- Email: [email protected]
Biliary micellar cholesterol nucleates via the vesicular pathway
Biochimica et Biophysica Acta, 1003 (1989) 246-249 Elsevier
246
BBA 53121
Biliary micellar cholesterol nucleates via the vesicular pathway Y o c h a n a n Peled, Zamir H a l p e r n , Benzion Eitan, G i d e o n G o l d m a n , Fred K o n i k o f f and Tuvia G i l a t Department of Gastroenterology, Tel.Aviv Medical Center, lchilov Hospital and the Sackler Faculty of Medicine, Tel Aviv University (Israel)
(Received 14 October 1988) (Revised manuscript received 6 February 1989)
Key words: Cholesterol; Nucleation; (Phospholipid vesicle); (Bile)
Blllary cholesterol nucleates pflmadly from phospl~ipld vesicles. In this study, we investigated the mode of nucleation of micellar ~oJesteroL Ten Idles (four human and six model) were examined. The vesicular and micellar fractions of each bile were separated by gel chromatography. The whole biles and their isolated carriers were incubated at 37°C until nudeatk)n time. In whole human Idles, the proportion of total cholesterol in vesicles rose throughout the I ~ (from zero time to nucleation time) from 15.5 ± 8,6~ to 28.0 4- 12.5%, and in model biles from 46.8 4" 22.4% to ?S.S :1:8 . ~ . The vesicular isolated fraction remained unchanged throughout Incubation. In isolated miceiles devoid of resides at zero time, new veddes fermed during ineulmtlea, carrying increasing proportions of cholesterol. At uadeatim time, these vesicles contained 11.0% of origlhally mleellar cholt~teml in human biles, and 41.2% in model bfles. The new vesicles formed In whole bile and In the micellar fraction were chromatographically and chemically alndlar to the vealeles eflOmlly present in bile. These data suggest that micellar cholesterol nucleates via the neofommtlon of p i ~ l p i d realties, which seem to he the final common pathway for ehelesterol nucleation in bile. Inueducflm Biliary cholesterol is carried by mixed micelles and phospholipid vesicles [1,2]. When the two carriers are separated and incubated at 37°C, cholesterol crystals first appear in the vesicular phase, while micellar cholesterol nucleates after a considerable delay [3,4]. The nucleation time (NT) of whole bile correlated signifieantly with the ~ of vesicles, whereas no correlation was found with the NT of micelles [3]. It has previously been shown that vesicle a ~ r e p t i o n in bile is morpholosieally and chronolo8ieally associated with cholesterol crystal formation [5]. All tl,~e data suggest that vesicles are the less stable carrier, and the preferred vehicle for cholesterol nucleation in bile. These data leave unanswered the question regarding the mode of nucleation of micellar cholesterol. Since miceHes carry a major portion of bifiary cholesterol which eventually nucleates even from the isolated micellar carrier, the question is of considerable importance in biliary pathophysiology.
~
: Y. Pek~ Departmentof Gastroenterology, Tel-Aviv Medical Centar, Ichilov Hospital, 6 Weizn~n St., Tel-Aviv 64239, Israel
In the present study, we have analyzed the events associated with the nucleation of micellar cholesterol, and in particular whether this occurs via the vesicular pathway. Materials and Methods
Two gallbladder biles, two hepatic biles and six model (artificial) biles were examined. The gallbladder bile specimens were obtained at operation: one from a gallstone patient at cholecystectomy, the other was aspirated from the gallbladder of a patient with no gallstone or gallbladder disease. The hepatic biles were obtained from gallstone patients with an indwelling open T-drain following cholecystectomy. The model biles were prepared as previously described [3]. In brief, measured aliquots of lipid stock solutions (Sigma Chemical Co.) were mixed to produce a mixture having the desired composition: 10 mol~ cholesterol, 72 mol~ bile salt (sodium taurocholate) and 18 mol~ phospholipid (egg yolk phosphatidylcholine). The total lipid concentration of the model bile thus obtained was 8.0 g/dl. The native biles were ultracentrifuged at 192 000 × g (average) for 40 min at 25°C. Samples of the supernatant bile were labeled with [3H]cholesterol (Amersham International, Buckinghamshire, U.K.) and applied to a
0005-2760/89,/$03.50 ~ 1989 Elsevier Science Publishers B.V. (Biomedical Division)
247 chromatographic column (Sephacryl S-300 gel) to separate the vesicular and micellar phases as previously described [6]. The pooled isolated vesicular and micellar fractions were immediately rechromatographed to confirm their composition and purity. The original whole bile samples (human as well as model) and the vesicular and micellar fractions of each bile were then incubated at 37°C under nitrogen. To determine the nucleation time, aliquots were taken daily from all samples and examined under polarized light, until cholesterol crystals were observed [7]. Model biles as well as their isolated vesicular fractions were rechromatographed after 3 days. Their respective micellar fractions were rechromatographed after 5 days of incubation. All incubated original biles and their vesicular and micellar fractions were also rechromatographed at nucleation time. The presence and proportions of vesicles and micelles were determined at each chromatography. Concurrently, the chemical composition of the isolated vesicles and miceiles was determined. The chemical composition of whole human biles was determined at zero time. To calculate the distribution of biliary cholesterol between the vesicular and micellar carriers, the areas under the radioactivity peaks (vesicles and micelles) were measured, and calculated as percentage of the total area under the cholesterol radioactivity curve eluted from the chromatographic column [3]. The bile acid concentration was determined enzymatically [8], cholesterol was determined by the method of Abbell et al. [9], and phospholipids as described by Bartlett [10]. The cholesterol saturation index (CSl) was calculated using the critical tables of Carey [11]. Some specimens of the incubated biles or isolated fractions were examined with a JEOL 100B electron microscope using the negative staining method. Statistical analysis. The paired Student's t-test was used to compare the changes in the distribution of cholesterol in whole bile and the micellar fractions. The signed rank t-test was used to compare the changes in
A 100 o*
C
9C
_~ ac 7C
~ 6c -=
5C
P- 2o
U
10 0
--13
*~1
NI'¢
-6
'3
NIT
0
.5
NIT
Time tday~)
Fig. 1. The proportion of cholesterol in vesicles. Model biles (n = 6) and their isolated vesicular and micellar fractions during incubation at 37 ° C from zero time to nucleation time. (A) Whole bile: (B) vesicular phase; (C) micellar phase.
tie cholesterol to phospholipid ratio of vesicles with time. Results
The chemical composition and tlle nucleation times of the original human biles, and the distribution of their cholesterol between the vesicular and micellar fractions at zero time, are given in Table I. The nucleation time of model biles ranged from 4 to 7 days (mean 5.8 days). The distribution of cholesterol in vesicles (expressed as percent cholesterol) in model hiles and in their isolated vesicular and micellar fractions during incubation is depicted in Fig. 1. It can be seen that the vesicular fraction remained devoid of micelles throughout the incubation time. In whole model bile, there was an increase in the prcpottion of cholesterol carried by vesicles from (mean ± S.D.) 46.8 ± 22.4% at zero time, to 69.0 + 9.7% (P < 0.05) after 3 days of incubation and 75.5 ± 8.2% at nucleation time (P < 0.05 vs. zero time, P < 0.05 vs. 3 days). In the isolated micellar fraction, new vesicles were formed, carrying 25.8 + 19.9% of the total cholesterol after 5 days of
TABLE I
Nucleation time, chemical composition and cholesterol distribution between vesicles and micelles of human biles B, bile; V, vesicles; M, micelles; CHOL, cholesterol; PL, phospholipids. BS, bile salts; CSI, cholesterol saturation index; G, gallbladder; T, "T" tube. * At zero time. Bile
Nucleation time (days) B
G G T T
5 4 24 8
V 7 5 28 14
M 7 11 30 15
CSI
Cholesterol distribution *
Total
Composition (mol~)
(%) V
CHOL
PL
B5
M
lipid (gm~,)
4.6 12.6 23.0 21.8
95.4 87.4 77.0 78.2
3.91 2.18 0.72 0.86
13.3 5.6 ,q.0 9.4
21.7 19.8 32.3 30.9
65.1 74.6 59.7 59.7
2.10 1.14 2.33 2.14
248 B
A
100
@...-.---@
C
Discussion
u
40
tC
N'T
-~
~T
T i m e (days)
Fill. 2. The proportion of cholesterol in vesicles. Human biles (n - 4) and their Itolated vesicular and micellar fraction at zero time and at nucleation time. (A) Whole bile; (B) vesicular phase; (C) micellar phase.
incubation and 41.2 4. 27.0~ at nucleation time (P < 0.05 vs. 5 days). In human biles, the findings were similar (Fig. 2). The vesicular fraction did not change with time. in whole human bile, the proportion of cholesterol in vesicles increased from 15.5 + 8.6~ at zero time to 28.0 4. 12.5~ (P < 0.05) at nucleation time. In the micellar fraction, there was a neoformation of vesicles which carried 11.0 + 2.0~ of the total cholesterol at nucleation time. The absence of vesicles in the micellar fraction at zero time was confirmed by immediate rechromatography in both model and native biles. The model biles contained more vesicles than the human biles included in this study (Figs. 1 and 2). The cholesterol/phosphofipid ratio of the new vesicles in the micellar fraction was much higher than the ratio in the original micelles (0.22 + 0.02) and was similar to that of the original vesicles in whole bile (Table I1). The cholesterol/phospholipid ratio of vesicles in the whole biles (native and model) rose during in~tbation, and was higher at nucleation than at zero time (P < 0.01). The pres~cc of vesicles was also demonstrated by electron microscopy in some biles and fractions examined. TABLE I!
~
/
~
Source of vesicles
nu~oo/ve~c/es Whole bile at tero
at nucleation
Micellar phase at nucleation
time
time
time
~iative'biles
1,24 4- 0.24
1.59 4-0.28
1,65 4- o.31
(m-3) Modalbiles (a-3)
1.62 4-0.21
1.79 4-0.21
1.17 4-0.19
All biles
1.434-0.28
1.694-0.27 *
* P < 0.01.
The mode of cholesterol nucleation from the micellar phase via transition to the vesicular phase has, to the best of our knowledge, not previously been investigated. Three groups have demonstrated that vesicular cholesterol precipitated faster than micellar cholesterol [3,4,12]. The present study indicates that micellar cholesterol also precipitates via the vesicular pathway through the spontaneous neoformation of vesicles and a progressive shift of micellar cholesterol to these newly formed vesicles. Thus, when native biles were incubated at 37°C, the proportion of vesicular cholesterol increased progressively from 15.5 :!: 18.6 at zero time to 28.0 + 12.5 at NT. When the two carriers, vesicles and micelles, were separately incubated, the amount of cholesterol in vesicles remained stable until nucleation and neoformation of micelles did not occur. However, in the isolated micellar fraction, a different sequence of events was demonstrated: vesicles which were not present at zero time began to appear and contained a progressively larger proportion of initially micellar cholesterol. This proportion was highest at nucleation time. Preliminary data not included in the present study show that following nucleation, a progressive decrease in vesicular cholesterol is seen in whole bile and in the isolated micellar fraction. This is compatible with other observations showing that cholesterol precipitates from vesicles [3,5]. Biochemical analysis confirmed that the newly formed vesicles in the isolated micellar fraction have a similar cholesterol/phosphofipid ratio (similar to 'old' vesicles present in the native bile. Conversely, the micellax fraction had a cholesterol/phospholipid ratio of approx. 0.3 at zero time as well as at nucleation time. These chemical data thus support the chromatographic data. Thus, it was shown that vesicles were initially absent in the micellar fraction and that they subsequently appeared in this fraction, particularly towards nucleation time. These newly formed vesicles were shown by two methods - chromatography and chemical analysis - to be similar to the 'old' vesicles initially present in bile. In some samples, it was also confirmed by electron microscopy. All these observations are compatible with the assumption that vesicles are the final common pathway for cholesterol precipitation in bile. Initial cholesterol nucleation occurs from vesicles originally present and as time goes on, micellar cholesterol is transferred into newly formed vesicles, and thus the process of cholesterol precipitation continues. Further experiments are needed to determine whether any biliary cholesterol precipitates directly from micelles. These data and interpretations are in agreement with previous observations using video-enhanced
249 microscopy, which showed that cholesterol crystals were intimately associated with vesicular aggregates and that rapid neoformation of vesicles was observed in biles of patients with gallstones [5]. These data are also compatible with early pioneering experiments in model systems [13-15], where liquid crystals (presumably vesicular aggregates) were observed in the metastable phase in which rapid cholesteroi crystal formation was observed.
Acknowledgments The authors appreciate the help and biochemical assistance of Mrs. R. Rosenberg, and the expert assistance of Mr. S. Dabush in electron microscopy. This study was supported by the Reka Foundation Tel-Aviv and the Germalfis-Kaufman Chair of Gastroenterology, Tel-Aviv University. References 1 Somjen, G.J. and Gilat, T. (1983) FEBS Lett. 156, 266-.268. 2 Pattinson, N.R. (1985) FEBS Lett. 181, 339b1342.
3 Peled, Y., Halpern, Z., Baruch, R. and Gilat, T. (1988) Hepatology 8, 914-918. 4 I-iarvey, P.R.C., Somjen, G.J., Lichtenberg, M.S., Petrunka, C., Gilat, T. and Strasberg, S.M. (1987) Biochim. Biophys. Acta 921, 198-204. 5 Halpern, Z., Dudley, M.A., Kibe, A., Lynn, M.P., Breuer, A.C. and Holzbach, R.T. (1986) Gastroenterology 90, 875-885. 6 Somjen, G.J. and Gilat, T. (1985) J. Lipid Res. 26, 699-704. 7 Holan, K.R., Holzbach, R.T., Hermann, R.E., Cooperman, A.M. and Claffey, N.J. (1979) Gastrco,,terology 77, 611-617. 8 Talalay, P. (1960) Methods Bioc,~,m. Anal. 8, 119-143. 9 Abbell, L.L., Levy, B.B., Brodie, B.B. and Kendall, F.E. (1952) J. Biol. Chem. 195, 357-366. 10 Bartlett, G.R. (1959) J. Biol. Chem. 234, 466-468. 11 Carey, M.C. (1979) J. Lipid Res. 19, 945-955. 12 Lee, S.P., Park, H.Z., Madani, H. and Kaler, E.W. (1987) Am. J. Physiol. 252, 6374-6383. 13 SmaU, D.M. and Bourges, M. (1966) Mol. Cryst. Liq. Cryst. 1541-1561. 14 Olszewski, M.F., Holzbach, R.T., Saupe, A. and Brown, G.H. (1973) Nature 242, 376-377. 15 Holzbach, R.T. and Corbusier, C. (1978) Biochim. Biophys. Acta 528, 436-444.
246
BBA 53121
Biliary micellar cholesterol nucleates via the vesicular pathway Y o c h a n a n Peled, Zamir H a l p e r n , Benzion Eitan, G i d e o n G o l d m a n , Fred K o n i k o f f and Tuvia G i l a t Department of Gastroenterology, Tel.Aviv Medical Center, lchilov Hospital and the Sackler Faculty of Medicine, Tel Aviv University (Israel)
(Received 14 October 1988) (Revised manuscript received 6 February 1989)
Key words: Cholesterol; Nucleation; (Phospholipid vesicle); (Bile)
Blllary cholesterol nucleates pflmadly from phospl~ipld vesicles. In this study, we investigated the mode of nucleation of micellar ~oJesteroL Ten Idles (four human and six model) were examined. The vesicular and micellar fractions of each bile were separated by gel chromatography. The whole biles and their isolated carriers were incubated at 37°C until nudeatk)n time. In whole human Idles, the proportion of total cholesterol in vesicles rose throughout the I ~ (from zero time to nucleation time) from 15.5 ± 8,6~ to 28.0 4- 12.5%, and in model biles from 46.8 4" 22.4% to ?S.S :1:8 . ~ . The vesicular isolated fraction remained unchanged throughout Incubation. In isolated miceiles devoid of resides at zero time, new veddes fermed during ineulmtlea, carrying increasing proportions of cholesterol. At uadeatim time, these vesicles contained 11.0% of origlhally mleellar cholt~teml in human biles, and 41.2% in model bfles. The new vesicles formed In whole bile and In the micellar fraction were chromatographically and chemically alndlar to the vealeles eflOmlly present in bile. These data suggest that micellar cholesterol nucleates via the neofommtlon of p i ~ l p i d realties, which seem to he the final common pathway for ehelesterol nucleation in bile. Inueducflm Biliary cholesterol is carried by mixed micelles and phospholipid vesicles [1,2]. When the two carriers are separated and incubated at 37°C, cholesterol crystals first appear in the vesicular phase, while micellar cholesterol nucleates after a considerable delay [3,4]. The nucleation time (NT) of whole bile correlated signifieantly with the ~ of vesicles, whereas no correlation was found with the NT of micelles [3]. It has previously been shown that vesicle a ~ r e p t i o n in bile is morpholosieally and chronolo8ieally associated with cholesterol crystal formation [5]. All tl,~e data suggest that vesicles are the less stable carrier, and the preferred vehicle for cholesterol nucleation in bile. These data leave unanswered the question regarding the mode of nucleation of micellar cholesterol. Since miceHes carry a major portion of bifiary cholesterol which eventually nucleates even from the isolated micellar carrier, the question is of considerable importance in biliary pathophysiology.
~
: Y. Pek~ Departmentof Gastroenterology, Tel-Aviv Medical Centar, Ichilov Hospital, 6 Weizn~n St., Tel-Aviv 64239, Israel
In the present study, we have analyzed the events associated with the nucleation of micellar cholesterol, and in particular whether this occurs via the vesicular pathway. Materials and Methods
Two gallbladder biles, two hepatic biles and six model (artificial) biles were examined. The gallbladder bile specimens were obtained at operation: one from a gallstone patient at cholecystectomy, the other was aspirated from the gallbladder of a patient with no gallstone or gallbladder disease. The hepatic biles were obtained from gallstone patients with an indwelling open T-drain following cholecystectomy. The model biles were prepared as previously described [3]. In brief, measured aliquots of lipid stock solutions (Sigma Chemical Co.) were mixed to produce a mixture having the desired composition: 10 mol~ cholesterol, 72 mol~ bile salt (sodium taurocholate) and 18 mol~ phospholipid (egg yolk phosphatidylcholine). The total lipid concentration of the model bile thus obtained was 8.0 g/dl. The native biles were ultracentrifuged at 192 000 × g (average) for 40 min at 25°C. Samples of the supernatant bile were labeled with [3H]cholesterol (Amersham International, Buckinghamshire, U.K.) and applied to a
0005-2760/89,/$03.50 ~ 1989 Elsevier Science Publishers B.V. (Biomedical Division)
247 chromatographic column (Sephacryl S-300 gel) to separate the vesicular and micellar phases as previously described [6]. The pooled isolated vesicular and micellar fractions were immediately rechromatographed to confirm their composition and purity. The original whole bile samples (human as well as model) and the vesicular and micellar fractions of each bile were then incubated at 37°C under nitrogen. To determine the nucleation time, aliquots were taken daily from all samples and examined under polarized light, until cholesterol crystals were observed [7]. Model biles as well as their isolated vesicular fractions were rechromatographed after 3 days. Their respective micellar fractions were rechromatographed after 5 days of incubation. All incubated original biles and their vesicular and micellar fractions were also rechromatographed at nucleation time. The presence and proportions of vesicles and micelles were determined at each chromatography. Concurrently, the chemical composition of the isolated vesicles and miceiles was determined. The chemical composition of whole human biles was determined at zero time. To calculate the distribution of biliary cholesterol between the vesicular and micellar carriers, the areas under the radioactivity peaks (vesicles and micelles) were measured, and calculated as percentage of the total area under the cholesterol radioactivity curve eluted from the chromatographic column [3]. The bile acid concentration was determined enzymatically [8], cholesterol was determined by the method of Abbell et al. [9], and phospholipids as described by Bartlett [10]. The cholesterol saturation index (CSl) was calculated using the critical tables of Carey [11]. Some specimens of the incubated biles or isolated fractions were examined with a JEOL 100B electron microscope using the negative staining method. Statistical analysis. The paired Student's t-test was used to compare the changes in the distribution of cholesterol in whole bile and the micellar fractions. The signed rank t-test was used to compare the changes in
A 100 o*
C
9C
_~ ac 7C
~ 6c -=
5C
P- 2o
U
10 0
--13
*~1
NI'¢
-6
'3
NIT
0
.5
NIT
Time tday~)
Fig. 1. The proportion of cholesterol in vesicles. Model biles (n = 6) and their isolated vesicular and micellar fractions during incubation at 37 ° C from zero time to nucleation time. (A) Whole bile: (B) vesicular phase; (C) micellar phase.
tie cholesterol to phospholipid ratio of vesicles with time. Results
The chemical composition and tlle nucleation times of the original human biles, and the distribution of their cholesterol between the vesicular and micellar fractions at zero time, are given in Table I. The nucleation time of model biles ranged from 4 to 7 days (mean 5.8 days). The distribution of cholesterol in vesicles (expressed as percent cholesterol) in model hiles and in their isolated vesicular and micellar fractions during incubation is depicted in Fig. 1. It can be seen that the vesicular fraction remained devoid of micelles throughout the incubation time. In whole model bile, there was an increase in the prcpottion of cholesterol carried by vesicles from (mean ± S.D.) 46.8 ± 22.4% at zero time, to 69.0 + 9.7% (P < 0.05) after 3 days of incubation and 75.5 ± 8.2% at nucleation time (P < 0.05 vs. zero time, P < 0.05 vs. 3 days). In the isolated micellar fraction, new vesicles were formed, carrying 25.8 + 19.9% of the total cholesterol after 5 days of
TABLE I
Nucleation time, chemical composition and cholesterol distribution between vesicles and micelles of human biles B, bile; V, vesicles; M, micelles; CHOL, cholesterol; PL, phospholipids. BS, bile salts; CSI, cholesterol saturation index; G, gallbladder; T, "T" tube. * At zero time. Bile
Nucleation time (days) B
G G T T
5 4 24 8
V 7 5 28 14
M 7 11 30 15
CSI
Cholesterol distribution *
Total
Composition (mol~)
(%) V
CHOL
PL
B5
M
lipid (gm~,)
4.6 12.6 23.0 21.8
95.4 87.4 77.0 78.2
3.91 2.18 0.72 0.86
13.3 5.6 ,q.0 9.4
21.7 19.8 32.3 30.9
65.1 74.6 59.7 59.7
2.10 1.14 2.33 2.14
248 B
A
100
@...-.---@
C
Discussion
u
40
tC
N'T
-~
~T
T i m e (days)
Fill. 2. The proportion of cholesterol in vesicles. Human biles (n - 4) and their Itolated vesicular and micellar fraction at zero time and at nucleation time. (A) Whole bile; (B) vesicular phase; (C) micellar phase.
incubation and 41.2 4. 27.0~ at nucleation time (P < 0.05 vs. 5 days). In human biles, the findings were similar (Fig. 2). The vesicular fraction did not change with time. in whole human bile, the proportion of cholesterol in vesicles increased from 15.5 + 8.6~ at zero time to 28.0 4. 12.5~ (P < 0.05) at nucleation time. In the micellar fraction, there was a neoformation of vesicles which carried 11.0 + 2.0~ of the total cholesterol at nucleation time. The absence of vesicles in the micellar fraction at zero time was confirmed by immediate rechromatography in both model and native biles. The model biles contained more vesicles than the human biles included in this study (Figs. 1 and 2). The cholesterol/phosphofipid ratio of the new vesicles in the micellar fraction was much higher than the ratio in the original micelles (0.22 + 0.02) and was similar to that of the original vesicles in whole bile (Table I1). The cholesterol/phospholipid ratio of vesicles in the whole biles (native and model) rose during in~tbation, and was higher at nucleation than at zero time (P < 0.01). The pres~cc of vesicles was also demonstrated by electron microscopy in some biles and fractions examined. TABLE I!
~
/
~
Source of vesicles
nu~oo/ve~c/es Whole bile at tero
at nucleation
Micellar phase at nucleation
time
time
time
~iative'biles
1,24 4- 0.24
1.59 4-0.28
1,65 4- o.31
(m-3) Modalbiles (a-3)
1.62 4-0.21
1.79 4-0.21
1.17 4-0.19
All biles
1.434-0.28
1.694-0.27 *
* P < 0.01.
The mode of cholesterol nucleation from the micellar phase via transition to the vesicular phase has, to the best of our knowledge, not previously been investigated. Three groups have demonstrated that vesicular cholesterol precipitated faster than micellar cholesterol [3,4,12]. The present study indicates that micellar cholesterol also precipitates via the vesicular pathway through the spontaneous neoformation of vesicles and a progressive shift of micellar cholesterol to these newly formed vesicles. Thus, when native biles were incubated at 37°C, the proportion of vesicular cholesterol increased progressively from 15.5 :!: 18.6 at zero time to 28.0 + 12.5 at NT. When the two carriers, vesicles and micelles, were separately incubated, the amount of cholesterol in vesicles remained stable until nucleation and neoformation of micelles did not occur. However, in the isolated micellar fraction, a different sequence of events was demonstrated: vesicles which were not present at zero time began to appear and contained a progressively larger proportion of initially micellar cholesterol. This proportion was highest at nucleation time. Preliminary data not included in the present study show that following nucleation, a progressive decrease in vesicular cholesterol is seen in whole bile and in the isolated micellar fraction. This is compatible with other observations showing that cholesterol precipitates from vesicles [3,5]. Biochemical analysis confirmed that the newly formed vesicles in the isolated micellar fraction have a similar cholesterol/phosphofipid ratio (similar to 'old' vesicles present in the native bile. Conversely, the micellax fraction had a cholesterol/phospholipid ratio of approx. 0.3 at zero time as well as at nucleation time. These chemical data thus support the chromatographic data. Thus, it was shown that vesicles were initially absent in the micellar fraction and that they subsequently appeared in this fraction, particularly towards nucleation time. These newly formed vesicles were shown by two methods - chromatography and chemical analysis - to be similar to the 'old' vesicles initially present in bile. In some samples, it was also confirmed by electron microscopy. All these observations are compatible with the assumption that vesicles are the final common pathway for cholesterol precipitation in bile. Initial cholesterol nucleation occurs from vesicles originally present and as time goes on, micellar cholesterol is transferred into newly formed vesicles, and thus the process of cholesterol precipitation continues. Further experiments are needed to determine whether any biliary cholesterol precipitates directly from micelles. These data and interpretations are in agreement with previous observations using video-enhanced
249 microscopy, which showed that cholesterol crystals were intimately associated with vesicular aggregates and that rapid neoformation of vesicles was observed in biles of patients with gallstones [5]. These data are also compatible with early pioneering experiments in model systems [13-15], where liquid crystals (presumably vesicular aggregates) were observed in the metastable phase in which rapid cholesteroi crystal formation was observed.
Acknowledgments The authors appreciate the help and biochemical assistance of Mrs. R. Rosenberg, and the expert assistance of Mr. S. Dabush in electron microscopy. This study was supported by the Reka Foundation Tel-Aviv and the Germalfis-Kaufman Chair of Gastroenterology, Tel-Aviv University. References 1 Somjen, G.J. and Gilat, T. (1983) FEBS Lett. 156, 266-.268. 2 Pattinson, N.R. (1985) FEBS Lett. 181, 339b1342.
3 Peled, Y., Halpern, Z., Baruch, R. and Gilat, T. (1988) Hepatology 8, 914-918. 4 I-iarvey, P.R.C., Somjen, G.J., Lichtenberg, M.S., Petrunka, C., Gilat, T. and Strasberg, S.M. (1987) Biochim. Biophys. Acta 921, 198-204. 5 Halpern, Z., Dudley, M.A., Kibe, A., Lynn, M.P., Breuer, A.C. and Holzbach, R.T. (1986) Gastroenterology 90, 875-885. 6 Somjen, G.J. and Gilat, T. (1985) J. Lipid Res. 26, 699-704. 7 Holan, K.R., Holzbach, R.T., Hermann, R.E., Cooperman, A.M. and Claffey, N.J. (1979) Gastrco,,terology 77, 611-617. 8 Talalay, P. (1960) Methods Bioc,~,m. Anal. 8, 119-143. 9 Abbell, L.L., Levy, B.B., Brodie, B.B. and Kendall, F.E. (1952) J. Biol. Chem. 195, 357-366. 10 Bartlett, G.R. (1959) J. Biol. Chem. 234, 466-468. 11 Carey, M.C. (1979) J. Lipid Res. 19, 945-955. 12 Lee, S.P., Park, H.Z., Madani, H. and Kaler, E.W. (1987) Am. J. Physiol. 252, 6374-6383. 13 SmaU, D.M. and Bourges, M. (1966) Mol. Cryst. Liq. Cryst. 1541-1561. 14 Olszewski, M.F., Holzbach, R.T., Saupe, A. and Brown, G.H. (1973) Nature 242, 376-377. 15 Holzbach, R.T. and Corbusier, C. (1978) Biochim. Biophys. Acta 528, 436-444.