma  85  cholesterol levels in 2 ways. First, if extrahepatic tissues produce their own cholesterol, the extrahepatic LDL receptor might be down-regulated causing plasma LDL-cholesterol levels to increase. Hepatic LDL receptors would subsequently be down-regulated in the suckling animal compared to the adult (as seen here). Second, increased cholesterol synthesis by the intestine could result in more cholesterol entering the plasma in lipoproteins.  4.1.2.2 Suckling versus adult LDL receptor relative mass The inverse relationship between LDL receptor mass and plasma LDL-cholesterol levels in adult compared to 4 or 8 day old guinea pigs is in agreement with the theory that these are normally inversely related (i.e. when plasma LDL-cholesterol levels are higher, as at 4 and 8 days of age, LDL receptor mass is lower). The above statement assumes that the levels of LDL receptors in the young guinea pig have decreased from fetal levels. It is not known if this is the case. Factors present in the newborn period which are known to reduce expression of LDL receptors in adult animals may be responsible. For instance, the LDL receptor is down-regulated in many species in response to increased dietary cholesterol or saturated fat (Spady and Dietschy, 1988). These conditions are present in the young guinea pig as guinea pig milk contains cholesterol (approximately 1 mM, Connor and Lin, 1967) and relatively high levels of saturated fat. Finally, neonatal guinea pigs experience "physiologic" cholestasis (Tuchweber et al., 1990). This may lead to down-regulation of the LDL receptor as intravenously infused bile acids have been shown to decrease LDL binding to the hepatic LDL receptor in dogs (Angel in et al., 1983). Direct comparison of these guinea pig results with most other studies on the development of the LDL receptor is not possible because of differences in methodology, receptor specificity, cell types and ages of animals. Some comparisons can be made, however, to a rat study done in this laboratory which also used ligand blotting to detect the hepatic LDL receptor (Auestad et ah, 1991). The results were opposite from those found for the guinea pig; mass of LDL receptor was greater in  86  neonatal (2-3 days old) than adult rat. It is not surprising that this species difference is present given the other differences in cholesterol metabolism between the rat and the guinea pig. These two animals have different developmental patterns of hepatic HMG CoA reductase activity, distributions between hepatic and extrahepatic cholesterol synthesis (Spady and Dietschy, 1983), dependence on the liver cholesterol synthesis compared to LDL receptor activity for maintenance of cholesterol homeostasis (Spady and Dietschy, 1983), and adult LDL- and HDL-cholesterol profiles. The only other study of LDL receptors in the early postnatal period was by Erickson et al. (1988), also in the rat. No differences between the neonate and the adult were found when binding to hepatic membranes was measured. However, there was greater uptake of iodinated lipoprotein by cultured hepatocytes from neonates than adults. Although this binding was not necessarily specific to the LDL receptor, these latter results are in agreement with above mentioned data of Auestad et al. (1991). The decrease in cholesterol synthesis and increase in LDL receptor number in the newborn rat support the hypothesis that HMG CoA reductase is inhibited after birth. Cholesterol metabolism in the neonatal rat may be analogous to the human administered Pravastatin (Reihner et al, 1990). This drug inhibits HMG CoA reductase activity and results in an increase in receptor activity. Because the decrease in rat cholesterol synthesis rates actually occurs before birth (Haave and Innis, 1991), it is unlikely that such an inhibitor is present only in rat milk (McNamara et al., 1972). Instead, some other products of squalene besides cholesterol, which are inhibitors of HMG CoA reductase (Bruenger and Rilling, 1986), may be responsible.  4.1.3 Guinea pig study summary The coincident increases in plasma cholesterol levels and the rate of hepatic HMG CoA reductase activity in the first 4 days after birth in the guinea pig support the hypothesis that newly synthesized cholesterol may contribute to elevated plasma cholesterol levels after birth. The increase in plasma lathosterol concentrations over the first 8 days of life further favours this suggestion.  87  According to the LDL receptor theory of cholesterol homeostasis, however, it seems paradoxical that plasma cholesterol levels and synthesis increase at the same time. The LDL receptor theory suggests that if there is sufficient LDL-cholesterol in the plasma to provide for the cell's cholesterol needs, cholesterol synthesis will be repressed. This is apparently what occurs in the rat. After birth, plasma cholesterol levels are high and rates of cholesterol synthesis are low. The rat, however, has much higher HMG CoA reductase activity than the guinea pig and regulates cholesterol homeostasis primarily by changing reductase activity rather than the LDL receptor number (Spady and Dietschy, 1983). Despite this, two reports indicate that receptor number is also increased in the neonatal rat (Erickson et al., 1988, Auestad et al., 1991). In the suckling guinea pig, it appears that hepatic LDL receptor number and hepatic HMG CoA reductase activity are not necessarily coordinate. The LDL receptor mass is only 50% of that found in the adult and does not seem to change in the first 8 days after birth. On the other hand, reductase activity was elevated at 4 days of age relative to all other neonatal ages studied, although still was 40% lower than in the adult. Perhaps early postnatal LDL receptor levels are decreased in response to the high plasma cholesterol levels. Why then, does HMG CoA reductase activity increase? Instances of uncoupling of changes in cholesterol synthesis and LDL receptor number have been reported in male versus female hamsters (Spady et al., 1985), with inhibition of HMG CoA reductase in humans (Reihner et al., 1990), with cholestyramine resin or cholesterol feeding of rats (Spady et al., 1985) and finally in adult guinea pigs fed diets rich in polyunsaturated or saturated fatty acids (Ibrahim and McNamara, 1988). In all these cases, however, LDL receptor number changed while cholesterol synthesis rates remained constant. In contrast, LDL receptor number was constant while hepatic HMG CoA reductase activity changed in the young guinea pig. A possible rationalization for the results from the guinea pig is that something after birth down-regulates the LDL receptor (perhaps hormones) or prevents receptor binding (perhaps insulin or free fatty acids). This would result in high plasma cholesterol levels that are "unavailable" to the cell because of the  88  decreased receptor binding. To meet the cholesterol requirements of growth, cholesterol synthesis would thus need to be up-regulated. Another possibility is that the early postnatal guinea pig is in an analogous state to the adult guinea pig fed a high fat diet. Fernandez and McNamara (1991) found that a high fat diet resulted in increased plasma LDL, cholesterol synthesis and receptor affinity in adult guinea pigs. Despite the low relative mass of receptor found here in the young guinea pig, the affinity of binding might have been increased.  4.2 Term infant study In term infants, plasma lathosterol levels decreased by approximately 10% between birth and 4 days of age. If plasma lathosterol levels reflect the rate of cholesterol synthesis in newborn humans, as they do in adults, this suggests that the early postnatal increase in plasma cholesterol was not coincident with increased rates of cholesterol synthesis. This is contrary to the indications of increased rates of cholesterol synthesis between 1 and 4 days of age in the guinea pig. Further, there were no differences in plasma lathosterol concentrations between neonatal and normal adult values. In agreement with this, Levin et al. (1990) found similar hepatic HMG CoA reductase mRNA levels in human newborns and adults in the few samples measured. The difference between data from humans and guinea pigs may be the result of the ages chosen for study. The ages selected for study may not be equivalent in these species. Conversely, these divergent data may indicate a species difference. There are a number of possible explanations for the changes in plasma sterol levels in term infants after birth. First, plasma LDL-cholesterol levels might increase because of the regression of the fetal zone of the adrenal as discussed in section 1.3.5. Alternatively, hormones, or other modulators may affect down-regulation of the LDL receptor and thus cause increased plasma LDLcholesterol levels. Both of these hypotheses rely on down-regulation of LDL receptors. There was no change in the relative mass of receptor protein in the first 8 days after birth in the guinea pig,  89  however, levels were lower than found in the adult. Second, decreased levels of plasma lathosterol, and perhaps cholesterol synthesis, in term infants might be in response to the high plasma cholesterol levels. If large amounts of LDL-cholesterol are available to the cell, both the receptor and cholesterol synthesis rates will be down-regulated according to LDL receptor theory. Finally, infants are in negative energy balance over the first 4 days after birth, not having regained their birth weight. Relatively low rates of cholesterol synthesis at this time could be explained by limiting acetyl CoA levels for synthetic pathways during the establishment of enteral feeding in the first days after birth. It is difficult to interpret the effect of diet on plasma cholesterol levels found at 4 days of age. Although multivariate analysis showed higher cholesterol levels in infants fed formula (at birth and 4 days of age), the actual increase in cholesterol levels from birth to 4 days of age was not different for the two diets. Many studies of older infants show that plasma cholesterol levels are lower in infants fed formula than human milk (for example, Darmady et al., 1972). Although there are many variables present, this could be the result of the different fatty acid and cholesterol content of the two diets. Perhaps, also, 4 days of age is too soon to see the effects of the different diets or the diet effects are masked by the more significant changes occurring during adaptation to extrauterine life. Potter (1977) reported a similar lack of diet effect on plasma cholesterol levels at this early age. Given the comparable plasma cholesterol levels of the infants fed formula and human milk, one might expect that cholesterol synthesis rates would be higher in the group fed formula, because of their minimal dietary cholesterol. The plasma lathosterol data, however, does not support this, as again there was no diet effect. This is contrary to data from 5 day old pigs where hepatic HMG CoA reductase activity was greater in animals fed formula than suckled (Jones et ah, 1990). It should again be noted that 4 days may not have been long enough to detect a diet effect and that plasma lathosterol concentrations may not be an accurate indicator of cholesterol synthesis at this age.  90  4.3 Plasma lathosterol as an indicator of cholesterol synthesis in the neonate As discussed for the guinea pig (section 4.1.1), there has been no prior validation of the use of plasma lathosterol as an indicator of cholesterol synthesis in the neonate. To test this, one would need to compare results to other potential measures of cholesterol synthesis. In human infants, it would not be feasible to assay hepatic HMG CoA reductase activity since this would require a liver biopsy. Again, such studies would only give an indication of potential hepatic synthesis. Plasma mevalonate analysis could be performed but would also require verification for use in the neonate. Finally, in order to compare lathosterol levels to pH]water incorporation into cholesterol, one would require extensive blood sampling, which is unsuitable for newborn infants. In addition, single point determinations, i.e. birth and day 4 of age, could not be performed. The data from the preterm infant study included in this thesis, supports the use of plasma lathosterol as an indicator of cholesterol synthesis in the neonate. Data from several animal studies (section 1.4) indicate that cholesterol synthesis increases with phopholipid infusion and, as reported here, plasma lathosterol levels increased with 10% Intralipid infusion. The potential of dietary lathosterol to contribute to the plasma lathosterol concentration was not considered in this study. Kallio et al. (1989) measured lathosterol in milk from women 2 months postpartum and found that milk contained approximately 1.1 fiM lathosterol. In contrast, lathosterol levels in formula have not been measured but are likely to be negligible because levels of cholesterol in formula are also very low. Infants fed formula consumed less that 1 litre of milk by 4 days of age. Assuming that similar volumes of human milk were ingested by the breast-fed infants and that milk produced at 4 days postpartum contains similar lathosterol concentrations as at 2 months postpartum, infants would have taken in approximately 1 /xM of lathosterol. Whether or not this is absorbed at the intestine or what affect it has on plasma lathosterol levels is unknown.  91  4.4 Speculation about increased plasma cholesterol levels after birth Despite the major species differences in guinea pig, rat and human cholesterol metabolism after birth, increased plasma LDL-cholesterol levels occur in all. The initiation of enteral feeding is a factor common to all species at birth. Although it seems that dietary cholesterol is not necessary for the postnatal rise in plasma cholesterol levels, dietary fat may be. A possible mechanism for the effect of dietary fat is increased cholesterol synthesis. As shown here, hepatic HMG CoA reductase activity and plasma lathosterol concentrations increased after birth in the guinea pig, but plasma lathosterol concentrations in the term human infant did not. Some authors have suggested that downregulation of LDL receptor number occurs with fat feeding (see section 1.3.2). Receptor numbers are lower in the suckling than adult guinea pig but again, not in the rat. There are a number of indications that cholesterol homeostatic mechanisms, which include the LDL receptor and cholesterol synthesis, are mature in the fetus. It seems unlikely, therefore, that a developmental change in these occurs at birth and is responsible for the postnatal increase in plasma cholesterol levels. If this change in plasma cholesterol levels is not of dietary origin, or caused by increased cholesterol synthesis, then where does it originate? The "adrenal" theory perhaps explains this as it hypothesizes a decreased demand for cholesterol at birth and an "overshoot" phenomenon (i.e. increased plasma LDL-cholesterol levels before homeostasis is again achieved). An equivalent phenomenon to the human regression of the fetal zone of the adrenal may also occur in other animals who do not have a fetal zone of the adrenal. Equally attractive is the concept that a subtle change in LDL receptor number or affinity of binding causes increased postnatal plasma LDL-cholesterol levels. This might be mediated by changes in hormone levels at birth which are probably common to all animals.  92  4.5 Preterm infant study 4.5.1 Effect of 10% Intralipid on plasma lathosterol Increased plasma lathosterol levels during Intralipid infusion, along with the fall in plasma lathosterol levels upon cessation of lipid infusion, provide the first experimental data to indicate increased rates of cholesterol synthesis in association with 10% Intralipid administration in humans. Although recent results from studies in other animals support this interpretation, there is the potential that differences in plasma lathosterol may not represent differences in cholesterol synthesis. The plasma lathosterol levels found in preterm infants given 10% Intralipid were 4 to 7-fold higher than found in enterally fed preterm infants, term infants or normal adult controls, and were as high as those found in hypercholesterolemic adults treated with cholestyramine resin (Table 3.4). As well, the percentage of plasma cholesterol present in the free form was greatly increased during 10% Intralipid infusion. This can be taken as presumptive evidence of the presence of LpX because LpX is known to be composed of mainly free rather than esterified cholesterol. These data lend support to the theory (disscussed in section 1.4) that excess phospholipid in 10% Intralipid results in free cholesterol being transferred from cell membranes into plasma to form LpX particles and that cholesterol synthesis must be increased to compensate for this loss. Potentially, there are long term implications of having high plasma cholesterol levels in the early postnatal period (as outlined in Table 1.2) and altering cholesterol metabolism (cholesterol synthesis rates) in such a profound way. The preterm infant infused with 10% Intralipid may be analogous to the guinea pig fed cholestyramine resin at 1 week of age for a period of 6 weeks (Li et al., 1980). As mentioned earlier, cholestyramine resin in adult animals increases the rate of cholesterol synthesis. When the treated young guinea pigs became adults, they were challenged with a high cholesterol diet. Animals who had been fed cholestyramine resin after birth had decreased plasma cholesterol levels, increased steroid excretion, bile acid pool size and hepatic 7a-hydroxylase and no change in hepatic HMG CoA reductase activity compared to control animals. This illustrates the effect that early manipulations of  93  cholesterol metabolism can have on later life. According to this guinea pig study, one might predict that early stimulation of cholesterol synthesis may reduce adult plasma cholesterol levels. One factor not considered in this study was the presence of plant sterols in Intralipid (2 mg/g triglyceride, Griffin et ah, 1979). These might accumulate in the plasma during lipid infusion. Because the cholesterol (3/S-hydroxysterol) enzymatic assay used does not discriminate between cholesterol and plant sterols, elevated plant sterol levels may give falsely high cholesterol values. Griffin et al. (1979), however, measured plasma sterols by gas chromatography and found that plant sterols were only slightly elevated in infants given Intralipid for either 24 hours or 2.5 months. It seems unlikely, therefore, that elevation of plasma cholesterol in this study was a result of accumulated infused plant sterol.  4.5.2 Effect of 10% Intralipid on plasma apoproteins The lack of increase in plasma apo AI levels seen here during parenteral nutrition has also been found in similarly treated adults (Weinberg and Singh, 1989). Possibly this may be explained by the absence of nutritional input at the intestine which is a site of apo AI synthesis. The lack of increase in apo B levels, which similarly occurs in adults during Intralipid administration, could be the result of perturbation of hepatic triglyceride synthesis and secretion of VLDL (Weinberg and Singh, 1989). Alternatively, the lack of increase in apo B levels during Intralipid infusion may be related to the expression of the LDL receptor. If there is a cellular membrane cholesterol deficit during lipid infusion, as hypothesized, one would expect the LDL receptor to be up-regulated. Decreased cellular free cholesterol is also thought to mediate LDL receptor proliferation in human hepatoma cells treated with estrogen (Semenkovich and Ostlund, 1987). This would, thus, result in greater cell uptake of LDL and a decrease in plasma LDL-cholesterol and apo B levels. LDL receptors could also be upregulated by the non-lipid components of total parenteral nutrition. Parenteral nutrition lacking lipid  94  in adult patients caused decreased LDL-cholesterol levels (Chait et al., 1981). Chait et al. (1981) hypothesized that this resulted from the hyperinsulinemia, which occurs with parenteral nutrition, upregulating the LDL receptor.  4.5.3 Hepatic versus extrahepatic cholesterol synthesis during 10% Intralipid infusion Although Jakoi and Quarfordt (1974) found increased hepatic HMG CoA reductase activity in rats infused with phospholipid, there is reason to believe that rates of extrahepatic cholesterol synthesis may be elevated as well in response to this treatment. First, phospholipid infusion-induced hypercholesterolemia has been found to occur in rats even in the absence of a functioning liver (Edwards, 1975). Also, Innis and Boyd (1983) found that rats infused with Intralipid had increased adipose tissue and skeletal muscle HMG CoA reductase activity and decreased hepatic HMG CoA reductase activity, contrary to the liver data of Jakoi and Quarfordt (1974). In the infants studied here, rates of extrahepatic and hepatic cholesterol synthesis could not be differentiated. Apoprotein data, however, reveal some information. Neither apo AI nor B increased during the lipid infusion implying that neither HDL nor LDL+VLDL are associated with the high levels of cholesterol in the plasma. Thus, the increased amount of cholesterol in the plasma is probably in the form of LpX. According to the theory of Griffin et al. (1979), cholesterol could be obtained from all cells in contact with the plasma phospholipid. One would expect, therefore, that enhanced rates of cholesterol synthesis would occur in all of these cells, including both hepatic and extrahepatic cells. It is possible that hepatic synthesis of VLDL is occurring at an enhanced rate but that the resultant LDL has a higher turnover rate in the plasma due to up-regulation of the LDL receptor. Circumstantial evidence that LpX accumulated in the plasma is that the amount of free cholesterol and the percentage of total cholesterol in the free form were both higher after Intralipid infusion.  95  4.5.4 Preterm versus term plasma cholesterol and lathosterol levels Cord plasma cholesterol levels are higher in very preterm than term infants (Sharma et al., 1983, Amr et al., 1988, Genzel-Boroviczeny et al., 1988). This is hypothesized to reflect either enhanced hepatic cholesterol synthesis related to a spurt in liver growth (Johnson et al., 1982) or relatively lower numbers of LDL receptors (Cai et al., 1991). Although the mean plasma cholesterol level was higher here for preterm than term infants, this was not significantly different. This is likely due to the large degree of variability especially among preterm infants as discussed below. It is perhaps not surprising, therefore, that cord lathosterol concentrations were also not different between preterm and term infants. The high degree of variability in preterm cord plasma sterol concentrations might result from the wide range of gestational ages among the infants studied. Plasma cholesterol levels are known to vary according to gestational age (Johnson et al., 1982). Also, cord plasma was from elective caesarian section births only in the term infant study but not in the preterm infant study. The many variables of labour and delivery that affect cord plasma cholesterol levels (Boulton, 1979) could have also influenced cholesterol synthesis rates in the preterm infants studied. One such variable is maternal dexamethasone administration, which is often given to the mother before preterm delivery to reduce neonatal respiratory distress (Liggins and Howie, 1972). In infants of 26-32 weeks gestation, administration of this drug results in increased total, HDL-, and LDL-cholesterol and apo AI levels in cord blood, possibly due to enhanced fetal cholesterol synthesis (Parker et al., 1987). Another potential mechanism is that dexamethasone causes decreased binding to the LDL receptor (Salter et al., 1988). Only four infants in this study were exposed to dexamethasone before birth. This was not enough to determine whether or not cord plasma lathosterol levels were elevated in infants whose mothers were treated with dexamethasone. Possibly, however, dexamethasone treatment contributed to the high degree of variability in the preterm cord plasma lathosterol data. Irrespective of gestational age, plasma lathosterol levels did not increase in humans in the  96  early postnatal period. Preterm infant plasma cholesterol concentrations increased about 20% over the first 3-4 days after birth while plasma lathosterol concentrations decreased by approximately 45% over the same time. In contrast, term infant plasma cholesterol levels increased 40% and lathosterol decreased 10%. The decrease in plasma lathosterol levels in these infants could be explained by the probability that newborn infants are in negative energy balance during the first 3-4 days after birth. This could be expected to result in decreased cholesterol synthesis because of limiting acetyl CoA levels for biosynthetic pathways. The difference in the change of plasma lathosterol levels between term and preterm infants might be because preterm infants are possibly in greater negative energy balance than term infants, especially when supported by non-lipid parenteral nutrition immediately after birth.  4.5.5 Summary 10% Intralipid administration to preterm infants (23-32 wk gestation) is accompanied by an increase in the plasma lathosterol concentration, which is a possible indicator of the rate of cholesterol synthesis. Several complications of administering intravenous lipid to preterm infants have been suggested (reviewed by Friedman et al., 1978, Stahl et ah, 1986), but the immediate and long term impact of potential changes in the rate of cholesterol synthesis is unknown.  4.6 Future studies It is hoped that this thesis will add knowledge to the complex field of perinatal cholesterol metabolism but perhaps more importantly serve as the impetus for other research. "Postnatal hypercholesterolemia" is difficult to study because neonatal animals are not in a steady state. Animals are still adapting to their extrauterine environment. Some metabolic pathways, such as fatty acid oxidation, are being used for the first time. Some enzyme systems, such as gluconeogenesis, are immature and are slow to develop.  97  Future animal studies might concentrate on isolating some of the factors involved in the development of cholesterol metabolism. For instance, cultured cells could be studied to determine the effects of some of the hormonal changes that occur at birth on LDL receptor binding and expression. Although many of these have been studied in isolation, the combined effect of hormonal changes is unknown. More interesting, perhaps, is the potential effect of the postnatal increase in free fatty acids on LDL receptors. The effect of inhibition of HMG CoA reductase activity on plasma cholesterol levels in the newborn guinea pig might be another interesting avenue of study and could give added insight into the importance of cholesterol synthesis during this period. Would repression of cholesterol synthesis cause deceased plasma cholesterol levels and up-regulation of the LDL receptor number? Perhaps "postnatal hypercholesterolemia" would occur regardless of this treatment. The possible divergence between hepatic and extrahepatic cholesterol synthesis, as suggested by the data in this thesis, could also be further studied. More intensive investigation using pHJwater incorporation into cholesterol to study cholesterol synthesis would be necessary. Perinatal treatments that have late effects on cholesterol metabolism might alter cholesterol synthesis in various organs or tissues differently. Different species might also have distinct developmental patterns of hepatic and extrahepatic contributions to cholesterol synthesis. The effect of cholesterol added to formula could also be studied in animals. Such controlled studies have been performed, but without examining the immediate postnatal increase in plasma cholesterol levels and cholesterol synthesis. It would be interesting to determine the actual effect of dietary cholesterol on the early elevation in plasma cholesterol levels after birth. An obvious extension of the term infant study conducted here would be to follow the infants through weaning. One might hypothesize that levels of lathosterol would be higher in breast fed infants than those fed formula at 2 months of age, a time when plasma cholesterol levels are significantly different.  98  Finally, further work with preterm infants might include long term retrospective studies of older children who were born prematurely. This would require a large population of children to control for the many variables present. It would be interesting to consider whether or not early 10% Intralipid infusion could be correlated with some parameters of cholesterol metabolism. Other possible investigations include the comparison of plasma lathosterol levels after administration of various intravenous lipid solutions. Recent studies have shown that intravenous lipid solutions which contain 20% triglyceride (versus 10%) or mixtures of medium-chain and long-chain triglycerides, minimize or do not elicit hypercholesterolemia in infants (Lima et al., 1988, Haumont et al., 1988, Wells et al., 1989) or adults (Messing et al., 1990). The lathosterol methodology described may be of value in defining the effect of these intravenous solutions on cholesterol synthesis. In addition, it would be interesting to measure plasma lathosterol levels in cord blood of preterm infants whose mothers had or had not received dexamethasone prior to delivery. As mentioned earlier, dexamethasone is thought to cause increased cholesterol synthesis.  4.7 Overall conclusions The findings of this thesis are as follows:  1) The large increase in plasma LDL-cholesterol levels between 1 and 4 days of age in the guinea pig was accompanied by an increase in hepatic HMG CoA reductase activity. The decrease in plasma LDL-cholesterol levels between 4 and 8 days of age was coincident with a fall in hepatic HMG CoA reductase activity. This suggests that in the guinea pig, increased hepatic HMG CoA reductase activity may contribute to the postnatal rise in plasma cholesterol levels. Plasma lathosterol concentrations also increased after birth but remained elevated at 8 days of age. This divergence between these two potential measures of cholesterol synthesis may show that only hepatic  99  cholesterol synthesis is correlated with plasma cholesterol levels. LDL receptor relative mass was constant over the first 8 days after birth indicating that postnatal changes in plasma cholesterol levels are not dependent on changes in the expression or maturation of the LDL receptor. It also appeared that liver cholesterol stores did not contribute to the plasma cholesterol pool during this time. Hepatic HMG CoA reductase activity and LDL receptor mass were lower in neonates than in adults while plasma lathosterol levels were higher in neonates than in adults. This may indicate that extrahepatic, relative to hepatic cholesterol synthesis is elevated in the suckling guinea pig.  2) In term human infants, plasma lathosterol concentrations slightly, but significantly, decreased between birth and 4 days of age. This showed that despite an increase in plasma cholesterol levels over this period, rates of cholesterol synthesis may possibly decrease. At 4 days of age, there were no differences in plasma lathosterol concentrations between infants fed human milk and those fed formula. These data might indicate that the relative lack of cholesterol in the diet of infants fed formula does not stimulate increased rates of cholesterol synthesis to achieve similar plasma cholesterol concentrations.  3) In preterm infants, it was observed that hypercholesterolemia, induced by 10% Intralipid, was associated with increased plasma lathosterol. If this can be assumed to be indicative of an increase in cholesterol synthesis in the neonate, as it is in the adult, these data support the hypothesis that cholesterol synthesis is elevated during 10% Intralipid infusion. Phospholipid from the infused Intralipid is thought to sequester  100  free cholesterol from cell membranes to form LpX particles. This can be expected to result in increased intracellular cholesterol synthesis to replace lost membrane cholesterol. Apoprotein analysis indicated that the increased plasma cholesterol concentrations in the preterm infants studied was not correlated with elevated levels of LDL, VLDL or HDL. It is reasonable to assume, therefore, that the increased free cholesterol in plasma was probably associated with LpX.  The mechanism for "postnatal hypercholesterolemia" remains elusive. Although the rise in plasma cholesterol levels after birth is a seemingly universal phenomenon, there does not appear to be an explanation applicable to all species. Differences in the cholesterol metabolism of the neonatal guinea pig, human and rat, lead one to suspect that distinct mechanisms might exist for each species. Thus, changes in plasma cholesterol in the guinea pig may be mediated by increased cholesterol synthesis, perhaps due to the synthesis of VLDL. In humans, this might be caused by regression of the fetal zone of the adrenal at birth, and in the rat, developmental changes in LDL receptor number might be of paramount importance. It seems improbable, though, that postnatal levels of plasma cholesterol increase in so many species, in a parallel fashion, yet by different mechanisms. It also seems doubtful that this postnatal phenomenon is totally independent of the onset of enteral feeding. Two of the methods used in these studies have great potential for use in studies of cholesterol metabolism under various physiological conditions. Both offer advantages over other methods. Most importantly, ligand blotting of the LDL receptor is not affected by the high hepatic nonspecific binding that plagues other procedures. Also, measurement of lathosterol is relatively simple and requires only plasma, not liver. It is often asked whether or not cholesterol should be added to infant formula to concentrations present in human milk. The importance of dietary cholesterol to young animals will not be known until the late effects of early nutrition are elucidated. Long term studies, such as those  101  recently performed with baboons fed either baboon milk or formula with varying amounts of added cholesterol (Mott et al., 1990), will perhaps answer this question. Further contributions to this field may result from studies in Britain of adult humans of approximately 60-70 years of age, for whom neonatal records exist (Barker, 1990). Finally, the situation of the very low birth weight preterm infant is an example of how knowledge of normal perinatal cholesterol metabolism is necessary for detection of abberations. The preterm infant administered 10% Intralipid has a profound hypercholesterolemia and elevation of plasma lathosterol levels (a potential indicator of cholesterol synthesis). The population of preterm infants will likely expand both in number and range of gestational age. Understanding the effect of early nutrition of these infants on their health in later life is crucial to determining their best nutritional care.  102  REFERENCES Amr S, Chowdry P, Hamosh P, Hamosh M 1988 Low levels of apolipoprotein Al are not contributors to the low lecithin-cholesterol acyl transferase activity in premature newborn infants. PediatrRes 24:191-193. Andersen GE 1985 Changes in plasma lipoproteins from first day to third week of human life. In: Detection and Treatment of Lipid and Lipoprotein Disorders of Childhood. Alan R Liss, Inc., NY, NY, pp 87-91. Andersen JM, Dietschy JM 1977 Regulation of sterol synthesis in 16 tissues of rat. J Biol Chem 253:9024-9032. Andersen GE, Johansen KB 1980 LDL receptor studies in term and pre-term infants: measurement of sterol synthesis in cord blood lymphocytes. ACTA Paediatr Scand 69:577-580. Angelin B, Raviola CA, Innerarity TL, Mahley RW 1983 Regulation of hepatic lipoprotein receptors in the dog. J Clin Invest 71:816-831. Auestad N, Hamilton JJ, Innis SM 1991 Identification of the LDL (B/E) receptor during early development in rat liver using a sensitive ligand blotting assay. FASEB J 5:A950. Averna MR, Barbagallo CM, Di Paulo G, Labisi M, Pinna G, Marino G, Dimita U, Notarbartolo A 1991 Lipids, lipoproteins and apolipoproteins AI, AH, B, CII, Oil and E in newborns. Biol Neonate 60:187-192. Bargen-Lockner C, Hahn P, Pendray M, Riddell G 1983 Effect of Intralipid on total and highdensity lipoprotein cholesterol levels in newborns and infants. Biol Neonate 44:272-277. Barker DJP 1990 The fetal and infant origins of adult disease. Br Med J 301:1111. Bartholow LC, Geyer RP 1982 Sterol efflux from mammalian cells induced by human serum albumin-phospholipid complexes. J Biol Chem 257:3126-3130. Bauer JE 1990 Increased serum and liver lipid mass and hepatic 3-hydroxy-3-methylglutaryl CoA reductase activities in rabbits fed soy protein saturated fat diets. Artery 17:176-188. Beisiegel U, Schneider WJ, Goldstein JL, Anderson RGW, Brown MS 1981 Monoclonal antibodies to the low density lipoprotein receptor as probes for study of receptor-mediated endocytosis and the genetics of familial hypercholesterolemia. J Biol Chem 256:11923-11931. Berenson GS, Frank GC, Hunter SM, Srinivasan SR, Voors AW, Webber LS 1982 Cardiovascular risk factors in children; should they concern the pediatrician. Am J Pis Child 136:855-862. Berkow SE, Spear ML, Stahl GE, Gutman A, Polin RA, Pereira GR, Olivecrona T, Hamosh P, Hamosh M 1987 Total parenteral nutrition with intralipid in premature infants receiving TPN with heparin: effect on plasma lipolytic enzymes, lipids, and glucose. J Pediatr Gastroenterol Nutr 6:581588.  103  Bihain BE, Deckelbaum RJ, Yen FT, Gleeson AM, Carpentier YA, Witte LD 1989 Unesterified fatty acids inhibit the binding of low density lipoproteins to the human fibroblast low density lipoprotein receptor. J Biol Chem 264:17316-17321. Bilheimer DW, Goldstein JL, Grundy SM, Starzl TE, Brown MS 1984 Liver transplantation to provide low-density-lipoprotein receptors and lower plasma cholesterol in a child with homozygous familial hypercholesterolemia. New Engl J Med 311:1658-1664. Bizzi A, Vereoni E, Garattini S 1963 Hypercholesterolemia in suckling rats. J Athero Res 3:121128. Bjorkhem I, Miettinen T, Reihner E, Ewerth S, Angelin 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. Blazquez E, Sugase T, Blazquez M, Foa PP 1974 Neonatal changes in the concentration of rat liver cAMP and of serum glucose, free fatty acids, insulin, pancreatic, and total glucagon in man and in the rat. J Lab Clin Med 83:957-967. Blum CB, Davis PA, Forte TM 1985 Elevated levels of apolipoprotein E in the high density lipoproteins of human blood cord plasma. J Lipid Res 26:755-760. Boguslawski W, Wrobel J 1974 An inhibitor of sterol biosynthesis present in cow's milk. Nature 247:210-211. Bohmer T, Havel RJ, Long JA 1972 Physiological fatty liver and hyperlipemia in the fetal guinea pig: chemical and ultrastructural characteristics. J Lipid Res 13:371-382. Bortz WM 1967 Fat feeding and cholesterol synthesis. Biochim Biophvs ACTA 137:533-539. Boulton, TJC 1979 Fetal, maternal and intrapartum factors and their effects on cord serum cholesterol and triglyceride. Aust NZ J Med 9:57-62. Brans YW, Andrew DS, Carrillo DW, Dutton EP, Menchaca EM, Puleo-Scheppke BA 1988 Tolerance of fat emulsions in very-low-birth-weight neonates. Am J Pis Child 142:145-152. Brans YW, Andrew DS, Carrillo DW, Dutton EP, Menchaca EM, Puleo-Scheppke BA 1990 Tolerance of fat emulsions in very low birthweight neonates: effect of birth weight on plasma lipid concentrations. Am J Perinatol 7:114-117. Brown MS, Goldstein JL, Vandenberghe K, Fryns JP, Kovanen PT, Eeckels R, van den Berghe H, CassimanJJ 1978 Prenatal diagnosis of homozygous familial hypercholesterolemia. Lancet i:526529. Brown MS, Kovanen, PT, Goldstein JL 1979 Receptor-mediated uptake of lipoprotein-cholesterol and its utilization for steroid synthesis in the adrenal cortex. Recent Prog Horm Res 35:215-257. Brown MS, Goldstein JL 1984 How LDL receptors influence cholesterol and atherosclerosis. Scientific American 251:58-66.  104  Brown MS, Goldstein JL 1985 The LDL receptor and HMG CoA reductase - two membrane molecules that regulate cholesterol homeostasis. Current Topics in Cellular Regulation 26:3-15. Brown M, Goldstein J 1986 A receptor-mediated pathway for cholesterol homeostasis. Science 232:34-47. Bruenger E, Rilling HC 1986 Prenyltransferase and squalene synthetase in livers of neonate rats. Biochim Biophvs ACTA 876:500-506. Cai H, Xie C, Chen X, Chen Y 1990 The relationship between hepatic low-density lipoprotein hepatic receptor activity and serum cholesterol level in the human fetus. Hepatology 13:852-857. Carr BR, Ohashi M, MacDonald PC, Simpson ER 1980 Human anenchephalic adrenal tissue: low density lipoprotein metabolism and cholesterol synthesis. J Clin Endocrinol Metab 53:406-411. Carr BR, Simpson ER 1981a Synthesis of cholesterol in the human fetus: 3-hydroxy-3methylglutaryl coenzyme A reductase activity of liver microsomes. J Clin Endocrinol Metab 53:810812. Carr BR, Simpson ER 1981b Lipoprotein utilization and cholesterol synthesis by the human fetal adrenal gland. Endocr Rev 2:306-326. Carr BR, Simpson ER 1982 Cholesterol synthesis in human fetal tissues. J Clin Endocrinol Metab 55:447-452. Carr BR, Simpson ER 1984 Cholesterol synthesis by human fetal hepatocytes: effect of lipoproteins. Am J Obstet Gynecol 150:551-557. Carroll KK 1964 Acetate incorporation into cholesterol and fatty acids by livers of fetal, suckling, and weaned rats. Can J Biochem 42:79-86. Carroll KK, Hamilton RMG, MacLeod GK 1973 Plasma cholesterol levels in suckling and weaned calves, lambs, pigs and colts. Lipids 8:635-640. Carroll KK, Huff MW 1977 Influence of dietary fat and protein on plasma cholesterol in the early postnatal period. Adv Med Biol 82:638-643. Chart A, Bierman EL, Albers JJ 1979a Low-density lipoprotein receptor activity in cultured human skin fibroblasts. J Clin Invest 64:1309-1319. Chait A, Bierman EL, Albers JJ 1979b Regulatory role of triiodothyronine in the degradation of low density lipoprotein by cultured human skin fibroblasts. J Clin Endocr in Metab 48:887-889. Chait A, Foster D, Miller DG, Bierman EL 1981 Acceleration of low-density lipoprotein catabolism in man by total parenteral nutrition. Proc Soc Exp Biol Med 168:97-104. Chevallier F 1964 Transfer et synthese du cholesterol chez le rat au cours de sa croissance. Biochim Biophvs ACTA 84:316-339.  105  Chong KS, Nicolosi RJ, Rodger RF, Arrigo DA, Yuan RW, MacKey JJ, Georas S, Herbert PN 1987 Effect of dietary fat saturation on plasma lipoproteins and high density lipoprotein metabolism of the rhesus monkey. J Clin Invest 79:675-683. Coates PM, Brown SA, Sonaware BR, Koldovsky O 1983 Effect of early nutrition on serum cholesterol levels in adult rats challenged with high fat diet. J Nutr 113:1046-1050. Cole JWL, Grizzle JE 1966 Applications of multivariate analysis to repeated measures experiments. Biometrics 22:810-828. Connor WE, Lin DS 1967 Placental transfer of cholesterol-4-MC into rabbit and guinea pig fetus. J Lipid Res 8:558-564. Cooke RJ, Yeh YY, Gibson D, Debo D, Bell GL 1987 Soybean oil emulsion administration during parenteral nutrition in the preterm infant: effect on essential fatty acid, lipid and glucose metabolism. J Pediatr 111:767-773. Cress HR, Shaher RM, Laffin R, Karpowicz K 1977 Cord blood hyperlipoproteinemia and perinatal stress. Pediatr Res 11:19-23. Csako G, Csernyanszky H, Bobok I, Ludmany K 1974 Lipoprotein fractions in maternal, cord and newborn serum. ACTA Paed Acad Sci Hungary 15:101-108. Dahms BB, Halpin TC 1980 Pulmonary arterial lipid deposit in newborn infants receiving intravenous lipid infusion. J Pediatr 97:800-805. Darmady JM, Fosbrooke AS, Lloyd JK 1972 Prospective study of serum cholesterol levels during first year of life. Br Med J 2:685-688. Decsi T, Molnar D, Klujber L 1990 Lipid levels in very low birthweight preterm infants. ACTA Paediatr Scand 79:577-580. Dhanireddy R, Hamosh M, Sivasubramanian K, Chowdhry P, Scanlon J, Hamosh P 1981 Postheparin lipolytic activity and Intralipid clearance in very low-birth-weight infants. J Pediatr 98:617-622. Drevon CA, Norum KR 1975 Cholesterol esterification and lipids in plasma and liver from newborn and young guinea pigs raised on milk and non-milk diets. Nutr Metab 18:137-151. Edwards PA 1975 Effect of plasma lipoproteins and lecithin-cholesterol dispersions on the activity of 3-hydroxy-3-methylglutaryl-coenzyme A reductase of isolated rat hepatocytes. Biochim Biophys ACTA 409:39-50. Erickson SK, Bruscalupi G, Conti Devirgiliis L, Leoni S, Mangiantini MT, Spagnuolo S, Trentalance A 1988 Changes in parameters of lipoprotein metabolism during rat hepatic devlopment. Biochim Biophys ACTA 963:525-533. Fainaru M, Havel RJ, Imaizumi K 1977 Apolipoproteincontent ofplasma lipoprotein of the rat separated by gel chromatography or ultracentrifugation. Biochem Med 17:347-355.  106  Farkkila MA, Miettinen TA 1988 Plasma lathosterol and campesterol in detection of ileal dysfunction. Scand J Gastroenterol 23:19-25. 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. Fielding CJ, Fielding PE 1985 Metabolism of cholesterol and lipoproteins. In: Vance DE, Vance JE (eds) Biochemistry of Lipids and Membranes, Benjamin/Cummings Publishing Company, Inc., Menlo Park, CA, pp. 404-471. Filler RM, Takada Y, Carraras T, Heim T 1980 Serum Intralipid levels in neonates during parenteral nutrition: the relation to gestational age. J Pediatr Surg 15:405-410. Folch J, Lees M, Sloane-Stanley, GH 1957 A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem 226:497-509. Fong LG, Bonney E, Kosek JC, Cooper AD 1989 Immunohistochemical localization of low density lipoprotein receptors in adrenal gland, liver and intestine. J Clin Invest 84:847-856. Forte TM, Genzel-Boroviczeny O, Austin MA, Kao LC, Scott C, Albers JJ, D'Harlingue AE 1989 Effect of total parenteral nutrition with intravenous fat on lipids and high-density lipoprotein heterogeneity in neonates. J Parenteral Enteral Nutr 13:490-500. Fox JC, McGill HC, Carey KD, Getz GS 1987 In vivo regulation of hepatic LDL receptor mRNA in the baboon. J Biol Chem 262:7014-7020. Franklin FA, Watkins JB, Heafitz L, Clowes AW, Breslow JL 1976 Serum lipids during total parenteral nutrition with intralipid. Pediatr Res 10:354. Friedman G, Wernette-Hammond ME, Hui DY, Mahley RW, Innerarity TL 1987 Characterization of lipoprotein receptors on rat Fu5AH hepatoma cells. J Lipid Res 28:1482-1494. Friedman M, Byers SO 1961 Effects of diet on serum lipids of fetal, neonatal, and pregnant rabbits. Am J Phvs 201:611-616. Friedman Z, Marks KH, Maisels J, Thorson R, Naeye R 1978 Effect of parenteral fat emulsion on the pulmonary and reticuloendothelial systems in the newborn infant. Pediatr 61:694-698. Genzel-Boroviczeny O, D'Harlingue AE, Kao LC, Scott C, Forte TM 1988 High-density lipoprotein subclass distribution in premature newborns before and after onset of enteral feeding. Pediatr Res 23:543-547. Gherardi E, Brugni N, Bowyer DE 1988 Purification of low density lipoprotein receptor from liver and its quantification by anti-receptor monoclonal antibodies. Biochem J 253:409-415.  107  Ginsberg BE, Zetterstrom R 1977 High density lipoprotein concentrations in newborn infants. ACTA Paediatr Scand 66:39-41. Ginsberg BE, Zetterstrom R 1980 Serum cholesterol concentrations in newborn infants with gestational ages of 28-42 weeks. ACTA Paediatr Scand 69:587-592. Glomset JA, Nichols AV, Norum KR, King W, Forte T 1973 Plasma lipoproteins in familial LCAT deficiency. J Clin Invest 52:1078-1092. Goldstein JL, Harrod, MJE, Brown MS 1974 Homozygous familal hypercholesterolemia: specificity of the biochemical defect in cultured cells and feasibility of prenatal detection. Am J Hum Genet 26:199-206. Goldstein JL, Kita T, Brown MS 1983 Defective lipoprotein receptors and atherosclerosis. Lessons from an animal conterpart of familial hypercholesterolemia. New En^l J Med 309:288-296. Goodwin CD, Margolis S 1976 Improved methods for the study of hepatic HMG CoA reductase: one step isolation of mevalonolactone and rapid preparation of endoplasmic reticulum. J Lipid Res 17:297-303. Greenberg LD 1970 Plasma cholesterol levels of monkeys maintained on various controlled and deficient diets. Am J Clin Nutr 23:1101-1104. Griffin E, Breckenridge WC, Kuksis A, Bryan MH, Angel A 1979 Appearance and characterization of lipoprotein X during continuous Intralipid infusions in the neonate. J Clin Invest 64:1703-1712. Guo LSS, Hamilton R, Ostwald R, Havel RJ 1982 Secretion of nascent lipoproteins and apoproteins by perfused livers of normal and cholesterol-fed guinea pigs. J Lipid Res 23:543-555. Gylling H, Vanhanen V, Miettinen TA 1989 Effects of acipimox and cholestyramine on serum lipoproteins, non-cholesterol sterols and cholesterol absorption and elimination. Eur J Clin Pharmacol 37:111-115. Haave NC, Innis SM 1991 Perinatal development of hepatic cholesterol synthesis in the rat. Biochim Biophvs ACTA 1085:35-44. Hahn P 1984 Effect of litter size on plasma cholesterol and insulin and some liver and adipose tissue enzymes in adult rodents. J Nutr 114:1231-1234. Hahn P, Koldovsky O 1966 Utilization of nutrients during postnatal development. Pergamon Press, London, p. 100. Hahn P, Kirby L 1973 Immediate and late effects of premature weaning and of feeding a high fat or carbohydrate diet to weanling rats. J Nutr 103:690^696. Hahn P, Walker BL 1979 Hepatic 3-hydroxy-3-methyl glutaryl CoA reductase response to litter size in suckling rats. Can J Biochem 57:1216-1219.  108  Hahn P, Innis SM 1984 Cholesterol oxidation and 7orhydroxylation during postnatal development of the rat. Biol Neonate 46:48-52. Hamilton RMG, Carroll KK 1977 Plasma cholesterol levels in suckling and weaned kittens, puppies, and guinea pigs. Lipids 12:145-148. Handley DA, Arbeeny CM, Witte LD, Chien S 1981 Colloidal gold-low density lipoprotein conjugates as membrane receptor probes. Proc Natl Acad Sci USA 78:368-371. Harris RA, MacNintch JE, Quackenbush FW 1966 Mechanism of suckling rat hypercholesterolemia: dietary and drug studies. J Nutr 90:40-46. Haumont D, Deckelbaum RJ, Richelle M, Dahlan W, Coussaert E, Bihain BE, Carpentier YA 1988 Plasma lipid and plasma lipoprotein concentrations in low birth weight infants given parenteral nutrition with twenty or ten percent lipid emulsion. J Pediatr 115:787-793. Havinga JR, Lohse P, Beigiegel U 1987 Immunoblotting and ligand blotting of the low-density lipoprotein receptor of human liver, HepG2 cells and HeLa cells. FEBS 216:275-280. Hegsted DM, McGandy RM, Myers ML, Stare RJ 1965 Quantitative effects of dietary fats on serum cholesterol in man. Am J Clin Nutr 17:281-295. Higgs SC, Malan AF, Heese HDV 1974 A comparison of oral feeding and total parenteral nutriton in infants of very low birthweight. South African Med J 48:2169-2173. Hollanders B, Aude X, Girard-Globa A 1985 Lipoproteins and apoproteins in fetal and newborn piglets. Biol Neonate 47:270-279. Ibrahim JBT, McNamara DJ 1988 Cholesterol homeostasis in guinea pigs fed saturated and polyunsaturated fat diets. Biochim Biophvs ACTA 963:109-118. Imrie CG, Graham SG 1920 The fat content of embryonic livers. J Biol Chem 44:243-254. Ingebritsen TS, Geelen MJH, Parker RA, Evenson KJ, Gibson DM 1979 Modulation of hydroxymethylglutaryl CoA reductase activity, reductase kinase activity and cholesterol synthesis in rat hepatocytes in response to insulin and glucagon. J Biol Chem 254:9986-9989. Innerarity TL, Bersot TP, Arnold KS, Weisgraber KH, Davis PA, Forte TM, Mahley RW 1984 Receptor binding activity of high-density lipoproteins containing apoprotein E from abetalipoproteinemic and normal neonate plasma. Metabolism 33:186-195. Innis SM 1983 Influence of maternal cholestyramine treatment on cholesterol and bile acid metabolism in adult offspring. J Nutr 113:2464-2470. Innis SM 1989 Changes in plasma cholesterol density distribution of young adult male rats fed a high fat and cholesterol diet following maternal cholestyramine treatment. J Nutr 119:373-379. Innis SM, Boyd MC 1983 Cholesterol and bile acid synthesis during total parenteral nutrition with and without lipid emulsion in the rat. Am J Clin Nutr 38:95-100.  109  Innis SM, Frohlich J, McLeod R, Allardyce DB, Hahn P 1985 Serial measurement of plasma cholesterol and LCAT activity in adults receiving TPN. J Parenter Ent Nutr 9:34-37. Jacobson NL, Richard M, Berger PJ 1973 Depression of plasma cholesterol in calves by supplementing a high cholesterol liquid diet with dry feed. J Nutr 103:1533-1536. Jain S 1985 Prematurity and lecithin-cholesterol acyl transferase deficiency in newborn infants. Pediatr Res 19:58-60. Jakoi L, Quarfordt SH 1974 The induction of hepatic cholesterol synthesis in the rat by lecithin mesophase infusions. J Biol Chem 249:5840-5844. Jensen RG 1989 Lipids in human milk - composition and fat soluble vitamins. In: Lebenthal E (ed) Textbook of Gastroenterology and Nutrition in Infancy, 2nd ed. Raven Press, New York, pp 157208. Johansson M 1983 Lipoproteins and lipids in fetal, neonatal and adult rat serum. Biol Neonate 44:278-286. Johansson M, Karlsson B 1982 Lipoprotein and lipid profiles in the blood serum of the fetal, neonatal and adult pig. Biol Neonate 42:127-137. Johnson HJ, Simpson ER, Carr BR, MacDonald PC, Parker CR 1982 The levels of plasma cholesterol in the human fetus throughout gestation. Pediatr Res 16:682-683. Jones PJH, Schoeller DA 1990 Evidence for diurnal periodicity in human cholesterol synthesis. J Lipid Res 31:667-674. Jones PJH, Hrboticky N, Hahn P, Innis SM 1990 Comparison of breast-feeding and formula feeding on intestinal and hepatic cholesterol metabolism in neonatal pigs. Am J Clin Nutr 51:979584. Kallio MJT, Siimes MA, Perheentupa J, Salmenpera L, Miettinen TA 1989 Cholesterol and its precursors in human milk during prolonged exclusive breast-feeding. Am J Clin Nutr 50:782-785. Kandutch AA, Chen HW 1977 Consequences of blocked sterol synthesis in cultured cells. J Biol Chem 252:409-415. Kao LC, Cheng MH, Warburton D 1984 Triglycerides, free fatty acids, free fatty acids/albumin molar ratio, and cholesterol levels in serum of neonates receiving long-term lipid infusions: controlled trial of continuous and intermittent regimes. J Pediatr 104:429-435. Kaplan A, Lee VF 1965 Serum lipid levels in infants and mothers at parturition. Clin Chim ACTA 12:258-263. Kempen HJM, Glatz JFC, Leuven JAG, van der Voort HA, Katan MB 1988 Serum lathosterol concentration is an indicator of whole-body cholesterol synthesis in humans. J Lipid Res 29:11491155.  110  Kirstein D, Johansen KB, Petersen M, Andersen GE 1985 Changes in plasma lipoproteins from first day to third week of life in healthy breast-fed infants I. Lipid and protein composition of lipoproteins. ACTA Paed Scand 74:733-737. Koldovsky O, Thornburg W 1987 Hormones in milk. J Ped Gastroenterol Nutr 6:172-196. Kovanen PT, Brown MS, Goldstein JL 1979a Increased binding of LDL to liver membranes from rats treated with 17aethinyl estradiol. J Biol Chem 254:11367-11373. Kovanen PT, Basu SK, Goldstein JL, Brown MS 1979b Low density lipoprotein receptors in bovine adrenal cortex II. Low density binding to membranes prepared from fresh tissue. Endocrinology 104:610-616. Kovanen PT 1987 Regulation of plasma cholesterol by hepatic low-density receptors. Am Heart J 133:464-469. Krause BR, Hartman AD 1984 Adipose tissue and cholesterol metabolism. J Lipid Res 25:97-110. Laemmli UK 1970 Cleavage of structural protein during the assembly of the head of bacteriophage T4. Nature 227:680-685. Lafeber HN, Rolph TP, Jones CT 1984 Studies on the growth of the fetal guinea pig. The effect of ligation of the uterine artery on organ growth and development. J Dev Phys 6:441-459. Laird CW, Fox RR 1970 The effect of age on cholesterol and PBI levels in the III/IIIc hybrid rabbit. Life Sciences 9:1243-1253. Lane DM, McConathy WJ 1983 Factors affecting the lipid and apoprotein levels of cord sera. PediatrRes 17:83-91. Lane DM, McConathy WJ 1986 Changes in the serum lipids and apolipoproteins in the first four weeks of life. Pediatr Res 20:332-337. Leat WMF, Kubasek FOT, Buttress N 1976 Plasma lipoproteins of lambs and sheep. Quarterly J Exp_Phys 61:193-202. Leoni S, Spanuolo S, Conti-Devergilis L, Dini L, Mangiantini MT, Trentalance A 1984 Cholesterogenesis and related enzymes in isolated rat hepatocytes during pre- and postnatal life. J Cell Phvs 118:62-66. Lester R 1980 Physiologic cholestasis. Gastroenterology 78:864-870. Levin MS, Pitt AJA, Schwartz AL, Edwards PA, Gordon JI 1989 Developmental changes in the expression of genes involved in cholesterol biosynthesis and lipid transport in human and rat fetal and neonatal livers. Biochim Biophvs ACTA 1003:293-300. Li JR, Dinh DM, Ellefson RD, Subbiah MT 1979a Sterol and bile acid metabolism during development. 3. Occurrence of neonatal hypercholesterolemia in guinea pig and its possible relation to bile acid pool. Metabolism 28:151-156.  Ill  Li JR, Bale LK, Subbiah MTR 1979b Effect of enhancement of cholesterol degradation during neonatal life of guinea pig on its subsequent response to dietary cholesterol. Atherosclerosis 32:9398. Li JR, Bale LK, Kottke BA 1980 Effect of neonatal modulation of cholesterol homeostasis on subsequent response to cholesterol challenge in adult guinea pig. J Clin Invest 65:1060-1068. Liggins GC, Howie RN 1972 A controlled trial of antepartum glucocorticoid treatment for prevention of the respiratory distress syndrome in premature infants. Pediatr 50:515525. Lima LAM, Murphy JF, Stansbie D, Rowlandson P, Gray OP 1988 Neonatal parenteral nutrition with a fat emulsion containing medium chain triglycerides. ACTA Paed Scand 77:332-339. Little ME, Hahn P 1990 Diet and metabolic development. FASEB J 4:2605-2611. Lombardi P, Norata G, Maggi FM, Canti G, Franco P, Nicolin A, Catapano AL 1989 Assimilation of LDL by experimental tumours in mice. Biochim Biophys ACTA 1003:301-306. Lowry OH, Rosebrough NS, Farr AL, Randall RJ 1951 Protein measurement with the folin phenol reagent. J Biol Chem 193:256-275. Maggi FM, Catapano AL 1987 A dot-blot assay for the low density lipoprotein receptor. J Lipid Res 28:108-111. Mahley RW, Hui DY, Innerarity TL, Weisgeraber KH 1981 Two independent lipoprotein receptors on hepatic membranes of dog, swine and man. J Clin Invest 68:1197-1206. Markwell MAK, Haas SM, Tolbert NE, Bieber LL 1981 Protein determination in membrane and lipoprotein samples: manual and automated procedures. In: Calowick SP, Kaplan NO (eds) Methods in Enzymology, vol. 72, Academic Press, New York, pp. 296-303. McConathy WJ, Lane DM 1980 Studies on the apolipoproteins and lipoproteins of cord serum. Pediatr Res 14:757-761. McConathy WJ, Blackett PR, Kling OR 1981 Studies on serum apolipoproteins and lipids in amniotic fluid and neonatal urine. Clin Chim ACTA 111:153-162. McNamara DJ 1984 Cholesterol homeostasis in the guinea pig. Biochim Biophys ACTA 796:5154. McNamara DJ, Quackenbush FW, Rodwell VW 1972 Regulation of hepatic 3-hydroxy-3-methylglutaryl coenzyme A reductase. J Biol Chem 247:5805-5810. Meijer GW, van der Palen JGP, de Vries H, Kempen HJM, van der Voort HA, van Zutphen LFM, Beynen AC 1992 Evaluation of the use of serum lathosterol concentration to assess whole-body cholesterol synthesis in rabbits. J Lipid Res 33:281-286.  112  Messing B, Peynet J, Poupon J, Pfeiffer A, Thuillier F, Chazouilleres O, Legrand A 1990 Effect of fat-emulsion phospholipids on serum lipoprotein profile during 1 mo cyclic total parenteral nutrition. Am J Clin Nutr 52:1094-1100. Miettinen T 1982 Diurnal variation of cholesterol precursors squalene and methyl sterols in human plasma lipoproteins. J Lipid Res 23:466-473. Miettinen TA, Tilvis RS, Kesaniemi YA 1990 Serum plant sterols and cholesterol precursors reflect cholesterol absorption and synthesis in volunteers of a randomly selected male population. Am J Epidemiol 131:20-31. Mills GL, Lane PA, Weech PK 1984 Chapter 1. The isolation and purification of plasma lipoproteins. In: A Guidebook to Lipoprotein Technique, Elsevier, Amsterdam, pp. 18-114. Mott GE, Jackson EM, McMahan CA, McGill, HC 1990 Cholesterol metabolism in adult baboons is influenced by infant diet. J Nutr 120:243-251. Naseem SM, Khan MA, Jacobson MS, Nair P, Heald FP 1980 The influence of dietary cholesterol and fat on the homeostasis of cholesterol metabolism in early life in the rat. Pediatr Res 14:10611066. Nervi FO, Dietschy JM 1975 Ability of six different lipoprotein fractions to regulate the rate of hepatic cholesterogenesis in vivo. J Biol Chem 250:8704-8711. Ness GC, Miller JP, Moffler MH, Woodsand LS, Harris HB 1979 Perinatal development of 3hydroxy-3-methylglutaryl coenzyme A reductase activity in rat lung, liver and brain. Lipids 14:447450. Nestel PJ, Poyser A, Boulton TJC 1979 Changes in cholesterol metabolism in infants in response to dietary cholesterol and fat. Am J Clin Nutr 32:2177-2182. Nikkila K, Miettinen TA 1988 Serum cholesterol precursors, cholestanol, and plant sterols in primary biliary cirrhosis. Scand J Gastroenterol 23:967-972. O'Brien BC, McMurray DN, Reiser R 1983 The influence of premature weaning and the nature of the fat in the diet during development on adult plasma lipids and adipose cellularity in pair-fed rats. J Nutr 113:602-609. Papadopoulos A, Hamosh M, Chowdhry P, Scanlon JW, Hamosh P 1988 Lecithin-cholesterol acyltransferase in newborn infants: low activity level in preterm infants. J Pediatr 113:896-898. Parker CR, Simpson ER, Bilheimer DW, Leveno K, Carr BR, MacDonald PC 1980 Inverse relation between low-density lipoprotein-cholesterol and dehydroisoandrosterone sulfate in human fetal plasma. Science 208:512-514. Parker CR, Carr BR, Simpson ER, MacDonald PC 1983 Decline in the concentration of lowdensity lipoprotein-cholesterol in human fetal plasma near term. Metabolism 32:919-923. Parker CR, MacDonald PC, Carr BR, Morrison JC 1987 The effects of dexamethasone and anencephaly on newborn serum levels of apoprotein A-I. J Clin Endocrinol Metab 65:1098-1101. 113  Parker TS, McNamara DJ, Brown CD, Kolb R, Ahrens EH, 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. Plonne" D, Schlag B, Winkler L, Dargel R 1990 Tracer kinetic studies of the low density lipoprotein metabolism in the fetal rat: an example for estimation of flux rates in the nonsteady state. J Lipid Res 31:747-752. Potter J 1977 Perinatal plasma lipid concentrations. Aust NZ J Med 7:155-160. Raghunathan R, Arinze D 1977 Perinatal development of gluconeogenesis in guinea-pig liver. Int J Biochem 8:737-743. Reihner E, Rudling M, Stahlberg D, Berglund L, Ewerth S, Bjorkhem I, Einarsson K, Angelin B 1990 Influence of pravastatin, a specific inhibitor of HMG CoA reductase, on hepatic metabolism of cholesterol. New Engl J Med 323:224-228. Reiser R, Henderson GR, O'Brien BC 1977 Persistence of dietary suppression of 3-hydroxy-3methylglutaryl coenzyme-A reductase during development in rats. J Nutr 107:1131-1138. Reiser R, O'Brien B, Henderson G, Moore R 1979 Studies on a possible function for cholesterol in milk. Nutr Reports Int 19:835-849. Roach PD, Zollinger M, Noel S-P 1987 Detection of the low density lipoprotein (LDL) receptor on nitrocellulose paper with colloidal gold-LDL conjugates. J Lipid Res 28:1515-1521. Roberts DCK, West CE 1974 Influence of rabbit milk on cholesterolemic response on offspring of rabbits. Lipids 9:495-497. Roberts DCK, Huff MW, Carroll KK 1979 Influence of plasma cholesterol concentrations in suckling and weanling rabbits. Nutr Metab 23:476-486. Rodwell VW, Nordstrom JL, Mitschelen JJ 1976 Regulation of HMG-CoA reductase. Adv Lipid Res 14:1-74. Rovamo LM 1985 Post-heparin plasma lipases and carnitine in infants during parenteral nutrition. Pediatr Res 19:292-296. Rovamo LM, Nikkila EA, Raivio KO 1988 Lipoprotein lipase, hepatic lipase, and carnitine in premature infants. Arch Pis Child 63:140-147. Rudling MJ, Peterson CO 1985 LDL receptors in bovine tissues assayed as the heparin-sensitive binding of 125I-labeled LDL in homogenates: relation between liver LDL receptor and serum cholesterol in the fetus and post term. Biochim Biophys ACTA 836:96-104. Salter AM, Fisher SC, Brindley DN 1987 Binding of low-density lipoprotein to monolayer cultures of rat hepatocytes is increased by insulin aqnd decreased by dexamethasone. FEBS 71:159-162.  114  Salter AM, Fisher SC, Brindley DN 1988 Interactions of triiodothyronine, insulin and dexamethasone on the binding of human LDL to rat hepatocytes in monolayer culture. Atherosclerosis 71:77-80. Sather SD, May KA, Cooper AD, Cushman SW, Kraemer FB 1988 Distribution of low density lipoprotein receptors in the isolated rat adipocyte. Clin Res 36:159a. Semenkovitch CF, Ostlund RE 1987 Estrogens induce low-density lipoprotein receptor activity and decrease intracellular cholesterol in human hepatoma cell line Hep G2. Biochemistry 26:4987-4992. Shakespeare V, Postle AD 1979 Regulation of cholesterol synthesis in skin fibroblasts derived from old people. Atherosclerosis 33:359-364. Shapiro DJ, Rodwell VW 1971 Regulation of hepatic 3-hydroxy-3-methylglutaryl coenzyme A reductase and cholesterol synthesis. J Biol Chem 246:3210-3216. Sharma SC, Misra PK, Tolani MK, Kaul R 1983 Lipid profile in the early neonatal period in normal and low birth weight infants. Indian Pediatr 20:179-183. Shennan A, Bryan M, Angel A 1977 The effect of gestation age on Intralipid tolerance in newborn infants. J Pediatr 91:134-137. Shepherd J, Packard CJ, Grundy SM, Yeshurum D, Gotto AM, 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. Soutar AK, Harders-Spengel K, Wade DP, Knight BL 1986 Detection and quantitation of low density lipoprotein (LDL) receptors in human liver by ligand blotting, immunoblotting and radioimmunoassay. J Biol Chem 261:17127-17133. Soutar AK, Knight BL 1990 Structure and regulation of the LDL-receptor and its gene. Br Med Bull 46:891-916. 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, Bilheimer DW, Dietschy JM 1983a Rates of receptor-dependent and -independent low density lipoprotein uptake in the hamster. Proc Natl Acad Sci USA 80:3499-3503. Spady DK, Turley SD, Dietschy JM 1983b Dissociation of hepatic cholesterol synthesis from hepatic low-density lipoprotein uptake and biliary cholesterol saturation in female and male hamsters of different ages. Biochim Biophvs ACTA 753:381-392. Spady DK, Bilheimer DW, Dietschy JM 1985 Rates of low density lipoprotein uptake and cholesterol synthesis are regulated independently in the liver. J Lipid Res 26:465-472. 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.  115  Stahl GE, Spear ML, Hamosh M 1986 Intravenous administration of lipid emulsions to premature infants. Clin Perinatol 13:133-162. Stanley CA, Gonzales E, Baker L 1983 Development of hepatic fatty acid oxidation and ketogenesis in the newborn guinea pig. Pediatr Res 17:224-229. Steinbrecher UP, Witztum JL, Kesaniemi YA, Elam RL 1983 Comparison of glucosylated low density lipoprotein with methylated or cyclohexanedione-treated low density lipoprotein in the measurement of receptor-independent low density lipoprotein catabolism. J Clin Invest 71:960-964. Stozicky F, Slaby P, Volenikova L 1982 The pattern of major serum apolipoproteins during the early neonatal period. ACTA Paed Scand 71:239-241. Strandberg TE, Til vis RS, Miettinen TA 1989 Variations of hepatic cholesterol precursors during altered flows of endogenous and exogenous squalene in the rat. Biochim Biophys ACTA 1001:150156. Strobl W, Widhalm K, Kostner G, Pollak A 1983 Serum apolipoproteins and lipoprotein (a) during the first week of life. ACTA Paed Scand 72:505-509. Subbiah MTR, Yunker RL, Menkhaus A, Poe B 1985 Premature weaning-induced changes of cholesterol metabolism in guinea pigs. Am J Phys 249:E251-E256. Tashiro T, Mashima Y, Yamamori H, Horibe K, Nishizawa M, Sanada M, Okui K 1991 Increased lipoprotein X causes hyperlipidemia during intravenous administration of 10% fat emulsion in man. J Parenteral Enteral Nutr 15:546-550. Tenenbaum D, Gambert P, Meunier S, d'Athis P, Nivelon J, Lallemand C 1988 Serum lipoproteins in venous blood serum from birth to the end of the week: feeding influences. Biol Neonate 53:126131. Thompson GR, Segura R, Hoff H, Gotto AM 1975 Contrasting effects on plasma lipoproteins of intravenous versus oral administration of a triglyceride-phospholipid emulsion. Eur J Clin Invest 5:373-384. Tuchweber B, Ducruet N, Yousef IM, Weber AM 1990 Development of bile secretory function in 2the neonatal guinea pig. Biol Neonate 58:279-290. Turley SD, West CE, Horton BJ 1976 Sterol synthesis in the liver, intestine and lung of the guinea pig. Lipids 11:281-286. Turley SD, Dietschy JM 1982 Cholesterol metabolism and excretion. In: Arias I, Popper H, Schaeter D, Shafritz DA (eds) Liver Biology and Pathobiology, Raven Press, NY NY, pp. 467-492. Untracht SH 1982 Intravascular metabolism of an artificial transporter of triacylglycerols; alterations of serum lipoproteins resulting from total parenteral nutrition with intralipid. Biochim Biophvs ACTA 711:176-192.  116  Van Biervliet JP, Vercaemst R, de Keersgieter W, Vinaimont N, Caster H, Rosseneu M 1980 Evolution of lipoprotein patterns in newborns. ACTA Paed Scand 69:593-596. Van Biervliet JP, Vinaimont N, Caster H, Vercaemst R, Rosseneu M 1981 Plasma apoprotein and lipid patterns in newborns: influence of nutritional factors. ACTA Paed Scand 70:851-856. Van Biervliet JP, Rosseneu M, Bury J, Caster H, Stul MS, Lamote R 1986 Apolipoprotein and lipid composition of plasma lipoproteins in neonates during the first month of life. Pediatr Res 20:324-328. Van Duyne CM, Havel RJ 1959 Plasma unesterified fatty acid concentration in fetal and neonatal life. Proc Soc Expt Biol Med 102:599-602. Viktora J, Fodor J, Grafnetter D, Hahn P, Koldovsky O, Lojda Z 1960 Cs Fvsiol 9:63. Wade DP, Knight BL, Soutar AK 1985 Detection of the low-density-lipoprotein receptors with biotin-low-density lipoprotein. Biochem J 229:785-790. Weinberg RB, Singh KK 1989 Short-term parenteral nutrition with glucose and Intralipid effects on serum lipids and lipoproteins. Am J Clin Nutr 49:794-798. Wells DH, Ferlauto JJ, Forbes DJ, Graham TR, Newell RW, Wareham JA, Wilson CA 1989 Lipid tolerance in the very low birth weight infant on intravenous and enteral feedings. J Parenter Enteral Nutr 13:623-627. Whyte HM, Yee IL 1958 Serum cholesterol levels of Australians and natives of New Guinea from birth to adulthood. Australian Annal of Medicine 7:336-339. Widdowson EM 1950 Chemical composition of newly born mammals. Nature 166:626-628. Widdowson EM, McCance RA 1955 Physiological undernutrition in the newborn guinea-pig. Br J Nutr 9:316-321. Wiggers KD, Jacobson NL, Getty R 1971 Experimental atherosclerosis in the young bovine. Atherosclerosis 14:375-378. Winter JSD, Hughes IA, Reyes FI, Faiman C 1976 Pituitary-gonadal relations in infancy: 2. Patterns of serum gonadal steroid concentrations in man from birth to two years of age. J Clin Metabol Endocrin 42:679-686. Witztum JL, Young SG, Elam RL, Carew TE, Fisher M 1985 Cholestyramine-induced changes in low density lipoprotein composition and metabolism. I. Studies in the guinea pig. J Lipid Res 226:92-103. Wixom RL, Sheng Y, Anderson HL, Yamanaka WK, Terry BE 1976 Some nutrient interrelations during total intravenous alimentation in adult man- a review. Lipids 11:299-306. Yau-Young AO, Rothblat GH, Small DM 1982 Mobilization of cholesterol from cholesterol esterenriched tissue culture cells by phospholipid dispersions. Biochim Biophys ACTA 710:181-187.  117  Yunker RL, Subbiah MTR 1985 Ontogeny of guinea pig liver cholesterol 7ahydroxylase. Biochem Biophvs Res Comm 132:702-707.  

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.831.1-0086529/manifest

Comment

Related Items