sign in sign up
<<
Home , Heme

THE FORMATION OF FROM IN VIVO

J. Donald Ostrow, … , James H. Jandi, Rudi Schmid

J Clin Invest. 1962;41(8):1628-1637. https://doi.org/10.1172/JCI104620.

Research Article

Find the latest version: https://jci.me/104620/pdf Journal of Clinical Investigation Vol. 41, No. 8, 1962 THE FORMATION OF BILIRUBIN FROM HEMOGLOBIN IN VIVO * By J. DONALD OSTROW,t JAMES H. JANDL AND RUDI SCHMID (From the Thorndikc Memorial Laboratory and the Second and Fourth [Harvard] Medical Services, Boston City Hospital, and the Departmcntt of Medicine, Harvard Medical School, Boston, Mass.)

(Submitted for publication March 7, 1962; accepted March 30, 1962)

About 80 to 85 per cent of the fecal pig- hepatic circulation of the bilirubin contained in mients are believed to be derived from the hemo- the administered bile (8). of senescent erythrocytes that are destroyed In the present investigation, these difficulties in the reticuloendothelial system (1). It is not have been surmounted by injecting C"4-labeled, known, however, whether the hemoglobin of se- antibody-sensitized red cells into rats with an ex- questered red cells is converted quantitatively to ternal bile fistula. This permitted both determina- bilirubin and eventually to , or is in tion of the excretory rate of bilirubin-C14 in the part metabolized to other products (2, 3). bile and estimation of the efficiency of conversion Earlier studies of this problem in man have re- of the injected hemoglobin to bilirubin, without vealed that the rate of fecal urobilinogen excretion regard for variations in endogenous bilirubin ex- falls short of the expected output as calculated cretion. Since the injected red cells were also la- from total red cell volume and mean erythrocyte beled with y-emitting Fe"s or Cr51, it was possible life span (4-6). This has been attributed to to follow their rate of removal from the circulation, incomplete intestinal conversion of bilirubin to and to determine the interval between sequestration urobilinogen (3, 4, 6, 7), destruction of urobili- of the cells and excretion of the resulting bilirubin- nogen in the large bowel (3, 4, 6), and loss of pig- C14 in the bile. ment through intestinal absorption (3, 4, 6, 8). At- Similar experiments were carried out with in- tempts to circumvent these difficulties by estimat- travenously injected solutions of hemoglobin la- ing bilirubin excretion through a T-tube in the beled with C14 and either Fe59 or Cr51. Urinary common bile duct have been unsatisfactory (9, loss of pigment was minimized or prevented by ad- 10) because this technique does not permit com- ministering only small amounts of hemoglobin or, plete bile collection. In animals with an external in some instances, by nephrectomizing the rats biliary fistula, this obstacle has been overcome, but prior to the experiment. the rate of endogenous bilirubin excretion was found to exhibit marked fluctuations (11, 12), MATERIALS AND METHODS which rendered difficult comparison of the ob- Preparation of red cells and hemioglobin labeled with served values with the endogenous production of C" and Fe59. Male Sprague-Dawley rats weighing ap- bilirubin calculated from hemoglobin turnover, or proximately 300 g were given, at 48-hour intervals, three with excess bilirubin formation after intravenous subcutaneous injections of 20 mg phenylhydrazine hydro- in neutralized 2 per cent aqueous solution. Be- injection of a hemoglobin load. In animals with ginning three days after the last phenylhydrazine injec- chronic bile fistulae, these acute fluctuations in pig- tion, 0.2 mc -2-C" was administered subcutaneously ment output tend to balance out, but the necessity twice daily to a total dose of 2.0 mc.1 Over the same of giving oral bile supplements to replace drainage period, a total dose of 100 Asc transferrin-bound Fe"Cl losses (11, 13) results in falsely high pigment ex- (14) was injected into a tail vein in three to five separate daily doses.2 Two weeks after the last isotope adminis- cretion due to intestinal absorption and entero- tration, the rats were exsanguinated with heparinized syringes, and the red cells were separated and washed * in Health Supported part by United States Public with cold isotonic saline. Service Grants A-1833 and RG-3507. This work was presented in part before the American Society for Clini- 1 Specific activity of glycine-2-C04, 17 to 20 mc per cal Investigation, Atlantic City, N. J., April 30, 1961. mmole; New England Nuclear Corp., Boston, Mass. t Postdoctoral Research Fellow, National Institutes of 2 Specific activity of Fe"Cl,, 3 to 10 ,tc per leg; Abbott Health, Bethesda, Md. Laboratories, Chicago, Ill. 1628 FORMATION OF BILIRUBIN FROM HEMOGLOBIN 1629

An aliquot of these red cells was sensitized in vitro or by incubation with heated rabbit serum containing "in- (dpm perj~g) X 0.976 = bilirubini (dpm per Mg). complete" antibodies against rat erythrocytes (15) Af- ter these sensitized red cells were washed and resuspended Clearance of injectted labeled red cells or hemizoglobinl in isotonic saline, they showed slight agglutination in front the circilationi. In 16 male Sprague-Dawley rats, saline, but strong agglutination in an appropriate Coombs weighing 450 to 600 g, an external bile fistula was pro- serum (14) and in polyvinylpyrrolidone (16). On incu- duced (20) and the animals were then placed in loose re- bation for 2 hours at 370 C in a large volume of fresh straining cages. Bile flow was maintained in excess of rat serum, no occurred. 0.7 ml per hour by appropriate hydration (21). After Solutions of labeled hemoglobin were prepared by collecting bile for 3 hours, the rats were briefly anesthe- lysing an aliquot of donor red cells in cold distilled wa- tized with ether and 1.6 to 2.0 ml of a suspension of the ter. After restoring isotonicity with 5 per cent saline, labeled red cells, or of the hemoglobin solution, xvas in- the stroma was removed by centrifugation in the cold for jected into the tail vein. From the total amount of he- 1 hour at 105,000 G. The clear supernatant was then ad- moglobin administered (Table I), the expected quantity justed to the desired hemoglobin concentration and to pH of bilirubin was calculated from the conversion ratio 1 g 7.8 by addition of appropriate volumes of isotonic saline hemoglobin = 34.5 mg bilirubin (22). and 0.15 M phosphate buffer, pH 8.1. The slightly alka- The rate of removal of the administered red cells or line pH was employed to obviate the slow, spontaneous hemoglobin from the circulation was determined by precipitation that occurred with rat hemoglobin in aque- following the gamma activity of serial samples col- ous solution. lected by incision of the tail. One-tenth ml blood was Isolation and radioassay of labeled heinin. From a pipetted into 1 ml water for determination of whole blood third aliquot of the C"-, Fe'-labeled red cells, hemin was radioactivity, and into 1 ml saline for measurement of isolated and crystallized (17). After recrystallization plasma radioactivity after removal of red cells by centrif- (18), the hemin was dissolved in pyridine and the con- ugation (15). The initial concentration of the inj ected in on centration estimated spectrophotometrically, with a stand- material the blood was estimated the basis of the predicted blood volume, which for male rats of this size ard curve prepared by dissolving weighed amounts of hemin in similar fashion. The pyridine solutions were was assumed to be 4.5 per cent of body weight (15). The rats were killed by cervical dislocation 22 hours then diluted with 1 M Hyamine-methanol 3 to yield a after the injection of the labeled heme pigment. The , final concentration of 10 Ag labeled hemin per ml. For 1 to 4 labeled hemin and 20 unlabeled car- , kidneys, lungs, and right femur were removed, counting, Ag fig 5' rier hemin, dissolved in a total of 1 ml of 1 M Hyamine- blotted with damp gauze, weighed, and the Fe activity determined on specimens not exceeding 2 g in weight methanol, were mixed with 17 ml scintillator solution 4 (14). Total Fe" in the marrow was roughly estimated in a 20 ml low potassium glass counting vial.5 The vials by multiplying the femur radioactivity by 14.4 (23 ). were counted in a scintillation liquid spectrometers using and bile Fe"' or Cr5' radioactivity were determined as the internal standard.7 After 6 to 8 toluene-C" months, on the total 22-hour specimens. had and the ac- most of the Fe" activity decayed, specific Because of the rapid reutilization and redistribution of of hemin a constant value attributable tivity the approached from the metabolized hemoglobin (24, 25), the or- to Absence of detectable emission due solely C". gamma gan content of Fe"' after 22 hours is not a valid meas- to remaining Fe"' was further ascertained in a well-type ure of the initial distribution of injected heme. scintillation counter. Allowing for the difference in mo- Therefore six of the animals (Table I) were injected with lecular weights and for the loss of a labeled methene-bridge sensitized red cells or hemoglobin solution labeled with carbon atom (19), the C"-specific activity of the biliru- Cr5, which is released more slowly from the tissues (25, bin derived from the injected hemoglobin-C' was calcu- 26). C`-labeled red cells, prepared as described above, lated as follows: were labeled imt vitro with Na..Cr5"04 (15) prior to in- cubation with antiserum.9 Hemoglobin was prepared sp. act. of hemin X 584.7 X 7 = sp. act. of bilirubin from these cells as described. Crystalline hemin-C" de- rived from these cells contained no Cr5', as determined in 3 Hydroxide of Hyamine 1OX, Packard Instrument Co., a well-type scintillation counter. Blood and tissue sam- La Grange, Ill. ples from these rats were processed as above. 4 2,5-Diphenyloxazole, 4 g per L, and 1,4-bis-2 (5-phenyl- Rats R and S (Table I) received injections of hemo- oxazolyl)-benzene, 50 mg per L, dissolved in dry toluene. globin solutions labeled only with C", which were pre- Prepared from Liquifluor, Pilot Chemical Co., Water- pared as above except that the donor rats were not given town, Mass. Fe'. Rats J and K were bilaterally nephrectomized im- Catalog no. 6001015, Packard Instrument Co., La mediately before insertion of the biliary catheter. Grange, Ill. IDeterinination of bilirtibini-C` excretion. All bile ex- 6 Tri-Carb model 314-X, Packard Instrument Co., La creted during the 22 hours of the experiment was collected Grange, Ill. Toluene-C", lot no. 61-12, New England Nuclear 8Specific activity of Na2Cr"04, 300 to 500 gc per Aug; Corp., Boston, Mass. Abbott Laboratories, N. Chicago, Ill. 1630 J. DONALD OSTROW, JAMES H. JANDL AND RUDI SCHMID over consecutive periods ranging from 30 to 120 minutes. load of unconjugated bilirubin were studied by injecting The collection tubes were kept on ice and protected from a 500-g male rat with crystalline bilirubin-C" (21). Bili- light, and then stored at - 20° C until analyzed. Under rubin-C", 0.93 mg with a specific activity of 168 dpm per these conditions, the bilirubin does not deteriorate over Asg, was dissolved in 0.2 ml 0.1 N NaOH, diluted to 1.0 periods of several weeks (21). For analysis, the indi- ml with 0.1 M phosphate buffer, pH 7.4, and slowly added vidual samples were thawed, the volume recorded, and to 2.5 ml pooled normal rat serum. Three ml of this solu- the bilirubin content determined by the diazo method tion, containing 0.80 mg bilirubin-C", was injected into (27). From each sample, bilirubin was crystallized (21) the femoral vein of an anesthetized rat with an external and the C" specific activity was determined (21). Ab- bile fistula. Bile was collected in 4- to 6-minute frac- sence of Fe'9 or Cr" activity in the crystalline pigment tions for the first half hour, and in 15- to 30-minute frac- was ascertained in the well-type scintillation counter. tions thereafter. Each bile sample was diluted with water For each collection period, total excretion of C" activity to a volume of 1.0 ml and then divided into two equal in bilirubin was calculated from the concentration and portions. One-half was used for determination of biliru- specific activity of the excreted bilirubin. The quantity bin content and total radioactivity. The other halves of of bilirubin derived from the injected hemoglobin-C" was the bile samples were pooled in such a way that each obtained by dividing the total C" activity of bilirubin in pool represented one-half of the bile excreted during each each sample by the calculated specific activity of the bili- hourly collection period. From each pool, bilirubin was rubin-C" derived from the injected heme-C". Excretion crystallized and its specific activity determined. of unlabeled bilirubin derived from endogenous sources was calculated as the difference between the total bili- RESULTS rubin output and the excretion of bilirubin-C". In rats R, S, T, U, V, and W (Table I), these determinations Rate of heme clearance and of bilirubin excre- were performed only on pooled bile samples collected over tion. For each experimental animal, data concern- the entire 22-hour period. ing the nature and quantity of the administered Related experiments. The fluctuations of endogenous bilirubin excretion under the conditions of this experi- hemoglobin load are given in Table I. In all in- ment were determined in three rats with a bile fistula to stances, cumulative curves of removal of labeled which no exogenous pigment load was given. Bilirubin red cells or hemoglobin from the blood, and of ex- excretion was measured during individual collection pe- cretion of labeled bilirubin in the bile, revealed riods ranging from 30 to 60 minutes, and the effect of similar patterns (Figure 1). Clearance of ad- anesthesia and surgery on pigment output was determined. The rate and efficiency of excretion of an intravenous ministered heme pigment is expressed as per cent

TABLE I Bilirubin-C"4 excretion in rat bite after administration of C'4-labeled heme pigment

Dose Isotopic injected (as Bilirubin- At5ot Bilirubin-C"4 Material injected Rat label hemoglobin) Heme-tbo* t ot recovered§ mg/kg min min min % Sensitized red cells A Cr5l, C14 245 54 237 183 72.1 B Crl C14 188 75 345 270 69.9 C Fe59, C14 57.1 43 240 197 66.4 D Fe'9, C14 63.3 13 185 172 63.3 T Crl, C'4 3.8 79.6 U Cr"l, C"4 3.5 65.0 Hemoglobin solution E Fe59, C'4 60.2 15 185 170 63.4 F Fe69, C'4 51.3 21 172 151 55.4 R C'4 62.5 58.3 P Fe69, C'4 11.7 16 170 154 98.6 Q Fe59, C'4 12.7 7 125 118 98.7 S C14 12.5 97.5 V Cr5l, C'4 4.2 98.6 W Cr5l, C'4 3.9 76.1 Jill Fe59, C14 55.3 45 260 215 69.7 K11 Fe59, C'4 50.0 77 260 183 62.8 * Heme-t5o = time for half the injected Fe59 or Cr5l to disappear from the circulation. t Bilirubin-t5o = time for half the total amount of bilirubin-C'4 excreted to appear in the bile. $ At5o = difference between heme-t5o and bilirubin-tio. § Recovery of bilirubin-C'4 is expressed as the per cent of the amount of bilirubin-C'4 expected in bile if all the ad- ministered heme pigment were converted to bilirubin. 11 Rats J and K were nephrectomized prior to hemoglobin-C'4 injection. FORMATION OF BILIRUBIN FROM HEMOGLOBIN 1631 SENSITIZED RED CELLS HEMOGLOBIN SOLUTION

I! ~~~~AE 20 / 245 mg/kg - 60mg/kg 100 XFFi

80 - /

Q~ 60 A 150 1so e 1= t,50 40 -43min. 240min. /6mm. 1/70 nmin.

20 57mg/kg A 12 mg/kg

QQ- -I-'-'A'. ,I. ~~ ~80~ 60;T 15,0 15=I-t4 =/0 '-4-/3m/n.7 -a--/5mrnn. ,-45m/n. - 20 n/n. 40 J D NEPHRECTOMIZED 20 63 mg/kg 55 mg/kg

2 4 6 8 10 2 4 6 8 10 HOURS AFTER INJECTION

FIG. 1. CUMULATIVE CURVES OF RED CELL OR HEMOGLOBIN CLEARANCE AND OF BILIRUBIN-C" EXCRETION IN SIX REPRESENTATIVE RATS. Sequestration of heme pigments is expressed in per cent of the injected dose of Cr5` or Fe59. Excretion of bilirubin-C" is expressed in per cent of the total quantity of labeled bilirubin that appeared in the bile during 22 hours. 1632 J. DONALD OSTROW, JAMES H. JANDI, AND RUDI SCHMID of the injected dose of Fe5 or Cr51, whereas bili- heme pigment (heme-t,5), the greater was the in- rubin-C'4 output is plotted as per cent of the total terval (A too) between heme sequestration and ex- amount of isotopic bilirul)in that was excreted in cretion of labeled bilirubin (r = 0.76; p < 0.01) 22 hours. The time required after injection of (28). heme pigment for half the labeled heme to be Figure 2 illustrates absolute rates of bilirubin-C14 cleared from the circulation (heme-t50) and for excretion in four animals, two of which were in- half the labeled bilirubin to be excreted in the bile jected with sensitized red cells (A and C) and two (bilirubin-t5o) was determined from these curves with hemoglobin solution (E and P). As ex- (Table I). Since the curves are not exponential, pected, the rate of bilirubin-C14 excretion de- these t50 values do not represent mathematically pended upon the quantity of heme pigment ad- derived half-lives. m-iinistered, the maximum being reached between Sequestration of sensitized red cells began 2 and 3 hours after heme injection. Since virtually within a few minutes after their injection, with 50 all injected red cells or hemoglobin had been per cent of the cells being removed in 15 to 75 cleared from the circulation during the period of minutes, and 70 to 80 per cent within 2 hours 22 hours, and since, in most instances, excretion (Figure 1; A, C, and D). Excretion of biliru- of bilirubin-C'4 was thereafter no longer detectable, bin-C'4 began within 30 to 50 minutes after the on- it was possible to compare directly the total amount set of sequestration, after 1 hour reached a nearly of heme pigment injected with the total amount constant rate lasting 3 to 4 hours, and was almost of labeled bilirubin excreted. completed within 7 to 8 hours. With all doses of Recovery of biliriibin-C'4 from the injected red cells injected, the bilirubin excretion increas- hen e-C14. Total excretion of bilirubin-C'4 during ingly lagged behind erythrocyte sequestration (Fig- the 22 hour period, as per cent of the amount ex- ure 1). At the t50 level, the interval between the pected from the administered heme-C'4, is given two curves (A t5,) ranged from 172 to 270 min- in Table I and in Figure 3. In the six animals utes (Table I). that received different doses of sensitized red cells, After injection of labeled hemoglobin in solu- recovery of bilirubin-C'4 ranged from 63 to 80 per tion, the patterns of heme clearance and bilirubin- cent. Comparable or slightly lower recovery of C14 excretion were similar to those obtained after labeled bilirubin was obtained in rats E, F, and R administration of red cells (Figure 1; E and P). In general, hemoglobin was cleared more rapidly than sensitized red cells (heme-t5, = 7 to 21 min- utes), resulting in an accelerated rate of bilirubin- C14 excretion (A t50 = 118 to 170 minutes) (Table I). As with red cells, excretion of labeled bili- rubin began within 30 to 50 minutes after the he- \t moglobin injection. In the nephrectomized rats th (J and K), as in the intact animals, initial re- moval of hemoglobin was rapid, with approxi- mately 40 per cent of the injected dose being i.4 cleared within 15 minutes (Figure 1). In the ab- sence of the kidneys, however, the remaining he- AFTER INJECTION OF FREE HEMOGLOBIN-C'4 moglobin was cleared more slowly, resulting in de- 3 iHb INJECTED layed excretion of labeled bilirubin (Table I). 2 5- 60 mg/kg With all doses both of sensitized red cells and of A-* /2mg/kg "I hemoglobin solution, irrespective of the presence I - --I ,3, I I V- 0 2 4 6 8 10 12 14 16 I8 20 22 or absence of the kidneys, the rate of bilirubin-C'4 HOURS AFTER INJECTION excretion appeared to depend upon the rate of FIG. 2. ABSOLUTE RATE OF BILIRUBIN-C" EXCRETION heme-C14 sequestration. Moreover, the longer the IN BILE AFTER ADMINISTRATION OF LABELED, SENSITIZED time required for the clearance of the administered RED CELLS OR HEMOGLOBIN SOLUTION TO RATS. FORMATION OF BILIRUBIN FROM HEMOGLOBIN11633

SENSITIZED RED CELLS HEMOGLOBIN SOLUTION (Figure 3), which were injected with an amount of hemoglobin in solution equivalent to that in the red cells given to rats C and D. Nephrectomy did not alter the recovery of bilirubin, as shown by animals J and K, which received the same dose of hemoglobin solution as rats E, F, and R (Figure 3). When smaller quantities of hemoglobin, cal- culated not to exceed the anticipated hemoglobin- binding capacity of rat plasma,9 were administered, recovery of bilirubin-C14 in the bile of 4 out of 5 the values if all in- 63 3.83 60 5163 12 13 13 4.2 animals approached expected DOSE INJECTED (as hemoglobin), mg / Kg jected hemoglobin were quantitatively converted to bilirubin (rats P. Q, S. and V; Figure 3). FIG. 3. RECOVERY OF BILIRUBIN-C14 IN RAT BILE AFTER labeled red cells anzd heimo- ADMINISTRATION OF LABELED, SENSITIZED RED CELLS OR HE- Organ uptake of MOGLOBIN SOLUTION. Results are expressed in per cent globin. Since recovery of Fe59 or Cr51 in the tis- of the amount of labeled bilirubin expected if all the ad- sues of the recipient rats was determined only 22 ministered heme had been converted to bilirubin. Each bar represents one animal, and the number below the bar 9 levels in rat plasma were kindly deter- indicates the dose of administered heme pigment ex- mined by Dr. Eloise Giblett, Seattle, Wash. In two groups pressed as milligrams of hemoglobin per kilogram of of normal rats, mean haptoglobin levels bound 34 and 45 body weight. mg hemoglobin per 100 ml plasma.

4.0

INJECTED DOSE EQUIVALENT INJECTED___|v. AS BILIRUBIN t50

Bilirubin-C'4 2.3mg/kg 420 min. t3.0 )I- |v Hemoglobin-C'4 2.1 mg/kg 3 hrs. ¢ - O C'4-Red Cells 2.0mg/kg 3 hrs. I. zc2.0 86 % of in/ected Si/irubin excreted within one hour 41 . £ _0

W- -~~~~~~~~~~~~~~~~~~~ 0 1 2 3 4 5 6 7 8 9 HOURS AFTER INJECTION FIG. 4. EXCRETION OF LABELED BILIRUBIN IN RAT BILE AFTER INTRAVENOUS INJECTION OF EQUIVALENT DOSES OF UNCONJUGATED BILIRUBIN-C", HEMOGLOBIN-C1 IN SOLUTION, AND SENSITIZED C -LABELED RED CELLS. The excretory rate of bilirubin-C"4 is expressed in per cent of the dose of pigment ad- ministered.T _ _ E | Y 16.34 J. DONALD OSTROW, JAMES H. JANDL AND RUDI SCHMID hours after injection of the labeled heme pigment, rate of bilirubin excretion after bile duct catheteri- the data are of limited value and will not be given zation first rose and then declined before leveling in detail. As expected (15, 29), virtually all sen- off at a constant rate after approximately 8 hours. sitized red cells were sequestered in approximately Pentobarbital-ether anesthesia sharply reduced equal proportions by spleen and liver, with no sig- bile volume and pigment output (12, 33), followed nificant uptake of Cr51 or Fe59 by the kidneys nor by a rebound which lasted for 4 hours after cessa- significant excretion of radioactivity in the urine. tion of ether administration. Injection of 20 mg After injection of comparable quantities of hemo- pentobarbital per kg alone did not affect bile and globin solution, clearance from the circulation also pigment excretion. was virtually complete, but the tissue distribution In all rats intravenous injection of labeled red of radioactivity was different (23, 30). Hepatic cells or hemoglobin solution was followed by a uptake greatly exceeded splenic uptake, which sharp rise and subsequent fall in the excretion of was never more than 3 per cent of the injected unlabeled endogenous bilirubin in rough propor- dose. Moreover, approximately half of the in- tion to the amount of administered heme. The jected Fe59 or Cr51 was recovered in the kidneys magnitude of this effect exceeded the variations in and urine, suggesting that the total plasma hemo- bilirubin excretion which resulted from anesthesia globin-binding capacity may have been exceeded alone. As outlined under Methods, this did not even with the smallest dose of hemoglobin adminis- interfere with the calculations involved in the tered (30, 31). However, estimation of urinary present experiments, since the C'4-label permitted hemoglobin excretion, by a quantitative benzidine identification of all bile pigment derived from the technique (32), revealed that less than 2 mg he- injected labeled heme. moglobin was lost in the urine, corresponding to less than 4 per cent of the largest dose of hemo- DISCUSSION globin injected. Thus, most of the urinary Fe59 must have been excreted after splitting of the heme In the present experiments rats with a bile fis- , leaving the moiety to be me- tula were given an intravenous injection of one of tabolized to bilirubin and excreted in the bile. the following heme pigment preparations: 1) sus- Related findings. After intravenous adminis- pensions of antibody-sensitized red cells which tration of -bound, unconjugated bilirubin- were primarily sequestered in the spleen and liver C14, labeled pigment appeared in the bile within 3 (15, 29); or 2) various amounts of hemoglobin in minutes (Figure 4). Of the injected pigment, solution, which were predominantly removed by 50 per cent was excreted as conjugated bilirubin the liver and the kidneys (23, 30, 31). In spite of in 20 minutes, 86 per cent in 1 hour, and 96 per these differences, the labeled heme pigment was cent in 4 hours. cleared from the circulation and excreted as radio- In the three rats that were not injected with active bilirubin in the bile at similar rates. This exogenous hemoglobin or red cells (Figure 5), the indicates that the antibody-sensitized erythrocytes were lysed almost immediately after their seques- tration. It further suggests that the reticuloendo- 4 AfATER BILE DUcr AFTRE NEMBAUAL- thelial cells of the liver, the spleen, and perhaps CANNULArION ETER ANAESTES the kidneys, are all capable of converting hemo- 3.0 3.0 globin to bilirubin.

2.0.R 2.0 Although a small portion of the injected heme- I.'Q4 a-I. C14 was converted to labeled bile pigment within 1.0 30 minutes (33, 34), a rapid, nearly constant rate of excretion of bilirubin-C'4 was not achieved until 0 2 4 6 8 20 22 24 26 28 30 32 approximately 50 minutes after the onset of se- HOURS AFTER OPERATION questration of the administered hemoglobin or red FIG. 5. RATES OF BILIRUBIN EXCRETION IN A 420 G cells This in bile MALE SPRAGUE-DAWLEY RAT AFTER BILE DUCT CATHETERI- (Figure 1). delay pigment for- ZATION UNDER PENTOBARBITAL-ETHER ANESTHESIA, AND mation is in agreement with previous observations DURING AND AFTER SUCH SUBSEQUENT ANESTHESIA. in man, wherein serum levels of bilirubin did not FORMATION OF BILIRUBIN FROM HEMOGLOBIN 1635 begin to rise until 1 hour after administration of ancy between the quantity of heme-C14 injected sensitized red cells (29) or hemoglobin (35). and the amount of bilirubin-C'4 excreted. For all doses of hemoglobin and red cells, the It also appears unlikely that significant amounts mean interval between removal of half the injected of heme-C'4, or bilirubin derived therefrom, would heme pigment from the circulation and the ap- be retained in the tissues of the recipient animal pearance of half the excreted bilirubin-C"4 in the 22 hours after administration. However, this pos- bile was approximately 3 hours (Table I). On sibility could not be definitely excluded, since the the other hand, a comparable intravenous load of C14 label was also present in the globin as well as bilirubin-C14 was almost completely excreted in the heme of the injected red cells or hemoglobin. within 1 hour, confirming prior reports (36-38). Thus, residual C14 activity in the tissues could not This indicates that the delay between heme se- be assumed to represent retained heme or bile questration and bilirubin excretion occurred in pigments. On the other hand, the virtual cessa- the reticuloendothelial system (39) after lysis of tion of bilirubin-C'4 excretion within 22 hours the red cells and before release of the bile pigment after injection of the labeled red cells or hemo- into the circulation. globin strongly suggests that the and This delay does not seem to be due to overload elimination of the administered pigment had been of the conversion mechanism, since widely dif- completed within this period. Indeed, it seems ferent doses of heme pigment were converted to improbable that labeled heme pigment or bilirubin bilirubin at similar rates. Indeed, the mechanism should accumulate in the tissues at a time when for bilirubin formation did not appear to be satu- excretion of unlabeled bilirubin, derived from en- rated by a load of heme pigment equivalent to ap- dogenous red cells, continued unabated. proximately 50 times the physiological turnover The failure to account for a significant fraction of rat red cells (25). Furthermore, the labeled of the administered heme-C14 in these instances red cells in any given injection were all generated is most likely due to its conversion to excretory within a 5-day period, and were metabolized to products other than bilirubin. Although earlier bile pigment at the same rate as a comparable dose investigators have suggested that dipyrrolic com- of extracorpuscular hemoglobin. They thus be- pounds, such as mesobilifuscin (7, 44) and pro- haved as a relatively uniform population of red pentdyopent (44, 45), may be normally occurring cells, which was sequestered and metabolized in excretory products of heme metabolism, recent random fashion. Consequently, the observed lag findings have failed to confirm this concept (46). may be considered consistent with the kinetics of a The present experiments constitute strong evidence precursor-product relationship involving an in- for the existence of alternate pathways of heme tervening rate-limiting step (40), although the catabolism, but they do not establish the identity nature of these conversion processes is not yet of the compounds involved. clear (41 ). Intravenously injected bilirubin-C'4 was al- Despite virtually complete removal from the cir- most completely recovered in the bile. This would culation, 20 to 45 per cent of the administered indicate that the postulated alternate excretory heme-C14 was not recovered as bilirubin-C14 in the products may only be derived from heme or per- bile of those animals which received sensitized red haps from bilirubin within the reticuloendothelial cells or large doses of hemoglobin in solution (42, system, but cannot be formed from bilirubin in the 43). Urinary loss of heme pigment never ex- circulation. Moreover, the greatest percentage ceeded 4 per cent of the administered dose of conversion to bilirubin was observed after the low- dissolved hemoglobin, and was absent in nephrec- est doses of hemoglobin in solution. Under these tomized animals and after injection of sensitized conditions the pigment is believed to be removed red cells. Moreover, less than 3 per cent of the ad- from the circulation as a complex with haptoglo- ministered Fe59 or Cr5' was excreted in the bile, bin (47). This suggests that the metabolic dis- and bilirubin was never detected in the urine. position of intracorpuscular hemoglobin and of Therefore, urinary and biliary loss of heme pig- unbound plasma hemoglobin may differ from that ment could not account for the significant discrep- of the hemoglobin-haptoglobin complex. 1636 J. DONALD OSTROW, JAMES H. JANDL AND RUDI SCHMID After inj ection of labeled red cells, biliary ex- accounted for was converted to excretory products cretion of unlabeled bile pigment was temporarily other than bilirubin. increased, possibly as a result of accelerated de- struction of endogenous red cells by the recipient ACKNOWLEDGMENT animal. Since this effect was also observed after The authors wish to acknowledge the continued inter- administration of dissolved hemoglobin, it does est and advice of Dr. William B. Castle and the expert not appear to be the result of the elution of anti- technical assistance of Louise Engel, Lydia Hammaker, body from the injected to the endogenous red cells, and Deanna Samuelson. nor of indiscriminate trapping of host cells by an actively sequestering spleen. These observations REFERENCES complicate interpretation of earlier experiments in 1. London, I. M., West, R., Shemin, D., and Rittenberg, which increments in bile pigment excretion were D. On the origin of bile pigment in normal man. J. biol. Chem. 1950, 184, 351. assumed to reflect quantitatively catabolism of ad- 2. Ottenberg, R. Bilirubin and bile salts in . ministered loads of heme pigment (42, 43, 48). J. Mt Sinai Hosp. 1943, 9, 937. The use of labeled hemoglobin obviates these diffi- 3. With, T. K. The bilirubin production of the human culties, and permits direct comparison of the quan- and its significance to the pathogenesis of jaundice. Acta med. scand. 1946, 123, 166. tity of injected heme pigment with that of the ex- 4. Bungenberg de Jong, W. J. H. Ueber Blutfarbstoff- creted bile pigment. wechsel und Urobilinogen. Dtsch. Arch. klin. Med. 1951, 198, 655. SUMMARY 5. With, T. K. Quantitative studies on the urinary ex- cretion of bilirubin and urobilinoids and on se- 1. Rats with a bile fistula were injected with rum bilirubin in diseases of the liver and jaundice antibody-sensitized red cells labeled with C14 and without liver lesion. Acta med. scand. 1944, 119, with either Fe59 or Cr51. Similar experiments 214. were performed with equivalent doses of labeled 6. Watson, C. J. Pyrrol pigments and hemoglobin ca- hemoglobin in solution. In all instances the re- tabolism. Minn. Med. 1956, 39, pp. 294, 403, 467. 7. Siedel, W., von Polnitz, W., and Eisenreich, F. moval of the administered heme pigment from Bilifuscin und Mesobilifuscin als natiirliche Ab- the circulation and the excretion of bilirubin-C14 bauprodukte des Blutfarbstoffes; uber Vorkommen in the bile were determined. und Bildung. Naturwissenschaften 1947, 34, 314. 2. After clearance from the circulation, intra- 8. Lester, R., Ostrow, J. D., and Schmid, R. Entero- and extracorpuscular hemoglobin were converted hepatic circulation of bilirubin. Nature (Lond.) 1961, 192, 372. to bilirubin at similar rates. Labeled bilirubin ap- 9. Eppinger, H. Die Hepato-Lienalen Erkrankungen peared in the bile within 30 minutes after the onset (Pathologie der Wechselbeziehungen zwischen of heme sequestration, and 50 per cent was ex- Milz, Leber und Knochenmark). Berlin, J. creted in approximately 3 hours. Springer, 1920, pp. 91-93. 3. A delay was observed between removal of 10. With, T. K. Bilirubin and urobilinoid contents of human bile. Acta med. scand. 1945, 122, 513. heme pigment from the circulation and excretion 11. McMaster, P. D., Broun, G. O., and Rous, P. Stud- of labeled bilirubin, which appeared to be due pri- ies on the total bile. I. The effects of operation, marily to the time required for the conversion of exercise, hot weather, relief of obstruction, inter- heme to bilirubin in the reticuloendothelial system. current disease, and other normal and pathological 4. With sensitized red cells, only 63 to 80 per influences. J. exp. Med. 1923, 37, 395. could be recovered 12. Wheeler, H. 0. The flow and ionic composition of cent of the administered heme bile. Arch. intern. Med. 1961, 108, 156. as bilirubin-C'4. Similar bile pigment recovery 13. Hawkins, W. B., and Whipple, G. H. The life cycle was obtained after injection of relatively large of the in the dog. Amer. J. Physiol. doses of hemoglobin in solution. In contrast, af- 1938, 122, 418. ter administration of small doses of hemoglobin, 14. Jandl, J. H. The agglutination and sequestration of calculated not to exceed the binding capacity of immature red-cells. J. Lab. clin. Med. 1960, 55, plasma haptoglobin, conversion to bilirubin-C' 4 663. 15. Jandl, J. H., and Kaplan, M. E. The destruction of was nearly complete. red cells by antibodies in man. III. Quantitative 5. Possible explanations for this difference are factors influencing the patterns of hemolysis in discussed, and it is suggested that the heme not zvzio. J. clin. Invest. 1960, 39, 1145. FORMATION OF BILIRUBIN FROM HEMOGLOBIN 1637 16. Jandl, J. H., and Castle, W. B. Agglutination of 33. Greene, C. H., and Snell, A. M. Studies in the me- sensitized red cells by large anisometric . tabolism of the bile. II. The sequence of changes J. Lab. clin. Med. 1956, 47, 669. in the blood and bile following the intravenous in- 17. Labbe, R. F., and Nishida, G. A new method of jection of bile or its constituents. J. biol. Chem. hemin isolation. Biochim. biophys. Acta 1957, 26, 1928, 78, 691. 437. 34. Tarchanoff, J. F. Ueber die Bildung von Gallenpig- 18. Shemin, D., London, I. M., and Rittenberg, D. The ment aus Blutfarbstoff im Thierk6rper. Pflug. synthesis of protoporphyrin in vitro by red blood Arch. ges. Physiol. 1874, 9, 53. cells of the duck. J. biol. Chem. 1950, 183, 757. 35. Duesberg, R. Ueber die biologischen Beziehungen 19. Fischer, H., and Orth, H. Die Chemie des Pyrrols. des Hamoglobins zu Bilirubin und Hamatin bei Leipzig, Akadem. Verlagsgesellschaft, 1937, vol. normalen und pathologischen Zustanden des 2, part 1, p. 626. Menschen. Arch. exp. Path. 1934, 174, 305. 20. Siperstein, M. D., and Chaikoff, I. L. C"4-cholesterol. 36. Weinbren, K., and Billing, B. H. Hepatic clearance III. Excretion of carbons 4 and 26 in , urine, of bilirubin as an index of cellular function in the and bile. J. biol. Chem. 1952, 198, 93. regenerating rat liver. Brit. J. exp. Path. 1956, 21. Ostrow, J. D., Hammaker, L., and Schmid, R. The 37, 199. preparation of crystalline bilirubin-C"4. J. dlin. In- 37. Bonomo, E., and Colombo, B. L'eliminazione per via vest. 1961, 40, 1442. biliare della bilirubina e dell'acido glucuronico nel 22. Stecher, P. G., Ed. The Merck Index of Chemicals ratto in corso di carico con bilirubina indiretta. and Drugs, 7th ed. Rahway, N. J., Merck and Arch. Mal. Appar. dig. 1959, 26, 162. Co., 1960. 38. Arias, I. M., Johnson, L., and Wolfson, S. Biliary 23. Garby, L., and Obara, J. Organ uptake and plasma excretion of injected conjugated and unconju- transportation kinetics of hemoglobin in rats. Blut gated bilirubin by normal and Gunn rats. Amer. 1960, 6, 143. J. Physiol. 1961, 200, 1091. 39. Mann, F. C., Sheard, C., Bollman, J. L., and Baldes, 24. Garby, L., and Noyes, W. D. Studies on . J. The formation of bile pigment from hemo- metabolism II. Pathways of hemoglobin iron me- globin. Amer. J. Physiol. 1926, 76, 306. tabolism in normal man. J. dlin. Invest. 1959, 38, 1484. 40. Zilversmit, D. B., Entenman, C., and Fishler, M. C. On the calculation of "turnover time" and "turn- 25. Hughes Jones, N. C., and Cheney, B. The use of '1Cr over rate" from experiments involving the use and 'Fe as red cell labels to determine the fate of labeling agents. J. gen. Physiol. 1943, 26, 325. of normal erythrocytes in the rat. Clin. Sci. 1961, 41. Lemberg, R. The chemical mechanism of bile pig- 20, 323. ment formation. Rev. Pure appl. Chem. 1956, 6, 26. Belcher, E. H., and Hughes-Jones, N. C. The 1. mathematical analysis of 51Cr deposition in organs 42. Sribhishaj, K., Hawkins, W. B., and Whipple, G. H. following the injection of '1Cr-labelled red cells. I. Bile pigment and hemoglobin interrelation in Clin. Sci. 1960, 19, 657. normal dogs. Amer. J. Physiol. 1931, 96, 449. 27. Malloy, H. T., and Evelyn, K. A. Determination of 43. Hawkins, W. B., Sribhishaj, K., Robscheit-Robbins, bilirubin with the photoelectric colorimeter. J. F. S., and Whipple, G. H. II. Bile pigment and biol. Chem. 1937, 119, 481. hemoglobin interrelation in anemic dogs. Amer. J. 28. Edwards, A. L. Statistical Methods for the Be- Physiol. 1931, 96, 463. havioral Sciences. New York, Rinehart, 1954, p. 44. Engel, H. Die heutigen Vorstellungen uber den 147. Blutfarbstoffabbau zu Gallenfarbstoff. Klin. 29. Jandl, J. H., Jones, A. R., and Castle, W. B. The de- Wschr. 1940, 19, 1177. struction of red cells by antibodies in man. I. Ob- 45. Bingold, K., and Stich, W. Fortschritte auf dem servations on the sequestration and lysis of red Gebiet des Blutfarbstoffes. Ergebn. inn. Med. cells altered by immune mechanisms. J. clin. In- Kinderheilk. 1954, 5, 707. vest. 1957, 36, 1428. 46. Gilbertsen, A. S., Lowry, P. T., Hawkinson, V., and 30. Murray, R. K., Connell, G. E., and Pert, J. H. The Watson, C. J. Studies of the dipyrrylmethene role of haptoglobin in the clearance and distribution ("fuscin") pigments. I. The anabolic significance of extracorpuscular hemoglobin. Blood 1961, 17, of the fecal mesobilifuscin. J. dlin. Invest. 1959, 38, 45. 1166. 31. Kawaguchi, M. Studies on iron metabolism of hemo- 47. Laurell, C. B., and Nyman, M. Studies on the se- globin by means of Fe"5. 2. On the distribution of rum haptoglobin level in hemoglobinemia and its loaded hemoglobin iron and its hourly variations influence on renal excretion of hemoglobin. Blood in the liver, , spleen, and . 1957, 12, 493. Igaku Kenkyfi 1957, 27, 1015. 48. Shen, S. C. Urobilinogen excretion after hemoglobin 32. Ham, T. H. A Syllabus of Laboratory Examinations infusion in patients with normal hematologic and in Clinical Diagnosis. Cambridge, Mass., Harvard hepatic findings. Proc. Soc. exp. Biol. (N. Y.) University Press, 1950, p. 100. 1960, 105, 264.