Bile Lipid Secretion in Isolated Peffused Rat Liver. A Model for Metabolic Studies J.C. Henrya,*, N. Domingob, F. ChanussotV, H. LafonW, J.C. Hautonb and J.P. C a n o a alNSERM U-278, Faculte de Pharmacie, Marseille Cedex 5, France, and blNSERMU-130, Marseille, France Isolated perfused rat liver was used to study the effects of constant taurocholate perfusion, with or without the addition of phosphatidylcholine unilamellar vesicles, upon both the bile salt-dependent and bile salt-independent secretion of bile. Taurocholate introduction increased bile flow and normalized the bile lipid secretion by restoring the bile salt-dependent secretion. A t a flow rate of 30 ml/min, the liver was perfused by a single-pass method. The perfusion medium contained 17.5 ~M taurocholate with or without 5.83 ~M phosphatidylcholine. In light of a recent quantitative dynamic concept on the interphase partition of lipids, it was calculated that more than 99% of the taurocholate reaches the liver as monomers and/or dimers. It was also deduced that the lipids were secreted in bile as small discoidal lipoprotein structures rather than unilamellar lipoproteic vesicles. During the course of the experiments (2 hr), the excellent criteria of viability of this model make it highly suitable for the investigation of hepatic metabolism. Furthermore, the addition of phosphatidylcholine unUamellar vesicles to the perfusate constitutes a potential vector for various liposoluble molecular species. Lipids 22, 1-5 (1987}.

Isolated perfused rat liver is a valuable model for the investigation of physiological, biochemical and pharmacological aspects of liver functions and has been widely used for this purpose by many investigators {1,2). Under physiological conditions, the rate of bile lipid secretion is dependent on the amount of bile salts (BS) returned in the portal blood (3). Most bile salts stored in the intestinal tract reach the liver in this way {4,5}, since the portal flow is 500 times higher than the intestinal lymph flow (6). Preservation of a physiological enterohepatic circulation of BS is a prerequisite for normal bile lipid secretion. Interruption of this circulation by biliary diversion induces a progressive decrease in the bile lipid secretion correlated with the disappearance of the intestinal BS pool. This disappearance is not compensated by a sufficient increase in BS synthesis in the liver. In isolated perfused liver without perfused BS, this phenomenon is rapidly and dramatically accentuated since the liver pool of BS is very small compared to the BS pool in the intact enterohepatic circulation. The BS-independent secretion of bile is correlated with secretion of water and electrolytes {7,8}. BS-dependent bile secretion is an active hepatic process indispensable for normal bile secretion of lipids {9,10), i.e., mainly phosphatidylcholine (PC) and free cholesterol. Most authors have assumed that bile lipids are mixed miceUes or unilamellar vesicles (11,12}. However, our results (13-15} show that bile lipids may occur as a bile lipoprorein complex (BLC), i.e., that micellar-like structures or

unilamellar vesicles are associated electrostatically with IgA fragments and hydrophobically with an anionic polypeptide fraction (APF}, which is different from the well-known plasma apolipoproteins (14) that are detected in trace amount in bile by radioimmunoassay {16}. Bile contains small amount of plasma, and immunoelectrophoresis of concentrated bile (obtained after delipidation and lyopbillzation) gives a protein pattern similar to plasma {17}, thus explaining the presence of apolipoproteins. Due to their own osmotic activity, BS molecularly dispersed in the aqueous polar phase actively taken up by hepatocytes in the sinusoidal pole {space of Disse) display a choleretic effect. Isolated perfused rat liver is a widely used model for the study of drug biotransformations and biliary excretion of their metabolites. Up to now, however, most of the isolated perfused liver models have taken into account only BS-independent bile secretion and neglected BS-dependent bile secretion {18-20}. A recent exception to this oversight is the data of Lowe et al. (21}, showing the effect of an intraportal BS bolus on the secretion of PC and free cholesterol. The purpose of the present study was to ascertain the effect of a constant BS perfusion. Under this condition, a stable biliary secretion of both the BS-dependent and BS-independent fractions was observed. MATERIALS AND METHODS

Liverperfusior~ To minimize anoxia of the liver, the portal

vein was rapidly catheterized (30 sec). Krebs-Henseleit buffer (22), pH 7.3, containing 2 mM of calcium chloride, 8 mM of glucose and 2% Dextran was infused at a constant flow rate of 30 ml/min. The perfusate was mainrained at 38 +_ 0.5 C and continuously oxygenated with an O2:CO2 (95:5, v/v) mixture at a flow rate of 2 l]min. Osmolarity was monitored with the Fiske 330 D osmometer {the values found were 300 +_ 5 mOsm/1). After this procedure, the bile duct was catheterized. The whole operating procedure lasted ca. 10 min. The equipment used was based on the model proposed by Brunengraber et al. (23} under nonrecirculating conditions. The liver left in situ was kept moist throughout the experiments. Test animals were divided into three groups, one control and two experimentals. The control group (group 1} had two animals and was perfused with buffer without taurocholate (TC) and PC. The two experimental groups {five rats per group} were per fused either with buffer containing 17.5 ~M of TC only {group 2) or with buffer containing 17.5 ~M of TC and 5.83 ~M of PC {group 3}. In the experimental groups, after stabilizing the system for 15 min, 14C-TC (1.28 ~Ci} in 20 ml of perfusion medium was also administered at a flow rate of 1 ml/min. Bile sampling was performed every 20 min in preweighed tubes. The volume of secretion was recorded in grams {bile density equal to 1 g/ml}. Alanine amino transferase *To whomcorrespondenceshouldbe addressed at INSERM U-278, {ALAT} and aspartate amino transferase {ASAT} transFacult6 de Pharmacie, 27 Bd Jean Moulin, 13005 MarseilleCedex aminase activity as well as creatinin phosphokinase {CPK) 5, France. activity in the outflowing perfusate were checked every LIPIDS, Vol. 22, No. 1 (1987)

J.C. H E N R Y E T AL.

10 man. The amount of radioactivity not bound by the liver was determined by sampling outflowing perfusate every 10 sec during '4C-TC perfusion. Chemicals. The following high grade chemicals were used: taurocholate A grade from Calbiochem Boehring Corp. (Meudon, France; purity by TLC 98%); phosphatidylcholine from Sigma (Paris, France; type III-E from frozen egg yolk in hexane, purity 99%); '4C-TC in aqueous solution containing 2% ethanol from Amersham (Les Ullis, France; sp act 64 mCi/mmol); ALAT, ASAT and ~-glucuronidase from Boehringer Mannheim Gmbh kit (Meylan, France); goat anti-rabbit IgG and alkaline phosphatase from Sigma; p-nitrophenol from Flow Lab. (Paris, France); NaC1, KC1, KH2PO4, CaCI~, MgSO4 and glucose from Farmiglia Carlo Erba (Paris, France); NaHCO3 from Prolabo (Paris, France), and Dextran (mol wt 70,000) from Fluka (Buchs, Switzerland). Dilution of mixed miceUes in the perfusate. Mixed micelles were prepared by carefully evaporating to dryness 17.86 mg of PC in hexane solution under nitrogen and solubilizing the resulting residue in 10 ml of a micellar solution containing 37.60 mg of TC. As calculated in mol. cm -3 according to the quantitative dynamic concept on the interphase partition of lipids applied to TC-PCwater ternary system (13,24), the concentration of TC and PC, incorporated after the dilution into unilamellar vesicles, was 0.05.10 -4 mol. cm-3 and 12.73.10 -4 mol. cm -~, respectively, in the mixed bilayer interracial lipid phase. Thus, since the PC/TC molar ratio in the vesicles averages 255, these vesicles can be considered pure PC unilamellar vesicles. On the other hand, the concentration of TC molecularly solubilized in the water polar phase is 17.48.10 -9 mol.cm -3, which is almost identical to the 17.50.10 -9 mol-cm -3 obtained with TC perfusate without PC. The advantage of adding TC-PC to the perfusate is that PC unilamellar vesicles can solubilize class 3 and 4 lipids (13,24) without appreciably affecting the concentration of TC molecularly solubilized in the water phase as monomers and dimers. The total inflow of PC through the liver (21/~mol) corresponds to ca. 8% of the total PC pool in the liver (averaging 260 tLrnol},while the amount of PC secreted in bile (5.1 #mol) corresponds to 2% of the PC content of the liver. Analytical methods. PC were assayed by the semiautomatic method of Amic et al. (25), TC by the automatic enzymatic method of Domingo et al. (26),/3-glucuronidase activity by the method of Fishman et al. (27) with a Sigma Kit and ALAT, ASAT, CPK activity in the perfusate by the Boehringer enzymatic method. 14C-TC was assayed with a liquid scintillation spectrophotometer (Beckman LS 2800) in 5 ml of PCS (Amersham) scintillation fluid. Detection of the anionic polypeptide fraction (APF) in rat bile. Anti-human APF specific polyclonal serum was prepared from rabbits by the procedure of Henry et al. (28). Work in progress using the ELISA method shows the presence of a common antigenic determinant in the amino acid sequence in human, rat, dog, calf and pig APF. The APF injected into the rabbits was purified using a procedure of zonal ultracentrifugation in 1-15 % sucrose density gradient containing glycodesoxycholate at a concentration of 1.5 mM (15). The principle of the ELISA assay (29,30) is the following: the wells in a polystyrene plate (LINBRO E.I.A. microtitration plate, Flow Laboratories) are coated with LIPIDS, Vol. 22, No. 1 (1987)

diluted rat bile (at a dilution of 1/200), or purified APF (0.5 to 20 ~g per 100 ~ according to automatic amino acid analysis) for the standard curve, in a carbonate buffer 0.1 M at pH 9.6, and then washed three times to remove the unbound components. The diluted polyclonal anti-APF serum (at a dilution of 1/200 in isotonic phosphate buffered saline, PBF) was added during 90 re_in at 37 C. The complex antigen-antibodies were washed three times. Bound antibodies are then detected using an anti-rabbit IgG serum (at a dilution of 1/200) conjugated to alkaline phosphatase (Sigma), and the activity was detected by the p-nitrophenylphosphate as the substrate (1 mg/ml). The enzymatic reaction was allowed to proceed for 30 re_in in the dark at 37 C. The absorbance was measured at 405 nm with a Multiskan Titertek MC (Flow Laboratories) against a blank. Statistical method. Statistical significance of results was determined by the Student t-test. RESULTS

Uptake of '4C-TC by the liver is almost complete, and the radioactivity of the exiting perfusate is negligible, its concentration in bile peaks 20 min after injection and then drops quickly during the following 60 min, regardless of whether the nonlabeled TC perfusion contains PC unilamellar vesicles (Fig. 1). The hepatic clearance of '4CTC is very high--95% (group 2, TC alone) and 98% (group 3, TC + PC vesicles). No significant difference was observed between these two groups. Figure 2 shows the concentration of PC and BS recorded in the bile collected from the three groups. When T C with or without PC vesicles is present in the perfusate, PC and BS concentrations and BS/PC molar ratio (10.4 _+ 0.3) remain within the physiological range. No significant

14c TC 0

ol o

o

%inbile o I

0 1

l

m

l

I

FIG. 1. Bile secretion of l'C-bile salts in function of time in rain from perfusion medium containing ~4C-taurocholate (TC) or ~4C-TC -t- PC unilamellar vesicles by isolated rat liver (see text). Each point represents the mean value • SEM (n -- 5}. Group 2 (TC), A--A; group 3 (TC + PC), A - - A .

ISOLATED RAT LIVER AND BILE LIPIDS TABLE 1

a I

I

I

I

I

Biliary Flow (~l/min)

I

'~ I =

Time of collection (rain)

5Z

Control (n = 2)

TC + PC vesicles (n = 5)

TC (n = 5)

-

0-20 20-40 40-60 60-80 80-100 100-120

~0

O

40

80 '

5.5 5.5 5.0 4.0 4.5 4.75

• • • • • •

1.5 1.5 1.0 1.0 1.5 1.75

11.6 10.8 8.6 8.4 10.1 10.1

• • • • • •

1.68 2.01 2.57 2.60 1.37 0.88

9.8 8.0 9.1 9.5 9.2 8.7

• • • • • •

1.70 2.37 1.47 1.44 1.11 0.91

120

TIME min

Each value represents mean • SEM in the three groups of rat.

C, 10

I

I

i

1

!

TABLE 2

I

/3-Glucuronidase Activity: Percent of Variation to Basal Value {0-20 rain) m O

20-40 min (n = 3) 60-80 rain In = 3) 100-120 min (n = 3)

0,,,

O,

I

i

i

I

40

0

t

I

I

100 175.2 -- 55.0 195.5 • 35.7

100 190.8 • 46.9 145.4 • 18.0

120

C I

TC + PC vesicles

I

80

TIME min

(D

TC

1

I

l

10

Q~ r

am

A l t h o u g h 13-glucuronidase activity did double during perfusion {Table 2), it never exceeded the values previously reported in isolated perfused liver {31), b u t was nevertheless five times lower t h a n physiological values in intact animals {32,33). To check the viability of the liver, the A L A T , A S A T and C P K activities were assayed in the exiting perfusate. These activities were found to be normal in all three groups {Table 3}. DISCUSSION

u_

a.

ol

i 0

I 40

I

1 80

I

i 120

TIME rain FIG. 2. Concentration of lipids in bile in function of time in min. (a) Bile salts (BS); (b) phosphatidylcholine (PC). Each point represents the mean value • SEM (n = 5). (c) Determination of the APF/PC weight percentage (see text) in one rat of group 2 and one rat of group 3. Control, D--El; group 2 (TC),/x--/x; group 3 (TC + PC), A - - A .

difference was noted between groups 2 and 3. In group 1 (control}, the concentration of PC and BS is v e r y low, b u t the molar ratio PC/BS remains normal for 40 rain. The A P F / P C weight ratio in one r a t of group 2 and one r a t of group 3 (Fig. 2) remains stable (0.045-0.050). As shown in Table 1, the bile flow from liver perfused with TC with or without PC unilamellar vesicles is 1.5 to 2 times higher t h a n in livers perfused without TC. In experimental groups 2 and 3, bile flow remains stable t h r o u g h o u t the 120-min t e s t period.

The greatest a d v a n t a g e of our experimental model is t h a t bile secretion is maintained at normal physiological levels. Also, perfusion with TC with or without PC unilamellar vesicles stabilized the concentration of BS and PC as well as the molar ratio BS/PC in bile (10.4 ___ 0.3) during the 2-hr sampling period. The A P F / P C weight ratio in bile studied in two r a t s remained constant, suggesting a normal formation of the bile lipoprotein complex. The absence of opalescence and the respective concentration of BS and PC in bile favor the presence of the bile lipoprotein complex as small discoidal lipoproteic s t r u c t u r e s r a t h e r t h a n vesicles {13,20}; it was previously observed {unpubhshed data} that, unlike h u m a n hepatic bile, the bile lipoproteic complex is present in r a t bile only as the low mol w t pseudomicellar family, and not as large discoidal s t r u c t u r e s or lipoproteic vesicles {the r a t is devoid of gallbladder with its concentrating activity}. In control liver perfused without TC, lipid levels decreased rapidly and dramatically during the first 40 min after isolation due to an u n c o m p e n s a t e d wash-out of BS. Interestingly, the low secretion rate continued to be observed in controls; this m a y indicate the occurrence of LIPIDS, VoL 22, No. 1 (1987)

4 J.C. HENRY ET AL. TABLE 3 Enzyme Activity in the Perfusion Medium a

Time of perfusion ~min) 0 10 20 30 40 50 60 70 80 90 100 110 120

ALAT {UI/1) 2.0 3.12 4.25 5.25 6.50 9.37 11.12 13.87 16.62 18.25 20.50 23.25 29.12

• 0.46 _+ 0.61 _+ 0.61 • 1.08 • 1.75 • 2.07 • 2.59 +_ 2.26 • 2.54 + 3.42 +_ 2.78 _ 1.97 • 3.05

ASAT {UI/I) 2.75 2.37 2.50 2.62 4.37 6.87 9.25 11.37 14.12 13.88 18.50 23.25 29.50

• 0.45 • 0.46 • 0.56 -- 0.98 -+ 1.48 • 2.29 • 2.56 _+ 2.77 _+ 2.87 __ 3.92 __ 2.89 • 3.25 - 4.89

CPK ~UI]I) 3.37 1.62 1.87 2.37 2.50 3.25 3.25 5.50 6.87 6.37 7.12 10.12 9.75

• 0.86 ___0.37 • 0.44 • 0.56 • 0.42 • 0.41 • 0.41 -4- 0.65 ___ 1.0 • 1.03 • 0.95 + 2.09 • 1.03

aThe activities of different enzymes have been tested from the exiting perfusate. ALAT, alanine amino transferase; ASAT, aspartate amino transferase; CPK, creatinin phosphokinase. Normal values ALAT - ASAT were 0-46 UI/1 (ref. IFFA-Credo}. There was no difference in the values obtained from isolated rat liver perfused with minimal perfusion medium, TC or TC + PC vesicles. Therefore, each value represents the mean + SEM of both livers tested (n = 12).

in situ BS synthesis. The m e a s u r e m e n t of nearly normal BS/PC molar ratios in control animals perfused for 40 min after isolation would m e a n t h a t BS-dependent bile secretion is sustained at a v e r y low level and t h a t the liver PC pool is far from exhausted. As s t a t e d in Materials and Methods, only a b o u t 2% of the total liver PC pool was eliminated b y animals in groups 2 and 3 during the 2-hr s t u d y period. Thus, within a short period, bile secretion in controls became largely BS-independent. The maintenance of this t y p e of secretion was formerly the only bile criteria for liver viability, B y simulating normal enterohepatic circulation, addition of BS to the perfusate activates secretion of PC and BS in the bile. In this respect, i.e., normalization of the lipid content of secreted bile, TC perfusion with or without PC unflamellar vesicles had the same effect in our concentration conditions. J u d g i n g from the levels of ALAT, A S A T and C P K measured in the hepatic perfusate effluent, enzyme activity remains in the normal r a n g e during the 2-hr perfusion period. Unlike Krell et al. (34) and Sugano et al. (35), who reported a 50% decrease in bile flow after 40 and 70 rain of perfusion, we observed a stable flow; this observation is strong evidence for the viability of our model. Given the level of the/~glucuronidase activity, the hepatic detoxification process would a p p e a r to remain intact. This finding was further s u p p o r t e d b y the m e a s u r e m e n t of normal osmolarity in exiting perfusate, which is an indication of limited cell damage. I t previously has been d e m o n s t r a t e d t h a t a hemoglobin-free p e r f u s a t e can be used without affecting the viability of the liver; Sugano et al. (35) showed t h a t perfusion of a saline bicarbonate mixture without macromolecules at a flow r a t e of 3-3.5 ml/min/g of liver was adequate to preserve the fine structure of the entire liver for 70 rain. To prolong viability b y m a i n t a i n i n g normal h y d r o s t a t i c pressure and p r e v e n t i n g cellular distension, the perfusate used in this s t u d y was supplemented with LIPIDS, Vol. 22, No. 1 (1987)

dextran (mol wt 70,000), which does not interfere with TC and PC. To avoid the depletion of the perfusate and thus better duplicate the consistent quality of the portal flow, the p e r f u s a t e was not recirculated. In addition to simplifying the experimental technique and avoiding artefacts due to perfusate variations, the unrecirculated mode has two main experimental a d v a n t a g e s . First, exiting perfusate can be collected and used to i n v e s t i g a t e any n u m b e r of p a r a m e t e r s . The second benefit of not recirculating the p e r f u s a t e is a g r e a t i m p r o v e m e n t in the reproducibility of results. In our experiments, clearance of TC by the liver was highly efficient since it was 98 and 95% in animals perfused with or without PC unilamellar vesicles. A s s a y s carried out with '4C-TC showed t h a t u p t a k e was almost immediate, since all the label was taken up within 20 sec. A t the end of the experiment, no label was even detected in the liver of experimental animals of groups 2 and 3, thus indicating t h a t clearance of TC was practically complete. A rapid transfer of TC from perfusate to bile is shown b y the fact t h a t 90% of the label of the perfused TC is secreted in the bile within 20 min. W i t h regard to uptake, excretion and kinetics, this model b e h a v e s identically to the in vivo model using desoxycholate (unpublished data). U p t a k e of BS, which is practically u n s a t u r a b l e under physiological conditions, seems to be independent of hydroxylation and conjugation {36). The addition of TC with or without PC unilamellar vesicles to the perfusate did not affect u p t a k e and secretion of BS. W i t h or without PC unilamellar vesicles in our conditions, BS reach the liver as m o n o m e r s and dimers; the incorporation of TC into the lipid bilayer interfacial phase of the PC unilamellar vesicles is negligible (13,24). The model proposed presents several a d v a n t a g e s for use in pharmacological and physiological studies. These a d v a n t a g e s include p r e s e r v a t i o n of active bile secretion, c o n s t a n t bile flow, normal bile lipid secretion and excellent organ viability. To furnish a vehicle for liposoluble molecular species, PC unilamellar vesicles can be added to the perfusate without altering the quality of bile secretion.

ACKNOWLEDGMENTS Mrs. Dicostanzo-Dufetel performed biological measurements, Mrs. Grosclaude gave immunological help, Mr. Bosano provided technical assistance, Mr. Corsini reviewed the text and Mrs. Bonneil typed the manuscript. REFERENCES 1. Bond, J.A., Medinsky, M.A., and Dutcher, J.S. {1984} ToxicoL AppL Pharmacol. 75, 531-538. 2. Pang, S.K., Cherry, W.F., Terrel, J.A., and Ulm, E.H. {1984} Drug Metab. Dispos. 12, 309-313. 3. Carey, M.C. (1982}in The Liver: Biology and Pathology {Arias, I., Popper, H., Schachter, D., and Shafritz, D.A., eds.} pp. 429-465, Raven Press, New York. 4. Thomsen, O.O. {1984} Scand. J. Gastroenterol. 19, 97-152. 5. Scott-Jones, A., and Meyer, N.C. {1979} Ann. Rev. Phys. 41, 67-82. 6. Ewerth, S., Bjorkem, I., Einarsson, K., and Osa, L. {1982}J. Lipid Res. 23, 1183-1186. 7. Erlinger, S. (1981) Hepatology 7, 352-359. 8. Boyer, J.L. {1971}Am. J. PhysioL 221, 1156-1163. 9. Kay, R.F., and Entenmann, C. {1961} Am. J. Physiol. 200, 855-859.

ISOLATED RAT LIVER AND BILE LIPIDS 10. Robins, J.S., and Brunengraber, H. {1982) J. Lipid Res. 23, 604-608. 11. Small, D.M. {1968} N. Engl. J. Meal 279, 588-593. 12. Mazer, A.M., Benedek, G.B., and Carey, M.C. (1980} Biochemistry 19, 601-615. 13. Hauton, J.C. (1986} in Molecular and Cellular Basis of Digestion IDesnueUe, P., SjostrSm, H., and Nor~n, O., eds.} pp. 147-170, Elsevier, Amsterdam. 14. Vigne, J.L., Lafont, H., Nalbone, G., Domingo, N., CharbonnierAugeire, M., Lairon, D., L~onardi, J., and Hauton, J.C. {1979} Biochem. Biophys. Res. Commun. 88, 1284-1287. 15. Martigne, M., Domingo, N., Lafont, H., Nalbone, G., and Hauton, J.C. {1985) Lipids 20, 884-889. 16. Sewell, R.B., Mao, J.T.S., Kawamoto, T., and La Russo, N.F. (1983) J. Lipid Res. 24, 391-401. 17. Hauten, J.C., Greusard, C., Perrot, M.J., Depieds, R., and Sarles, H. (1965} Acta GastroenteroL Belg. 28, 408-413. 18. Faed, F.M., Dobbs, B.R., and Lee, D. {1984} Arch. Int. Pha~ macodyn. Ther. 272, 4-16. 19. Christensen, C.B., Jansen, J.A., Raun-Jonsen, A., and Reiff, L. (1984} Acta Pharmacol. Toxicol. 54, 134-140. 20. Ehninger, G.E., Proksh, B., Hartmann, F., Garter, H.V., and Wilms, K. {1984} Cancer Chemother. PharmacoL 12, 50-52. 21. Lowe, P.J., Barnwell, S.G., and Coleman, R. {1984}Biochem. J. 222, 631-637. 22. Krebs, H.A., and Henseleit, K. (1932) Hopp~Seyler's 7,. PhysioL Chem. 210, 33-66. 23. Brunengraber, H., Boutry, M., and Lowenstein, J.M. {1973} J. Biol. Chem. 248, 2656-2669.

24. Hauton, J.C., Domingo, N., Martigne, M., Lafont, H., Nalbone, G., Chanussot, F., and Lairon, D. (1986) Biochimie 68, 275-285. 25. Amic, J., Lairon, D., and Hauton, J.C. {1972) Clin. Chim. Acto~ 40, 107-114. 26. Domingo, N., Amic, J., and Hauton, J.C. (1972) Clin. Chim. Acta 37, 399-404. 27. Fishman, W.H., Kato, K., Anstiss, C.L., and Green, S. {1967} Clin. Chim. Acta 15, 435-447. 28. Henry, X., Bernard, D., and Depieds, R. {1969}An~ Inst. Pasteur 114, 395-399. 29. Engwall, R., and Perlmann, P. 11972}J. Immunol. 109, 129-135. 30. Scherrer, R., and Bernard, S. {1977} Ann. Microbiol. (Institut Pasteur] 128A, 499-510. 31. Sewell, R.B., Barham, S.S., Zimzeister, A.R., and La Russo, M. {1984} Am. J. Physiol. 246, G8-G15. 32. Lafont, H., Chanussot, F., Dupuy, C., Lech~ne, P., Lairon, D., Charbonnier-Augeire, M., Chabert, C., Portugal, H., Pauli, A.M., and Hauton, J.C. {1984} Lipids 19, 195-201. 33. Chanussot, F., Lafont, H., Dupuy, C., Charbonnier-Augeire, M., Chabert, C., Portugal, H., Panli, A.M., and Hanton, J.C. {1985} Proc. Soc. Exp. Biol. Meal 179, 13-20. 34. Krell, H., Jaschke, H., Hoke, H., and Pfaff, E. ~1984) HoppeSeyler's Z. Physiol. Chem. 365, 1115-1122. 35. Sugano, T., Suda, K., Shimada, M., and Oshino, M. {1978} Biochem. J. 83, 995-1007. 36. Reichen, J., and Panmgartner, G. {1976} Am. J. Physiol. 231, 734-742. [ R e c e i v e d F e b r u a r y 21, 1986]

LIPIDS, Vol. 22, No. t (t987)