" " .11 An“ . J --\\\\ .x ‘ .4 (Ilium ‘ w , F "x; “any OVERDUE FINES: 25¢ per day per item RETUMIMS LIBRARY MATERIALS: Place in book return to remove charge from circulation records COMPARATIVE EFFECTS OF FIREMASTER BP-6, 2,2',4,4',5,5'-HEXABROMOBIPHENYL (HEB) AND 3,3',4,4',5,5'-HBB ON LIPOPROTEINS AND SELECTED SERUM ENZYMES AND HEPATIC MICROSOMAL DRUG-METABOLIZING ENZYMES IN RATS BY Morrow Bradford Thompson A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pathology 1981 ABSTRACT COMPARATIVE EFFECTS OF FIREMASTER BP-S, 2,2',4,4',5,5'-HEXABROMOBIPHENYL (HEB) AND 3,3',4,4',5,5'-HBB ON LIPOPROTEINS AND SELECTED SERUM ENZYMES AND HEPATIC MICROSOMAL DRUG-METABOLIZING ENZYMES IN RATS BY Morrow Bradford Thompson Young male rats were fed diets containing 0, 0.1, l, 10 or 100 ppm of Firemaster (FM) BP-6, 2,2',4,4',5,5'-hexabromobiphenyl (HEB) or 3,3',4,4',5,5'-HBB for 10 and 30 days. Effects on serum lipoproteins, serum enzymes and hepatic microsomal drug-metabolizing enzymes were compared. At dietary levels of 100 ppm, FM BP-6 and 2,2',4.4',5,5'- HBB caused elevations in total serum cholesterol at 10 and 30 days, occurring mainly in the high density lipoprotein (HDL) fraction. Similarly, an intermediate dietary level (1 ppm) of 3,3',4,4',5,S'-HBB produced a significant elevation in serum cholesterol which occurred in the HDL fraction. Rats given 100 ppm 3,3',4,4',S,5'-HBB for 10 days had dramatic decreases in total serum cholesterol, as did those fed 10 ppm for 30 days. Decreases were mainly in the high density fraction. Rats fed dietary levels of 100 ppm 3,3',4,4',5,5'-HBB died within 20 days. Mild to moderate decreases in serum triglycerides resulted from the higher dietary concentrations of PM BP-6 and both congeners. Decreases were mostly in the very low density lipoprotein and to a lesser degree in the low density fractions. Measurement of serum alanine aminotransferase, aspartate aminotransferase (AST), gamma Morrow Bradford Thompson glutamyl transpeptidase, alkaline phosphatase (ALP) and sorbitol dehydrogenase (SDH) indicated that only SDH was consistently elevated at the higher dietary concentrations of all 3 compounds. This indi- cates that SDH may be a sensitive indicator of altered hepatocellular permeability in the rat. Serum AST values were significantly elevated from control values by the highest dietary concentrations of both the 10- and 30-day 3,3',4,4',5,5'*HBB treatments. Serum ALP levels in each 10-day treatment and the pair-fed controls to the 10-day, 100 ppm 3,3',4,4'.5,5'-HBB group were significantly lower than those of control rats. Results of assays for hepatic microsomal drug-metabolizing enzymes clearly indicated that 2,2',4,4',5,5'-HBB is a phenobarbital- type inducer, 3,3',4,4',5.S'-HBB is a 3-methylcholanthrene-type inducer and FM BP-6 is a mixed-type inducer. The possible relationships between the lipid alterations and the induction patterns are discussed. DEDICATION To My Wife Dolores J. Kunze ii INTRODUCTION OBJECTIVES TABLE OF CONTENTS LITERATURE REVIEW. . . . . . . . . . . . . . . Historical Perspective. . . . . . . . . Environmental Contamination . . . . . . Structure of the Compounds. . . . . . Hepatic Microsomal Drug-Metabolizing Enzymes. . . Metabolism and Structure-Activity Relationships . . Specific Effects of the PHB Compounds . Physical . . . . . . . . . . . . Histopathology . . . . . . . . . Biochemistry of PHB Toxicity. . . . . . Hepatic Microsomal Drug-Metabolizing Serum Enzymes. . . . . . . . . . Lipoproteins . . . . . . . . . . MATERIALS AND METHODS. . . . . . . . . . . . . Experimental Design . . . . . . . . . . Animals, Housing, Feed. . . . . . . . . Chemicals . . . . . . . . . . . . . . Anesthesia and Blood Collection . . . Determinations. . . . . . . . . . . . . Body and Organ Weights . . . . . Serum Cholesterol, Triglyceride. Serum Enzymes. . . . . . . . . . Lipoproteins . . . . . . . . . . Gas-Liquid Chromatography. . 7 Hepatic Microsomal Enzymes . . . Statistical Analysis. . . . . . . . . . RESULTS. Serum Enzymes . . . . . . . . . . . . . Serum Cholesterol . . . . . . . . . . . lipoprotein Cholesterol . . . . . . . . Serum Triglyceride. . . . . . . . . . . iii Page 33 33 35 36 36 37 37 37 38 40 42 42 43 44 44 44 51 58 Lipoprotein Triglyceride . Lipoprotein Electrophoresis. Microsomal Enzymes . . Serum and Lipoprotein GLC Analysis Body Weights . Organ weights. DISCUSSION. . . Serum Enzymes. Serum Cholesterol. . . Lipoprotein Cholesterol. Lipoprotein Triglyceride . Serum and Lipoprotein GLC Analysis Lipoprotein Electrophoresis. Microsomal Enzymes . . Organ Weights, SMARY O C I O BIBLIOGRAPHY. . VITA O O O I O O APPENDICES. . . A B 0 I U I El I Body Weights an SERUM ENZYME O O . STUDIES ORGAN WEIGHTS. . . BODY WEIGHT GAINS. SERUM LIPID STUDIES. iv d Toxicity . HEPATIC MICROSOMAL ENZYME STUDIES. Page 58 58 63 63 67 67 73 73 77 81 83 86 87 88 88 92 95 105 107 107 111 118 121 124 Table 10 11 12 13 14 15 LIST OF Experimental design. . . . . Free cholesterol assay . . . Summary of treatment effects TABLES on serum enzyme activity. . . Serum alkaline phosphatase activity in lO-day experiments. Serum aspartate aminotransferase activity in 10- and 30-day 3,3',4,4',5,5'-HBB experiments. . . . . . . . . . . Summary of treatment effects on serum total, free and percent free/total cholesterol . . . . . . . . . . . . . . Serum free cholesterol levels in 10- and 30-day experiments. . . . . . . . . Summary of treatment effects cholesterol levels . . . . . Summary of treatment effects Serum triglyceride levels in Summary of treatment effects density and very low density levels . . . . . . . . . . . Very low density lipoprotein and 30-day experiments . . . on HDL, LDL and VLDL on serum triglyceride levels. 10- and 30-day experiments. . on high density, low lipoprotein triglyceride triglyceride levels in 10- Treatment effects on relative concentrations of high density/low density lipoprotein determined by lipoprotein electrophoresis. . . . . . . Cytochrome P-450 content and shift in wavelength maximum of hepatic microsomes for carbon monoxide difference spectrum . . . . . . . . . . Serum, lipoprotein and albumin associated concentrations Of 2’2. 14:4"S,5'-I{BB and 3'3'l4'4'1515'-HBB from 30"day experiments. . . . . . . . . 45 48 49 50 54 55 59 61 62 64 65 68 Table 16 17 18 Summary of treatment effects on body weight gains. . . . Treatment effects on liver weights in 10- and 30-day experiments. . . . . . . . . . . . . . . . . . . . . . . Summary of treatment effects on thymic, thyroid and splenic weights. . . . . . . . . . . . . . . . . . . . . Appendices Sorbitol dehydrogenase and alkaline phosphatase enzyme ac ti Vi ti es 0 O O O O O O O O O O O O O O O O O O O O O O Alanine aminotransferase and aspartate aminotransferase enzm actiVi ties O O O O O O O O O O O O C I C I O I O O Gamma glutamyltranspeptidase enzyme activities . . . . . Serum cholesterol and triglyceride values. . . . . . . . Serum free cholesterol and percent free/total cholesterol values . . . . . . . . . . . . . . . . . . . Serum high density and high density + low density lipo- protein cholesterol values . . . . . . . . . . . . . . . Serum low density and very low density lipoprotein cholesterol values . . . . . . . . . . . . . . . . . . . Serum high density and high density + low density lipo- protein triglyceride values. . . . . . . . . . . . . . . Serum low density and very low density lipoprotein tri- glyceride values . . . . . . . . . . . . . . . . . . . . Microsomal enzyme assays - cytochrome P-450 and wave- length maximum . . . . . . . . . . . . . . . . . . . . . Microsomal enzyme assays - aminopyrine demethylation and benzo[a]pyrene hydroxylation . . . . . . . . . . . . . . Hepatic and thymic weights . . . . . . . . . . . . . . . Thyroid and splenic weights. . . . . . . . . . . . . . . Body weight gains. . . . . . . . . . . . . . . . . . . . vi Page 69 7O 71 108 109 110 112 113 114 115 116 117 119 120 122 123 125 LIST OF FIGURES Figure 1 Basic formula for polyhalogenated biphenyls. . . 2 $03 activity in lO-day experiments . . . . . . . 3 SDH activity in 30-day experiments . . . . . . . 4 Serum cholesterol levels in lO-day experiments . 5 Serum cholesterol levels in 30-day experiments . 6 HDL cholesterol levels in lO-day experiments . . 7 HDL cholesterol levels in 30-day experiments . . 8 Hepatic microsomal drug-metabolizing enzyme activity 30-day experiments . . . . . . . . . . . . . . . vii in Page 46 47 52 53 56 57 66 INTRODUCTION There is a definite need for concern about the production and subsequent release of man-made substances into the environment. Regardless of whether the release of an agent is deliberate or acci- dental, the end result is often the same. Animals, including human beings, are exposed to a variety of substances either through direct contact or frequently through their food chains. Unfortunately, the short- and long-term effects of many agents are simply not known. In 1973, livestock throughout Michigan were given feeds that had been accidentally contaminated with a commercial mixture of poly- brominated biphenyls (PBB). At the time that the compound was finally identified in the feeds, only a small amount of very inadequate research had been attempted to determine the toxic potential of the brominated biphenyls. Now, 8 years later, much has been and is still being done to correct for these inadequacies. Many of the initial studies concentrated on describing the various physical, gross and histopathologic effects of the parent compound, Firemaster BP-6 (FM), in both farm and laboratory animals. With the identification of some of the individual congeners in FM, efforts were started to determine which components are responsible for the various toxic effects attributed to the parent mixture. Additionally, techniques such as electron microscopy, microsomal enzyme assays and gas chroma- tography were employed to further describe and understand the actions of the compounds in biologic systems. 1 2 Several research areas have received only peripheral attention in most PBB experiments. Included in these are evaluations of the effects of the compounds on various serum enzymes and lipoproteins. The objective of this study was to evaluate the effects of FM and 2 PBB congeners on serum enzymes, lipoproteins and hepatic microsomal enzymes in rats. These determinations were selected for several reasons. The P885 are known to produce hepatic alterations in rats. The bank of serum enzymes was chosen to determine which, if any, would best reflect changes in hepatocellular integrity or cholestasis. Furthermore, since the liver is the most important organ in the metabolism of lipoproteins and since lipid-soluble compounds such as PBB have to be transported in the blood either bound to proteins or in lipoprotein particles, the composition and relative amounts of the lipoprotein fractions were carefully evaluated. The selection of the compounds was also an important aspect of the experiment. The parent compound, FM, was used as the standard against which the effects of the other 2 congeners were compared. The congener 2,2',4,4',5,5'-hexabromobiphenyl (HBB) is quantitatively the most important component of the mixture and is a phenobarbital(PB)-type hepatic microsomal enzyme inducer. The third compound, 3,3',4,4',S,5'¥ HBB, does not occur in PM but was used because of its structural char- acteristics and known 3-methylcholanthrene(MC)-type microsomal enzyme induction. - Through these experiments it was hoped that some additional knowledge and understanding about the biochemical effects of these compounds would be gained. An attempt was made to evaluate and compare certain biochemical determinations from rats fed PM with those from rats fed 3 a congener with either PB- or MC-type effects. Specific alterations detected by these assays in the FM-treated animal hopefully could be attributed to the actions of the PB- or MC-type compound. OBJECTIVES Rats were fed diets containing various concentrations of either Firemaster BP-6 (FM), 2,2',4,4',5,5'-hexabromobipheny1 (HBB) or 3, 3',4,4',5,5'-HBB. These coupounds are mixed-, phenobarbital- and 3-methylcholanthrene(MC)-type hepatic microsomal enzyme inducers, respectively. The objectives of this research project were: 1. To evaluate the effects of the different compounds on selected serum enzymes and lipids. 2. To evaluate the effects of the treatments on the relative amounts and composition of the major lipoprotein fractions. 3. To attempt to identify the mode of serum transportation for the 2 purified compounds. 4. To describe the pattern of hepatic microsomal enzyme induction for each compound. 5. To identify future areas of investigation needed to describe the mechanisms associated with altered lipid metabolism. LITERATURE REVIEW Historical Perspective With the controversy surrounding the accidental introduction of polybrominated biphenyls (PBB) into the food chains of livestock and, subsequently, that of human beings still raging, it is often difficult to think of the problem in terms other than those of a current crisis. If one considers the use of PBB to be an extension of the industrial use of polychlorinated biphenyls (PCB), these compounds, the polyhalo- genated biphenyls (PHB), have been around for at least 100 years. In a review article discussing polychlorinated polycyclic compounds, Kimbrough (1974) cited a reference which attributed the first descrip- tion of the synthesis of PCBs to Liebig’s Annalen in 1881. It was not until the 19305, however, that PCB, because of its unique physical properties, was used extensively for various industrial applications. The compounds are very stable, nonflammable and have excellent dielectric and plasticizing properties (Hammond, 1972; Kimbrough, 1974). They have been used as dielectric fluids in capacitors and transformers, as hydraulic and heat exchange fluids, as plasticizers, adhesives, textile coatings and components of paints and varnishes, and as components in insulation coatings around electric wires and cables (Hammond, 1972: Kimbrough, 1974). Production of PCB peaked in 1970 and began declining in 1971 after the only United States producer voluntarily limited sales of the compound (Nisbet and Sarofim, 1972). 6 Industrial production of PBB in the United States began in 1970 and, like the PCB, was limited to one company (Kerst, 1974). The PBB compounds are relatively inert, water insoluble and highly heat stable. The commercial product, Firemaster BP-6 (FM), was incorporated into materials because of its flame-retarding properties. Greater than 80% of the product was incorporated into housings and components for business machines and industrial and electric equipment (Kerst, 1974). Production of PM was stopped in 1974 (DiCarlo et al., 1978). Environmental Contamination Both PCB and PBB are present in the environment. For PCB the major routes of entry are suspected to be leaks from transformers, heat exchangers and hydraulic systems, spills and losses during manufacturing, vaporization or leaching from PCB-containing formulations and disposal of PCB-containing fluids (Nisbet and Sarofim, 1972). Higher concentra- tions occur in waterways and seas (and in their associated animal life and sediments) around industrialized areas than in undeveloped areas of the world (Hammond, 1972). In addition to these important routes, several well documented incidences resulting in the high level contamina- tion of waterways, animal feeds and human food have occurred (Nisbet and Sarofim, 1972; Kimbrough, 1974; Van Houweling et al., 1977). Perhaps the most serious accident occurred in Japan in 1968, in which more than 1000 people ingested PCB-contaminated rice oil (Kuratsune et al., 1972). Polybrominated biphenyls are ubiquitous, having been iden- tified in ecosystems around the globe (Risebrough and deLappe, 1972) and in approximately 1/3 of the samples of adipose tissue taken from people throughout the United States (Yobs, 1972). 7 Unlike the widespread problems with PCB, the release of high levels of PBB into the environment has been essentially confined to one inci- dent. The events and their consequences have been described by numerous authors (Dunckel, 1975; Carter, 1976; Getty et al., 1977; Kay, 1977). During the summer of 1973, 500 to 1000 pounds of FM were accidentally Shipped to a large feed mixing mill in Michigan. Firemaster was inad- vertently mixed into feed in place of magnesium oxide. From this mill highly contaminated feeds were shipped to numerous distribution sites in Michigan and subsequently sold to local farmers. These feeds not only contaminated animals that ate it and meat, eggs and milk they produced, but also bins in which it was stored and barns and pastures in which the animals were kept. Contamination of farm families and the general public that consumed tainted products was inevitable. Approximately 8 to 9 months elapsed before PBB were identified in samples of contaminated feed. An extensive sampling, quarantine and disposal program was begun. One report (Van Houweling et al., 1977) indicated that 30,000 cattle, 6000 swine, numerous sheep and poultry, and several hundred tons of feed and several tons of dairy products were destroyed. Incidents such as this should not recur, since the production of PBB in the United States has been stopped. But because of the environ- mental stability of the compound (Jacobs et al., 1978) and its tendency to accumulate and persist in adipose tissue (Fries, 1978; Tuey and Matthews, 1980), research examining chronic effects of PBB will continue. Structure of the Compounds Polychlorinated and polybrominated biphenyls share the same basic formula, as illustrated in Figure 1. For either PCB or PBB, substitutions with their respective halogens on rings A and B can produce 210 different 3,m 2,o 6',o 5',m 5,m 6,0 2',o 3',m Figure 1. Basic formula for polyhalogenated biphenyls. Note numbered positions on biphenyl rings and letters indicating ortho (0), meta (m) and para (p) positions. 9 possible congeners for each class of compounds (Cook, 1972). Commer- cial preparations of PCB are formulated based upon their average chlorine content by weight. Examples of the preparations contain 21, 42, 48, S4, 60, 62 and 68% chlorine (Nisbet and Sarofim, 1972). Each preparation consists of numerous different congeners. Analyses of preparations containing 42, 54 and 60% chlorine have been shown to have 45, 69 and 78 congeners, respectively (Cook, 1972; Kflmbrough, 1974). The Firemaster mixture is not as complex as the different PCB mixtures. It contains approximately 30 different compounds, of which . 13 are major PBB congeners (Moore et al., 1979; Moore et al., 1980). If each congener had similar biologic effects, then the complexity of - these compounds would be interesting but not too significant. As will be discussed later, however, the various congeners differ greatly in their-toxic potential. Hepatic Microsomal DrugéMetabolizing Enzymes Before the functional characteristics of the various compounds can be discussed, it is necessary to understand the biologic systems responsible for their disposition. In addition to its many homeostatic functions, the liver has evolved as the major organ for the detoxifica- tion and excretion of drugs and foreign compounds (xenobiotics) (Kappas and Alvares, 1975). This system is physically located in the membrane of the endoplasmic reticulum (ER) (Ullrich, 1978). Through disruption of the ER, microsomes can be produced and isolated in vitro. Variously called the monooxygenase, mixed function oxidase (MPG) and hepatic drug-metabolizing microsomal enzyme system, one of its important biologic functions is to convert foreign, hydrophobic compounds into hydrophilic 10 forms which can be more easily excreted by the body (Ingelman-Sundberg, 1980). To accomplish this, various reactions are catalyzed by the MEG system, including epoxidation, hydroxylation, oxidation, dealkyla- tion, and desulfuration (Ullrich, 1978). While most xenobiotics are less toxic after being metabolized by the MEG system, some may form reactive intermediates, compounds which, if not detoxified by endogenous mechanisms, may react with proteins and nucleic acids and predispose to mutagenesis and carcinogenesis (Ullrich, 1978). The microsomal enzymes can be induced by many compounds. There are, however, 2 basic patterns of microsomal enzyme induction. One pattern is produced by administration of the drug phenobarbital (PB) and the other by 3-methylcholanthrene (MC). Drugs that induce micro- somal enzymes similar to those induced by one of these 2 compounds are classified as either PB—type or MC-type enzyme inducers (Conney, 1967). In 1967, the induction characteristics of over 200 drugs, insecticides, carcinogens and other chemicals had been described. The only common denominator that vaguely unites this diverse group of compounds is their solubility in lipid at a physiological pH (Conney, 1967). The effects of PB and the compounds with similar induction char- acteristics are extensive and can be evaluated by various sophisticated techniques. The most obvious physical effect of the administration of PB-type compounds is a marked increase in liver size and weight. Indi- vidual hepatocytes are swollen and there is usually an increase in the smooth endoplasmic reticulum (SER). Phenobarbital has an anabolic effect on the liver with a dramatic increase in the content of micro- somal protein per gram of liver (Conney, 1967). All of these changes, however, are but reflections of important biochemical alterations occurring in the ER in response to these PB-type 11 compounds. There can be marked increases in the enzymes responsible for the metabolism of these compounds. There are at least 7 forms of the terminal enzyme, cytochrome P-450, with varying degrees of substrate specificity. These primarily catalyze oxidation reactions (Ingelman- Sundberg, 1980). Induction of the MFO system by PB-type compounds is reflected in increases in microsomal protein, cytochrome P-450, cyto- chrome P-450 reductase and microsomal enzyme activities, including aminopyrine N-demethylase, epoxide hydratase and ethylmorphine-N- demethylase (Dent et al., 1976a; Dannan et al., 1978b). How PB-type compounds increase the protein and enzyme content of hepatic ER is not known. Researchers speculate that a cytosolic receptor for these compounds may be required (Poland and Glover, 1977). Suspected mechanisms of action include de novo synthesis of protein, decreased catabolism of existing microsomal enzymes and enhanced forma- tion of enzymes at the ribosomal level. The time required for daily doses of PB to exert its maximal effects on microsomal enzymes in the rat is 3 days (Conney, 1967). A very important characteristic of those compounds that have PB- type induction patterns is that, in spite of their enzyme induction and ability to increase the size of the liver, they generally are considered to be nontoxic (Moore et al., 1980). This is in sharp contrast to polycyclic aromatic hydrocarbon compounds that are MC-like and induce AHH activity. Toxicity of these latter compounds is characterized by lethality, chloracne, hyperkeratosis, lymphoid organ involution, hepatic lesions, teratogenicity and species-specific diseases such as chick edema syndrome (Poland et al., 1979). Those compounds with MC-type activity include polycyclic aromatic hydrocarbons, which by definition are compounds with "3 or more fused 12 benzene rings in linear, angular or cluster arrangements and contain only carbon and hydrogen“ (Zedeck, 1980). Other compounds which have MC-type activity are certain halogenated aromatic hydrocarbons, which include the biphenyls, dibenzo-p-dioxins, dibenzofurans, azo- and azoxybenzenes and naphthalenes (Poland et al., 1979). The MC-type compounds do not stimulate the incorporation of protein into the ER as actively as PB-type compounds (Conney, 1967). Histopathologic changes in the livers of animals exposed to MC-type compounds have included cellular swelling, vacuolization, lipid accumulation, necrosis and increases in the rough ER (RER) with little change in the SER (Kociba et al., 1978; Gasiewicz et al., 1980). Compounds with MC-type activity induce the production of a unique terminal cytochrome, cytochrome P-448. This enzyme derives its name from the wavelength of absorption for the carbon monoxide (CO) difference spectrum maximum (Ullrich, 1978). Microsomal enzyme activities thought to be catalyzed by this cytochrome include benzo[a]pyrene hydroxylase (aryl hydrocarbon hydroxylase, AHH) and ethoxycoumarin-o-deethylase (Dent et al., 1976a; Dannan et al., 1978b). Glutathione-s-transferase, DT-diaphorase, ornithine decarboxylase and 6-amino levulinic acid synthetase are examples of other hepatic enzymes that are induced by MC-type compounds (Poland and Glover, 1980). Enzyme activity after the administration of MC-type compounds can double within 3 to 6 hours and can be maximally induced within 24 hours (Conney, 1967). This is in contrast to the 3 to 5 days for maximum induction cited by the same author for PB-type compounds. The mechanism for induction by MC-type compounds has been extensively researched and reviewed (Poland and Glover, 1977; Poland et al., 1979; Poland and Glover, 1980). They observed that some strains of mice do 13 not respond to MC while others do but that both respond to 2,3,7,8- tetrachloro-p-dioxin (TCDD). If given in sufficient quantities, TCDD has approximately 30,000 times the potency of MC. Poland and his associates hypothesized that there is a cytosolic receptor for MC- type compounds. Apparently, through mutations this receptor varies in its affinity for its substrate. If binding does occur, the substrate- receptor complex is thought to move to the nucleus and initiate the expression of genes that regulate AHH and numerous other enzyme acti- vities through de novo synthesis of protein. The structure of the receptor and, therefore, its affinity for a substrate is controlled by the Ah locus in mice. This locus may not only regulate the produc- tion of the putative TCDD receptor and, because of this, AHH activity, but also numerous other hepatic enzymes. This model predicts that the toxicity of MC-type compounds depends upon their ability to bind to a receptor and to initiate the expression of a group of genes. These genes control the de novo synthesis of AHH and numerous other enzymes. How enzyme induction is related to toxicity is not known. The basic configuration of the polyhalogenated biphenyls (PHB) is not unlike that of TCDD. It is not surprising that PHB are known inducers of AHH activity (Poland and Glover, 1977). Also, because of the structural similarities between PCB and PBB, these 2 classes of compounds are generally considered to have similar biologic effects (Dent et al., 1976a; Kay, 1977). The order of potency for halogen sub- stitutions is bromine > chlorine > fluorine (Poland and Glover, 1977). Metabolism and Structure-Activity Relationships Mixtures of PCB and PBB that have both PB- and MC-type effects are called "mixed" inducers (Goldstein et al., 1977; Dannan et al., 1978b). 14 With numerous different compounds in the commercial mixtures and the possibility that potent contaminants might also be included, attempts to understand the effects of individual congeners had to await their identification, isolation and purification. In 1976 and 1977, the comercial PBB mixture, FM, was clearly shown to be a mixed-type inducer (Dent et al., 1976a; Dent et al., 1976b; Dent et al., 1977). As the induction pattern of FM was being described, other researchers were isolating and identifying the major congener as 2,2',4,4',5,S'-HBB (Sundstrom et al., 1976). Approximately 54 to 68% of the FM mixture by weight consists of this congener. Presently, 13 major congeners have been detected in FM and the structures of 9 of these are known (Moore et al., 1980). The obvious task is to determine which compounds are responsible for the numerous treatment effects produced. Fortunately, through observations with both PCB and PBB, some understandings of the relationships between structure and activity have evolved. The structural features that permit the metabolism of a PBB congener have been reviewed (Moore et al., 1980). Rapid, in vitro metabolism of PBB congeners occurs if both or one of the para_positions (4,4') is unoccupied and there is an adjacent unoccupied carbon (Dannan et al., 1978a). Of the 13 major FM congeners, only 2 meet these requirements. Of the congeners which are metabolized, the rates of metabolism are generally increased by grthg_substitutions and decreased by increasing the number of bromines on the molecules. A recent report by Purdy and Safe (1980), however, indicated that there is some metabolism of 2,2',4,4',5,5'-HBB in vitro. Metabolism of a compound may or may not be a desirable fate. If the metabolite is less toxic than the original compound, then metabolism is desirable. If the metabolite is a reactive molecule that, for example, 15 binds to DNA or produces membrane destruction through lipid peroxidation, then metabolism may be biologically detrimental. Few attempts have been made to identify possible PBB metabolites and fewer still have been undertaken to determine the possible biologic effects of these products. Metabolites of PCB and PBB have generally been shown to be mono- and dihydroxylated forms of their corresponding parent congeners (Safe et al., 1975; Kohli et al., 1978; Safe at al., 1978). Potentially harmful metabolites from PHB congeners have been identified, however. An arene oxide was identified as an intermediate molecule in the metabolism of 4-bromobiphenyl. This metabolite could possibly bind to cellular mole- cules and produce toxic effects (Kohli et al., 1978). In this same study, 4—bromobiphenyl was shown to be mutagenic in the Salmonella typhimurium TA 1538 assay using microsomes from rats treated with PCB. Another study using FM and 2,2',4,4',5,5'-HBB failed to demonstrate either initiating or promoting activity for the compounds using a mouse strain sensitive to skin tumors (Haroz and Aust, 1979). The metabolism of polycyclic aromatic hydrocarbons was recently reviewed (Zedeck, 1980). These compounds are generally metabolized to dihydrodiols, phenols and glutathione conjugates. These products require the formation of epoxide intermediates. Certain epoxides are potent carcinogenic and mutagenic compounds. These compounds are electrophilic and are known to bind with sites on DNA and RNA (Ingelman-Sundberg, 1980). The PHB compounds are presumed to produce their toxic effects through mechanisms similar to those attributed to the polycyclic aromatic hydrocarbons (Poland and Glover, 1977). A thorough understanding of the enzyme induction patterns produced by the PHB compounds and of the effects of the parent compounds and their metabolites on cellular components is imperative. 16 The ability of specific PHB compounds to produce either PB- or MC-type effects follows certain structural guidelines. PHB compounds that have at least 1 unsubstituted pagg_position either do not induce or are very poor inducers of microsomal enzymes (Goldstein et al., 1977; Moore et al., 1980). The rapid metabolism of these compounds may pre- clude significant enzyme induction (Goldstein et al., 1977). In FM, the only compounds identified with less than 2 pg£a_bromines are 2,2',4,5,5'- H38 and 2,2',3,4',S,S'-HBB. The structural requirements needed for PHB molecules to have MC- type activity have been extensively investigated. One important require- ment for this activity is that the molecule be able to assume a planar configuration (Poland and Glover, 1977; Goldstein et al., 1977; Moore et al., 1980). This appears to be essential if the compound is to bind with the putative TCDD-receptor in the cytosol of target cells (Poland and Glover, 1977). Nonplanarity, or at least energy barriers to planarity, result from increased halogen substitutions at ortho (2,2',6,6') positions (Moore et al., 1980). The ability to assume planarity, however, is not by itself adequate for MC-type activity. Biphenyl, for example, has no MC-type activity (Poland and Glover, 1977). Para or meta substitutions are obviously required. Surprisingly, the compounds 3,3';5,5'-tetrabromo- and tetrachlorobiphenyl do not have MC-type activity (Poland and Glover, 1977). Halogenated biphenyl compounds that have been identified as having MC-type activity include 3,3',4,4',S-pentachlorobiphenyl, 3,3',4,4'-tetrachlorobiphenyl and 3,3',4,4',5,5'-hexabromo- and hexachlorobiphenyl (Goldstein et al., 1977; Poland and Glover, 1977; Yoshimura et al., 1979). From these studies some observations have been made. For a compound to be an MC-type inducer, the ability to I? assume planarity and the presence of 2 para halogens may be essential (Moore et al., 1980). The para halogens do not appear to be adequate by themselves, however, since 4,4'-dichlorobiphenyl failed to induce MC-type activity (Poland and Glover, 1977). Mg£a_substitutions may be required in addition to the pa£a_ones to impart MC-activity. Pure MC-type inducers have not been identified in the FM mixture (Moore et al., 1980). However, several components of FM have been investigated for both MC- and PB—type activities. Of these, 2,3',4,4',5,5'-HBB (Dannan et al., 1978b) and 2,3,3',4,4',5-HBB (Robertson et al., 1981) have mixed-type effects and 2,3',4,4',5- pentabromobiphenyl is suspected of being a mixed inducer (Moore et al., 1980). The ability of a biphenyl molecule with 1 ggthg substitution to have MC-type activity requires that the energy barrier to rotation be small enough to allow the molecule to at least briefly assume a planar configuration and bind to the TCDD receptor. Several general comments can be made about the structural char- acteristics needed for a compound to be a PB-type inducer. The PHB compounds have at least 1 and usually 2 ggthg_halogens. These by them- selves are not sufficient, however, since 2,2'-dibromobipheny1 is not a microsomal enzyme inducer (Moore et al., 1979). Mg£g_and pg£a_substitu- tions are apparently also needed for PB-type induction (Moore et al., 1980). Specific Effects of the PHB Compounds Physical The earliest indication that a compound may be toxic is often reflected by the general appearance of the exposed animal, by alterations in body weight gains, or by gross changes in specific organs. Polyhalo— genated biphenyls are rm) exceptions and their physical and gross effects 18 have been described in various laboratory animals, livestock and human beings. These will be briefly reviewed. Toxic effects produced by these compounds are associated with their ability to induce AHH activity. Compounds that are PB-type microsomal enzyme inducers are generally considered to be nontoxic (Moore et al., 1980). Of the numerous compounds known to be MC-type microsomal inducers, TCDD is the most potent. Toxic effects associated with it include slow wasting and death, chloracne and hyperkeratosis, chick edema disease, hepatic lesions, teratogenicity and embryotoxicity and lymphoid involu- tion (Poland et al., 1979). In a 2-year study involving the feeding of TCDD to male and female rats at various dietary levels, the major gross findings included increased mortality, emaciation and decreased weight gain, liver toxicity, thymic and splenic atrophy, icterus, some focal hemorrhages and pulmonary congestion and edema (Kociba et al., 1978). Total parenteral nutrition (TPN) can prevent the weight loss in rats associated with TCDD toxicity, but it (TPN) does not prevent and may increase the toxicity of the com- pound (Gasiewicz et al., 1980). Gross findings in this study again included icterus, thymic atrophy and enlarged livers. For PCB, numerous experiments have evaluated the effects of the commercial mixtures on various animal species. In rats, the predominant gross finding produced by chronic exposure to 100 parts per million (ppm) of several comercial mixtures was an increase in liver weights (Allen et al., 1976). These animals gained weight similarly to controls and appeared healthy. At a higher dietary concentration of a commercial PCB mixture, feed intake and weight gain were decreased while liver weights were increased in rats (Garthoff et al., 1977). Other animal species are more susceptible to the potential toxic effects of the PCB 19 compounds. Mink, for exanple, can be killed by less than 4 ppm PCB in the diet with the physical and gross changes being similar to those described in the rat at high levels of exposure (Platonow and Karstad, 1973; Aulerich and Ringer, 1977). Monkeys fed PCB mixtures at different concentrations and for different time periods had physical and gross alterations that included death, weight loss, alopecia, eyelid edema, chloracne, enlarged livers, ulceration of the gastric mucosa and thymic atrophy (Abrahamson and Allen, 1973; Allen et al., 1974). In 1968 in Japan, more than 1000 people consumed various quantities of a rice oil contaminated with approximately 2000 to 3000 ppm of a commercial PCB mixture (Kuratsune et al., 1972). The clinical symptoms and physical findings included chloracne, swelling of eyelids, increased pigmentation of the skin, jaundice, malaise, fever and various neuro- logical signs. Ten years after the exposure most of the initial symptoms, many of which were related to disorders of the skin and mucous membranes, had improved. Unfortunately, other complaints, including dullness, headache, joint swelling and pain, retarded growth in children and bronchitis-like symptoms had replaced the initial symptoms (Urabe et al., 1979). For PBB, the clinical signs and physical and gross findings are generally similar to those reported for PCB. This should be expected based on the shared structural and chemical characteristics of these classes of compounds. Since the major release of PBB into the environ- ment occurred in livestock feed, the first important description of the effects of the PBB mixture was in cattle (Jackson and Halbert, 1974). Physical effects attributed to the compound were death, anorexia, weight loss, abortion, decreased milk production, abnormal hoof growth, lameness, hematomas, abscesses, enlarged livers and thickening and 20 wrinkling of the skin. Other studies with cattle have confirmed many of these findings (Moorehead et al., 1977; Cook et al., 1978). Moorhead et a1. (1977) also noted thymic atrophy as an important gross finding in their cattle given FM. A herd health study by Mercer et a1. (1976) failed to identify any significant physical or clinical effects between control and exposed herds. Many of the apparent differences between studies could be related to the amount and duration of exposure to the PBB compounds. In rats exposed to the FM mixture, the major physical and gross findings have included decreased weight gain, increased liver weight, mottled copper-colored appearance to the liver and atrophy of the thymus and spleen (Sleight and Sanger, 1976; Garthoff et al., 1977; Harris et al., 1978; Gupta and Moore, 1979). The most consistent finding in these studies was the increased liver weights. At oral doses of 100, 300 and 1000 mg/kg/day, the mortality rate in rats during a chronic study was high (Gupta and Moore, 1979). Studies in rats with specific PBB-congeners have also been conducted. Moore et a1. (1978) examined the effects of 2,2',4,4',5,5'-HBB on rats. The only reported gross findings were swollen livers with rounded lobes. Render (1980) fed FM and 2 purified congeners to rats for 10 days. These were the PB-type inducer 2,2',4,4',5,S'-HBB and the MC-type inducer 3,3',4,4',5,5'-HBB. Physical signs of toxicity were confined to rats fed high levels of 3,3',4,4',5,5'-HBB. Rats fed 100 ppm became depressed, anorectic, emaciated and died within 20 days. Within the 10-day feeding period, rats fed 10 and 100 ppm ate less, gained less weight, had decreased thymic weights and, as in animals fed the other compounds, had increased liver weights when compared to controls. Other experimental animals have been used in studies with PBB. Important physical findings cited in these studies are as follows: 21 monkeys - weight loss, alopecia, subcutaneous edema and liver enlarge- ment (Allen et al., 1978); mink - death, anorexia, weight loss, unthrifti- ness and fatty livers (Aulerich and Ringer, 1979); Quail - subcutaneous edema of neck and shoulders in chicks and increased liver and thyroid weights(Ringer and Polin, 1977); pigs - decreased weight gain and increased liver weights (Ku et al., 1978). Histopathology In rats exposed to TCDD, the most consistent histopathologic changes have been confined to the thymus and liver (Gasiewicz et al., 1980). Thymic atrophy was related to a decrease in the number of cortical lymphocytes. In the livers of treated rats there was diffuse swelling and enlargement of hepatocytes with some loss of architectural pattern. vacuolization and necrosis of cells in the midzonal and centrilobular regions. More severe microscopic changes included extensive necrosis, vacuolization, cystic space formation filled with inflammatory debris, bile duct proliferation, fatty change and an increased incidence of foci of hepatocellular alterations (Kociba et al., 1978; Gasiewicz et al., 1980). Ultrastructural changes in the livers of rats given TCDD have included an increase in the rough endoplasmic reticulum (RER) with an inconsistent change (occasionally increased) in the smooth endoplasmic reticulum (SER), an increase in cytoplasmic lipid droplets and an increase in lysosomal activity with residual body formation (Kociba et al., 1978). At high levels of exposure, the RER formed concentric arrays and degenerating mitochondria were present (Gasiewicz et al., 1980). ' Microscopic and ultrastructural changes reported for animals exposed to PHB are generally similar to those described for TCDD. Rats that were given various levels of PCB had hepatocellular 22 swelling, focal areas of degeneration and necrosis, cytoplasmic vacuolization and cyst formation, fatty infiltration and bile duct proliferation (Kimbrough et al., 1972; Allen et al., 1976; Kasza et al., 1976). All of these studies, including another by Norback and Allen (1972), noted important ultrastructural alterations in the livers. These included a proliferation of the SER with a decrease and disorgani- zation of the RER, the presence of granular cytoplasmic inclusions and a loose "motheaten" appearance to the cytoplasm. Concentric membrane arrays developed in the cytOplasm and smooth membranes encircled mito- chondria and lipid droplets. Histopathologic and ultrastructural changes in livers from monkeys and mink fed PCB mixtures included necrosis, cellular swelling, lipid accumulation and proliferation of the SER.(Platonow and Karstad, 1973; Allen et al., 1974; Allen, 1975). Thymic atrophy due to cortical hypo- plasia has also been observed in monkeys fed PCB (Abrahamson and Allen, 1973). In liver biopsy samples of Japanese victims of the "Yusho" incident, the amount of RER.was reduced while the SER.was increased. Mitochondria were reported to be heterogeneous and contained filamentous inclusions in the matrix (Kuratsune, 1972). Histopathologic descriptions of abnormalities produced by PBB are extensive and this review, concentrating on the primary organs affected, will be confined to changes described in the liver and thymus. Micro- scopic changes in the livers of rats and mice fed the FM compound have included cellular swelling and vacuolization (frequently oil red 0 positive), increased cellular pleomorphism, areas of inflammation and necrosis and, at very high levels, bile duct proliferation and associated fibrosis (Sleight and Sanger, 1976; Kimbrough et al., 1978; Gupta and Moore, 1979; Kimbrough et al., 1980). Thymic alterations consisted of 23 marked atrophy, cortical hypoplasia and a loss of demarcation between cortical and medullary regions (Gupta and Moore, 1979). Rats given intraperitoneal (IP) injections of the congeners 2,2',4,4',5,5'-HBB and 2,2',3,4,4',5,5'-heptabromobiphenyl had swollen and vacuolated hepa- tocytes (Moore et al., 1978; Moore et al., 1979). Rats fed 3,3',4,4',S,5'-HBB had cortical atrophy of the thymus and an associated cortical infiltration of macrophages. Livers of these rats had swollen hepatocytes with prominent nucleoli, loss of sinusoidal spaces and midzonal to centrilobular vacuolization (Render, 1980). Hypercellularity of portal areas related to proliferation of bile duct cells was present in rats fed 100 ppm of this congener for 20 days. Ultrastructural evaluations of livers from rats and mice given FM and octabromobiphenyl have detected mitochondrial swelling and degeneration, increases in SER and peripheral displacement and decreases in RER, myelin body formation (paired, smooth membrane arrays often surrounding lipid droplets), cytoplasmic vacuolization and increases in lysosomes and a reduction in glycogen (Lee et al., 1975; Sleight and Sanger, 1976; Corbett et al., 1978; Kimbrough et al., 1980). Rats fed 2,2',4,4',5,5'-HBB had an increase in hepatocellular SER and those fed 3,3',4,4',5,5'-HBB had a marked increase in SER and an increase in individualized, double membrane RER often located around mitochondria (Render, 1980). Other ultrastructural changes noted by Render (1980) in hepatocytes of rats fed 3,3',4,4',5,5'-HBB were increased amounts of lipid droplets, disorganization of the BER and numerous free ribosomes in the cytoplasm. Similar hepatic changes have been reported in monkeys (Allen et al., 1978) and mink (Aulerich and Ringer, 1979). Although histopathologic 24 alterations in cattle fed PBB have been described in the liver (pri- marily fatty change), significant renal changes also occur. These include extreme dilatation of collecting ducts and convoluted tubules with pronounced epithelial degeneration (Jackson and Halbert, 1974; Moorhead et al., 1977; Cook et al., 1978). Biochemistry of PHB Toxicity Hepatic Microsomal Drug-Metabolizing Enzymes Earlier in this review the induction of hepatic microsomal enzymes. by various compounds was briefly discussed. It was stressed that certain compounds are similar to phenobarbital in their ability to induce microsomal enzymes while others are similar to 3-methylcholan- threne. Compounds which induce AHH activity are considered to be toxic. Numerous variables can be measured which reflect the type of induction a compound may have. These include measurement of microsomal protein, amount of cytochrome P-450, measurement of the carbon monoxide (CO) difference spectrum and metabolism of numerous substrates to evaluate the induction of microsomal enzyme activities. The effects of TCDD, PCB and PBB on these variables will be summarized. Compounds which have PB-type activity as opposed to MC-type are reported to have a greater positive effect on the incorporation of protein into microsomes (Conney, 1967). This is evident in the data of Dannan et al. (1978b) by comparing the microsomal protein content from livers of PB- and MC-treated rats. Different microsomal proteins are induced by these treatments, as shown by polyacrylamide gel electrophoresis patterns. These proteins are thought to be either the various induced species of cytochrome P-450 or their apoproteins (welton and Aust, 1974; Haugen et al., 1976; Toftgard et al., 1980). 25 Often paralleling the changes in microsomal protein content, therefore, is the amount of cytochrome P-450 in the hepatic microsomes. The effects of TCDD on microsomal protein content have been reported to be negligible while the cytochrome P-450 content has been increased (Gasiewicz et al., 1980; Poland and Glover, 1977). Parent PCB and PBB mixtures, however, can produce dramatic increases in both microsomal protein content and cytochrome P-450 (Babish and Stoewsand, 1977; Garthoff et al., 1977; Poland and Glover, 1977; Moore et al., 1979; Dannan et al., 1978b). The 2 major congeners in FM, 2,2',4,4',5,S'-HBB and 2,2',3,4,4',5,5'- heptabromobiphenyl, are strictly PB—type inducers and both produce marked increases in microsomal protein and P-450 content. These responses are similar to those caused by the FM mixture but greater than those produced by injections of MC (Moore et al., 1978; Moore et al., 1979). The 2,3' ,4,4' ,5,5'-HBB congener, which has mixed-type induction effects, also increases microsomal protein and P-450 content (Dannan et al., 1978b). Another method of evaluating the type of enzyme induction in micro- somes is to measure the wavelength of the carbon monoxide (CO) difference spectrum for the reduced microsomes (Ullrich, 1978). Substances that have PB-type activity induce terminal P-450 enzymes which have a maximal difference at 450 nm. The MC-type inducers, however, stimulate produc- tion of different terminal cytochromes that have a maximal absorption at 448 nm. Thus, a shift in the absorption spectrum from 450 nm towards 448 nm would indicate exposure to an MC-type compound and could be associated with overt signs of toxicity. For the PBB congeners 2,2',4,4',5,5'-HBB and 2,2',3,4,4',5,5'-heptabromobiphenyl, the cyto- chrome P-450 spectral maximum is not shifted from 450 nm (Moore et al., 26 1978; Moore et al., 1979). The FM mixture, which is a mixed-type inducer, however, has been shown to shift the CO difference spectrum towards or to 448 nm (Dent et al., 1976a; Babish and Stoewsand, 1977). Both TCDD and 3,3',4,4'-tetrachlorobiphenyl shift the CO difference spectrum towards 448 nm (Poland and Glover, 1977). Different forms of cytochrome P-450 catalyze different reactions and have varying degrees of substrate specificity (Ullrich, 1978). Different substrates can be used to detect the ability of microsomal mixed function oxidase (MFO) systems to catalyze a given type of reaction and this ability can be used to categorize the induction as PB- or MC-like. The ability to demethylate aminopyrine (amiopyrine demethylase activity, AD) and to hydroxylate benzo[o]pyrene (arylhydro- carbon hydroxylase activity, AHH) are frequently used to reflect PB- and MC-type activities, respectively (Conney, 1967). The most potent MC-type inducer known, TCDD, has been shown to have AHH and not AD activity (Poland and Glover, 1977; Gasiewicz et al., 1980). Both FM and 2,3',4,4',5,5'-HBB will induce both activities (Dannan et al., 1978b) and 2,2',4,4',5,5'-HBB and 2,2',3,4,4',5,S'-heptabromobiphenyl primarily induce AD activity (Moore et al., 1978; Moore et al., 1979). By using these methods to categorize compounds as PB- or MC-type inducers, extremely practical information about their mechanisms of action and suspected toxicity can be obtained. Serum Enzymes Several serum enzyme determinations are often included in the battery of tests used to evaluate the effects of a suspected toxic compound. Most are selected because an elevation in their serum level tends to reflect either hepatic or cholestatic disorders. Changes 27 reported for these enzymes in animals exposed to PCBs or PBBs are inconsistent. Perhaps much of the variability is related to use of different congeners or mixtures, duration and level of exposure and different species and sexes of animals employed in the experiments. Alanine aminotransferase (ALT, SGPT) and aspartate aminotransferase (AST, SGOT) are enzymes that catalyze the interconversion of amino groups from their respective amino acids to a-oxoacids. Both occur in relatively high concentrations in hepatic tissue but high levels of AST are also present in cardiac and skeletal muscle (Kachmar and Moss, 1976). Intracellular distribution of these 2 enzymes differs in that AST is located in both the cytoplasm and mitochondria while ALT is found primarily within the cytoplasm. While species differences in tissue specificity occur, elevated levels of either enzyme within the serum are frequently interpreted as an indication of altered hepato- cellular integrity. Sorbitol dehydrogenase (SDH) is a cytosolic enzyme that is reported to be very liver specific (Kachmar and Moss, 1976). It functions in the catalysis of the interconversion of sorbitol and fructose. Elevated serum levels should reflect altered hepatocellular integrity and should not be associated with primary dysfunctions of other organ systems. Gamma-glutamyltranspeptidase (GGT) is an enzyme found in high concentrations on the renal brush border of the kidney and in plasma membranes enriched in bile canaliculi and biliary duct epithelial cells of the liver (Shaw and Newman, 1979; Huseby, 1979). While the enzyme was once thought to function in the translocation of amino acids into cells via the Y-glutamyl cycle, the suggested function now is extra- cellular catabolism of glutathione to L-glutamate and L-cysteinylglycine 28 (McIntyre and Curthoys, 1979; Shaw and Newman, 1979). Species dif- ferences in organ levels of GGT are marked. Human liver and kidney have approximately 10 times the tissue levels as do the same organs in rats (Shaw and Newman, 1979), and guinea pigs are reported to have much higher hepatic GGT levels than rats (Huseby, 1979). Certain drug treatments, such as PB, increase GGT levels in guinea pigs and rats, although in the latter species prolonged treatment was required. The elevated serum levels are thought to involve the induction of micro- somal drug-metabolizing enzymes (Huseby, 1979). In clinical situations, the enzyme can be elevated in many types of liver disease, but the highest levels result from intra- or posthepatic biliary obstruction (Kachmar and Moss, 1976). In the dog, for example, GGT appears to be more specific and sensitive than alkaline phosphatase for biliary obstruction (Noonan and Meyer, 1979). Serum alkaline phosphatase (ALP) is a mixture of various isoenzymes primarily derived from liver, bone, intestinal tract and placenta. The enzyme is apparently involved in the transportation of metabolites across cell membranes, including lipid transportation in the intestines and calcification processes in bone (Kachmar and Moss, 1976). In the liver, ALP is located in hepatic cell sinusoidal membranes, microvilli of bile canaliculi and in the endothelial cells of the portal and central veins (WOlf, 1978). Righetti and Kaplan (1971) reported that the major source of ALP in the fasted rat is the bone isozyme. These rats, however, were young, weighing approximately 200 grams, and a high serum level of the bone isoenzyme could be expected in young animals. Fishman et a1. (1962) indicated that ALP in the serum of a normal, well nourished rat is almost all of intestinal origin. Fishman observed a decline in serum 29 ALP in the rat related to a decrease in the intestinal isoenzyme following bile duct ligation. He reasoned that the lack of bile pre- vented adequate intestinal fat emulsification leading to a decrease in fatty acid uptake by the intestinal mucosa and a corresponding decreased release of intestinal ALP into the lymphatics. Another study showed that bile duct ligation produced an elevation in the serum of a high weight isoenzyme similar to that induced on the cana- licular surfaces of the hepatocytes (Toda et al., 1980). These authors reasoned that the induced isoenzyme is solubilized from hepatic mem- branes by bile salts. The bile with this elevated ALP activity cannot flow through the obstructed biliary tract and may seep into sinusoidal blood. After feeding, sharp increases in serum ALP occur in rats because of the increased activity of the intestinal isoenzyme (Saini and Posen, 1969). The lifespan of injected intestinal ALP in rats is very short, with most disappearing from the circulation within 2.5 hours (Saini and Posen, 1969). Elevated ALP levels in the absence of bone disease are usually associated with hepatic disease. The enzyme is said to be a sensitive indicator of intra- or extrahepatic cholestasis (wolf, 1978). Induction of the enzyme with resulting increased serum levels can occur with cholestasis, glucocorticoid therapy and with certain drugs, including primidone, phenobarbital and dieldrin (Hoffmann, 1977). In 2 separate experiments during which rats were fed TCDD, increases occurred in serum values of ALP, GGT,ALT and AST (Gasiewicz et al., 1980; Kociba et al., 1978). These changes were compatible with both hepatocellular leakage and cholestasis. There were no increases reported in ALP, ALT, AST or lactate dehydrogenase values from 2 studies 30 in which rats were fed commercial preparations of PCB at high levels (up to 500 ppm) for as long as 5 weeks (Garthoff et al., 1977; Kasza et al., 1976). The "Yusho" victims previously described were subjected to very extensive clinical examinations. Even though the average amount of PCB ingested by each victim was 2 grams (Kuratsune et al., 1972), only slight increases in ALP were detected in the serum of severely affected individuals. Lactate dehydrogenase, ALT and AST values were normal in all (Kuratsune, 1972). Even though histopathologic, ultra- structural and microsomal enzyme alterations were demonstrated in liver samples from most of these studies, the changes were apparently not sufficient to produce or sustain hepatocellular leakage. In one study in which rats were fed FM, there was no reported increase in ALT or AST (Garthoff et al., 1977). similarly, there were no increases in serum levels of AST or ALP in the serum of pigs fed the commercial PBB mixture (Ku et al., 1978). An increase in AST was reported in cows fed high levels of PBB for 2 months (Moorhead et al., 1977). While it would be difficult to make conclusions from these sparse data, there does not appear to be any firm evidence that moderate levels of PHB exposure produce consistent liver changes that can be detected via serum enzyme determinations. Lipoproteins The basic knowledge of lipoprotein metabolism will not be discussed in this review. Recent review articles have been written by Small (1977), Kane (1977), Tall and Small (1978), Miller (1979), Witztum and Schonfeld (1979) and Albers and Warnick (1981). Relatively few of the numerous research projects examining the effects of PHB have studied possible alterations in lipoprotein metabolism. Many exogenous and 31 endogenous factors have been identified that stimulate lipoprotein synthesis. These include exercise (Hartung et al., 1980), alcohol (Belfrage et al., 1977), estrogenic hormones, insulin (Nikkila, 1978) and numerous drugs and compounds, including chlorinated hydrocarbon pesticides such as lindane, DDT, Aroclor (PCB) and Kepone (Carlson and Kolmodin-Hedman, 1972; Kato et al., 1978; Ishikawa et al., 1978), TCDD (Poli et a1. , 1980) , phenobarbitone (Durrington, 1979) , glutethimide (Bolton et al., 1980) and clofibrate (Nikkila, 1978). Although there are rare exceptions, the consistent trend for these factors is to increase cholesterol concentrations through elevations in the high density lipoprotein fraction (HDL). Theories advanced for the HDL cholesterol increases include enhanced hepatic cholesterol synthesis and incorporation into HDL particles(Ishikawa et al., 1978; Durrington, 1979; Bolton et al., 1980; Poli et al., 1980), decreased cholesterol catabolism in the liver (Poli et al., 1980) and increased very low density lipoprotein (VLDL) catabolism via lipoprotein lipase with elevated HDL particle formation (Belfrage et al., 1977; Nikkila, 1978). Kato and Yoshida (1980) recently explored the theory that hypercholesterolemia produced by PCB administration to rats was related to enhanced hepatic microsomal cholesterol synthesis. They were able to demonstrate an increased in vivo rate of cholesterol synthesis and an elevated level of B-hydroxy-B-methylglutaryl Coenzyme A reductase (HMG-CoA), the rate limiting enzyme in cholesterol synthesis, in livers of rats given PCB. The possibility that other factors, such as decreased catabolism, might be involved was not pursued. As indicated, significant increases in serum cholesterol have been reported in rats fed TCDD (Gasiewicz et al., 1980; Poli et al., 1980). In the study by Poli et a1. (1980) , increases were localized to the 32 HDL fraction. The total amount and distribution of serum triglyceride were not affected in this study (Poli et al., 1980) but serum tri- glyceride values were decreased in the other study (Gasiewicz et al., 1980). In addition to those experiments already cited, increased serum cholesterol levels in response to PCB administration to rats have been reported by Allen et a1. (1976) and Garthoff et a1. (1977). Total serum triglyceride values were only measured in one of these studies (Allen et al., 1976) and was reported to transiently increase. Relatively few studies have examined the possibility of serum lipid alterations in association with PBB administration. While one study using rats reported an increase in total serum cholesterol (Garthoff et al., 1977) and another detected altered lipoprotein electrophoresis patterns (Sleight et al., 1978), an experiment with monkeys (Allen et al., 1978) and a survey of contaminated dairy herds (Mercer et al., 1976) revealed decreased serum cholesterol levels. No changes in serum cholesterol were detected in swine fed PBB, although altered lipoprotein values were reported (Howard et al., 1980). As indicated earlier, these inconsistencies may be species, dose or time related. Additional studies are needed to define lipid alterations resulting’ from such environmental contaminants. MATERIALS AND METHODS Experimental Design_ The experimental design is depicted in Table 1. Rats were fed diets containing various amounts (0, 0.1, l, 10 or 100 ppm) of either FM, 2,2',4,4',5,5'-HBB or 3,3',4,4',5,5'-HBB. There were 6 rats in each group. Separate lO-day and 30-day exposure periods were conducted for each compound. Each experiment involved feeding one compound at the 4 concentrations for one time period. This resulted in 6 separate experiments, each with its own set of controls. The 30-day, 100 ppm group for the compound 3,3",3,3',4,5'-HBB could not be included because of the toxicity of the compound. During a trial experiment with 2 rats at this exposure level, one rat died on day 20 and the other was moribund and was euthanatized the same day. A paired-feeding experiment was included for the rats fed 100 ppm of 3,3',4,4',5,S'-HBB for 10 days. This was needed to determine if changes in the rats fed this compound at this concentration were dose related _ or associated with reduced feed intake. Animals fed this concen- tration (100 ppm) for 10 days ate an average of 13 g of feed/day/rat. Their controls (0 ppm) ate 23 g of feed/day/rat. Accordingly, during the paired-feeding experiment, 6 rats were fed 13 g of feed/day/rat, 6 were fed 23 g/day/rat and 6 were given feed ad libitum. 33 34 Table 1. Experimental design Exposure Periods (days) Conc. in Diet 10 30 Compound (PPm) O 0.1 1 10 100 0 0.1 1 10 100 FM -a - - - - - - - - - 212.1414'1515"HBB - - - - — - — a - - 3’3'1414'1535'-HBB - - - - +b - - - - NDC a . . Six animals per group b . Four animals per group CND = not determined 35 Animals, Housing, Feed Outbred, male, Sprague-Dawley rats were purchased from Spartan Research Animals, Haslett, MI. They weighed between 250 and 300 g. All animals were acclimated for at least 2 to 3 days before an experi- ment was begun. During this period, the rats were fed a commercial rat feed (Wayne Lab-Blox, Allied Mills, Inc., Chicago, IL). The same feed, ground into fine granules, was also used in all experimental trials. During the experiments, the rats were kept in plastic cages, 3 animals per cage. Animals were housed in separate cages during the paired feeding experiment. Water was available during all phases of an experiment. Bedding consisted of heat-treated wood chips (Northeastern Products Corporation, Warrensburg, NY). Once an experiment was started, the cages were placed in a laminar flow, filter chamber (Contamination Control, Inc., Lansdale, PA). This was necessary to prevent dissemina- tion of the PBB compounds throughout the animal room. The chamber was cleaned and all prefilters were changed between experiments. Before each experiment was begun, the total amount of feed and quantity of compound needed for that trial were calculated. Each com- pound was weighed and placed in warm (40 C) corn oil (MazolaR). Both FM and 2,2',4,4',5,5'-HBB dissolved with stirring--the former within 1 to 2 hours and the latter overnight. The congener 3,3',4,4',5,5'-HBB never dissolved but dissipated into a fine suspension. These solu- tions were added to a calculated amount of ground commercial feed to produce the 100 ppm diet. Fresh corn oil was added to this mixture for a final concentration of 10 ml/kg. One part of a higher dietary concentration added to 9 parts of stock ground feed produced the next lower dietary concentration. Corn oil content was adjusted for the 36 desired final concentration. Control diets contained only corn oil vehicle. All feed preparations were tumbled for 15 minutes on a rotary mixer. Animals were fed from porcelain cup feeders with stainless steel caps and floating perforated discs. Chemicals Firemaster BP-6 used in this experiment was produced by Michigan Chemical Corporation, St. Louis, MI. The 2,2',4,4',5,5'-HBB congener was separated and purified from FM in the laboratory of Steven D. Aust, Department of Biochemistry, Michigan State University. Firemaster was dissolved in hexane, applied to a column of alumina in hexane, eluted with hexane and speCific fractions were pooled, dried and recrystallized to greater than 99.9% purity (Moore et al., 1978). The 3,3',4,4',5,5'- HBB congener was purchased from RFR Corporation, Hope, RI. The congener was purified >99% in Dr. Aust's laboratory by repeated alumina chromatography. Anesthesia and Blood Collection Eighteen to 24 hours prior to the termination of an experiment feed was removed from all treatment groups. Each animal was anesthetized with carbon dioxide (CO , dry ice) in a closed plastic container. Blood 2 (10-12 ml) was aspirated from the heart by cardiac puncture. With the animal in dorsal reumbency, an 18 gauge, 1.5 inch needle attached to a 10 m1 syringe was inserted just lateral to the xiphoid cartilage. Upon aligning the needle in an anteroventral direction and applying slight negative pressure on the syringe plunger, the needle was advanced until the tip was located within a chamber of the heart and blood flowed freely into the syringe.' After sample collection, rats were left in the CO2 chamber until they died, which usually occurred within 5 minutes. 37 Within 10 minutes of collection, blood samples were placed in a refri- gerator at 4 C and left for 2 hours. They were then removed and centrifuged for 10 minutes at 3,000 revolutions per minute (rpm). Serum was removed from the clot and placed in storage at 4 C. All enzyme and lipoprotein assays were performed within 5 days of serum collection. Beyond that time, samples were frozen (-10 C) and for additional determinations, such as the measurement of the congeners associated with specific lipoprotein fractions, these serum samples were used. Determinations Body and Organ Weights The body weight of each rat was obtained at the beginning of an experiment, every other day during a trial, and immediately before blood collection and euthanasia. After each animal was euthanatized, the following organs were inmediately removed and weighed on a top- loading balance (Mettler Series P, Model 163, Mettler Instrument Corp., Hightstown, NY): thymus, liver, thyroids, and spleen. Serum Cholesterol, Triglyceride All cholesterol, triglyceride and serum enzyme assays were per- formed with centrifugal analyzers in the veterinary clinical pathology laboratory at Michigan State University. Cholesterol, triglyceride, GGT and SDH were assayed on a GeminiR analyzer (Electro-Nucleonics, Inc., Fairfield, NJ) and serum AST, ALT and ALP on a GemsaecR centrifugal autoanalyzer (Electro-Nucleonics, Inc., Fairfield, NJ). All assays except for free cholesterol were available as commercial kits. 38 Total serum cholesterol was measured using a kinetic enzymatic procedure (cholesterol oxidase) first described by Allain et a1. (1974). Free cholesterol was determined by modification of techniques described by Allain et a1. (1974), Nagasaki and Akanuma (1977) and worthington Diagnostics (1978). Modifications were made to eliminate cholesterol esterase from the reaction mixture, thus preventing the conversion of cholesterol esters to unesterified cholesterol and the inclusion of the former in final determinations. Table 2 depicts reagents and their concentrations used in free cholesterol assays. Amounts of reagent and serum used in each free cholesterol test were 0.7 ml and 10 ml, respectively. Aqueous cholesterol standards prepared by the method of Abele and Khayam-Bashi (1979) were used for both total and free cholesterol assays. Crystalline cholesterol (99+% pure for chromatography) was obtained from Sigma Chemical Company, St. Louis, MO) and powdered sodium desoxycholate from Fisher Scientific Company, Fair Lawn, NJ. Triglyceride concentrations were determined by a kinetic enzymatic assay using a commercial kit (Worthington Diagnostics, Freehold, NJ). Serum Enzymes Serum alkaline phosphatase (ALP), gamma glutamyltranspeptidase (GGT), alanine aminotransferase (ALT), aspartate aminotransferase (AST) and sorbitol dehydrogenase (SDH) were determined using standard kinetic assays available in commercial kit form (Spin Chem, Sunnyvale, CA). All serum samples were stored at 4 C and enzyme assays were performed within 4 days. 39 Table 2. Free cholesterol assay Reconstituted Reagent Reagent Concentration Source Cholesterol oxidase 150 U/l ICN Nutritional Bio- chemicals, Cleveland, OH Phenol 21.2 mol/l Mallinckrodt Chemical Works, St. Louis, MO 4-Aminoantipyrine 1.6 mmol/l Sigma Chemical Co., St. Louis, MO Peroxidase 5390 U/l Sigma Chemical Co. Tris-HCl buffer (pH 7.5) 50 mmol/l Sigma Chemical Co. Triton X-100 0.5 ml/l Rohm and Hass, Phila- delpha, PA 40 Lipoproteins Preparation and analysis of serum lipoprotein fractions were con- ducted within 5 days of sample collection. Serum lipoproteins were separated into 3 major fractions, very low density lipoproteins (VLDL), low density lipoproteins (LDL) and high density lipoproteins (HDL). Portions of each serum sample (175 ul) were placed in polyallomer tubes and spun in an ultracentrifuge (AirfugeR, Beckman Instruments, Inc., Palo Alto, CA) for 2.5 hours at 100,000 rpm (106,000 to 165,000 x 9, minimum to maximum). After centrifugation, each tube was carefully removed and placed in a hole drilled within a metal block such that 125 ul of the serum sample was below the surface of the block and 50 ul above. Using a razor blade, the tube was sliced flush with the block and each section of tube was placed in a separate test tube. The VLDLs (density 0.95 to 1.006 g/ml) were located in the top section of the tube while LDLs (density 1.006 to 1.063 g/ml) and HDLs (density 1.063 to 1.21 g/ml) were isolated in the bottom section. Each fraction was then reconstituted with 0.195 molar NaCl to the initial sample volume (175 pl). These were stored at 4 C. Isolation of HDLs was accomplished through selective chemical precipitation using sodium phosphotungstate (NaPhT) and magnesium chloride (Matheson, Coleman and Bell, Cincinnati, OH) by the method of Burstein et al. (1970). Into 50 ml of distilled water, 4.0 g of phosphotungstic acid (Sigma Chemical Company, St. Louis, MO) and 16 ml of 1.0 mol/l sodium hydroxide (Mallinckrodt Chemical works, St. Louis, M0) were added. Distilled water was added to produce a final volume of 100 ml. A 2.0 m1 /1 solution of MgCl was also prepared. To 1 ml of 2 serum, 100 pl of the NaPhT and 25 ul of the MgCl2 solutions were added. After brief mixing, the VLDLs and LDLs formed precipitates with the 40 LiEproteins ' I Preparation and analysis of serum lipoprotein 9m '2" -' I- ducted within 5 days of sample collection. Sonia 11W. 4 ‘f separated into 3 major fractions, very low density WI ._-.,. V .0 low density lipoproteins (LDL) and high density “WI my Portions of each serum sample (175 Lll) were p1” iinpo and spun in an ultracentrifuge (AirfugeR ' Beck: Ins um: Palo Alto, CA) for 2. 5 hours at 100,000 1pm (105,000 to 155 .“I minimum to maximum). After centrifugation, «Ch tub. II removed and placed in a hole drilled Within . ,4. . a ”m M “I 125 111 of the serum sample was below the ”the; of 111 above. Using a razor blade VLDLs (density 0.95 to 1.006 g/ml) Were the tube while LDLs (density 1-006 to 1. 063 gm, um hotm. Wu.- MI m the. volume (175 L11) . These were stored It 4 c- 1.063 to 1.21 g/ml) were isolated in m. was then reconstituted with 0.195 '01“ Isolation of HDLS was “NJ-181133 precipitation using sodium “03PM: “flute chloride (Matheson, Coleman and Ben Burstein et a1. (1970). Into 50 ‘1 “dis . _ “Ila phosphotungstic and (5191113 Chen-teal My: St?”- E . M0) were added. Distilled water "I: - then of 1.0 mol/l sodium hydroxide (”-11 100 m1. A 2. 0 ml /1 solution “,1.” '0: serum, 100 pl of the NaPhT and 25 "10 0f After brief 11" "TDL , 5% “11;. 41 reagents and were sedimented by centrigugation at 6,000 rpm for 10 minutes. The supernatant fluid, which contained the HDLs, was aspirated from the tube and stored at 4 C. In addition to the serum samples, total cholesterol and triglyceride concentrations were determined for the 3 lipoprotein fractions using the techniques previously described. Since the VLDL and HDL-LDL fractions were previously restored to their original serum concentrations, no adjustment was needed for the lipid values. For similar quantitative assays in the HDL fraction, all values were multiplied by a factor of 1.125 to correct for dilution. The LDL values were obtained by sub- tracting corrected HDL fractions from HDL-LDL fractions for the same animal. Lipoprotein electrophoresis was conducted on serum samples from all experiments and on lipoprotein fractions from the 30-day experiments during which 2,2',4,4',5,5'-HBB and 3,3',4,4',5,5'-HBB were fed. All electrophoresis procedures were performed within 5 days of sample col- lection. Equipment and reagents used for electrophoresis procedures were produced by Helena Laboratories, Beaumont, TX. Lipoprotein samples were applied to cellulose acetate plates (Titan III XWR) that had been presoaked for 24 hours in Tris-barbital-sodium barbital buffer (ElectraR HR Buffer) diluted to 650 111. These were immediately placed in an electrophoresis chamber containing the same buffer and voltage was adjusted to 180 volts, maintained for 20 minutes. Plates were then removed and stained for 2.5 to 3.5 hours in a 0.2% solution of oil red O in methanol (Oil Red OmR). Plates were then washed in water, soaked with glycerol and scanned with a densitometer (Quick Quant II, Helena Laboratories, Beaumont, TX). 42 Gas-Liguid Chromatography Pooled serum, albumin and lipoprotein samples from 30-day, 100 ppm 2,2',4,4',5,5'-HBB and 10 ppm 3,3',4,4',5,5'-HBB experiments were analyzed by electron capture, gas-liquid chromatography (GLC). This was an attempt to measure the concentration of these specific congeners in serum and their pattern of association with different serum components. Liddle et a1. (1976) described an extraction technique which involved an ethyl-hexane extraction of methanol-treated serum. To increase yield of 3,3',4,4',5,S'-HBB congener, ethyl ether-hexane solution was replaced with toluene during the extraction procedure. Extraction solutions were eluted through florisil columns, condensed and subjected to GLC analysis. The gas chromatograph (G.C. Model 3700, Varian Instru- ment Division, Palo Alto, CA) was operated by personnel in the Department of Pathology, Michigan State University. A brief description of the technique has been provided by Render (1980). Albumin fractions for GLC analysis were prepared from pooled serum samples from the 2 experimental groups previously cited. These were treated with a solution of sodium sulfite (Na SO , Mallinckrodt Chemical 2 3 werks, St. Louis, MO) such that a final concentration of 26.9% Na2803 was produced (Cannon et al., 1974). Precipitated globulins were sedi- mented by centrifugation at 3,000 rpm for 10 minutes and supernatant fluid containing the albumin fraction was aspirated and stored at -10 C until extraction could be performed. Hepatic Microsomal Enzymes Immediately after euthanasia, liver samples from rats in various treatment and control groups were placed in cold potassium chloride- nicotinamide solution. These pooled samples were analyzed in the 43 laboratory of Steven D. Aust, Department of Biochemistry, Michigan State University. Microsomes were isolated, washed and stored according to methods previously described (Pederson and Aust, 1970; Welton and Aust, 1974). The following assays were performed on the microsomes: protein content, cytochrome P-450 content, maximum wavelength of the carbon monoxide difference spectrum and aminopyrine demethylase and benzola]PYrene hydroxylase activities. These techniques have been previously described or referenced (Moore et al., 1978). Statistical Analysis Data were analyzed for significant treatment effects by a one- way analysis of variance (ANOVA). Significant differences between means were detected by Student-Newman-Keul's (SNK) test. RESULTS Serum Enzymes Treatment effects on serum enzyme values are summarized in Table 3. Serum levels of SDH for 10- and 30-day periods were consistently increased by all treatments except the paired feeding experiment. These data are presented in Figures 2 and 3, respectively. Serum values for ALP decreased with all lO-day treatments, including the paired feeding experiment. There were no significant treatment effects during the 30-day experiments on serum ALP levels. Data for the lO-day experiments are summarized in Table 4. Only 3,3',4,4',5,5'-HBB, during both the 10- and 30-day experiments, significantly altered (increased) serum levels of AST. None of the other treatments significantly changed AST values. These data are presented in Table 5. There were no significant treatment effects on serum levels of ALT or GGT. Serum Cholesterol Treatment effects on serum total cholesterol, free cholesterol, and percent free to total cholesterol are presented in Table 6. Serum total cholesterol values were significantly affected by all treatments except the 10-day FM and paired feeding experiments. During the 30-day FM and both feeding periods for 2,2',4,4',5,5'-HBB, serum cholesterol values increased, while the lO-day feeding of 3,3',4,4',5,5'~ HBB significantly decreased cholesterol levels, the 30-day feeding 44 45 Table 3- Summary of treatment effects on serum enzyme activity Enzyme Treatment and Period Days SDH ALP AST ALT GGT FM 10 + +‘ - - - 30 + - . - > - '- 2,2',4,4',5,5'-HBB 10 + + - - - 30 + - - - - 3,3',4,4',5,5'-HBB 10 f + + - - 30 i - + - - Pair-fed controls 10 - + - - - +Values significantly increased from control (p<0.05) 4'Values significantly decreased from control (p<0.05) -No significant treatment effect 46 cm 23.: noon. 5 5322360 653860 0.. pd 36 v a .3560 :3: 222.3 2.50556 * 0 anxlamumaacaV Omicmfinvé. $530 3.2 5 Ease Em .~ use”. mu...— ..n.n .\\\\ hug ED m 3 93:833.? we .\ 2. cu mm on 47 3. mp 94. mm on 8' mm \ \. \\\\\\ “W“ or scans too... a. cozozcooeoo ocaanoo P , no \E .. 'iza‘ézizzsiszisiziai55252".:5252 no... v a .3536 ES. Essen 2.52.2.5 * 8.1.. _ nmzcamumauQacuaflan \\\\ carcassxxaag ED 85:53 room 2. 52h: Em .m was: OF 2 3.55on.; we _\ 2. cu . mm on 48 Table 4. Serum alkaline phosphatase activity in lO-day experiments Dairy Compound Concentration 2,2',4,4',5,5'- 3,3,',4,4',5,5'- (ppm) FM (IU/l) HBB (IU/l) HBB (IU/l) ' 0 155 i 29 112 i 13 93 r 20 0.1 137 i 29 97 i 16 88 i 13 1 127 i 29 100 i 9 . 76 i 11 10 116 i 25 78 i 14b 82‘: 9 100 92 t 19b 80 i 8b 60 1 13b —i_ 3Data represent mean f SD bValues significantly different from control (p<0.05) 49 Table 5. Serum aspartate aminotransferase activity in 10- and 30-day 3,3',4,4',5,5'-HBB experimentsa Dietary Concentration Period (days) I#_ (ppm) 10 (Iv/1) 3o (IU/l) 0 _ 51 a 4 64 i s 0.1 66 i 13 68 a 7 1 58 z 9 69 z 10 b 10 65 a 8 98 a 26 b 100 83 i 5 NE aData represent mean 1 SD bValue significantly different from control (p<0.05) NE = no experiment 50 Table 6. Summary of treatment effects on serum total, free and percent free/total cholesterol Determinations Treatment and Period Days Total Cho. Free Cho. % Free/Total FM 10 - - - 30 f + - 2,2',4,4',5,5'-HBB 10 + + + 30 f + + 3,3',4,4',5,5'-HBB 10 + + + 30 + - + Pair-fed controls 10 — + + -No significant treatment effect +Values significantly increased from control (p<0.05) + Values significantly decreased from control (p<0.05) 51 increased serum total cholesterol at moderate dietary concentrations and drastically decreased it at higher dietary concentrations. These data are presented for 10- and 30-day experiments in Figures 4 and 5, respectively. Free cholesterol levels in serum were increased by 30-day FM, 10- and 30-day 2,2',4,4',5,5'-HBB and paired feeding experiments. Serum levels of free cholesterol for animals in the lO-day 3,3',4,4',5,5'-HBB experiment were significantly decreased, reflecting a similar decrease in total cholesterol. These changes in free cholesterol are presented in Table 7. Percent free to total cholesterol ratios generally reflected significant changes in free cholesterol content of serum samples. Lipoprotein Cholesterol Table 8 contains a summary of treatment effects on HDL, LDL and VLDL cholesterol concentrations. Ten- and thirty-day data for HDL cholesterol concentrations are presented in Figures 6 and 7, respec- tively. Both FM and 2,2',4,4',5,5'-HBB significantly elevated HDL cholesterol during both feeding periods. For both feeding periods 3,3',4,4',5,5'-HBB tended to increase HDL cholesterol at a moderate dietary level (although this was only significant in the 30-day experi- ment) and tended to decrease HDL cholesterol at a higher level (although only the values from the lO-day experiment were significantly different from controls). LDL and VLDL cholesterol changes were few and unremarkable. They will not be presented in detail. Similarly, pair-fed control rats did not have any significant changes in cholesterol content of various lipoprotein fractions. 52 2... cu an. ac an. cm as: an ca. 2: a: cup 23 E55 coo...— c. cozoseoocoo 2:69:60 9 , ...o 36 v a .3366 EB. 22...... 2.52.285 * omH mergemsxxaa V\\\. unruemsxxawg Egg bang. z— magma 48m»? gm .3 ”Bo—“— EU I .3. ca .2" av cm 8 3323632. we . 2 .222 .om cop . a: cup ._ cap 53 2.55 too... :. cozozeoocoo oesanoo on on on on em cow c: on— car o: omw \\ \ ‘ T p £5.53 5.8 2. 39m: .8ng 72% .m use: mmI-.m.m..v.v caresseésmug 1 \\ omH .. 8 . 8. .392 mod v a ._o::oo So: I o: .coaoEo 2:322:55 * on law In: low 3305339. my To“? 122. non-M § 4 ED . 3: amp 54 . a Table 7. Serum free cholesterol levels in 10- and 30-day experiments Period (days) Dietary Con- Com ound centration ( m) 10 30 P - PP (mg/100 ml) (mg/100 ml) FM 0 10 r 2 12 r 4 0.1 9 r 1 14 i 2 l 10 r 1 12 r 1 10 10 r 2 14 1 5b 100 12 i 5 30 r 6 2,2',4,4',5,5'-HBB 0 8 r 2 9 r l 0.1 8 i 2 10 i l 1 11 r 2 10 i 1b 10 10 i 2b 13 r 3b 100 14 r 2 18 r 1 3,3',4,4',5,S'-HBB 0 11 r 1 13 r 1 0.1 12 r 3 14 r 2 1 12': 2 14 r 1 10 10 r 2b 100 1 r 1 ND Pair-fed controls ad libitum 12 i 2 NE 23 g/day 12 r 2 NE 13 g/day 15 1 1b NE a Data represent means f SD b . . . . Value Significantly different from control (p<0.05) ND = not determined NE = no experiment 55 Table 8. Summary of treatment effects on HDL, LDL and VLDL cholesterol levels Period Cholesterol Determinations Compound (days) HDL LDL VLDL FM 10 + I - - 30 f - - 2,2',4,4',5,5'-HBB 10 + - - 30 + ' + + 3,3',4,4',5,5'-HBB 10 + - - 30 + + - Pair-fed controls 10 - - - +Values significantly increased from controls (p<0.05) -No significant treatment effect + Values significantly decreased from control (p<0.05) 56 :53 poem 5 5:223:00 meson—coo op. owl on: owl omI ow... an- col 2: I. o: r amp I of. r whzwtmmmxm >39 :— mod v a 22:30 So... 322.5 22.85.35 * cm H morissxxaa 7/4 morsemnxxawg ED 395 Saguaro .5: .m 52“. L c— on on ov om cm on on cm 2: or w cup of. 3.65322. 5 .22.. 57 code too“. a. 5:223:60 oesanoo S 8 on 8 8 8 2 8 8 2: .o. P P... cap meg—mus EGAN z— 39”.. Beam—5505 d: modv a .3530 Ea... 222:9 2235.55 * unrumfise maximise N go; .cw low .cn [cw .om 1 8 3.6.33.2. my . E .292 l on .om 12: 1°: 18. a8. 58 Serum Triglyceride A summary of treatment effects on serum triglyceride concentrations is presented in Table 9. A more detailed presentation of results from the triglyceride assays is given in Table 10. The predominant effect of feeding these 3 compounds to rats was to lower serum triglyceride levels.' Even with the 10-day 2,2',4,4',5,5'-HBB group there was a trend, although not at significant levels, for the serum triglyceride values to be decreased with treatment. The reduced feed intake enforced during the paired feeding experiment also significantly depressed serum triglyceride values. Lipoprotein Triglyceride A summary of the HDL, LDL and VLDL triglyceride concentrations during the various experiments is presented in Table 11. The treatment effects on triglyceride content of a specific lipo- protein class occurred primarily in the VLDL fraction. These data are presented in more detail in Table 12. All treatments decreased tri- glyceride content of the VLDLs and these changes were significant in 4 treatment groups, including the paired feeding experiment. Lipoprotein Electrgphoresis Electrophoretic separation of serum lipoproteins was performed on all serum samples and on the lipoprotein fractions from the 30-day 2,2',4,4',5,5'-HBB and 3,3',4,4',5,5'-HBB studies. Electrophoretic patterns of various fractions in all cases were pure and showed no evidence of incomplete isolation. Electrophoretic patterns were evaluated by measuring the relative percentage of each lipoprotein band on densitometer tracings. Since rat VLDLs tend to migrate with HDLs, it was usually impossible to detect 59 Table 9. Summary of treatment effects on serum triglyceride levels Time Period Compound (days) Serum Triglyceride FM 10 + 30 + 2,2',4,4',5,5'-HBB 10 -' 30 + 3,3',4,4',5,5'-HBB 10 l 30 + Pair—fed controls 10 + Values significantly decreased from control (p<0.05) -No significant treatment effect 60 Table 10- Serum triglyceride levels in 10- and 30-day experimentsa Dietary Con- PerlOd (days) Co ound centratio ( m) 10 30 mp n pp (mg/100 ml) (mg/100 ml) FM 0 56 i 7 69 i 22 0.1 44 r 7 57 r 11 1 50 r 4 64 a 8 10 54 r 9b 50 r 10b 100 39 r 10 32 r 6 2,2',4(4. ,5,5'-HBB O 60 i 10 74 i 17 0.1 76 i 12 75 r 20 1 68 i 8 72 :t '8 10 71 a 12 63 r l6b 100 51 i 7 40 r 8 3,3',4,4',5,5'-HBB O 50 i- 10 61 it 10 0.1 55 r 9 60 r 10 l 56 r 11 57 r 7b 10 52 r 8 35 r 9 100 16 i 4 NE Pair-fed controls ad libitum 76 r 11 NE 23 g/day 57 i 14 NE 13 g/day 37 r 6 NE a Data represent means.: SD bValue significantly different from control (p<0.05) NE = no experiment 61 Table 11. Summary of treatment effects on high density, low density and very low density lipoprotein triglyceride levels Period Determinations Compound (days) HDL LDL VLDL FM 10 - - + 30 - - - 2,2',4,4l,5,5'-HBB 10 - - - 30 - - + 3,3',4,4',5,5'-HBB 10 - I - i 30 - + - Pair-fed controls 10 - - + -No significant treatment effect Values significantly decrease from control (p<0.05) 62 Table 12. Very low density lipoprotein triglyceride levels in 10- and 30-day experimentsa ' d Dietary Con- 10 Period ( ays) 30 Compound centration (ppm) (mg/100 ml) (mg/100 ml) FM 0 28 r 4 36 i 14 0.1 20 r 4 35 r 11 1 26 r 2 30 r 8 10 22 r 9b 26 i 9 100 14 r 4 16 r 6 2,2',4,4',5,5'-HBB 0 34 r 7 32 r 14 0.1 42 r 6 27 i 12 1 44 i 9 23 r 16 10 43 r 10 16 1 8b 100 29 r 8 3 r 2 3,3',4,4',5,5'-HBB O 22 i- 5 11 i 5 0.1 21 i 6 9 r 4 1 28 r 8 9 r 3 10 26 r 4b 6 r 2 100 5 r 2 NE Pair-fed controls ad libitum 23 i 10 NE 23 g/day 10 i 4b NE 13 g/day 2 r 2b NE a Data represent means i SD b . . . . Values Significantly different from control (p<0.05) NE = no experiment 63 this band and, therefore, to assess treatment effects on VLDLs with this technique. Effects of various treatments on relative percentages of HDL and LDL fractions are presented in Table 13, with the sum of HDL and LDL bands expressed as 100%. Both 10- and 30-day treatment periods for FM and 3,3',4,4',5,5'-HBB increased the relative percentage of HDL in relationship to LDL. Neither animals fed 2,2',4,4',5,5'-HBB nor pair-fed controls had an obvious change in the ratio of HDL and LDL. Microsomal Enzyme§_ Table 14 contains a summary of changes in data for cytochrome P-450 content and wavelength maximum of carbon monoxide (CO) difference spectrum of hepatic microsomes from each treatment group. All treat- ments except pair-fed controls produced 2- to 6-fold increases in cyto- chrome P-450 content of hepatic microsomes. Shifts in wavelength maximums of microsomes were towards 448 for both FM and 3,3',4,4',5,5'- HBB and essentially unchanged for 2,2',4,4',5,5'-HBB and paired feedings. In Figure 8 the effects of treatments on aminopyrine demethylase (AD) and benzola]pyrene hydroxylase activities for 30-day experiments are compared. The data were converted to means and then represented as increases over controls. FM increased both enzyme activities, while 2,2',4,4',5,5'-HBB primarily increased aminopyrine demethylase activity and 3,3',4,4',5,5'-HBB, only benzo[a]pyrene hydroxylase activity. Data for lO-day experiments were similar to these findings. Serum and Lipoprotein GLC Analysis for PBB Pooled serum, lipoprotein and albumin fractions from the 30-day experiments at the 100 ppm dietary level of 2,2',4,4',5,5'-HBB and 10 ppm level of 3,3',4,4',5,5'-HBB were analyzed for PBB by gas-liquid 64 Table 13. Treatment effects on relative concentrations of high density/ low density lipoprotein determined by lipoprotein electro- phoresisa Dietary Con- Periods (days) Compound centration (ppm) 10(%) 30(%) FM 0 81/19 82/18 0.1 77/23 - 1 82/18 89/11 10 86/14 93/7 100 94/6 97/3 212.1414. (S'S'-HBB 0 82/18 87/13 0.1 87/13 84/16 1 84/16 88/12 10 85/15 90/10 100 87/13 87/13 3,3',4’4' ,5,5'-HBB 0 75/25 80/20 0.1 76/24 87/13 1 70/30 92/8 10 72/28 98/2 100 92/8 NE Pair-fed controls ad libitum 80/20 NE 23 g/day 82/18 NE 13 g/day 79/21 NE aData represent mean for group NE = no experiment 65 Table 14. Cytochrome P-450 content and shift in wavelength maximum of hepatic microsomes for carbon monoxide difference spectrum Determinations Period Cytochrome P-450, Shift in A Compound (days) Fold Increasea Maximumb (nm) PM 10 ND ND 30 2.2 - 450 to 449 2,2',4,4',5,5'-HBB 10 6.0 . 449 to 450 30 2.8 450 to 450 3,3',4,4',5,5'-HBB 10 3.0 450 to 448 30 3.6 450 to 449 Pair-fed controls 10 0.9 449 to 450c a . Data calculated as nmol P-450/mg protein represent mean of results from highest treatment group % mean of control Data represent mean for control versus mean for highest treatment group cData represent mean for group fed ad libitum versus group fed 13 g/day ND = not determined 66 man—.2... .98.. 8”. 3958 $6 $993. .2 8883 (Sr 2. w 2. o . n or u an 8' p p ...o ._ an o unrumdneésna o 0.. ca 8 25.: .83. c. not..m.m..v.v..«.u .230 652.800 Sp 3 w a or .35 0E .55 zo.h<._>xO¢o>I l_________ \208: 33: E 323 a . on zo.._.<4>:._.m Ema magnum». s. u L on u z _ c >a O z . 2 < 2.8333” (mpg—g >3AH 5 >53; £5 023—35.; 4383—: u: