METABOLISM OF STEROID AND AMINO ACID MOIETIES OF CONJUGATED BILE ACIDS IN MAN

Transcription

1 GASTROENTEROLOGY 72: , 1977 Copyright 1977 by The Williams & Wilkins Co. Vol. 72, No.1 Printed in U.S.A. METABOLISM OF STEROID AND AMINO ACID MOIETIES OF CONJUGATED BILE ACIDS IN MAN V. Equations for the perturbed enterohepatic circulation and their application NEVILLE E. HOFFMAN, M.B., B.S., PH.D., AND ALAN F. HOFMANN, M.D. Gastroenterology Unit, Mayo Clinic and Mayo Foundation, Rochester, Minnesota Equations previously developed to describe the enterohepatic circulation of the major biliary bile acids in man (Gastroenterology 67:887, 1974) were modified in order to predict the effect on biliary bile acid composition and pattern of amino acid conjugation after prototypic perturbations of the enterohepatic circulation in man. For the steroid moiety, the effects of bile acid feeding, increased recycling frequency, decreased intestinal conservation, and increased dehydroxylation were simulated. For the glycine or taurine moiety, the effect of increased deconjugation or preferential loss of one of the amino acid moieties was simulated. For the steroid moiety, the steady state biliary bile acid composition reflects the balance between input and conservation for each bile acid. Similarly, the distribution of bile acids between glycine and taurine conjugates reflects the balance between conjugation and conservation fot each amino acid moiety. Because these values may vary widely and independently, analysis of biliary bile acid composition in terms of the steroid moiety or the glycine-taurine ratio per se cannot be used to infer the relative rates of input or conjugation. The first three papers in this series described the behavior in healthy persons of the steroid and amino acid moieties of the four major conjugated bile acids present in man. 1-3 In the fourth paper, we presented a set of definitions and equations to describe the enterohepatic circulation of the major biliary bile acids under the usual steady state conditions. In this paper we apply these equations to predict the response of the enterohepatic circulation to various prototypic perturbations of this steady state for the major biliary bile acids. We test sequentially the effect of varying input or intestinal conservation on pool size, as well as on bile acid secretion-the product of pool size and the cycling frequency; other independent variables are treated as constants in each case. Then we scrutinize the determinants of composition of the pool under steady state conditions. The second portion of the paper deals with the amino acid moiety of the conjugated bile acids. Requirements for daily synthesis and effect of varying input on decreased conservation, including increased deconjugation, are assessed. Finally, the determinants of Received February 17, Accepted July 12, Address reprint requests to Dr. Alan F. Hofmann, Gastroenterology Unit, Mayo Clinic and Mayo Foundation, Rochester, Minnesota This study was supported in part by Grant AM from the National Institutes of Health and a grant-in-aid from the Mead Johnson Company. Doctor Hoffman was a Fulbright-Hays Research Fellow. His present address is: Department of Medicine, Division of Gastroenterology, University of Texas Medical School, Houston, Houston, Texas Hl the steady state amino acid composition of conjugated bile acids are probed. Steroid Moiety Variable Input to the Pool (P) The first perturbation to be considered is that of increased input of bile acid, as occurs in bile acid feeding. The recycling frequency and tabs (fractional absorption of bile acid per cycle, i.e., the intestinal conservation fraction) will be held constant as the effect of varied input on pool size is tested. It is assumed that the new input mixes with the bile acid pool distal to the site where it is sampled and that the tabs of the expanded pool remains constant. For the purpose of this discussion, let the daily input be presented in equal amounts per cycle. As the new steady state is approached: (la) where Pi is the pool size at the end ith cycle, Pi- 1 is the pool size at the end of the previous or i-lth cycle, and Ie is the input per cycle-this is the amount fed; synthesis is assumed to be zero or negligible relative to the amount fed. The value of Pi with successive cycles is shown in figure 1, taking the initial value for pool size, Po, as 6 mmoles, and tabs of It can be seen that a plateau value of PI is attained after about 60 cycles-approxi- mately 10 days-and the plateau or steady state value, p., is given by the equation: P, = (P, + Ie) fabs (lb)

2 142 HOFFMAN AND HOFMANN Vol. 72, No.1 16 Input/cycle mmols The obvious case is where the maximal absorptive capacity limits pool size. During attainment of the new steady-state: -... '" <:>., '" 12 <:> 4 o Number of enterohepatic cycles FIG. 1. Change of pool size with enterohepatic cycling for variable bile acid inputs; the simulation assumes that synthesis is zero or negligible relative to the input. In this simulation, the initial value of the pool size was 6 mmoles and the percentage of the pool absorbed per cycle (fab') was taken as which rearranges to: Ie = (P. + Icl (1 - fa.,) (Ie) Thus, the plateau value for P is attained when input per cycle is equal to loss per cycle. It must be emphasized that the concept of per cycle is really an averaging over many cycles. From cycle to cycle there may be deviation from the predicted behavior, but the averaging concept allows description of the system. Equation 1a considers that the bile acid pool is sampled before input, i.e., immediately beyond the liver and before mixing with dietary bile acid; it also assumes that dietary input is the only input of bile acid. In fact, there is probably continuing hepatic synthesis of bile acid during bile acid feeding and the extent may depend on the rate of bile acid feeding. To allow for this, equation 1a could be extended to: (2a) where Q is the rate of synthesis per cycle in millimoles, which is assumed to be constant per cycle. Therefore, in the steady state, equation 2a would become: or P, = (P. + Q + Ie) fab' (2b) (2c) The modification of equation la necessitates only the addition of a constant to each point in figure 1. The magnitude of that constant, (Q. labs), must be estimated experimentally. A second possible limitation for expansion of the bile acid pool is labs' In the previous simulation, labs was held constant and the intestinal absorptive capacity was assumed to be unlimited. The former assumption may be true only within limits and the latter is certainly untrue. The maximal intestinal absorptive capacity is not known, but it may be valuable to consider the case where this capacity is relatively small. and P will be limited when: Pi = (Pi.' + Q + Icl fab' (2a) where M is the maximal absorptive capacity of the intestine per cycle. But in the steady state, according to equation 2c, output per cycle = (P. + Q + Ie> (1 - labs)' Adding equations 3 and 2c one obtains: (3) M= P, (4) Thus, in the new steady state the pool size is limited by the maximal absorptive capacity and is in fact equal to the maximal absorptive capacity per cycle. It does not depend on labs' Again the assumption of this treatment must be stressed. The concept of maximal absorptive capacity in the intestine requires carefully defined conditions and is another "average" concept. Variable Loss Irom the Pool Variable recycling rate. If daily input and efficiency of intestinal absorption remain constant, then the rate of recycling may determine pool size. In previous papers, 4,5 it was shown that: I. = CP (1 - fa.,) (5a) where Id is input per day, e.g., in millimoles, and C is the apparent frequency of recycling of the bile acid pool, in cycles per day. Thus: P = I./IC (1 - fab') 1 (5b) Pool was thus inversely proportional to the rate of recycling; this relationship is plotted in figure 2 for a constant daily input of Id mmole of bile acid for labs equal to 0.95 or But daily bile acid secretion is given by CP and from (5b) we obtain: 20.., 16.l!! <:) e: 12 e: 'II'.!::!.., <:) CP = 1./(1 - fab') Enterohepatic cycles/day (5c) FIG. 2. Effect of variable recycling frequency on pool size. In the simulation, input is held constant and two values are used for percentage absorption per cycle. In this model the hyperbolas describe a line of constant secretion.

3 January 1977 PERTURBED BILE ACID CIRCULATION EQUATIONS 143 Thus, if pool size is treated as the dependent variable, we see that daily bile acid secretion is dependent on the rate of synthesis (when synthesis is the only input) and the efficiency of intestinal absorption; however, secretion is independent of the rate of recycling. Similarly, bile acid returning to the liver is independent of the recycling rate and is given by: return to the liver/day CP iabs (/. iabs)/(l - i ab.) Variable tabs. Equation 5 also allows us to examine the effect of a variation in tabs on the pool size. For this case, the daily input will not be held constant but will be allowed to change to match loss until some maximal synthetic capacity is attained. The recycling rate will be held constant and we will consider the effect on steady state pool size of different values for tabs. It can be seen in figure 3 that as tabs falls, pool size can be compensated for by increased synthesis up to the point of maximal synthesis. At this point, the pool size falls rapidly with a small change in tabs. From the pool size, number of cycles, and synthesis rate, we may plot bile acid secretion from the liver, return to the liver, and hepatic synthesis as a function of tabs (fig. 4). In figure 5A we have plotted bile acid synthesis as a function of bile acid return and have produced a curve of similar shape to the experimental curve of Dowling, Mack, and Small as plotted by Small (6) et a1. 8 Figure 5B shows bile acid synthesis (lowest curve in figure 4) assuming in addition a basal synthetic rate. With this assumption and with rounding of the angles, the change in synthetic rate appears as a dose-response curve. Both of these simulations have neglected changes in pool size that might occur during the day, because meals are ingested only during the daytime, whereas bile acid synthesis may well be constant throughout the day. If one sets the fasting state pool size at an arbitrary value, assumes values for tabs and synthesis rate, and further assumes that three meals are ingested at conventional times, one can then calculate the time course of 6 Synthetic rate mmoles/day LOO O.B 4.0 fobs FIG. 3. Effect of variable intestinal conservation (fabs) on pool size for varying synthetic rates, assuming that iabs remains constant for each cycle. 6 A Bile acid return,mmoles <;..... '" " <> 1.00 " 0.50 iii B Bile acid return, mmol fobs FIG. 4. Bile acid secretion, return of bile acids to the liver, and bile acid synthesis for variable intestinal conservation. The model assumes that secretion cannot exceed 6 mmoles, and that synthesis has an upper limit (1.7 mmoles). FIG. 5. A, bile acid synthesis in relation to bile acid return assuming that there is a compensating increase in synthesis to maintain the bile acid pool and that synthesis has a maximal value. B, bile acid synthesis in relation to bile acid return assuming that there is both a maximal and a basal synthesis rate; the curve has been rounded at its angles to give it the appearance of a dose-response curve.

4 144 HOFFMAN AND HOFMANN Vol. 72, No.1 pool size throughout the day: Pi Pi-! + Q - [(Pi.') (1 - fabs) 1 where abbreviations are defined in equations (1) and (2). Figure 6 shows that pool size declines progressively during the day and is restored by nocturnal synthesis to the fasting state size. The preceding discussion has dealt with only one bile acid, and it must be stressed that input fabs and recycling frequency are not necessarily the same for individual bile acids. Steady State Pool composition. Biliary bile acid composition is determined by chromatographic analysis of biliary bile acids. For determination of pool size, one uses a conventional isotope dilution technique and a one-pool model. 7, 8 The daily fractional turnover rate, k, is equal to (1 - fabs)c.9 In the previous paper,4 we derived an equation: where N' e is the mole fraction of cholic acid in the bile acid pool, Ne is the mole fraction of cholic acid in the daily primary bile acid synthesis, and ke, k de, kede are fractional turnovers for cholic, deoxycholic, and chenodeoxycholic acids, respectively. fdehydrox is the fraction of the daily cholic acid loss which is conserved as deoxycholic acid. In the analysis to follow, we will consider effect of allowing one factor to vary while the other two remain constant. These factors are: (1) the fractional turnovers, (2) N e, and (3) fdehydrox' The relative magnitude of the fractional turnovers-kc:kcdc:kdc, An examination of equation 7 shows that it is not the absolute magnitude of the fractional turnover rates but rather their relative magnitude which determines biliary bile acid composition at equilibrium. In figure 7 A, we have plotted the ratio of N' e against the ratio of ke:kedc. On the basis of experimental data, we have taken the kede to be equal to k de. It can be seen that the relative composition of bile is not a good index of the relative rate of synthesis when the fractional turnovers differ appreciably. (One particular case of some interest... (:> EO EO q,"... 0;... (:> Meal Meal Meal t t t Synthesis = 0.3 mmol/hr 5 fabs = Time of day FIG. 6. Pool size in relation to time in a simulation of bile acid malabsorption (fa., 0.76). The pool becomes depleted during the day and repleted during overnight fasting. is when Ne = 0.5, ke = kede = k de, and fdehydrfrx = 1. Under (7) those conditions N'e = N'ede = N'de = 0.33.) Variations of fdehydrox' Figure 7 B demonstrates how increasing fdehydrox may reduce N'e in bile. Simply stated, the greater the amount of cholic acid that is dehydroxylated and reabsorbed, the smaller the proportion of cholic acid in bile. However, again the special case in the previous paragraph should be noted. In that case, one-half of the bile acid synthesized in the liver was cholic acid, but two-thirds of the bile acid in bile was derived from that cholic acid synthesis. An examination of bile composition alone may be very mi!>leading. Amino Acid Moiety The mass of unconjugated bile acids entering the hepatocyte, from synthesis, feeding, or intestinal absorption, determines the amount of amino acids required for conjugation; in health, all bile acids secreted into bile are conjugated. The only influence of amino acid metabolism on bile acid conjugation is related to the pattern of conjugation, -" " f dehydrox = 0.3 k cdc = kde 0.5 ' " ; : : : - = ' - - : : :.. : : - - '.. L.. J L L : ' Ne = 0.5 kde -kede ke/kede ke/kde f deh1drox FIG. 7. A, the effect of variable ratio of the fractional turnover rate of cholic acid (k e ) to the fractional turnover rate of chenodeoxycholic acid (kede) on the proportion of cholic acid in biliary bile acids (N'e). The lines indicate varying ratios of cholic acid and chenic acid synthesis by the liver, i.e., for Ne 0.9,90% of bile acid synthesis is composed of cholic acid. B, the effect Of a variable fractional turnover rate for cholic acid (k e ) in relation to that of deoxycholic acid (kde) on the proportion of deoxycholic acid in bile (N'de) for varying degrees of fdehydrox. In this computer simulation, hepatic synthesis of cholic acid and chenic acid is assumed to be equal (Ne 0.5) and the fractional turnover rate of deoxycholic acid and chenodeoxycholic acid are assumed to be identical

5 January 1977 PERTURBED BILE ACID CIRCULATION EQUATIONS 145 because the greater the size of the readily exchangeable taurine pool, the greater the proportion of bile acids conjugated with taurine. 3, 10, 11 Variable Input into the Pool An expression may be derived for the total daily molar reguirement of amino acid for the conjugation of bile acid: Variation in tabs' In figure 8 we have plotted the requirement for acid utilization-bile acid conjugation as the efficiency of intestinal conservation of the steroid moiety of bile acids was varied. We have assumed that the liver has a maximal synthesis capacity for bile acid but that there is no limit to the ability of the liver to conjugate free bile acids. It can be seen that as tabs declines, the amino acid requirement for conjugation rises to a maximum, but then falls as daily de novo synthesis reaches its maximum, and the amount of unconjugated steroid actually presented to the hepatocyte for conjugation-cp tabs (1 - Nconj)-declines. Variation in Neon)' For any given value of P and tabs, the amino acid requirement for conjugation is a linear., N conj = !! <:> 10.0 ",- " 7.5 '" <:>. 5.0 'I: 2.5 """""'"''... de novo r t h e s i s " -'.. :-: -. - R e c o n j u g a... t. i o n O.BO fabs FIG. 8. Effect of variable intestinal conservation (fabs) (abscissa) on the requirement for total amino acid conjugation (ordinate). In this simulation, N' conj (the mass of bile acids absorbed after deconjugation divided by the total mass secreted (in conjugated form» has been taken as 0.8. function of N conj : the lower the value of N conj, the greater the requirement of amino acid for reconjugation. Steady State Pool Composition: the Glycine-Taurine Ratio In the previous paper,4 we defined a fraction N I which is the mole fraction of glycine in the daily a m i acids (glycine and taurine) utilized for bile acid conjuga (9) tion. This fraction, together with the daily fractional where N conj is the mole fraction of that bile acid turnover rate for glycine and taurine, could be used to returning to the liver which is in the conjugated form. derive an expression for N'gIY, the mole fraction of This expression is the sum of the two terms, one of glycine conjugated bile acids in bile. Taurine adminiswhich was the amino acid requirement to conjugate de tration causes a decrease in N g1y, which in turn results in novo synthesis, given by CP (1 - tabs), and the second a decrease in N'gIY' the molar requirement of amino acid to conjugate free We have chosen to use N'glY rather than the more steroid moiety returning from the intestine after bacte- familiar analogue, the glycine-taurine ratio, for the rial deconjugation, given by CP tabs (1 _ N )' conj following reasons. First, the mole fraction gives a direct Clearly, the daily amino acid requirement, given that indication of the fraction of bile acids conjugated with a C and P are constant, depends on the values of tabs and given amino acid. Second, the mole fraction facilitates N ' conj We propose to examine the effect of each in turn algebraic computations, and third, the glycine-taurine on the total amino acid requirement and on the relative ratio is a poor term because the smaller value is usually requirements for conjugation of bile acid from each of its in the denominator, tending to magnify differences: i.e., two possible origins. a marked increase in the glycine-taurine ratio indicates only a small increase in N'gIY; e.g., an increase in the Variable Loss trom the Pool glycine-taurine ratio from 5 to 20 indicates a change in the mole fraction of only 13%, from 0.83 to We previously derived the following equation 4: (10) where kg1y and ktau were the fractional turnover rates of the amino acid moieties of a given bile acid. In figure 9, N'glY and the glycine-taurine ratio are shown as functions of kgly/ktau for varying values of N g1y. Clearly, the relative composition of the amino acid moiety of the conjugated bile acids in bile is the result of a hepatic process represented by N g1y and an intestinal process kgly/ktau. The multicompartmental model of the bile acid pools described previously allows us to examine the distribution of bile acid label between the two forms of amino acid conjugation. In figure 10, it is assumed that the <: '= -"- 1.0 " 04 '" 0.2 kgly/k'au a FIG. 9. The effect of a varying ratio of the fractional turnover rate of the glycine moiety (k gly ) to the fractional turnover rate of the taurine moiety (k tau ) on the proportion of biliary bile acids conjugated with glycine (N'gIY) (left ordinate) or the glycine-taurine ratio of biliary bile acids (right ordinate) for variable hepatic conjugation with glycine (N gly ). The values for N gly are indicated. \!)

6 146 HOFFMAN AND HOFMANN Vol. 72, No. 1 label was introduced as the free steroid moiety. The mole fraction of bile acids conjugated with glycine in bile (N' gly) is taken as 0.75, and the mole fraction of glycine used for bile acid synthesis is N g1y set at 0.89 or The figure shows how N' gly for label varies with time and how it may differ from N' gly mass when k g1y does not equal k tau. Discussion In the discussion, we shall relate our simulations to experimental data. Variable input to the pool. Ingestion of the major primary bile acids has consistent and predictable effects on biliary bile acid composition which are summarized in figure 11. When chenic ll - 2o or deoxycholic acid is fed to human subjects, the proportion of the respective bile acids increases markedly in association with an b 0.60 i Time, days FIG. 10. Time course of mode of conjugation in bile injected into the bile acid as the un conjugated steroid moiety when hepatic conjugation is 80% with glycine (Ng,y 0.80) or 89% with glycine (Ng,y 0.89). The figure shows that the glycine-taurine ratio of bile may not be identical to the proportion of bile acids conjugated with glycine or taurine in the liver. The shaded ribbon indicates the usual N'g,y for biliary bile acids (about 0.75). 4 Cholic. % FIG. 11. Effect of bile acid feeding on biliary bile acid composition, shown using triangular coordinates. Data shown are those of LaRusso et al. for gallstone patients ingesting cholic or chenic acid 15 or healthy volunteers ingesting deoxycholic acid. 22 However, similar results have been obtained by others after feeding of chenodeoxycholic acid,12-19 cholic acid,'2. 15 or deoxycholic acid.22 expansion of the pool of that bile acid. Expansion of the pool of the administered bile acid is moderate by equation 1b, but this equation deals only with the administered bile acid. When a bile acid is administered, synthesis of both primary bile acids is inhibited, based on isotope dilution measurements, and in addition, the fractional turnover rate of all bile acids increases. Thus, biliary bile acid reflects increased input of the administered bile acid and not only decreased input of other bile acids, but also increased turnover. When cholic acid is fed, biliary bile acids become comprised of equal parts of cholic acid and -its bacterial metabolite, deoxycholic acid The reason for the disproportionate increase in the deoxycholic acid proportion in bile is unclear. The simplest explanation is that the increased amount of cholic acid entering the colon facilitates the retention of deoxycholic acid. The fractional turnover rate of deoxycholic acid is always less than that of cholic, so that the input of deoxycholic acid can be considerably less than that of cholic, yet they can have equal proportions in bile (fig. 7A). A second exception is hyodeoxycholic acid. When this bile acid is fed, its proportion in biliary bile acids does not increase. 12 The reason for this is unclear. The development in equations 2a, 3, and 4 assumes that expansion of the bile acid pool is limited. by maximal intestinal absorptive capacity. Two lines of indirect evidence support this assumption: (1) with bile acid feeding, there is only a modest increase in the total bile acid pool, 15 whereas efficiency of intestinal conservation falls, indicated by the higher k values measured; 13 and (2) patients with small intestinal bypass have a smaller bile acid pool despite an increased input of bile acids into the pool owing to increased hepatic synthesis. 24 However, gallbladder contractibility may also control the pool size, because vagotomy causes a greatly expanded bile acid pool size in dogs. 25 If pool size is the dependent variable, that is, secretion is held constant, then the recycling frequency must vary inversely with pool size. Good experimental evidence for this has recently been published. 15 Variable tabs, Figure 4 confirms the prediction that increased bile acid synthesis can compensate for mild degrees of bile acid malabsorption, but that secretion will fall when increased hepatic synthesis cannot compensate for decreased absorption. The secretion curve will eventually meet the synthesis curve at the limiting case, a total biliary fistula, that is, when tabs becomes zero. We also showed that bile acid secretion should decline progressively during the day in patients with bile acid malabsorption. Indeed, in patients with ileal resection and documented bile acid malabsorption, measurement of intraluminal concentrations of bile acids during digestion,2& as well as radioimmunoassay of serum cholates,27 confirms this prediction. Amino acid moiety. As discussed, the partition of amino acid conjugation is related to N g1y (which in turn appears related to the size of the rapidly exchangeable

7 January 1977 PERTURBED BILE ACID CIRCULATION EQUATIONS 147 taurine pool) and the relative rates of intestinal conservation (which varies reciprocally with deconjugation). Therefore, for a defined state of taurine metabolism, the N' gly gives indirect information on relative intestinal conservation, which is related to the site of absorption and bacterial deconjugation. The greater the degree of bile acid malabsorption, the greater will N' gly resemble N g1y ; i.e., bile will reflect hepatic events. If this argument is correct, the increased N g1y of patients with ileal disease could reflect bile acid malabsorption or bile acid deconjugation, and need not imply any abnormality in taurine metabolism. Indeed, patients with small intestinal bypass have recently been shown to have normal serum taurine levels. 31 Why N'gIY is decreased in liver disease 32 remains unclear, and neither is it clear why the fraction falls in myxedema. 33 Both may reflect abnormalities in taurine metabolism, about which relatively little is known in man. Conclusions The over-all conclusions from these calculations is to confirm an assumption implicit in these simulations, i.e., that the size of the exchangeable bile acid pool for any circulating bile acid (steroid moiety) is related to the balance between input-from hepatic or bacterial formation-and intestinal loss. We are not the first to reach this conclusion, although the present paper provides a more detailed theoretical basis for what appears virtually axiomatic. Both synthesis and the efficiency of intestinal absorption may vary by an order of magnitude. Input can be varied by increased hepatic synthesis, by increased intestinal absorption of secondary bile acids, or by bile acid feeding. Intestinal conservation varies from very efficient absorption in health to total loss in the bile fistula patient. The size of the bile acid pool for any given bile acid reflects the dynamic equilibrium between input and loss, and gives no information on the absolute rate of either process. Measure of biliary bile acid composition alone, i.e., the proportion of individual bile acids, is even more ambiguous, and not even relative rates of synthesis or intestinal conservation can be inferred. For the amino acid moiety, these calculations also reinforce our previous conclusion, namely that the amino acid composition of biliary bile acids does not necessarily reflect the proportion of amino acids used for conjugation (input) but rather, as with the steroid moiety, the balance between input and loss. The present paper and our initial treatment did not consider the metabolism of lithocholic acid, its conjugation with glycine or taurine, or its subsequent sulfation. 34, 35 A multicompartmental model has been prepared, but insufficient information is available on lithocholic acid metabolism in man from which to infer rate constants, which in turn could be used to predict the proportion of the various lithocholic acid species in bile. Bile acid kinetics appear to belong to the area of pharmacokinetics relating to maintenance blood levels, but bile acid kinetics differ in at least two important ways: first, one is concerned with the total body load (pool size) rather than blood or tissue concentration, and second, one is concerned with the relative amounts of three or four bile acids, whereas most pharmacokinetic treatments deal with only a single drug. REFERENCES 1. Hepner GW, Hofmann AF, Thomas PJ: Metabolism of steroid and amino acid moieties of conjugated bile acids in man. 1. Cholyl glycine (glycocholic acid). J Clin Invest 51: , Hepner GW, Hofmann AF, Thomas PJ: Metabolism of steroid and amino acid moieties of conjugated bile acids in man. II. Glycine conjugated dihydroxy acids. J Clin Invest 51: , Hepner GW, Sturman JA, Hofmann AF, et al: Metabolism of steroid and amino acid moieties of conjugated bile acids in man. III. Cholyl taurine (taurocholic acid). J Clin Invest 52: , Hoffman NE, Hofmann AF: Metabolism of steroid and amino acid moieties of conjugated bile acids in man. IV. Description and validation of a multicompartment model. Gastroenterology 67: , Dowling RH, Mack E, Small DM: Effects of controlled interrup; tion of the enterohepatic circulation of bile salts by biliary diversion and by ileal resection on bile salt secretion, synthesis, and pool size in the Rhesus monkey. J Clin Invest 49: , Small DM, Dowling RH, Redinger RN: The enterohepatic circulation of bile salts. Arch Intern Med 130: , Lindstedt S: The turnover of cholic acid in man. Acta Physiol Scand 40:1-9, Hofmann AF, Hoffman N Measurement of bile acid kinetics by isotope dilution in man. Gastroenterology 67: , Hofmann AF: The syndrome of ileal disease and the broken enterohepatic circulation: cholerheic enteropathy. Gastroenterology 52: , Sturman JA, Hepner GW, Hofmann AF, et al: Metabolism of C"S) taurine in man. J Nutr 105: , Schersten T: Bile acid conjugation. In Metabolic Conjugation and Metabolic Hydrolysis, vol 2. Edited by WH Fishman. New York, Academic Press, 1970, p Thistle JL, Schoenfield LJ. Induced alterations in composition of bile of persons having cholelithiasis. Gastroenterology 61: , Danzinger RG, Hofmann AF, Schoenfield LJ, et al: Effect of oral chenodeoxycholic acid on bile acid kinetics and biliary lipid composition in women with cholelithiasis. J Clin Invest 52: , Northfield TC, LaRusso NF, Hofmann AF, et al: Biliary lipid output during three meals and an overnight fast. II. Effect of chenodeoxycholic acid treatment in gallstone subjects. Gut 16:12-17, LaRusso NF, Hoffman NE, Hofmann AF, et al: Effect of primary bile acid ingestion on bile acid metabolism and biliary lipid secretion in gallstone patients. Gastroenterology 69: , Coyne MJ, Bonorris GG, Chung A, et al: Treatment of gallstones ith chenodeoxycholic acid and phenobarbital. N Engl J Med 292: , Adler RD, Bennion LJ, Duane WC, et al: Effects of low dose chenodeoxycholic acid feeding on biliary lipid metabolism. Gastroenterology 68: , James 0, Cullen J, Bouchier lad: Chenodeoxycholic acid therapy for gallstones: effectiveness, toxicity and influence on bile acid metabolism. Q J Med 44: , Kallner M: The effect of chenodeoxycholic acid feeding on bile acid kinetics and fecal neutral steroid excretion in patients with hyperlipoproteinemia types II and IV. J Lab Clin Med 86: , 1975

8 148 HOFFMAN AND HOFMANN Vol. 72, No Stiehl A, Raedsch R, Regula M, et al: Treatment of patients with cholesterol gallstones with chenodeoxycholic acid. Alterations in the bile acid metabolism. Inn Med 2:13-18, Pomare EW, Low-Beer TS: The selective inhibition of chenodeoxycholate synthesis by cholate metabolites in man. ClinSci Mol Med 48: , LaRusso NF, Szczepanik PA, Hofmann AF, et al: Effect of deoxycholic acid ingestion on bile acid metabolism and biliary lipid secretion in. normal subjects. Gastroenterology 72: , Dietschy JM, Salomon HS, Siperstein MD: Bile acid metabolism. I. Studies on the mechflj;ljsms of intestinal transport. J Clin Invest 45: , Wise L, Stein T: Cholic and chenodeoxycholic acid metabolism in massively obese. patients before and after small bowel bypass (abstr). Gastroenterology 68:1014, White TT, Tournut RA, Scharplatz D, et al: Effect of vagotomy on biliary secretion and bile salt pools in dogs. Ann Surg 179: , Poley JR, Hofmann AF: Role of fat maldigestion in pathogenesis of steatorrhea in ileal resection. Fat digestion after two sequential test meals with and without cholestyramine. Gastroenterology 71:38-44, LaRusso N F, Korman MG, Hoffman NE, et al: Dynamics of the enterqhepatic circulation of bile acids. N Engl J Med 2 9 l : , McLeod GM, Wiggins HS: Bile salts in small intestinal contents after ileal resection and in other malabsorption syndromes. Lancet 1: , Garbutt JT, Heaton KW, Lack L, et al: Increased ratio of glycine to taurine-conjugated bile salts in patients with ileal disorders. Gastroenterology 56: , Bruusgaard A, Thaysen EH: Increase,d ratio of glycine taurine conjugated bile acids in the early diagnosis of terminal ileopathy. Acta Med Scand 188: , Sherr HW, Nair PP, White JJ, et al: Bile acid metabolism and hepatic disease following small bowel bypass for obesity. Am J Clin Nutr 2 7 : , 32. Sjovall J: Bile acids in man under normal a'nd pathological conditions: bile acids and steroids 73. Clin Chim Acta 5:33-41, Hellstrom K, Sjovall J: Conjugation of bile acids in patients with hypothyroidism. J Atheroscler Res 1: , Cowen AE, Korman MG, Hofmann AF, et al: Metabolism of lithocholate in man, I. Biotransformation and biliary excretion of intravenously administered lithocholate, lithocholylglycine, and their sulfates. Gastroenterology 69:59-66, Cowen AE, Korman MG, Hofmann AF, et al: Metabolism of lithocholate in man, II. Enterohepatic circulation. Gastroenterology 69:67-76, 1975