Physiology in Medicine
Formation, Transport, Metabolism and Excretion of Bilirubin
N Engl J Med 1969; 280:1339-1345June 12, 1969DOI: 10.1056/NEJM196906122802409
- Article
THE formation of bilirubin and its elimination from the body as a waste product of heme catabolism require a series of metabolic alterations and transport processes. Partial or complete failure at any point in this sequence can result in jaundice. This review will place particular emphasis on alterations of bilirubin metabolism and transport that are important in the pathogenesis of hyperbilirubinemia.
Bilirubin Synthesis
Although approximately 85 per cent of the bilirubin formed in the body is derived from senescent erythrocytes by conversion of the heme of hemoglobin to biliverdin and subsequent reduction to bilirubin within reticuloendothelial cells, little is known of the precise steps of this conversion or of the mechanism by which newly formed pigment enters the circulation. Controversy has centered largely on the question of whether the alpha-methene bridge of the protoporphyrin ring of heme is split before or after the removal of the globin and iron moieties from hemoglobin. Opening of the alpha-methene bridge results in the loss of one carbon atom and the formation of one molecule of carbon monoxide. Eighty per cent of the expired carbon monoxide is believed to be derived from the catabolism of heme from mature circulating erythrocytes; therefore, its rate of expiration may be used clinically as an index of heme catabolism.
Fifteen per cent of newly synthesized bilirubin is derived from sources other than mature circulating erythrocytes. Evidence that hemoglobin of mature circulating erythrocytes is not the only source of bilirubin came from the observation that during the first few days after administration of 15N-labeled glycine in vivo, and before its incorporation in circulating erythrocytes, the isotope was detected in stercobilin, an intestinal oxidation product of bilirubin. Incorporation of isotope into stercobilin also increased at about 120 days, which is the average survival of circulating erythrocytes formed when the 15N-labeled glycine was administered. The early labeling peak suggests that glycine is incorporated into bilirubin by a mechanism far more rapid than the synthesis and destruction of erythrocytes. Early-labeled bilirubin could arise from several sources, including heme or porphyrins not used during the synthesis of hemoglobin, intracorpuscular degradation of hemoglobin during erythrocyte maturation in the bone marrow, destruction of newly formed erythrocytes in the bone marrow, direct synthesis of bile pigment from porphyrins without degradation of the porphyrin ring, and turnover of nonhemoglobin heme-containing proteins. Recent studies using carbon-labeled precursors indicate that the early-labeled pigment fraction consists of at least two components, one associated with erythropoiesis and the other originating primarily in the liver.
The rate of synthesis of early-labeled bilirubin from nonerythropoietic sources can be estimated by measurement of the incorporation of isotopically labeled delta amino levulinic acid (ALA) into bile bilirubin. Radioactive ALA is more readily incorporated into early-labeled bile pigment than into hemoglobin heme. Nonhemoglobin-heme-containing proteins such as myoglobin, catalase, tryptophan pyrrolase and cytochromes are found in all tissues of the body. As with hemoglobin derived from erythrocytes, degradation of these proteins results in bilirubin synthesis and the release of carbon monoxide. The liver contains the greatest quantity of these proteins and is probably the largest source of bilirubin other than circulating erythrocytes.
The erythropoietic component of early-labeled bilirubin is increased in hematologic disorders characterized by ineffective erythropoiesis, such as thalassemia, pernicious anemia, dyserythropoietic jaundice and congenital porphyria. The nonerythropoietic component as well as the amount of carbon monoxide expired is increased after administration of phenobarbital, which enhances the synthesis and turnover of microsomal cytochrome P450 but not cytochrome B5.
Bilirubin-Albumin Binding
Unconjugated bilirubin, which is highly lipid soluble, has limited aqueous solubility from 0.1 to 5.0 mg per 100 ml at physiologic pH and tonicity. Binding of bilirubin by albumin increases its solubility in plasma; each mole of human albumin is capable of binding at least two moles of bilirubin, permitting a theoretical maximal serum unconjugated bilirubin concentration in an adult of approximately 80 mg per 100 ml. In addition, some studies have demonstrated binding of small amounts of bilirubin to plasma alpha1, alpha2 and beta globulins when the plasma bilirubin concentration exceeds 20 to 30 mg per 100 ml. Because of the small quantity of these proteins in plasma as compared to albumin, such binding appears to be of little quantitative importance. The binding of the first mole of bilirubin to albumin may be stronger than the binding of the second mole. Thus, there may be a gradient in binding affinity, resulting in greater dissociability of bilirubin from plasma proteins with increasing concentrations. Clinical interest in binding of bilirubin by albumin has primarily related to the development of bilirubin encephalopathy, kernicterus. In this condition, seen only rarely outside the newborn period, unconjugated bilirubin enters the neurons of the basal ganglions, hippocampus, cerebellum and medulla, causing necrosis of nerve cells, probably by interfering with cellular respiration.
The concentration of unconjugated bilirubin in plasma is determined by the relative rates at which it enters and leaves the circulation, the plasma albumin concentration, and the availability of binding sites on albumin. The normal bilirubin concentration is between 0.5 and 1.0 mg per 100 ml; however, the upper limit of normal is uncertain, for the distribution curve of values obtained from supposedly normal adults is asymmetrical. On the basis of a statistical analysis of a large population, the upper limit of normal has been designated as 1.5 mg per 100 ml. In other species, such as the dog, rat and sheep, the plasma is virtually devoid of bilirubin; yet in horses the plasma bilirubin concentration is considerably higher than in man. Virtually all bilirubin in normal plasma is unconjugated; however, none of the technics for analysis of bilirubin fractions permit meaningful detection of small amounts of conjugated bilirubin.
Alterations of Albumin-Bilirubin Binding
The binding capacity of albumin for bilirubin can be modified by a variety of physical and chemical alterations. Several anionic drugs, such as sulfonamides, are bound by plasma albumin and compete with bilirubin for binding sites. For example, the administration of sulfisoxazole to an infant with unconjugated hyperbilirubinemia may markedly reduce the serum bilirubin concentration, particularly if the anionic binding sites on albumin are saturated or nearly saturated with bilirubin. The displaced bilirubin enters tissues having a high lipid content, or may crystallize extracellularly in organs such as the gastrointestinal tract and kidney. Administration of sulfonamides to pregnant women or neonates increases the risk of kernicterus in the jaundiced infant. Free fatty acids are believed to behave similarly in competing with bilirubin for albumin binding, and elevated free fatty acid levels in plasma during the newborn period may be important in the neonate's increased susceptibility to kernicterus. Asphyxia, hypoxia and acidosis are also associated with increased risk of kernicterus in neonates, possibly by interfering with bilirubin-albumin binding, although they may also increase the permeability of the brain for unconjugated bilirubin. With recognition of the relation between the bilirubin binding capacity of plasma albumin and kernicterus there has been considerable interest in developing clinical methods for quantitating the capacity of albumin to bind bilirubin. Several technics have been devised, some of which use dyes such as 2-(4′-hydroxybenzeneazo) benzoic acid (HBABA) that have binding characteristics similar to the binding of bilirubin to albumin. However, extensive laboratory and clinical experience is required before the results of these tests can be relied upon as indications for exchange transfusion. Theoretical considerations support the concept that estimation of albumin binding capacity for bilirubin is a better indication of the need for exchange transfusion than is the serum bilirubin concentration.
Conjugated pigment is bound predominantly to plasma albumin and to a smaller extent to beta and alpha globulins. The question of whether conjugated bilirubin competes with unconjugated bilirubin for albumin binding is not as yet resolved. Elevated concentrations of conjugated bilirubin have not been associated with kernicterus, probably because conjugated bilirubin has little lipid solubility.
Transfer of Bilirubin from Plasma
The liver appears to have a selective capacity to remove unconjugated bilirubin and other organic anions from plasma. For example, after intravenous injection of a tracer dose of radioactive bilirubin into a rat, within five minutes 65 per cent of the dose is in the liver, and 85 per cent of this is found in the supernatant fraction of the liver homogenate. The mechanism for this rapid transfer and the hepatic selectivity are unknown; however, they are not primarily determined by hepatic blood flow. The relative specificity of the hepatic cell to remove unconjugated bilirubin from plasma could theoretically reside in the plasma membrane facing the sinusoid or cytoplasmic proteins with affinity for unconjugated bilirubin and other organic anions.
The lateral extension of the plasma membrane facing the hepatic sinusoid may have this unique functional property because of a specific molecular composition as compared with plasma membranes of other cells and other portions of the plasma membrane of the parenchymal liver cell. Liver-cell plasma membranes have been isolated by differential centrifugation. Their lipid and fatty acid composition is remarkably similar to that observed in similar membranes isolated from other tissues; however, heterogeneity in enzymatic activity in different portions of the hepatic-cell plasma membrane has been observed using cytochemistry. Rat-liver plasma membranes bind sulfobromophthalein (BSP) and other organic anions in vitro. Equimolar amounts of organic anions such as taurocholate, unconjugated bilirubin and fluorescein, which do not compete for hepatic uptake with BSP when simultaneously injected in vivo, show little or no interference with BSP binding by plasma membranes in vitro. By contrast, other organic anions, such as indocyanine green, flavaspidic acid glucaminate and iodipamide methyl glucamine, which compete with BSP for hepatic uptake in vivo, considerably reduce BSP binding by isolated plasma membranes in vitro. These findings suggest that the plasma membrane of the liver cell is not molecularly homogeneous and that specific receptor sites for various organic anions may be located in the lateral extension of the plasma membrane facing the sinusoid. This possibility may be open to experimental investigation because of the rediscovery of mutant Southdown sheep with an apparent functional defect in the transfer of several organic anions from plasma into the liver cell.
A second theory is that unconjugated bilirubin may be dissociated from albumin at the plasma membrane, which is generally permeable to nonpolar molecules, and an intracellular protein (or proteins) acts as an acceptor and facilitates the transfer of bilirubin into liver cells. Recently, two nonalbumin proteins, designated as Y and Z, have been isolated from liver cytoplasm and account for most of the intracellular binding of bilirubin and BSP observed after injection of these anions in vivo or their addition to liver supernatant in vitro. Y and Z have been isolated and partially characterized chemically and physiologically. Y is a basic protein and binds more bilirubin and BSP when compared to Z. Z is a basic protein that binds bilirubin or BSP when their concentrations on Y reach a critical level. Y and Z proteins are found in liver from all mammalian species studied. It is postulated that Y and Z may maintain the organic anion flux between plasma and liver that has been demonstrated in single circulation studies. Drugs such as flavaspidic acid and bunamiodyl, which regularly cause retention of unconjugated bilirubin and BSP in man and animals, interfere stoichiometrically with bilirubin and BSP binding to Z but not Y protein. The role of these cytoplasmic proteins in mild unconjugated hyperbilirubinemias such as the Gilbert syndrome, in neonatal jaundice, in the mutant sheep and in other states characterized by unconjugated hyperbilinibinemia remains to be determined.
Conjugation of Bilirubin with Glucuronic Acid
Mammalian liver contains an enzyme, referred to as glucuronyl transferase or uridine diphosphate glucuronic acid transglucuronylase, that catalyzes the transfer of glucuronic acid from uridine diphosphate glucuronic acid to various phenolic, carboxylic and amine receptors. Substrates for this enzyme include steroids (such as tetrahydrocortisone), metanephrine, bile pigments (such as bilirubin), drugs (such as chloramphenicol, salicylates and sulfonamides) and chemicals (such as menthol). Bilirubin forms a carboxylic (ester) glucuronide and its biosynthesis occurs via the glucuronide pathway:
In this system glucose serves as the source of glucuronic acid, and this mechanism represents the major if not the exclusive pathway for glucuronide formation in man. The reaction catalyzed by glucuronyl transferase appears to be rate limiting in the overall pathway. Glucuronyl transferase activity is found in skin, ovary, adrenal glands, kidneys, testes and synovial membrane; however, the greatest activity occurs in liver, where it is associated with the endoplasmic reticulum membrane portion of parenchymal liver cells. Glucuronyl transferase activity is relatively unstable, and the enzyme has been partially purified but not isolated or characterized. As a result, it is uncertain whether one or more glucuronyl transferases exist. This uncertainty has clinical implications in that most studies of glucuronyl transferase activity have been performed with the use of substrates other than bilirubin, largely for methodologic reasons. The enzyme catalyzing the formation of N-linked glucuronides has been separated from that catalyzing the formation of phenol and acyl glucuronides, and Gunn rats form N glucuronides in vitro and in vivo, suggesting that this glucuronyl transferase may be a separate enzyme. The situation regarding ester and ethereal glucuronides is less clear. For example, multiplicity of glucuronyl transferase has been suggested because of different rates of development with the use of acyl and ethereal substrates, varied effects of hypophysectomy or thyroidectomy on the capacity to form acyl or ethereal glucuronides, different effects of ions, pH and other factors on the formation of different glucuronides and different distribution of glucuronyl transferase activity in submicrosomal fractions of liver homogenates. The view that a single glucuronyl transferase catalyzes the formation of ester and acyl glucuronides is supported by the observation that rats and human beings with chronic unconjugated hyperbilirubinemia and glucuronyl transferase deficiency have defective formation of both types of glucuronides. The data are consistent with a multiplicity of glucuronyl transferases or greatly varied affinity of a single enzyme for different substrates.
Although glucuronyl transferase is found in several nonhepatic tissues the role of the enzyme in bilirubin glucuronide formation by these tissues is uncertain. Homogenates and slices of kidney and intestine form direct-reacting bilirubin in vitro, but the product has not been demonstrated to be bilirubin glucuronide, nor has uridine diphosphate glucuronic acid dependency been shown with these tissues. It has been proposed that pigment I, the presumable bilirubin monoglucuronide, is formed in nonhepatic tissues and converted to pigment II (or bilirubin diglucuronide) in the liver. However, in recent years the existence of pigment I as a naturally occurring substance has been questioned, and it may be a complex of pigment II and unconjugated bilirubin in the Chromatographic systems used for their separation. Similarly, the presence of nonglucuronide conjugates, particularly bilirubin sulfate, has been questioned, and the role of such conjugates in the excretion of bilirubin and in the production of jaundice is uncertain.
Inhibition and Stimulation of Glucuronyl Transferase
Glucuronyl transferase activity is inhibited in vitro by several drugs, dyes and steroids, some of which (such as novobiocin and pregnane-3 alpha, 20 beta-diol) are associated with prolonged nonhemolytic unconjugated hyperbilirubinemia in neonates. Pregnane-3 alpha, 20 beta-diol is excreted in the breast milk by a small portion of nursing mothers. Infants ingesting this milk, but not normal human milk, manifest prolonged benign unconjugated hyperbilirubinemia until the third to the 12th week of life unless breast feeding is temporarily terminated, after which hyperbilirubinemia disappears usually within three to 12 days. These mothers secrete this unusual isomer of pregnanediol in their urine during lactation, appear clinically normal and presumably have an inheritable defect in steroid metabolism that affects the course of hyperbilirubinemia in their neonates. Other healthy-appearing women regularly give birth to infants with severe nonhemolytic unconjugated hyperbilirubinemia and a high risk of kernicterus. During the last trimester of pregnancy serum from these mothers contains an unidentified factor that inhibits glucuronyl transferase activity in vitro, increases substantially in titer immediately before term and disappears rapidly after delivery from the maternal and the infant's serum. This factor may also be a progestational steroid and is absent from the mothers' serum specimens when they are not pregnant. The syndrome has been called transient familial neonatal hyperbilirubinemia or the Lucey-Driscoll syndrome.
Hepatic glucuronyl transferase activity is increased after administration of benzpyrene, chlorcyclizine aminiquinolines and phenobarbital to normal adult and neonatal animals. Administration of these drugs results in proliferation of smooth endoplasmic reticulum and increases the synthesis of microsomal protein including a variety of oxidative TPNH requiring microsomal enzymes associated with drug hydroxylation.
Glucuronyl Transferase Deficiency
Hepatic glucuronyl transferase deficiency is associated with several lifelong as well as transient disorders characterized by chronic nonhemolytic unconjugated hyperbilirubinemia in man and animals. Because bilirubin must be conjugated to be excreted in the bile or urine, glucuronyl transferase deficiency results in reduced or absent formation of bilirubin glucuronide, and unconjugated bilirubin accumulates in plasma and tissues. There is no apparent increase in the formation of nonglucuronide conjugates of bilirubin. In each of these disorders the formation of various glucuronides in vivo and in vitro is considerably reduced whereas bilirubin glucuronide formation appears to be absent. A mutant Wistar rat (Gunn strain) has lifelong chronic nonhemolytic unconjugated hyperbilirubinemia resulting from glucuronyl transferase deficiency in liver. However, the capacity to excrete administered conjugated bilirubin is retained. As in infants with erythroblastosis fetalis, the mutant rat is not jaundiced at birth probably because fetal bilirubin is transported across the placenta and subsequently conjugated and excreted by the maternal liver. Gunn rats, in which kernicterus frequently develops, constitute an animal model for study of the pathogenesis of this disorder. The accumulated bilirubin is unconjugated and the serum concentration is 10 to 15 mg per 100 ml. There is no bilirubinuria, and the bile is virtually colorless, containing only a trace of unconjugated bilirubin. The liver is morphogically normal on light microscopy, and on electron microscopy frequently shows large areas of agranular endoplasmic reticulum. Heterozygous rats are anicteric and have an intermediate defect in glucuronide formation. The defect is transmitted as an autosomal recessive character. Homozygous Gunn rats maintain normal production of bile pigment and relatively constant serum and tissue unconjugated bilirubin concentrations despite inability of the liver to conjugate bilirubin. This suggests that alternate pathways of pigment disposition are present. The following patterns of bile-pigment excretion have been observed in Gunn rats: the major portion is catabolized to diazo-negative, polar bilirubin derivatives that are excreted in the bile and urine; a small amount of unconjugated bilirubin is excreted in the bile; and a substantial amount of unconjugated bilirubin is transferred across the intestinal mucosa into the intestine. Presumably, these combined excretory rates balance the rate of unconjugated bilirubin formation, and a steady state in pigment turnover is attained. After intravenous injection of human albumin, which has a stronger binding capacity for unconjugated bilirubin than rat albumin, the plasma unconjugated bilirubin concentration increased threefold. This is analogous to results obtained by Odell, who injected human albumin into jaundiced newborn infants with hemolytic disease before performing exchange transfusions for the purpose of removing unconjugated bilirubin and preventing kernicterus.
A disorder comparable to that of the Gunn rat was described in man by Crigler and Najjar in 1952. Approximately 40 cases of this rare disorder have been presented. The serum bilirubin is unconjugated, and its concentration ranges from 20 to 45 mg per 100 ml. Kernicterus is frequently observed during the neonatal period. Several patients have survived into adolescence without signs of kernicterus, and others have survived into adult life despite gross kernicterus. At autopsy the latter patients, like the adult Gunn rat, do not reveal bilirubin staining of the central nervous system but show neuronal damage and gliosis, which initially took place during neonatal life, when the brain seems to be more susceptible to damage from hyperbilirubinemia.
Recent studies of patients with chronic nonhemolytic unconjugated hyperbilirubinemia associated with defective glucuronide formation in vivo and markedly reduced hepatic glucuronyl transferase activity in vitro suggest that this syndrome occurs in two forms that appear clinically similar but are genotypically heterogenous. Most patients described as having the Crigler — Najjar syndrome belong in Type 1, in which hyperbilirubinemia is usually more severe and kernicterus is frequent, as noted. As in the Gunn rat, their bile is virtually colorless and contains only a trace of unconjugated bilirubin; the defect is transmitted as an autosomal recessive character, and hyperbilirubinemia is unaffected by prolonged phenobarbital administration. Patients with Type 2 disease generally have less severe hyperbilirubinemia (serum bilirubin range of 8 to 22 mg per 100 ml) without kernicterus. Their bile is pigmented and contains abundant amounts of bilirubin glucuronide. The conjugation defect is transmitted in a different fashion, which is most consistent with an autosomal dominant character with variable expressivity. These patients respond dramatically to administration of phenobarbital, with complete amelioration of jaundice within five to 12 days and decline in serum unconjugated bilirubin concentration to approximately 1.6 to 2.2 mg per 100 ml. After cessation of phenobarbital administration, the observed hyperbilirubinemia before treatment recurs in approximately 10 days. The mechanism for the response to phenobarbital in these cases is uncertain although current evidence supports the view that the glucuronyl transferase activity may be induced by the drug. Other possibilities include increased formation of a hepatic bilirubin-binding protein, metabolism of heme or bilirubin into unrecognized pathways and enhanced biliary excretion of bilirubin. Studies with radioactive bilirubin indicate that phenobarbital administration results in removal of bilirubin from the body and not merely shifting it into another compartment. It is unlikely that patients of Type 1 are homozygous for the same gene for which those of Type 2 are heterozygous, for no family has been described in which both types of the disease are seen. The difference between Type 1 and Type 2 probably reflects differences in the structure of a single glucuronyl transferase or in the control of protein synthesis.
Neonatal Physiologic Hyperbilirubinemia
Normally occurring delayed development of glucuronyl transferase is generally believed to be responsible for the transient unconjugated hyperbilirubinemia ("physiologic jaundice") that occurs in nearly every human neonate although few studies using liver from human infants, particularly with bilirubin as a substrate, have been performed. Most studies of the development of glucuronyl transferase have used substrates other than bilirubin and have been performed in animals that do not manifest neonatal hyperbilirubinemia. The pattern and rate of development of glucuronyl transferase activity varies with the species as well as the substrate used as a glucuronide receptor. Consequently, the role of glucuronyl transferase in "physiologic jaundice" has not been definitively established. Other theoretical possibilities include increased bilirubin production resulting from reduced erythrocyte life-span in neonates, inhibition of glucuronyl transferase by steroids of fetal or placental origin and reduced transfer of bilirubin from plasma into the liver cells. Reduced hepatic uptake of bilirubin has been demonstrated in guinea pigs less than two weeks of age.
Stimulation of the synthesis of several hepatic endoplasmic reticulum enzymes by various sedatives, antihistaminics and other drugs suggested their use as possible therapy for "physiologic jaundice of the newborn." Chloroquine, which stimulated hepatic glucuronyl transferase activity in newborn rats when administered to the mother, failed to alter the course of "physiologic jaundice" in human neonates when the drug was administered to mothers during the last trimester of pregnancy. In another study, phenobarbital was given to women before delivery, and the maximum serum bilirubin concentrations in their infants were reduced by approximately 50 per cent. The mechanism of this effect and its therapeutic usefulness require further study.
Gilbert Syndrome
Not all unconjugated hyperbilirubinemia in the absence of overt hemolysis in adolescents and adults results from glucuronyl transferase deficiency. Chronic unconjugated hyperbilirubinemia with serum bilirubin concentrations of 1.5 to 5 mg per 100 ml is observed in the Gilbert syndrome, which has multiple causes, including compensated hemolytic states, increased production of bile pigment from sources other than mature circulating erythrocytes and viral hepatitis. It also occurs as an inherited abnormality transmitted as an autosomal dominant character. In some families with Type 2 hepatic glucuronyl transferase deficiency mild jaundice is also observed. In most cases the pathogenesis of the Gilbert syndrome is unknown, and studies of glucuronide formation in vivo and of glucuronyl transferase activity in vitro are usually normal. Impaired transfer of bilirubin from plasma into liver cells has been postulated and supported by indirect evidence but not demonstrated directly. Theoretically, disorders of the plasma membrane or organic anion binding by cytoplasmic proteins may account for this commonly observed syndrome. Jaundice is usually accentuated by exercise, fever, pregnancy, alcohol and various drugs, including oral contraceptives, and the mechanism for these effects is unknown.
Excretion of Conjugated Bilirubin into Bile and Urine
After conjugation with glucuronic acid, bilirubin is rapidly excreted by the liver cell. Although the concentration of bilirubin glucuronide has never been measured directly in the bile canaliculus, it is generally assumed that the observed high ratios of bile bilirubin to plasma result from energy-dependent hepatic cell excretory processes and not from selective solute reabsorption by the biliary epithelium. The excretory process is therefore considered to be an active transport system although its energy source and the mechanism for coupling energy to transport are unknown. Hepatic subcellular distribution studies with Radio-labeled bilirubin suggest that lysosomes and microbodies do not participate in excretory transport. The plasma membrane seems anatomically suited as the site for excretory transport because its surface area is greatly increased by virtue of microvillous formation, and several phosphatases, thought to be important in active ion transport in other cells, are associated with this membrane.
The excretion of conjugated bilirubin is believed to be rate limiting in the normal transfer of bilirubin from blood to bile. In animals the capacity to excrete bilirubin has been shown to increase with age during the neonatal period and appears to be rate limiting at all ages. In adult rats the maximal excretory rate is approximately 60 μg of bilirubin excreted per 100 gm per minute. Comparable data are not available for man; however, clinical observations suggest that there is a large functional reserve so far as hepatic excretory transport is concerned. These studies probably explain the observation that severe hemolysis in patients with normal liver function is manifested by modest unconjugated hyperbilirubinemia whereas hemolysis in patients with coexisting liver damage is associated with substantial conjugated hyperbilirubinemia.
Conjugated bilirubin is found in plasma only in hepatic disease, and there is little information regarding the capacity of the human liver for its excretion. Observations from cross circulation of jaundiced patients with liver failure and nonicteric recipients suggest that conjugated bilirubin is rapidly removed from the circulation by the recipient's normal liver, but the rate is unknown. In rats infused with conjugated bilirubin the removal rate from plasma is rapid, and the hepatic excretion rate is similar to that observed after intravenous administration of unconjugated bilirubin.
In dogs and man with conjugated hyperbilirubinemia a portion of the conjugated pigment in plasma is ultrafilterable and is bound to a low-molecular-weight material that migrates as an alpha or beta globulin on electrophoresis and is filtered by the glomerulus. Evidence for tubular reabsorption or secretion by the kidney has not been found in man, animals or aglomerular fish (lophius americanus). The low-molecular-weight carrier has not been identified; however, its concentration in plasma increases with the degree of renal damage in jaundiced patients with hepatitis or cirrhosis. In such patients progressive impairment in glomerular filtration rate is often associated with an increase in the serum bilirubin concentration largely in the conjugated fraction.
Bilirubin glucuronide is one of many organic anions found in bile, and it is uncertain if these substances share a common excretory mechanism. Studies involving competition between bilirubin and other organic anions, particularly bile salts, which are the major organic anion found in bile, suggest there may be a single organic anion excretory mechanism that has many similarities to that found in the kidney. On the other hand, possible multiplicity of excretory mechanisms is suggested by studies in human beings and sheep with the Dubin-Johnson syndrome, an inheritable disorder of hepatic excretory function. This disorder is characterized by reduced hepatic capacity to excrete various organic anions including conjugated bilirubin, BSP and iodopanoic acid. The liver is black because of accumulation of a black-brown polymer in hepatic lysosomes. This pigment has been termed lipofuscin on the basis of staining reactions in tissue section; however, the pigment has been isolated and chemically identified as a melanin possibly resulting from impaired biliary excretion of epinephrine metabolites. In a human patient with the Dubin—Johnson syndrome and a virtually complete biliary fistula and in the mutant sheep, the biliary excretion of taurocholate, the major organic anion in sheep and human bile, appears to be normal, suggesting possible multiplicity of hepatic excretory mechanisms. Plasma bile salt concentrations are normal, and cholestasis is absent both chemically and morphologically. For bile secretory failure (cholestasis) to occur, there must be altered bile salt formation or secretion or both.
Various drugs such as C-17 alkylated anabolic steroids, estrogens and oral contraceptive agents regularly reduce the capacity of the liver to excrete organic anions such as BSP; however, hyperbilirubinemia does not occur unless there is prior reduction in excretory capacity produced by liver damage or occurring in an inheritable defect such as the Dubin–Johnson syndrome. Women who manifest chemical and morphologic signs of cholestasis during the last trimester of pregnancy usually have a recrudescence of this syndrome when they receive oral contraceptive drugs. Similarly, patients with benign familial recurrent cholestasis experience exacerbation on administration of estrogens.
Although advances have been made in the understanding of bilirubin production, transport in plasma, transfer into liver, conjugation, excretion and metabolism, little is known about the mechanisms involved.
*From the departments of Pediatrics and Medicine, Albert Einstein College of Medicine (address reprint requests to Dr. Gartner at the Department of Pediatrics, Albert Einstein College of Medicine, New York, N.Y. 10461). (Dr. Gartner is the recipient of a career-development award from the National Institute for Child Health and Human Development, National Institutes of Health, United States Public Health Service.)
- References
References
1
Arias
CrossRef | Web of Science | Medline2
Idem. Formation of bile pigment. In Handbook of Physiology: A critical comprehensive presentation of physiological knowledge and concepts. Section 6. Alimentary Canal. Vol. 5. Bile; Digestion; Ruminal Physiology. Section editor: C. F. Code. Washington, D.C.: American Physiological Society, 1968. pp. 2347–2374.
3
Bilirubin Metabolism: (Papers presented at an international symposium held at the Royal Free Hospital in 1966). Edited by I. A. D. Bouchier and B. H. Billing. Oxford: Blackwell Scientific Publications, 1967.
4
Gartner, L. M., and Arias, I. M. Liver. In Pediatrics. Fourteenth edition. Edited by H. L. Barnett. New York: Appleton, 1968. pp. 1490–1518.
5
Schmid, R. Hyperbilirubinemia. In The Metabolic Basis of Inherited Disease. Edited by J. B. Stanbury, J. B. Wyngaarden and D. S. Fredrickson. Second edition. New York: McGraw-Hill, 1966. pp. 871–902.
6
With, T. K. Bile Pigments: Chemical, biological and clinical aspects. New York: Academic Press, 1968.
- Citing Articles (11)
Citing Articles
1
Francesca Gullotta, Alessandra di Masi, Massimo Coletta, Paolo Ascenzi. (2011) CO metabolism, sensing, and signaling. BioFactorsn/a-n/a
2
Dennis D. Black, E. Amy Suttle, Peter F. Whitington, Gene L. Whitington, Sheldon D. Korones. (1981) The effect of short-term total parenteral nutrition on hepatic function in the human neonate: A prospective randomized study demonstrating alteration of hepatic canalicular function. The Journal of Pediatrics 99:3, 445-449
3
D.J. Back, A.M. Breckenridge, Francesca E. Crawford, Karen J. Cross, M.L'E. Orme, P.H. Rowe, Eileen Smith, Kathryn Todd. (1978) Biliary excretion of synthetic sex steroids in heterozygous and homozygous gunn rats. Life Sciences 23:10, 1053-1056
4
Joel M. Andres, Richard K. Mathis, W. Allan Walker. (1977) Liver disease in infants. The Journal of Pediatrics 90:5, 686-697
5
D. J. Back, Jagmohan K. G. Singh. (1976) The biliary excretion of [3H] lysergic acid diethylamide in Wistar and gunn rats. Experientia 32:5, 616-617
6
Johnson , John D. , . (1975) Neonatal Nonhemolytic Jaundice. New England Journal of Medicine 292:4, 194-197
Full Text7
Berk , Robert N. , Loeb , Peter M. , Goldberger , Lawrence E. , Sokoloff , Joel , . (1974) Oral Cholecystography with Iopanoic Acid. New England Journal of Medicine 290:4, 204-210
Full Text8
(1972) Failure of Enzymatic Induction. New England Journal of Medicine 287:3, 148-149
Full Text9
H. Reyes, A. J. Levi, R. Levine, Z. Gatmaitan, I. M. Arias. (1971) BILIRUBIN: A MODEL FOR STUDIES OF DRUG METABOLISM IN MAN. Annals of the New York Academy of Sciences 179:1 Drug Metaboli, 520-528
10
Hans Popper, Fenton Schaffner. (1970) Pathophysiology of cholestasis. Human Pathology 1:1, 1-24
11
Coburn , Ronald F. , . (1970) Endogenous Carbon Monoxide Production. New England Journal of Medicine 282:4, 207-209
Full Text
