- Email: [email protected]
Effects of ethanol on the secretion of hepatic secretory protein in rat alcoholic liver injury
Alcohol, Vol. 8. pp. 433-437. PergamonPress plc. |991. Printed in the U.S.A.
0741~329/91 $3.00 + .00
Effects of Ethanol on the Secretion of Hepatic Secretory Protein in Rat Alcoholic Liver Injury YOSHIRO MATSUDA,
AKIRA TAKADA,
SHUJIRO TAKASE AND MINORU YASUHARA
Division o f Gastroenterology, Department o f Internal Medicine, Kanazawa Medical University Uchinada, Ishikawa, 920-02, Japan R e c e i v e d 29 A u g u s t 1990; A c c e p t e d 8 M a y 1991 MATSUDA, Y., A. TAKADA, S. TAKASE AND M. YASUHARA. Effects of ethanol on the secretion of hepatic secretory protein in rat alcoholic liver injar3'. ALCOHOL 8(6l 433--437. 1991.--It has been pointed out that one of the pathogenetic causes of alcoholic liver injury is the hepatocytic accumulation of exportable proteins due to a decrease in hepatic microtubules caused by acetaldehyde. To confirm and extend this secretory protein accumulation in the hepatocytes, the effects of alcohol treatment on the intracellular transport of secretory protein in the hepatocyte was studied using radioisotope-labeled leucine and fucose. Acute ethanol administration to rats did not show any effects on intrahepatocytic transport and secretion of transferrin. In alcohol pyrazole hepatitis rats, the secretion of transferrin labeled with both radioactive leucine and fucose into the serum was signi.ficantly delayed. Delaying in the secretion of fucose-labeled transferfin was more prominent than in leucine-labeled transfemn. This secretory inhibition was accompanied by a corresponding increase in the hepatic retention of both leucine- and fucose-labeled transferrin. At the time of the maximum inhibition of secretion, radioisotope labeled transferrin mainly retained in the Golgi apparatus. These results indicated that movement of secretory proteins along the secretory pathway impaired in alcoholic liver injury and that accumulation of the secretory proteins might play an important role in the development of alcoholic liver injury. Alcoholic liver injury
Impaired protein secretion
Golgi apparatus
At 15, 30, 45, 60, 90 and 120 minutes after injection of the labels, blood samples were collected from the retroorbital sinus. Each 5 rats of ethanol- and glucose-treated groups were killed by decapitation at 15 and 120 minutes respectively, after injection of the labels. Their livers were then perfused with ice-cold saline to remove contaminating serum proteins.
IT has been shown that the ballooning of the hepatocytes, which is one of the histological characteristics of alcohol liver injury, was caused by the accumulation of the exportable proteins, such as albumin or transferrin, in the hepatocytes (I, 5, 6). This protein accumulation in the hepatocytes was secondary to acetaldehyde-induced impairment of microtubular function (7,9). In our previous report, transferrin was retained in the Golgi apparatus of the hepatocytes (10), and desialoglycoproteins were found in the serum in both human and rat alcoholic liver injuries (10,11). These results suggested that the impaired Golgi functions, especially the impaired glycosylation of the glycoproteins, were primarily involved in the ethanol-induced secretory defect in the liver. In the present study, the effects of acute and chronic alcohol treatment on the intracellular transport of the secretory protein in the rat hepatocytes were studied using radioisotopelabeled amino acid and sugar to confirm and extend the previous observations.
Animal Procedures in the Chronic Experiment Wistar strain male rats (weighing about 150 g) were divided into two groups according to the diets. The detail of the diets was described in the previous paper (9). One group of 30 rats was fed with an alcohol (36% of total calories) liquid diet containing 35% of total calories as fat (high-fat diet) and the other group of 30 rats was fed with an alcohol liquid diet containing 15% of total calories as fat (low-fat diet). As the control groups, 30 rats were fed with a control (isocaloric sucrose) diet containing high fat. Each group of rats was further subdivided into two subgroups with or without pyrazole treatment. Two mmole of pyrazole was added to 1000 ml of the liquid diet in one group, but not in the other group. Since in our previous experiment the rats fed control diet containing either high or low fat showed no abnormal histological and biochemical findings in the liver (9), rats fed with the high-fat, control diet were used as a control in this study. Consequently, six experimental groups of 5 rats were studied as follows: alcohol-pyrazole high fat (A1.Py-HF), alcohol alone high fat (AI-HF), corresponding 2 low-fat groups (A1.Py-LF, A1-LF), pyrazole alone high fat (Py-HF) and control high fat (C-HF). All groups of rats were pair-fed with AI.Py-HF group for 12 weeks.
METHOD
Animal Procedures in the Acute Experhnent Wistar strain rats (Sankyo Lab, Tokyo), weighing about 150 g, were maintained on a standard laboratory diet with water ad lib. The fasted rats were given ethanol (3 g/kg body weight) by gastric intubation, whereas the control rats received an isocaloric dose of glucose. Two hours after ethanol or glucose administration, 3H-leucine (50 ixCi/100 g body weight, Amersham, USA) and 14C-fucose (8 txCi/100 g body weight, Amersham, USA) were injected simultaneously into the dorsal vein of the penis.
433
434
MATSUDA ET AL.
:)lt-leucinc
dpm/rnl
aC-fucose
t,O
1 4C_fucose
3H-leucine
dpm/ml
20'
n=6
n~6
dprn/nl .'.0
20
16
30
30
~,.-'1
12 20 :'0
8 1o
~
¢
15 30
60
90
t20 "'"
15 30
o
60
control
go
group
~2o .,. 15 30
FIG. I. Incorporation of 3H-leucine and "~C-fucose into serum transferrin in the rats treated with acute ethanol administration. There are no significant differences in incorporation between the ethanol-treated and control groups.
At the end of the 12th week, rats were given one-third volume of daily consumption of their own diets in the early morning (ethanol content was about 3 g/kg body weight) by gastric intubation. Two hours after diet intubation, radioisotope-labeled leucine and fucose were injected intravenously with the same dose as the acute experiment. Blood samples were collected from 5 rats of each group at 15, 30, 45, 60, 90, 120 minutes after injection of the labels and the livers of the rats were removed at 120 minutes after injection of the labels. Another 5 rats of each group were killed by decapitation at 15 and 30 minutes after injection of the labels. All livers were perfused with ice-cold saline.
Subcelhdar Fractionation of the Hepatocytes The subcellular fractions of the hepatocytes were obtained by the method of Volentine et al. (12) from the rat livers of the chronic experiment. Livers were homogenized in 5 volumes of ice-cold 0.25 M sucrose-10 mM Tris-HCI (pH 7.4) buffer solution. After cell debris, nuclei and mitochondria were cleared by centrifugation at 10,000 × g for ten minutes, total microsomal fractions (endoplasmic reticulum and Golgi apparatus) were obtained by centrifugation at 105,000 × g for 90 minutes. The re-
60
90
• .lll',lh l",ll {) ) h q h F h t , I','l c ,I."
:20 rnln • how F.,t. •
Hiqh ~ 1 .
~5 ,Xloz.hoI Ah~l
a H - l e u c i n e x
suiting 105,000 x g supernatants represented the cytosolic fraction. The microsomal pellet was resuspended and the concentration of sucrose was increased to 1.3 M. An aliquot of this suspension was used as a load fraction under the layers of 1.10 M and 0.25 M sucrose with a bottom cushion of 2.0 M sucrose. Upon centrifugation in a swing rotor (Beckman SW 271 for 3 hours at 82,000 x g. a band formed at the 0.25-1.10 M sucrose interface. This band was collected and centrifuged at 105,000 × g for 90 minutes. The pellet of this centrifugation represented the fraction enriched in Golgi-derived elements and vesicles. To obtain the smooth endoplasmic reticulum (SERI and rough endoplasmic reticulum (RER) fractions, the load fraction from the previous Golgi separation was adjusted with TKM buffer (10 mM Tris-HC1 pH 7.4, 25 mM KCI and 5 mM MgCI_,) to a concentration with less than 1.0 M sucrose. The pellet was resuspended in 12 ml of 2.0 M sucrose and TKM was used as a load fraction under a 12-ml layer of 0.44 M sucrose and a bottom layer of 2.0 M sucrose. This gradient was then subjected to centrifugation for 24 hours at 82,000 × g in a swing bucket rotor (Beckman SW27). The SER was located at the 0.#4--I.25 M sucrose interphase and the RER was located 1.25-2.00 M sucrose inter-
I":_'::} ~ttgh rat l l i g h Fat, Py bow Fat, A1 fflllIlll nigh Fat, m ~ ]
MC--fucose dp~?-H~.r x
1z~l t'~n
FIG. 3. Incorporation of 3H-leucine and "C-fucose into the serum transferrin in rats of the chronic experimental groups. Incorporation rates of both labels into the serum transferrin are significantly lower in the AI.Py-HF group than those in the other groups.
3
Io 4
90
P a L ,~lcT~l.l,yra:.,l.~ ~ Illnh ~ t , A I ~ o l . P ' t W a ~ I q
dpm/g-liver x 104 3.5 dllm/q.lt~
60
)0
• :1~
* 19
Io 4
Low r a t , ~ t . ~
35
W2.5 3.
125
* l,<0.05
F
S~
,-4
~11.5
2, ol
$.
1
~t
0.5 g 15mi
120m&n []
1~ t d . n
12G.d.n
control group .olcohol group
FIG. 2. Changes of ~H-leucine and ~'~C-fucoseof the hepatic transferfin in the rats treated with acute ethanol administration. There are no significant differences in activities between the ethanol and control groups.
0 15mi_n
1 20rain 30rain A1 : Alcohol Py:
[~zole
FIG. 4. Sequential changes of 3H-activity of the hepatic transferrin in the chronic experimental groups. The activity is significantly higher in the AI.Py-HF group than those in the other groups during experimental period. *p<0.05 compared to the other groups.
E F F E C T S O F E T H A N O L ON H E P A T I C S E C R E T O R Y P R O T E I N
435
TABLE I all-ACTIVITIES OF TRANSFERRIN IN TIlE SUBCELLULAR FRACTIONS AT 15 MINUTES AFTER INJECTION OF LABELS IN THE CHRONIC EXPERIMENTAL GROUPS (dpm/g LIVER z. 10"8) Alcoholic
Nonalcoholic
Pyrazole Fraction Total homogenate Cytosol Rough ER Smooth ER Golgi
High Fat 3.25 0.60 1.32 1.00 1.18
± s-'± ±
Nonpyrazole Low Fat
0.30* 0.09 0.05* 0.31 0.09
2.43 0.55 0.93 0.91 1.02
High Fat
m 0.22 "4- 0.08 ± 0.17 -'- 0.27 ± 0.07
2.24 0.32 0.69 0.71 0.73
- 0.11 +-- 0.10 - 0.20 +__ 0.27 __. 0.12
Low Fat 2.20 0.26 0.53 0.77 0.81
± 0.12 ± 0.08 -'- 0.18 ± 0.29 __. 0.08
Pyrazole
Nonpyrazole
High Fat
Low Fat
2,05 0.36 0.63 0,59 0.83
± 0.13 ± 0.14 ± 0.21 _+ 0.25 __. 0.10
1.92 0.32 0.56 0.51 0.67
- 0.12 --- 0.11 ~ 0.12 -'- 0.13 _+ 0.11
Values are mean ± SD of each 5 rats. *p<0.05 compared to the other groups. ER: endoplasmic reticulum.
labeled transferrin in the rat s e r u m appeared at 30 m i n u t e s and increased until 120 minutes after injection of the labels, which is the end o f the experiment. Incorporation rates o f 3H-leucine and ~'~C-fucose into the s e r u m transferrin tended to be lower in the ethanol-treated group than in the control group through 30 to 120 minutes after injection o f labels. However, the difference during experimental period was not statistically significant. Incorporation rates o f SH-leucine and ~aC-fucose into the liver transferrin were detected at 15 m i n u t e s after injection o f labels and the rates decreased at 120 m i n u t e s after the injection. T h e incorporation rates of both labels at 15 m i n u t e s after injection o f labels were significantly higher than those at 120 minutes after injection o f labels. However, the incorporation rates o f 3H-leucine and 14C-fucose into the liver transferrin at 15 and 120 minutes after injection o f labels were not different between the ethanol and control groups.
phase. T h e bands at the two interphases were diluted respectively with sucrose to a concentration of 0.25 M and centrifuged at 1 0 5 , 0 0 0 × g for 60 m i n u t e s to isolate pellets o f the SER and RER.
Determination of Radiolabeled Secretory Proteins Radiolabeled hepatic transferrin in the s e r u m and in 0 . 5 % s o d i u m deoxycholate extracts o f the liver and isolated subcellular fractions were m e a s u r e d by quantitative immunoprecipitation, u s i n g a goat antirat transferrin s e r u m by the method o f Volentine et al. (12). The activities o f 3H- and t'~C- of the immunoprecipitated transferrin were determined s i m u l t a n e o u s l y by the m e t h o d o f K o b a y a s h i et al. (4).
Statistics Chronic Alcoholic Liver hzjul3, in Rats
T h e results are e x p r e s s e d as m e a n ± standard deviation (SD). C o m p a r i s o n s were evaluated using S t u d e n t ' s t-test.
At the end o f the 12th week, ballooned hepatocytes were found at the centrolobular areas with a few hepatocytic necrosis in the A I . P y - H F group, but these c h a n g e s were very slight in the A I . P y - L F group and no abnormal c h a n g e s were found in the other groups, except s o m e fat droplets in the hepatocytes as reported previously (9). T h e serial c h a n g e s o f incorporation of 3H-leucine and ~4C-fucose into the s e r u m transferrin are s h o w n
RESULTS
Acute Experiment Acute effects o f ethanol administration on hepatic protein secretion are s h o w n in Figs. 1 and 2. T h e leucine- and fucose-
TABLE 2 3H-ACTIVITIES OF TRANSFERRIN IN THE SUBCELLULAR FRACTIONS AT 30 MINUTES AFTER INJECTION OF LABELS IN THE CHRONIC EXPERIMENTAL GROUPS (dpm/g LIVER x 10"~) Nonalcoholic
Alcoholic Pyrazole Fraction Total homogenate Cytosol Rough ER Smooth E R Golgi
High Fat 2.85 0.41 0.68 0.64 1.72
--. ± ± ± ±
0.22* 0.05 0.12 0.05 0.28*
Values are mean ± SD of each 5 rats. *p<0.02 compared to the other groups. ER: endoplasmic reticulum.
Nonpyrazole Low Fat
2.18 0.35 0.49 0.59 1.21
--- 0.19 ± 0.08 ± 0.13 ± 0.04 ± 0.17
High Fat 2.00 0.38 0.51 0.38 0.85
--. ± ± --. --.
0.19 0.05 0.09 0.07 0.09
Low Fat 2.16 0.28 0.50 0.42 0.80
-+ ± --. --. ---
0.14 0.05 0.05 0.09 0.08
Pyrazole
Nonpyrazole
High Fat
Low Fat
1.92 0.31 0.59 0.34 0.75
--+ 0.17 +-- 0.10 * 0.05 ± 0.04 ± 0.05
1.85 0.29 0.43 0.33 0.64
--- 0.18 ± 0.08 ± 0.10 ± 0.03 __. 0.09
436
MATSUDA ET AL.
~. !j,.:~
:/
3.5-
cantly higher than those in the older groups. At 15 minutes after the injection of labels, when no radioactivities of 3H and ~4C were detected in the serum transferrin, 3H-activity was detected in transferrin of all subcellular fractions of the liver (Table 1). In the total homogenate the activity in transferrin of the A1.Py-HF group was significantly higher than those in the other groups. The activities in transferrin of the cytosolic fractions were lower than those in the other fractions in all groups. The activities in transferrin of the rough endoplasmic reticulum, smooth endoplasmic reticulum and Golgi fractions showed no significant difference among the groups, nor between those fractions, except that 3H-activity in transferrin of the rough endoplasmic reticulum fraction, but not in the Golgi fraction, was significantly higher in the AI.Py-HF group than those in the other groups. At 30 minutes after the injection of labels, 3Hactivities in the Golgi fraction were higher than those in the other subcellular fractions in all groups. Among the experimental groups, the AI.Py-HF group showed significantly higher 3Hactivity of transferrin only in the total homogenate and Golgi fraction than those in the other groups (Table 2). Changes of ~4C-activity in the transferrin of the subcellular fractions of the hepatocytes were the same as those of the labeled leucine (Table 3. Table 4).
High Fat ~ High Fat, Py [''~ Low Fat, A1
dpm/g.liver x 10 4
:,}
High ~t, ~t
to~ Fat, m+~ ~_~
~ .:/
.:~oo5 **: 0<0.02
0.5] 15rain
i
i
30ndn
120min
Ai: ALcohol
py: p.yrazole
FIG. 5. Sequential changes of L4C-acitivity of the hepatic transferrin in the chronic experimental groups. The activity is significantly higher in the AI.Py-HF group than those in the other groups during experimental period. *p<0.05, **p<0.02 compared to the other groups.
in Fig. 3. 3H- and ~4C-activities were not found in the serum transferrin at 15 minutes after the injection of labels, then both activities appeared at 30 minutes in the serum transferrin and reached a plateau at 90 minutes in all groups. Incorporation rates of 3H-leucine and ~4C-fucose into the serum transferrin in the AI.Py-HF group were significantly lower than those in the other groups during the observation periods except for at 60 and 90 minutes in 3C-leucine. Incorporation rate of lac-fucose into the serum transferrin in the A1.Py-LF group was significantly lower than those in nonalcoholic groups during the observation period except at 60 minute, however, incorporation rate of SH-leucine was not different from those in the nonalcoholic group. The inhibition of incorporation of the labels in the A1.Py-HF group was more prominent in 14C-fucose than in 3H-leucine. The serial changes of 3H- and '4C-activities in hepatic transferrin are shown in Figs. 4 and 5. Both activities showed high values at 15 and 30 minutes, and at 120 minutes the activities clearly decreased in all groups. 3H-Activity in hepatic transferrin was significantly higher in the AI.Py-HF group than those in the other groups during the observation periods. ~4C-Activity in the hepatic transferrin in the AI.Py-HF group was also signifi-
DISCUSSION
In the present study, acute ethanol administration did not have any effects on the glycoprotein secretion from the hepatocytes. Volentine et al. (12) reported that acute ethanol administration inhibited the incorporation of the radiolabeled leucine and fucose in the serum transferrin. However, the dose of ethanol in their experiment was 6 g/kg, which was larger than that of the present study (3 g/kg). This may account for the difference between the two studies. Incorporation rates of the radiolabeled leucine and fucose into the serum transferrin were significantly decreased in the A1.Py-HF group. These results indicated that secretion of protein, especially glycoprotein, from the hepatocytes was inhibited in alcoholic liver injury. Glycoproteins destined for export from the hepatocytes are synthesized in the rough endoplasmic reticulum and then transported through the smooth endoplasmic reticulum to the transitional element of the Golgi complex. These newly synthesized glycoproteins are added to the terminal carbohydrate chain (glycosylation) in the Golgi apparatus and then are secreted into the blood (2,3). Fucose is a terminal carbohydrate
TABLE 3 t4C-ACTIVITIES OF TRANSFERRIN IN THE SUBCELLULAR FRACTIONS AT 15 MINUTES AFTER INJECTION OF LABELS IN THE CHRONIC EXPERIMENTAL GROUPS (dpm/g LIVER × 10'~)
Alcoholic
Nonalcoholic
Pyrazole Fraction Total homogenate Cytosol Rough ER Smooth ER Golgi
High Fat 3.78 0.66 1.48 1.02 1.72
--- 0.20* +_ 0.25 ± 0.22* ± 0.25 +- 0.38
Values are mean __. SD of each 5 rats. *p<0.05 compared to the other groups. ER: endoplasmic reticulum.
Nonpyrazole Low Fat
3.32 0.62 0.91 0.99 1.51
__. 0.12 ___ 0.18 +_ 0.24 _+ 0.22 ___ 0.27
High Fat 2.81 0.38 0.71 0.71 1.40
--- 0.23 ± 0.13 ± 0.32 +- 0.27 ± 0.22
Low Fat 2.85 0.31 0.61 0.77 1.38
+-- 0.24 - 0.12 --- 0.23 - 0.29 ± 0.38
Pyrazole
Nonpyrazole
High Fat
Low Fat
2.68 0.33 0.67 0.62 1.34
__. 0.33 _ 0.10 ± 0.22 ± 0.24 _ 0.22
2.72 0.30 0.52 0.67 1.15
_ 0.32 ± 0.08 ± 0.28 ± 0.28 --- 0.20
EFFECTS OF ETHANOL ON HEPATIC SECRETORY PROTEIN
437
TABLE 4 taC-ACTIVITIES OF TRANSFERRIN IN THE SUBCELLULARFRACTIONS AT 30 MINUTES AFTER INJECTION OF LABELS IN THE CHRONIC EXPERIMENTALGROUPS (dpm/g LIVER x 10'~) Alcoholic
Nonalcoholic
Pyrazole Fraction Total homogenate Cytosol Rough ER Smooth ER Golgi
High Fat 3.42 0.48 0.71 1.01 2.20
--- 0.14" +-- 0.12 --- 0.22 --- 0.39 ± 0.41"
Nonpyrazole Low Fat
3.12 0.30 0.50 0.85 1.45
± 0.15 ± 0.09 ± 0.10 ±- 0.21 ± 0.25
High Fat 2.95 0.23 0.66 0.66 1.16
• ~ ~ _ ~
0.27 0.11 0.09 0.11 0.23
Low Fat 2.99 0.17 0.47 0.44 1.02
--- 0.24 __. 0.13 _+_0.10 -,- 0.35 --- 0.20
Pyrazole
Nonpyrazole
High Fat
Low Fat
2.81 0.19 0.53 0.63 1.22
± 0.33 ___ 0.09 __. 0.10 __. 0.15 ± 0.18
2.78 0.16 0.50 0.50 1.01
± 0.41 ±- 0.12 +_ 0.13 --- 0.22 ± 0.13
Values are mean __. SD of each 5 rats. *p<0.02 compared to the other groups. ER: endoplasmic reticulum.
chain of secretory glycoproteins. In the present study, incorporation of fucose was inhibited more prominently than that of leucine in A1.Py-HF group. This result indicated that the glycosylation of glycoprotein in the hel~atocytic Golgi apparatus was impaired in alcoholic liver injury. The results of the present study also indicated that newly synthesized glycoproteins move from rough endoplasmic reticulum to Golgi apparatus and secrete into the blood taking about a 2-hour process. In the present study, both 3H-leucine and ~'~C-fucose activities in transferrin of the endoplasmic reticulum in the early period of the experiment and in transferrin of the Golgi apparatus in the late period were significantly higher in the AI.Py-HF group than those in the other groups. These results indicated that the intracellular movement of glycoprotein was delayed, and the terminal glycosylation in the Golgi apparatus and finally secretion of glycoprotein were impaired in alcoholic liver injury. These results also indicated that the appearance of carbohydrate-deficient transferrin in
the human alcoholics and experimental rat alcoholic liver injury (10,1 l) might he due to an impaired glycosylation of transferrin in the hepatocytic Golgi apparatus. The mechanisms of inhibition of glycoprotein secretion from hepatocytes following an impairrnent of glycosylation of glycoprotein in heptocytic Golgi apparatus are still unclear. Recently, it has been reported that microtubules play an important role in the function of the Golgi apparatus, including glycosylation of glycoprotein (8). We have reported that hepatic microtubules were decreased and transferrin was retained in the hepatocytes in experimental and human alcoholic liver injuries (5,6). Moreover, we have reported that a decrease in hepatic microtubules could be due to hepatic acetaldehyde accumulation (7). These results suggested that the retention of glycoprotein in the hepatocyte, which is a precursor of hepatocytic necrosis in alcoholic liver injuries, attributed to the impairment of glycosylation and secretion of glycoproteins in the Golgi apparatus due to a decrease in microtubules.
REFERENCES 1. Baraona, E.; Leo, M. A.; Borovsky, S. A.; Lieber, C. S. Alcoholic hepatomegaly: accumulation of protein in the liver. Science 190: 794--795; 1975. 2. Griffiths, G.; Simons, K. The trans Golgi network: Sorting at the exit site of the Golgi complex. Science 234:438--443; 1986. 3. Hanover, J. A.; Lennar, W. J. Transmembrane assembly of membrane and secretory proteins. Arch. Biochem. Biophys. 211:1-19; 1981. 4. Kobayashi, Y.; Maudsley, D. V. Practical aspects of double isotope counting. In: Bransome. E. D.. ed. The current status of liquid scintillation counting. New York: Grune & Stratton; 1970:76-85. 5. Matsuda, Y.; Takase, S.; Takada, A.; Sato, H.; Yasuhara, M. Comparison of ballooned hepatocytes in alcoholic and nonalcoholic liver injury in rats. Alcohol 2:303-308; 1985. 6. Matsuda, Y.; Takada, A.; Sato, H.; Yasuhara. M.; Takase, S. Comparison between ballooned hepatocytes occurring in human alcoholic and nonalcoholic liver diseases. Alcohol.: Clin. Exp. Res. 9:366-370; 1985. 7. Matsuda, Y.; Baraona, E.; Salaspuro, M.; Lieber, C. S. Effects of ethanol on liver microtubules and Golgi apparatus. Lab. Invest. 41:
455--463; 1979. 8. Mitranic, M. M.; Boggs, J. M.; Moscallello, M. A. An effect of colchicine on galactosyl and sialyltransferase of rat Golgi membrane. Biochem. Biophys. Acta 672:57--64; 1981. 9. Takada, A.; Matsuda, Y.; Takase, S. Effect of dietary fat on alcohol-pyrazole hepatitis in rats: The pathogenetic role of the nonalcohol dehydrogenase pathway in alcohol-induced hepatic cell injury. Alcohol.: Clin. Exp. Res. 10:403--41I; 1986. 10. Takada, A.; Matsuda, Y.; Sato, H.; Takase, S. Alcoholic liver injury and hepatic secretory proteins: Roles of microtubules and Golgi apparatus. In: Oda, T.; Okuda, K., eds. Falk Symposium in Tokyo, Japan. New trends in hepatology. Tokyo: Medical Tosho Co. Ltd.; 1986:33.-42. 11. Takase, S.; Takada, A.; Tsutsumi, M.; Matsuda, Y. Biochemical markers of chronic alcoholism. Alcohol 2:405--410; 1985. 12. Volentine, G. D.; Tuma, D. J.; Son'ell, M. F. Subcellular location of secretory proteins retained in the liver during ethanol-induced inhibition of hepatic protein secretion in the rat. Gastroenterology 90: 158-165; 1986.
0741~329/91 $3.00 + .00
Effects of Ethanol on the Secretion of Hepatic Secretory Protein in Rat Alcoholic Liver Injury YOSHIRO MATSUDA,
AKIRA TAKADA,
SHUJIRO TAKASE AND MINORU YASUHARA
Division o f Gastroenterology, Department o f Internal Medicine, Kanazawa Medical University Uchinada, Ishikawa, 920-02, Japan R e c e i v e d 29 A u g u s t 1990; A c c e p t e d 8 M a y 1991 MATSUDA, Y., A. TAKADA, S. TAKASE AND M. YASUHARA. Effects of ethanol on the secretion of hepatic secretory protein in rat alcoholic liver injar3'. ALCOHOL 8(6l 433--437. 1991.--It has been pointed out that one of the pathogenetic causes of alcoholic liver injury is the hepatocytic accumulation of exportable proteins due to a decrease in hepatic microtubules caused by acetaldehyde. To confirm and extend this secretory protein accumulation in the hepatocytes, the effects of alcohol treatment on the intracellular transport of secretory protein in the hepatocyte was studied using radioisotope-labeled leucine and fucose. Acute ethanol administration to rats did not show any effects on intrahepatocytic transport and secretion of transferrin. In alcohol pyrazole hepatitis rats, the secretion of transferrin labeled with both radioactive leucine and fucose into the serum was signi.ficantly delayed. Delaying in the secretion of fucose-labeled transferfin was more prominent than in leucine-labeled transfemn. This secretory inhibition was accompanied by a corresponding increase in the hepatic retention of both leucine- and fucose-labeled transferrin. At the time of the maximum inhibition of secretion, radioisotope labeled transferrin mainly retained in the Golgi apparatus. These results indicated that movement of secretory proteins along the secretory pathway impaired in alcoholic liver injury and that accumulation of the secretory proteins might play an important role in the development of alcoholic liver injury. Alcoholic liver injury
Impaired protein secretion
Golgi apparatus
At 15, 30, 45, 60, 90 and 120 minutes after injection of the labels, blood samples were collected from the retroorbital sinus. Each 5 rats of ethanol- and glucose-treated groups were killed by decapitation at 15 and 120 minutes respectively, after injection of the labels. Their livers were then perfused with ice-cold saline to remove contaminating serum proteins.
IT has been shown that the ballooning of the hepatocytes, which is one of the histological characteristics of alcohol liver injury, was caused by the accumulation of the exportable proteins, such as albumin or transferrin, in the hepatocytes (I, 5, 6). This protein accumulation in the hepatocytes was secondary to acetaldehyde-induced impairment of microtubular function (7,9). In our previous report, transferrin was retained in the Golgi apparatus of the hepatocytes (10), and desialoglycoproteins were found in the serum in both human and rat alcoholic liver injuries (10,11). These results suggested that the impaired Golgi functions, especially the impaired glycosylation of the glycoproteins, were primarily involved in the ethanol-induced secretory defect in the liver. In the present study, the effects of acute and chronic alcohol treatment on the intracellular transport of the secretory protein in the rat hepatocytes were studied using radioisotopelabeled amino acid and sugar to confirm and extend the previous observations.
Animal Procedures in the Chronic Experiment Wistar strain male rats (weighing about 150 g) were divided into two groups according to the diets. The detail of the diets was described in the previous paper (9). One group of 30 rats was fed with an alcohol (36% of total calories) liquid diet containing 35% of total calories as fat (high-fat diet) and the other group of 30 rats was fed with an alcohol liquid diet containing 15% of total calories as fat (low-fat diet). As the control groups, 30 rats were fed with a control (isocaloric sucrose) diet containing high fat. Each group of rats was further subdivided into two subgroups with or without pyrazole treatment. Two mmole of pyrazole was added to 1000 ml of the liquid diet in one group, but not in the other group. Since in our previous experiment the rats fed control diet containing either high or low fat showed no abnormal histological and biochemical findings in the liver (9), rats fed with the high-fat, control diet were used as a control in this study. Consequently, six experimental groups of 5 rats were studied as follows: alcohol-pyrazole high fat (A1.Py-HF), alcohol alone high fat (AI-HF), corresponding 2 low-fat groups (A1.Py-LF, A1-LF), pyrazole alone high fat (Py-HF) and control high fat (C-HF). All groups of rats were pair-fed with AI.Py-HF group for 12 weeks.
METHOD
Animal Procedures in the Acute Experhnent Wistar strain rats (Sankyo Lab, Tokyo), weighing about 150 g, were maintained on a standard laboratory diet with water ad lib. The fasted rats were given ethanol (3 g/kg body weight) by gastric intubation, whereas the control rats received an isocaloric dose of glucose. Two hours after ethanol or glucose administration, 3H-leucine (50 ixCi/100 g body weight, Amersham, USA) and 14C-fucose (8 txCi/100 g body weight, Amersham, USA) were injected simultaneously into the dorsal vein of the penis.
433
434
MATSUDA ET AL.
:)lt-leucinc
dpm/rnl
aC-fucose
t,O
1 4C_fucose
3H-leucine
dpm/ml
20'
n=6
n~6
dprn/nl .'.0
20
16
30
30
~,.-'1
12 20 :'0
8 1o
~
¢
15 30
60
90
t20 "'"
15 30
o
60
control
go
group
~2o .,. 15 30
FIG. I. Incorporation of 3H-leucine and "~C-fucose into serum transferrin in the rats treated with acute ethanol administration. There are no significant differences in incorporation between the ethanol-treated and control groups.
At the end of the 12th week, rats were given one-third volume of daily consumption of their own diets in the early morning (ethanol content was about 3 g/kg body weight) by gastric intubation. Two hours after diet intubation, radioisotope-labeled leucine and fucose were injected intravenously with the same dose as the acute experiment. Blood samples were collected from 5 rats of each group at 15, 30, 45, 60, 90, 120 minutes after injection of the labels and the livers of the rats were removed at 120 minutes after injection of the labels. Another 5 rats of each group were killed by decapitation at 15 and 30 minutes after injection of the labels. All livers were perfused with ice-cold saline.
Subcelhdar Fractionation of the Hepatocytes The subcellular fractions of the hepatocytes were obtained by the method of Volentine et al. (12) from the rat livers of the chronic experiment. Livers were homogenized in 5 volumes of ice-cold 0.25 M sucrose-10 mM Tris-HCI (pH 7.4) buffer solution. After cell debris, nuclei and mitochondria were cleared by centrifugation at 10,000 × g for ten minutes, total microsomal fractions (endoplasmic reticulum and Golgi apparatus) were obtained by centrifugation at 105,000 × g for 90 minutes. The re-
60
90
• .lll',lh l",ll {) ) h q h F h t , I','l c ,I."
:20 rnln • how F.,t. •
Hiqh ~ 1 .
~5 ,Xloz.hoI Ah~l
a H - l e u c i n e x
suiting 105,000 x g supernatants represented the cytosolic fraction. The microsomal pellet was resuspended and the concentration of sucrose was increased to 1.3 M. An aliquot of this suspension was used as a load fraction under the layers of 1.10 M and 0.25 M sucrose with a bottom cushion of 2.0 M sucrose. Upon centrifugation in a swing rotor (Beckman SW 271 for 3 hours at 82,000 x g. a band formed at the 0.25-1.10 M sucrose interface. This band was collected and centrifuged at 105,000 × g for 90 minutes. The pellet of this centrifugation represented the fraction enriched in Golgi-derived elements and vesicles. To obtain the smooth endoplasmic reticulum (SERI and rough endoplasmic reticulum (RER) fractions, the load fraction from the previous Golgi separation was adjusted with TKM buffer (10 mM Tris-HC1 pH 7.4, 25 mM KCI and 5 mM MgCI_,) to a concentration with less than 1.0 M sucrose. The pellet was resuspended in 12 ml of 2.0 M sucrose and TKM was used as a load fraction under a 12-ml layer of 0.44 M sucrose and a bottom layer of 2.0 M sucrose. This gradient was then subjected to centrifugation for 24 hours at 82,000 × g in a swing bucket rotor (Beckman SW27). The SER was located at the 0.#4--I.25 M sucrose interphase and the RER was located 1.25-2.00 M sucrose inter-
I":_'::} ~ttgh rat l l i g h Fat, Py bow Fat, A1 fflllIlll nigh Fat, m ~ ]
MC--fucose dp~?-H~.r x
1z~l t'~n
FIG. 3. Incorporation of 3H-leucine and "C-fucose into the serum transferrin in rats of the chronic experimental groups. Incorporation rates of both labels into the serum transferrin are significantly lower in the AI.Py-HF group than those in the other groups.
3
Io 4
90
P a L ,~lcT~l.l,yra:.,l.~ ~ Illnh ~ t , A I ~ o l . P ' t W a ~ I q
dpm/g-liver x 104 3.5 dllm/q.lt~
60
)0
• :1~
* 19
Io 4
Low r a t , ~ t . ~
35
W2.5 3.
125
* l,<0.05
F
S~
,-4
~11.5
2, ol
$.
1
~t
0.5 g 15mi
120m&n []
1~ t d . n
12G.d.n
control group .olcohol group
FIG. 2. Changes of ~H-leucine and ~'~C-fucoseof the hepatic transferfin in the rats treated with acute ethanol administration. There are no significant differences in activities between the ethanol and control groups.
0 15mi_n
1 20rain 30rain A1 : Alcohol Py:
[~zole
FIG. 4. Sequential changes of 3H-activity of the hepatic transferrin in the chronic experimental groups. The activity is significantly higher in the AI.Py-HF group than those in the other groups during experimental period. *p<0.05 compared to the other groups.
E F F E C T S O F E T H A N O L ON H E P A T I C S E C R E T O R Y P R O T E I N
435
TABLE I all-ACTIVITIES OF TRANSFERRIN IN TIlE SUBCELLULAR FRACTIONS AT 15 MINUTES AFTER INJECTION OF LABELS IN THE CHRONIC EXPERIMENTAL GROUPS (dpm/g LIVER z. 10"8) Alcoholic
Nonalcoholic
Pyrazole Fraction Total homogenate Cytosol Rough ER Smooth ER Golgi
High Fat 3.25 0.60 1.32 1.00 1.18
± s-'± ±
Nonpyrazole Low Fat
0.30* 0.09 0.05* 0.31 0.09
2.43 0.55 0.93 0.91 1.02
High Fat
m 0.22 "4- 0.08 ± 0.17 -'- 0.27 ± 0.07
2.24 0.32 0.69 0.71 0.73
- 0.11 +-- 0.10 - 0.20 +__ 0.27 __. 0.12
Low Fat 2.20 0.26 0.53 0.77 0.81
± 0.12 ± 0.08 -'- 0.18 ± 0.29 __. 0.08
Pyrazole
Nonpyrazole
High Fat
Low Fat
2,05 0.36 0.63 0,59 0.83
± 0.13 ± 0.14 ± 0.21 _+ 0.25 __. 0.10
1.92 0.32 0.56 0.51 0.67
- 0.12 --- 0.11 ~ 0.12 -'- 0.13 _+ 0.11
Values are mean ± SD of each 5 rats. *p<0.05 compared to the other groups. ER: endoplasmic reticulum.
labeled transferrin in the rat s e r u m appeared at 30 m i n u t e s and increased until 120 minutes after injection of the labels, which is the end o f the experiment. Incorporation rates o f 3H-leucine and ~'~C-fucose into the s e r u m transferrin tended to be lower in the ethanol-treated group than in the control group through 30 to 120 minutes after injection o f labels. However, the difference during experimental period was not statistically significant. Incorporation rates o f SH-leucine and ~aC-fucose into the liver transferrin were detected at 15 m i n u t e s after injection o f labels and the rates decreased at 120 m i n u t e s after the injection. T h e incorporation rates of both labels at 15 m i n u t e s after injection o f labels were significantly higher than those at 120 minutes after injection o f labels. However, the incorporation rates o f 3H-leucine and 14C-fucose into the liver transferrin at 15 and 120 minutes after injection o f labels were not different between the ethanol and control groups.
phase. T h e bands at the two interphases were diluted respectively with sucrose to a concentration of 0.25 M and centrifuged at 1 0 5 , 0 0 0 × g for 60 m i n u t e s to isolate pellets o f the SER and RER.
Determination of Radiolabeled Secretory Proteins Radiolabeled hepatic transferrin in the s e r u m and in 0 . 5 % s o d i u m deoxycholate extracts o f the liver and isolated subcellular fractions were m e a s u r e d by quantitative immunoprecipitation, u s i n g a goat antirat transferrin s e r u m by the method o f Volentine et al. (12). The activities o f 3H- and t'~C- of the immunoprecipitated transferrin were determined s i m u l t a n e o u s l y by the m e t h o d o f K o b a y a s h i et al. (4).
Statistics Chronic Alcoholic Liver hzjul3, in Rats
T h e results are e x p r e s s e d as m e a n ± standard deviation (SD). C o m p a r i s o n s were evaluated using S t u d e n t ' s t-test.
At the end o f the 12th week, ballooned hepatocytes were found at the centrolobular areas with a few hepatocytic necrosis in the A I . P y - H F group, but these c h a n g e s were very slight in the A I . P y - L F group and no abnormal c h a n g e s were found in the other groups, except s o m e fat droplets in the hepatocytes as reported previously (9). T h e serial c h a n g e s o f incorporation of 3H-leucine and ~4C-fucose into the s e r u m transferrin are s h o w n
RESULTS
Acute Experiment Acute effects o f ethanol administration on hepatic protein secretion are s h o w n in Figs. 1 and 2. T h e leucine- and fucose-
TABLE 2 3H-ACTIVITIES OF TRANSFERRIN IN THE SUBCELLULAR FRACTIONS AT 30 MINUTES AFTER INJECTION OF LABELS IN THE CHRONIC EXPERIMENTAL GROUPS (dpm/g LIVER x 10"~) Nonalcoholic
Alcoholic Pyrazole Fraction Total homogenate Cytosol Rough ER Smooth E R Golgi
High Fat 2.85 0.41 0.68 0.64 1.72
--. ± ± ± ±
0.22* 0.05 0.12 0.05 0.28*
Values are mean ± SD of each 5 rats. *p<0.02 compared to the other groups. ER: endoplasmic reticulum.
Nonpyrazole Low Fat
2.18 0.35 0.49 0.59 1.21
--- 0.19 ± 0.08 ± 0.13 ± 0.04 ± 0.17
High Fat 2.00 0.38 0.51 0.38 0.85
--. ± ± --. --.
0.19 0.05 0.09 0.07 0.09
Low Fat 2.16 0.28 0.50 0.42 0.80
-+ ± --. --. ---
0.14 0.05 0.05 0.09 0.08
Pyrazole
Nonpyrazole
High Fat
Low Fat
1.92 0.31 0.59 0.34 0.75
--+ 0.17 +-- 0.10 * 0.05 ± 0.04 ± 0.05
1.85 0.29 0.43 0.33 0.64
--- 0.18 ± 0.08 ± 0.10 ± 0.03 __. 0.09
436
MATSUDA ET AL.
~. !j,.:~
:/
3.5-
cantly higher than those in the older groups. At 15 minutes after the injection of labels, when no radioactivities of 3H and ~4C were detected in the serum transferrin, 3H-activity was detected in transferrin of all subcellular fractions of the liver (Table 1). In the total homogenate the activity in transferrin of the A1.Py-HF group was significantly higher than those in the other groups. The activities in transferrin of the cytosolic fractions were lower than those in the other fractions in all groups. The activities in transferrin of the rough endoplasmic reticulum, smooth endoplasmic reticulum and Golgi fractions showed no significant difference among the groups, nor between those fractions, except that 3H-activity in transferrin of the rough endoplasmic reticulum fraction, but not in the Golgi fraction, was significantly higher in the AI.Py-HF group than those in the other groups. At 30 minutes after the injection of labels, 3Hactivities in the Golgi fraction were higher than those in the other subcellular fractions in all groups. Among the experimental groups, the AI.Py-HF group showed significantly higher 3Hactivity of transferrin only in the total homogenate and Golgi fraction than those in the other groups (Table 2). Changes of ~4C-activity in the transferrin of the subcellular fractions of the hepatocytes were the same as those of the labeled leucine (Table 3. Table 4).
High Fat ~ High Fat, Py [''~ Low Fat, A1
dpm/g.liver x 10 4
:,}
High ~t, ~t
to~ Fat, m+~ ~_~
~ .:/
.:~oo5 **: 0<0.02
0.5] 15rain
i
i
30ndn
120min
Ai: ALcohol
py: p.yrazole
FIG. 5. Sequential changes of L4C-acitivity of the hepatic transferrin in the chronic experimental groups. The activity is significantly higher in the AI.Py-HF group than those in the other groups during experimental period. *p<0.05, **p<0.02 compared to the other groups.
in Fig. 3. 3H- and ~4C-activities were not found in the serum transferrin at 15 minutes after the injection of labels, then both activities appeared at 30 minutes in the serum transferrin and reached a plateau at 90 minutes in all groups. Incorporation rates of 3H-leucine and ~4C-fucose into the serum transferrin in the AI.Py-HF group were significantly lower than those in the other groups during the observation periods except for at 60 and 90 minutes in 3C-leucine. Incorporation rate of lac-fucose into the serum transferrin in the A1.Py-LF group was significantly lower than those in nonalcoholic groups during the observation period except at 60 minute, however, incorporation rate of SH-leucine was not different from those in the nonalcoholic group. The inhibition of incorporation of the labels in the A1.Py-HF group was more prominent in 14C-fucose than in 3H-leucine. The serial changes of 3H- and '4C-activities in hepatic transferrin are shown in Figs. 4 and 5. Both activities showed high values at 15 and 30 minutes, and at 120 minutes the activities clearly decreased in all groups. 3H-Activity in hepatic transferrin was significantly higher in the AI.Py-HF group than those in the other groups during the observation periods. ~4C-Activity in the hepatic transferrin in the AI.Py-HF group was also signifi-
DISCUSSION
In the present study, acute ethanol administration did not have any effects on the glycoprotein secretion from the hepatocytes. Volentine et al. (12) reported that acute ethanol administration inhibited the incorporation of the radiolabeled leucine and fucose in the serum transferrin. However, the dose of ethanol in their experiment was 6 g/kg, which was larger than that of the present study (3 g/kg). This may account for the difference between the two studies. Incorporation rates of the radiolabeled leucine and fucose into the serum transferrin were significantly decreased in the A1.Py-HF group. These results indicated that secretion of protein, especially glycoprotein, from the hepatocytes was inhibited in alcoholic liver injury. Glycoproteins destined for export from the hepatocytes are synthesized in the rough endoplasmic reticulum and then transported through the smooth endoplasmic reticulum to the transitional element of the Golgi complex. These newly synthesized glycoproteins are added to the terminal carbohydrate chain (glycosylation) in the Golgi apparatus and then are secreted into the blood (2,3). Fucose is a terminal carbohydrate
TABLE 3 t4C-ACTIVITIES OF TRANSFERRIN IN THE SUBCELLULAR FRACTIONS AT 15 MINUTES AFTER INJECTION OF LABELS IN THE CHRONIC EXPERIMENTAL GROUPS (dpm/g LIVER × 10'~)
Alcoholic
Nonalcoholic
Pyrazole Fraction Total homogenate Cytosol Rough ER Smooth ER Golgi
High Fat 3.78 0.66 1.48 1.02 1.72
--- 0.20* +_ 0.25 ± 0.22* ± 0.25 +- 0.38
Values are mean __. SD of each 5 rats. *p<0.05 compared to the other groups. ER: endoplasmic reticulum.
Nonpyrazole Low Fat
3.32 0.62 0.91 0.99 1.51
__. 0.12 ___ 0.18 +_ 0.24 _+ 0.22 ___ 0.27
High Fat 2.81 0.38 0.71 0.71 1.40
--- 0.23 ± 0.13 ± 0.32 +- 0.27 ± 0.22
Low Fat 2.85 0.31 0.61 0.77 1.38
+-- 0.24 - 0.12 --- 0.23 - 0.29 ± 0.38
Pyrazole
Nonpyrazole
High Fat
Low Fat
2.68 0.33 0.67 0.62 1.34
__. 0.33 _ 0.10 ± 0.22 ± 0.24 _ 0.22
2.72 0.30 0.52 0.67 1.15
_ 0.32 ± 0.08 ± 0.28 ± 0.28 --- 0.20
EFFECTS OF ETHANOL ON HEPATIC SECRETORY PROTEIN
437
TABLE 4 taC-ACTIVITIES OF TRANSFERRIN IN THE SUBCELLULARFRACTIONS AT 30 MINUTES AFTER INJECTION OF LABELS IN THE CHRONIC EXPERIMENTALGROUPS (dpm/g LIVER x 10'~) Alcoholic
Nonalcoholic
Pyrazole Fraction Total homogenate Cytosol Rough ER Smooth ER Golgi
High Fat 3.42 0.48 0.71 1.01 2.20
--- 0.14" +-- 0.12 --- 0.22 --- 0.39 ± 0.41"
Nonpyrazole Low Fat
3.12 0.30 0.50 0.85 1.45
± 0.15 ± 0.09 ± 0.10 ±- 0.21 ± 0.25
High Fat 2.95 0.23 0.66 0.66 1.16
• ~ ~ _ ~
0.27 0.11 0.09 0.11 0.23
Low Fat 2.99 0.17 0.47 0.44 1.02
--- 0.24 __. 0.13 _+_0.10 -,- 0.35 --- 0.20
Pyrazole
Nonpyrazole
High Fat
Low Fat
2.81 0.19 0.53 0.63 1.22
± 0.33 ___ 0.09 __. 0.10 __. 0.15 ± 0.18
2.78 0.16 0.50 0.50 1.01
± 0.41 ±- 0.12 +_ 0.13 --- 0.22 ± 0.13
Values are mean __. SD of each 5 rats. *p<0.02 compared to the other groups. ER: endoplasmic reticulum.
chain of secretory glycoproteins. In the present study, incorporation of fucose was inhibited more prominently than that of leucine in A1.Py-HF group. This result indicated that the glycosylation of glycoprotein in the hel~atocytic Golgi apparatus was impaired in alcoholic liver injury. The results of the present study also indicated that newly synthesized glycoproteins move from rough endoplasmic reticulum to Golgi apparatus and secrete into the blood taking about a 2-hour process. In the present study, both 3H-leucine and ~'~C-fucose activities in transferrin of the endoplasmic reticulum in the early period of the experiment and in transferrin of the Golgi apparatus in the late period were significantly higher in the AI.Py-HF group than those in the other groups. These results indicated that the intracellular movement of glycoprotein was delayed, and the terminal glycosylation in the Golgi apparatus and finally secretion of glycoprotein were impaired in alcoholic liver injury. These results also indicated that the appearance of carbohydrate-deficient transferrin in
the human alcoholics and experimental rat alcoholic liver injury (10,1 l) might he due to an impaired glycosylation of transferrin in the hepatocytic Golgi apparatus. The mechanisms of inhibition of glycoprotein secretion from hepatocytes following an impairrnent of glycosylation of glycoprotein in heptocytic Golgi apparatus are still unclear. Recently, it has been reported that microtubules play an important role in the function of the Golgi apparatus, including glycosylation of glycoprotein (8). We have reported that hepatic microtubules were decreased and transferrin was retained in the hepatocytes in experimental and human alcoholic liver injuries (5,6). Moreover, we have reported that a decrease in hepatic microtubules could be due to hepatic acetaldehyde accumulation (7). These results suggested that the retention of glycoprotein in the hepatocyte, which is a precursor of hepatocytic necrosis in alcoholic liver injuries, attributed to the impairment of glycosylation and secretion of glycoproteins in the Golgi apparatus due to a decrease in microtubules.
REFERENCES 1. Baraona, E.; Leo, M. A.; Borovsky, S. A.; Lieber, C. S. Alcoholic hepatomegaly: accumulation of protein in the liver. Science 190: 794--795; 1975. 2. Griffiths, G.; Simons, K. The trans Golgi network: Sorting at the exit site of the Golgi complex. Science 234:438--443; 1986. 3. Hanover, J. A.; Lennar, W. J. Transmembrane assembly of membrane and secretory proteins. Arch. Biochem. Biophys. 211:1-19; 1981. 4. Kobayashi, Y.; Maudsley, D. V. Practical aspects of double isotope counting. In: Bransome. E. D.. ed. The current status of liquid scintillation counting. New York: Grune & Stratton; 1970:76-85. 5. Matsuda, Y.; Takase, S.; Takada, A.; Sato, H.; Yasuhara, M. Comparison of ballooned hepatocytes in alcoholic and nonalcoholic liver injury in rats. Alcohol 2:303-308; 1985. 6. Matsuda, Y.; Takada, A.; Sato, H.; Yasuhara. M.; Takase, S. Comparison between ballooned hepatocytes occurring in human alcoholic and nonalcoholic liver diseases. Alcohol.: Clin. Exp. Res. 9:366-370; 1985. 7. Matsuda, Y.; Baraona, E.; Salaspuro, M.; Lieber, C. S. Effects of ethanol on liver microtubules and Golgi apparatus. Lab. Invest. 41:
455--463; 1979. 8. Mitranic, M. M.; Boggs, J. M.; Moscallello, M. A. An effect of colchicine on galactosyl and sialyltransferase of rat Golgi membrane. Biochem. Biophys. Acta 672:57--64; 1981. 9. Takada, A.; Matsuda, Y.; Takase, S. Effect of dietary fat on alcohol-pyrazole hepatitis in rats: The pathogenetic role of the nonalcohol dehydrogenase pathway in alcohol-induced hepatic cell injury. Alcohol.: Clin. Exp. Res. 10:403--41I; 1986. 10. Takada, A.; Matsuda, Y.; Sato, H.; Takase, S. Alcoholic liver injury and hepatic secretory proteins: Roles of microtubules and Golgi apparatus. In: Oda, T.; Okuda, K., eds. Falk Symposium in Tokyo, Japan. New trends in hepatology. Tokyo: Medical Tosho Co. Ltd.; 1986:33.-42. 11. Takase, S.; Takada, A.; Tsutsumi, M.; Matsuda, Y. Biochemical markers of chronic alcoholism. Alcohol 2:405--410; 1985. 12. Volentine, G. D.; Tuma, D. J.; Son'ell, M. F. Subcellular location of secretory proteins retained in the liver during ethanol-induced inhibition of hepatic protein secretion in the rat. Gastroenterology 90: 158-165; 1986.